E N C Y C L O P E D I A of

ENCYCLOPEDIA
of SCIENTIFIC DATING
METHODS

EARTH SCIENCES SERIES

Edited by
W. Jack Rink and Jeroen W. Thompson
 ENCYCLOPEDIA of
SCIENTIFIC DATING
METHODS
 Encyclopedia of Earth Sciences Series
ENCYCLOPEDIA OF SCIENTIFIC DATING METHODS

Volume Editors
William Jack Rink is Professor of Earth Sciences at McMaster University, Hamilton, Ontario, Canada. He has a B.Sc. and Ph.D. in Geology from
Florida State University, Tallahassee, Florida, USA. His research encompasses both fundamental studies of radiation exposure dating methods and
applications of electron spin resonance, uranium series dating, and luminescence dating of archaeological sites in the Middle East, Asia, Europe,
Africa, South America, and North America and geochronology of nearshore and ancient coastal geomorphologic features. He was elected to the
New York City Explorer’s Club in 2005.
Jeroen W. Thompson is Adjunct Assistant Professor in the Department of Medical Physics and Applied Radiation Sciences at McMaster
University. He holds a B.Sc. in Physics and a B.Sc. in Anthropology (Michigan State University) as well as an M.Sc. in Physics (University of
Connecticut) and a Ph.D. in Medical Physics and Applied Radiation Sciences (McMaster University). His research is highly interdisciplinary,
including diverse applications of radiation dosimetry and detection that span laboratory and fieldwork. In particular, he has applied electron
paramagnetic resonance and optically stimulated luminescence dosimetry both forensic dosimetry and geochronology. Dr. Thompson is currently
a research management consultant, with specific focus on radiation and environmental sciences.

Associate Editors Editorial Board

Larry M. Heaman Robert A. Creaser, University of Alberta, Edmonton, AB, Canada
Department of Earth and Atmospheric Sciences
University of Alberta Don Davis, University of Toronto, Toronto, ON, Canada
Edmonton, AB, Canada
Geoffrey A. T. Duller, Aberystwyth University, Aberystwyth, UK
A. J. Timothy Jull
NSF-Arizona AMS Laboratory John Gosse, Dalhousie University, Halifax, NS, Canada
University of Arizona
Tucson, AZ, USA Simon Y. W. Ho, University of Sydney, Sydney, NSW, Australia
James B. Paces James K.W. Lee, Queen’s University, Kingston, ON, Canada
U.S. Geological Survey Macquarie University, Sydney, NSW, Australia
Denver Federal Center
Denver, CO, USA Josep M. Pares, National Research Center on Human Evolution,
Burgos, Spain
Peter W. Reiners, University of Arizona, Tucson, AZ, USA

Ross Stevenson, University of Quebec, Montreal, QB, Canada

John F. Wehmiller, University of Delaware, Newark, DE, USA

Aims of the Series

The Encyclopedia of Earth Sciences Series provides comprehensive and authoritative coverage of all the main areas in the Earth Sciences. Each volume
comprises a focused and carefully chosen collection of contributions from leading names in the subject, with copious illustrations and reference lists.
These books represent one of the world’s leading resources for the Earth Sciences community. Previous volumes are being updated and new
works published so that the volumes will continue to be essential reading for all professional earth scientists, geologists, geophysicists,
climatologists, and oceanographers as well as for teachers and students. Go to http://link.springer.com to visit the Encyclopedia of Earth Sciences
Series online.

About the Series Editor

Professor Charles W. Finkl has edited and/or contributed to more than eight volumes in the Encyclopedia of Earth Sciences Series. For the past
25 years, he has been the Executive Director of the Coastal Education and Research Foundation and Editor-in-Chief of the international Journal of
Coastal Research. In addition to these duties, he is Professor at Florida Atlantic University in Boca Raton, Florida, USA. He is a graduate of the
University of Western Australia (Perth) and previously worked for a wholly owned Australian subsidiary of the International Nickel Company of
Canada (INCO). During his career, he acquired field experience in Australia, the Caribbean, South America, SW Pacific Islands, Southern Africa,
Western Europe, and the Pacific Northwest, Midwest, and Southeast USA.

Founding Series Editor

Professor Rhodes W. Fairbridge (deceased) has edited more than 24 encyclopedias in the Earth Sciences Series. During his career, he has worked
as a petroleum geologist in the Middle East, been a World War II intelligence officer in the SW Pacific, and led expeditions to the Sahara, Arctic
Canada, Arctic Scandinavia, Brazil, and New Guinea. He was Emeritus Professor of Geology at Columbia University and was affiliated with the
Goddard Institute for Space Studies.
 ENCYCLOPEDIA OF EARTH SCIENCES SERIES

ENCYCLOPEDIA of
SCIENTIFIC DATING
METHODS
edited by

W. JACK RINK
JEROEN W. THOMPSON
McMaster University, Canada

with Associate Editors

LARRY M. HEAMAN
University of Alberta, Edmonton

A. J. TIMOTHY JULL
University of Arizona, Tucson

JAMES B. PACES
U.S. Geological Survey, Denver
Library of Congress Control Number: 2015935943

ISBN: 978-94-007-6303-6
This publication is available also as:
Electronic publication under ISBN 978-94-007-6304-3 and
Print and electronic bundle under ISBN 978-94-007-6306-7

Springer Dordrecht, Heidelberg, New York, London

Printed on acid-free paper

Cover photo: Red onyx. istock Photo 10492317 © Missing35mm – iStock

Every effort has been made to contact the copyright holders of the figures and tables which have been reproduced
from other sources. Anyone who has not been properly credited is requested to contact the publishers, so that due
acknowledgement may be made in subsequent editions.

All rights reserved for the contributions Amino Acid Racemization, Eolianites; Biostratigraphy; Clays and Glauconites
(K-Ar/Ar-Ar); Gene Sequencing; Groundwater Dating with Atmospheric Halogenated Compounds; Hydrocarbons/
Rhenium-Osmium (Re-Os): Organic-rich Sedimentary Rocks; Isua Supracrustal Belt, West Greenland: Geochronology;
Luminescence Dating, Shell Rich Deposits; Potassium-Argon (Argon-Argon), Structural Fabrics; Radiocarbon Dating
of Terrestrial Carbonates; Uranium Series, Opal; Uranium Series, Rates of Basaltic Melt Generation and Transport;
Uranium Series, Volcanic Rocks; Uranium-Lead Dating, Opal; Uranium-Lead, Detrital Zircon

© Springer Science+Business Media Dordrecht 2015

No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means,
electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission from the
Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a
computer system, for exclusive use by the purchaser of the work.
 Contents

Contributors xi Amino Acid Racemization, Fluvial and Lacustrine

Sediments (AAR) 40
Preface xxvii John F. Wehmiller
Acknowledgments xxix
Amino Acid Racemization, Marine Sediments 44
Acasta Gneiss Complex 1 Darrell Kaufman
Samuel Bowring, Tom Chacko, Larry M. Heaman
and Jesse Reimink Amino Acid Racemization, Paleoclimate 47
Matthew Collins and Beatrice Demarchi
Accelerator Mass Spectrometry 3
A. J. Timothy Jull and George S. Burr Ancient Inks: A Forensic Art Historical Perspective 48
Howell G. M. Edwards
Age of the Earth 6
Peter Barry and Larry Taylor Apatite 53
W. Jack Rink
Alpha Spectroscopy 7
Aquifer Characteristics (U-Series) 54
Jeroen W. Thompson
Don Porcelli
Alpine Terranes (K–Ar/Ar–Ar) 7 Ar–Ar and K–Ar Dating 58
Jan R. Wijbrans James K. W. Lee
Amino Acid 12 Archaeomagnetic Dating 73
John F. Wehmiller Cathy Batt
Amino Acid Racemization Dating 13 Band Structure 81
Beatrice Demarchi and Matthew Collins Jeroen W. Thompson
Amino Acid Racemization, Biostratigraphy 26 Beta Counter 82
Matthew Kosnik Jeroen W. Thompson

Amino Acid Racemization, Coastal Sediments 28 Big Bang 82

John F. Wehmiller Gary R. Huss

Amino Acid Racemization, Eolianites 35 Biostratigraphy 83

Colin V. Murray-Wallace Marie-Pierre Aubry
vi CONTENTS

Bivalve Sclerochronology 108 Dendrochronology, Fire Regimes 204

Donna M. Surge and Bernd R. Schöne Peter M. Brown

Bomb Carbon 115 Dendrochronology, Progress 207

George S. Burr Laia Andreu-Hayles and Caroline Leland

Bones (U-Series) 120 Dendrochronology, Surficial Processes 213

Alistair W. G. Pike Markus Stoffel, Juan A. Ballesteros-Cnovas and
Christophe Corona
14
C in Plant Macrofossils 127
Christine Hatté and A. J. Timothy Jull Dendrochronology, Volcanic Eruptions 221
Franco Biondi
Carbonates, Lacustrine (U-Series) 132
Mordechai Stein Early Life on Earth 229
David Wacey
Carbonates, Marine Carbonates (U-Series) 136
Claudine H. Stirling Electron Spin Resonance (ESR) Dating of Coral 234
Gerhard Schellmann and Ulrich Radtke
Carbonates, Pedogenic (U-Series) 141 Electron Spin Resonance (ESR) Dating of Fossil
Warren D. Sharp Tooth Enamel 239
Mathieu Duval
Carbonates, Speleothem Archaeological (U-Series) 144
Dirk Hoffmann Electron Spin Resonance (ESR) Dating,
General Principles 246
Carbonates, Speleothem Climatic (U-Series) 147 Anne Skinner
David A. Richards
Electron Spin Resonance Spectrometer 255
Chemical Weathering (U-Series) 152 Jeroen W. Thompson
Anthony Dosseto
Environmental Releases 256
Chert 169 Robert Morrison
L. Paul Knauth
Exhumation (Thermochronology) 261
Chromatography 170 Pieter van der Beek
John F. Wehmiller
Extraterrestrial Materials (K–Ar/Ar–Ar) 264
Clays and Glauconites (K–Ar/Ar–Ar) 171 Timothy Swindle
Horst Zwingmann
Fault Zone (Thermochronology) 269
Ben van der Pluijm and Chris Hall
Continental Drift (Paleomagnetism) 177
Trond H. Torsvik, Pavel V. Doubrovine and Faults (U-Series) 274
Mathew Domeier Perach Nuriel
Corals (Sclerochronology) 187 Feldspar, Infrared-Stimulated Luminescence 279
Kristine L. DeLong Mayank Jain
Crustal Sulfide Minerals (Re–Os) 191 Feldspars 284
Robert A. Creaser Michel Lamothe
Dendrochronology, Dwellings 197 Fission Track Dating and Thermochronology 285
Stephen E. Nash Andrew J. W. Gleadow and Christian Seiler

Dendrochronology, Entomology 202 Gene Sequencing 297

Barry Cooke Jessica A. Thomas
 CONTENTS vii

Geomagnetism 298 Lu–Hf Dating: The Lu–Hf Isotope System 379

Vadim A. Kravchinsky Jeff Vervoort

Glacial Landscape (Cosmogenic Nuclide) 301 Luminescence Dating 390

Joerg M. Schaefer Geoffrey A. T. Duller

Groundwater Dating with Atmospheric Luminescence Dating of Archaeological

Halogenated Compounds 308 Sediments 404
Karl B. Haase and Eurybiades Busenberg James Feathers

Historical Development of Dating Methods 319 Luminescence Dating, Deep-Sea Marine and
James M. Mattinson Lacustrine 409
Helen M. Roberts
Hominid Evolution Timescale 329
Antoine Balzeau Luminescence Dating, Dose Rates 414
Guillaume Guérin
Hydrocarbons/Rhenium–Osmium (Re–Os):
Organic-Rich Sedimentary Rocks 330 Luminescence Dating, History 417
David Selby, Vivien M. Cumming, Ludwig Zöller and Günther A. Wagner
Alan D. Rooney and Alexander J. Finlay
Luminescence Dating, Instrumentation 422
Hydrothermal Ores (Thermochronology) 334 Kristina Jørkov Thomsen
István Márton
Luminescence Dating, Loess 425
Ice Cores 341 Helen M. Roberts
Anders Svensson
Luminescence Dating, Meteorites 430
Impact Glass (Fission Tracks) 348 Derek W. G. Sears
Giulio Bigazzi and Maria Laura Balestrieri
Luminescence Dating, Shell-Rich Deposits 431
Index Fossil 353 Robert Hendricks and Alex Hodson
Peter Harries
Luminescence Dating, Single-Grain Dose
Isua Supracrustal Belt, West Greenland: Distribution 435
Geochronology 354 Richard G. Roberts and Zenobia Jacobs
Vickie C. Bennett and Allen P. Nutman
Luminescence Dating, Uncertainties and
Jack Hills Zircon 359 Age Range 440
Simon A. Wilde Jakob Wallinga and Alastair C. Cunningham
Kimberlites (K–Ar/Ar–Ar) 361 Luminescence, Biogenic Carbonates 445
David Phillips Geoffrey A. T. Duller
Lacustrine Environments (14C) 365 Luminescence, Coastal Sediments 446
Susan Zimmerman and Amy Myrbo Barbara Mauz
Laser Ablation Inductively Coupled Mass
Spectrometer (LA ICP-MS) 371 Luminescence, Colluvial Sediments 450
Paul Sylvester Andreas Lang

Lichenometry 372 Luminescence, Desert Dunes 452

William B. Bull Matt Telfer

Lucy 378 Luminescence, Earthquake and Tectonic Activity 456

William H. Kimbel Morteza Fattahi
viii CONTENTS

Luminescence, Flints and Stones 460 Meteorites (Lu–Hf) 555

Daniel Richter Audrey Bouvier

Luminescence, Fluvial Sediments 465 Meteorites (U–Pb) 559

Amanda Keen-Zebert Yuri Amelin
Luminescence, Geomorphological Processes 470 Meteorites, Rubidium–Strontium, and
Stephen Tooth Samarium–Neodymium Chronology 562
Lars Borg
Luminescence, Glacial Sediments 475
Geoffrey A.T. Duller Minerals (40Ar–39Ar) 569
Simon Kelley, Clare Warren and
Luminescence, Martian Sediments 478 Camilla Wilkinson
Regina DeWitt
Model Ages (Sm–Nd) 573
Luminescence, Pottery and Bricks 481 Alan P. Dickin
Ian Bailiff
Molecular Clock Calibration 576
Luminescence, Rock Surfaces 485 Rachel Warnock
Reza Sohbati

Luminescence, Soils 489 Molecular Clocks 583

Arjun M. Heimsath Simon Y. W. Ho

Luminescence, Volcanic Rocks 493 Molecular Clocks, Human Evolution 588

Sumiko Tsukamoto Simon Y. W. Ho and Phillip Endicott

Magnetic Anomalies 497 Molecular Clocks, Relaxed Variant 591

Roi Granot Simon Y. W. Ho

Magnetic Chronology 500 Molecular Dating of Evolutionary Events 593

Manuel Calvo-Rathert David Duchene and Lindell Bromham

Magnetometer 507 Molecular Rate Variation (Molecular Clocks) 596

Vicente Soler Javaloyes Robert Lanfear

Magnetostratigraphic Dating 507 Molluscs, Foraminifera, and Other

Miguel Garces Carbonate Fossils 597
Bonnie A. B. Blackwell
Marine Isotope Stratigraphy 517
Galen P. Halverson Neutron Activation Analysis 607
Ron Hancock
Marine Varves 528
Konrad A. Hughen Noble Gas Mass Spectrometer 608
Leah E. Morgan
Mass Spectrometry 533
James B. Paces, Dominique Weis and Obsidian Hydration Dating 609
Trevor R. Ireland Ioannis Liritzis

Metamorphic Terranes (K–Ar/40Ar/39Ar) 542 Paleosol 625

Jan R. Wijbrans Gary E. Stinchcomb
210
Meteoric 10Be 547 Pb Dating 626
Jane Willenbring Peter W. Swarzenski

Meteorites (36Cl) 548 Peat (14C) 632

Kees Welten Philippa Ascough
 CONTENTS ix

Planetary Surfaces (Cratering Rate) 636 Sea Level Change (U-Series) 727
Stephanie C. Werner and Wolf Uwe Reimold Christina Gallup

Polymerase Chain Reaction DNA Amplification 641 Seawater Sr Curve 733

Martyna Molak Galen P. Halverson and Lucie Hubert-Théou
Potassium–Argon (Argon–Argon), Structural Secondary Ion Mass Spectrometry (SIMS) 739
Fabrics 642 Trevor R. Ireland
Michael A. Cosca
Sediment Mixing Rate, 210Pb and 234Th 740
Quartz 649 Joseph M. Smoak
Peter J. Heaney
Sediment, ESR 743
Quartz Defects, Optically Stimulated Helene Tissoux
Luminescence and Thermoluminescence 650
Marco Martini Sedimentary Rocks (Rb-Sr Geochronology) 748
Tod Waight
Radiation Defect 657
Jeroen W. Thompson
Sediments, Terrestrial (Paleomagnetism) 752
Radiation Dose Rate 658 Wout Krijgsman and Gillian Turner
Grzegorz Adamiec
Single-Crystal Laser Fusion 760
Radiation and Radioactivity 660 James K. W. Lee
Regina DeWitt
Skeletal Remains (14C) 763
Radioactive Decay Constants: A Review 666 Gregory W. L. Hodgins
W. Jack Rink and Larry M. Heaman
Sm–Nd Dating 768
Radiocarbon Dating 669 Richard W. Carlson
A. J. Timothy Jull and George S. Burr
Stellar Chronology 780
Radiocarbon Dating of Marine Carbonates 676 Trevor R. Ireland
Quan Hua
Tephrochronology 783
Radiocarbon Dating of Terrestrial Carbonates 680 David J. Lowe and Brent Alloway
Jeffrey S. Pigati
Terrestrial Cosmogenic Nuclide Dating 799
Radioluminescence (RL) 685 John Gosse and Jeff Klein
Tobias Lauer
Thermal Ionization Mass Spectrometer (TIMS) 813
Rb–Sr Dating 686 Roland Mundil
Oliver Nebel
Thermochronology, Detrital Zircon 814
Rb–Sr Geochronology (Igneous Rocks) 698 John I. Garver
Tod Waight
Thermochronology, Landform Evolution 818
Rhenium–Osmium Dating (Meteorites) 703 Thibaud Simon-Labric
Richard J. Walker
Thermochronology, Meteorites 824
Rhenium–Osmium Geochronology: Sulfides, Kyoungwon Min
Shales, Oils, and Mantle 707
Holly Stein and Judith Hannah Tsunamigenic Sediments 827
Gloria I. López
Rubidium–Strontium Dating, Hydrothermal
Events 723 Ujaraaluk Unit (Nuvvuagittuq Greenstone Belt) 833
Shu’nichi Nakai Jonathan O’Neil
x CONTENTS

Uranium Series, Ice 834 Uranium–Lead, Metamorphic Rocks 898

Stephanie A. Ewing Daniela Rubatto

Uranium Series, Opal 837 Uranium–Lead, Ore Deposits 903

James B. Paces Rolf L. Romer
Uranium Series, Rates of Basaltic Melt Uranium–Lead, Rubidium-Strontium, Kimberlite 907
Generation and Transport 843 Larry M. Heaman
Aaron J. Pietruszka

Uranium Series, Volcanic Rocks 845 Uranium–Lead, Zircon 914

Jorge A. Vazquez Fernando Corfu

Uranium–Lead Dating 848 U-Series Dating 918

Randall Parrish Bernard Bourdon

Uranium–Lead Dating, Opal 858 U–Th/He Dating 932

Leonid Neymark Peter K. Zeitler

Uranium–Lead, Chemical Isochron U–Pb Method Volcanic Glass (Fission Track) 941
(CHIME) 863 John A. Westgate
Kazuhiro Suzuki and Daniel J. Dunkley
Volcanic Rocks (Ar/Ar) 947
Uranium–Lead, Detrital Zircon 869 Ajoy K. Baksi
Keith Sircombe
Volcanogenic Sedimentary Rocks
Uranium–Lead, Diagenetic Processes 882 (K/Ar, 40Ar/39Ar) 950
E. Troy Rasbury Ian McDougall
Uranium–Lead, Extraterrestrial, Early
Solar System 885 Walther’s Law of Facies 957
Yuri Amelin Gloria I. López

Uranium–Lead, Extraterrestrial, Planetary Zircon 959

Materials 890 John M. Hanchar
Alexander Nemchin
Author Index 963
Uranium–Lead, Igneous Rocks 894
Donald W. Davis Subject Index 965
 Contributors

Grzegorz Adamiec Philippa Ascough

Institute of Physics, Centre for Science and Education AMS Laboratory
GADAM Centre, Silesian University of Technology Scottish Universities Environmental Research Centre
ul. Konarskiego 22B Ranking Ave
44-100 Gliwice East Kilbride G75 0QF
Poland Scotland, UK
grzegorz.adamiec@polsl.pl philippa.ascough@gla.ac.uk

Marie-Pierre Aubry
Brent Alloway Department of Earth and Planetary Sciences
School of Geography, Environment and Earth Sciences Rutgers University
Victoria University of Wellington 610 Taylor Road
6140 Wellington Piscataway, NJ 08854-8066
New Zealand USA
brent.alloway@vuw.ac.nz aubry@rci.rutgers.edu

Ian Bailiff
Yuri Amelin Department of Archaeology
Research School of Earth Sciences University of Durham
The Australian National University South Road
0200 Canberra DH1 3LE Durham
Australia UK
yuri.amelin@anu.edu.au ian.bailiff@durham.ac.uk

Ajoy K. Baksi
Laia Andreu-Hayles Department of Geology and Geophysics
Tree-Ring Laboratory Louisiana State University
Lamont-Doherty Earth Observatory of Columbia University E-315 Howe-Russell
Palisades, NY Baton Rouge, LA 70803
USA USA
lah@ldeo.columbia.edu akbaksi@yahoo.com
xii CONTRIBUTORS

Maria Laura Balestrieri Bonnie A. B. Blackwell

CNR- Institute of Geosciences and Earth Resources Department of Chemistry
UOS Florence Williams College
Italy Williamstown, MA 01267
balestrieri@igg.cnr.it USA
bonnie.a.b.blackwell@williams.edu
Juan A. Ballesteros-Cnovas
Institute for Geological Sciences, University of Bern Lars Borg
Baltzerstrasse 1 + 3 Chemical Sciences Division
3012 Bern Lawrence Livermore National Laboratory
Switzerland Livermore, CA 94550
juan.ballesteros@dendrolab.ch USA
borg5@llnl.gov
Antoine Balzeau
Department of Prehistory
UMR 7194, CNRS Bernard Bourdon
Musée National d’Histoire Naturelle Laboratoire de Géologie de Lyon
Paris ENS Lyon, CNRS, and UCBL
France 46 Allée d’Italie
abalzeau@mnhn.fr 69364 Lyon cedex 7
France
bernard.bourdon@ens-lyon.fr
Peter Barry
Department of Earth and Planetary Science
University of Tennessee Audrey Bouvier
1412 Circle Drive Department of Earth Sciences
Knoxville, TN 37996-1410 The University of Western Ontario
USA 1151 Richmond Street N
peter.barry@utk.edu London, ON
Canada, BGS 1026
Cathy Batt audrey.bouvier@uwo.ca
Archaeological Sciences
University of Bradford
Bradford BD7 1DP Samuel Bowring
UK Department of Earth, Atmospheric and Planetary Sciences
c.m.batt@bradford.ac.uk Massachusetts Institute of Technology
77 Massachusetts Ave., Building 54-1126
Cambridge, MA 02139
Vickie C. Bennett USA
Research School of Earth Sciences sbowring@mit.edu
The Australian National University
Canberra, ACT 0200
Australia Lindell Bromham
vickie.bennett@anu.edu.au Centre for Macroevolution and Macroecology
Division of Evolution, Ecology, and Genetics
Giulio Bigazzi Research School of Biology
CNR- Institute of Geosciences and Earth Resources Australian National University
Pisa Building 116
Italy Canberra, ACT 0200
bigazzi.pisa@gmail.com Australia
lindell.bromham@anu.edu.au
Franco Biondi
DendroLab Peter M. Brown
University of Nevada Rocky Mountain Tree-Ring Research, Inc.
Mailstop 154 2901 Moore Lane
Reno, Nevada 89557-0154 Fort Collins, CO 80526
USA USA
fbiondi@unr.edu pmb@rmtrr.org
 CONTRIBUTORS xiii

William B. Bull Matthew Collins

Department of Geosciences BioArCh, Department of Archaeology
University of Arizona University of York
1040 E. 4th St BioArCh, Biology S Block, Wentworth Way
Tucson, AZ 85721 YO10 5DD York
USA UK
bill@activetectonics.com matthew.collins@york.ac.uk

Barry Cooke
George S. Burr Government of Canada
NSF Arizona Accelerator Mass Spectrometry Laboratory Northern Forestry Centre, Canadian Forest Service,
Department of Geosciences and Physics Natural Resources Canada
University of Arizona Edmonton, AB
1118 East Fourth St Canada
Tucson, AZ 85721 barry.cooke@nrcan-rncan.gc.ca
USA
and Fernando Corfu
Department of Geosciences Department of Geosciences
National Taiwan University University of Oslo
Taipei 0316 Oslo
Taiwan Norway
burr@u.arizona.edu fernando.corfu@geo.uio.no

Christophe Corona
Eurybiades Busenberg
Institute for Geological Sciences, University of Bern
CFC Laboratory
Baltzerstrasse 1 + 3
U.S. Geological Survey
3012 Bern
MS 432, 12201 Sunrise Valley Drive
Switzerland
Reston, VA 20192
christophe.corona@univ-bpclermont.fr
USA
ebusenbe@usgs.gov
Michael A. Cosca
United States Geological Survey (USGS)
Manuel Calvo-Rathert Denver Federal Center
Departamento de Física Box 25046 MS 963
Escuela Politécnica Superior Denver, CO 80225-0046
Universidad de Burgos USA
Avenida de Cantabria s/n mcosca@usgs.gov
09006 Burgos
Spain Robert A. Creaser
mcalvo@ubu.es Department of Earth and Atmospheric Sciences
University of Alberta
Edmonton, AB
Richard W. Carlson Canada T6G2E3
Department of Terrestrial Magnetism robert.creaser@ualberta.ca
Carnegie Institution of Washington
5241 Broad Branch Road Vivien M. Cumming
Washington, DC 20015 Department of Earth Sciences
USA University of Durham
rcarlson@ciw.edu Science Labs
Durham DH1 3LE
UK
Tom Chacko and
Department of Earth and Atmospheric Sciences Department of Earth and Planetary Sciences
University of Alberta Harvard University
Edmonton, AB Cambridge, MA
Canada T6G 2E3 USA
tom.chacko@ualberta.ca v.m.cumming@durham.ac.uk
xiv CONTRIBUTORS

Alastair C. Cunningham Anthony Dosseto

Centre for Archaeological Science Wollongong Isotope Geochronology Laboratory
School of Earth and Environmental Sciences School of Earth and Environmental Sciences
University of Wollongong University of Wollongong
Wollongong Wollongong, NSW 2522
Australia Australia
acunning@uow.edu.au tonyd@uow.edu.au

Pavel V. Doubrovine
Donald W. Davis
Centre for Earth Evolution and Dynamics (CEED)
Department of Earth Sciences
University of Oslo
Earth Sciences Centre, University of Toronto
Oslo
22 Russell St
Norway
Toronto, ON
pavel.dubrovin@fys.uio.no
Canada
dond@es.utoronto.ca
David Duchene
Centre for Macroevolution and Macroecology
Kristine L. DeLong Division of Evolution, Ecology, and Genetics
Department of Geography and Anthropology Research School of Biology
Lousiana State University Australian National University
227 Howe-Russell Geoscience Complex Building 116
Baton Rouge, LA 70803 Canberra, ACT 0200
USA Australia
kdelong@lsu.edu david.duchene@anu.edu.au

Geoffrey A. T. Duller
Beatrice Demarchi
Department of Geography and Earth Sciences
BioArCh, Department of Archaeology
Aberystwyth University
University of York
Llandinham Bldg., Penglais Campus
BioArCh, Biology S Block, Wentworth Way
SY23 3DB Aberystwyth, Wales
YO10 5DD York
UK
UK
ggd@aber.ac.uk
beatrice.demarchi@york.ac.uk
Daniel J. Dunkley
Regina DeWitt Department of Applied Geology, Western Australian
Department of Physics School of Mines
East Carolina University Curtin University
Howell Science Complex, Tenth Street Perth, WA 6845
Greenville, NC 27858-4353 Australia
USA daniel.dunkley@curtin.edu.au
dewittr@ecu.edu
Mathieu Duval
ESR dating laboratory, Geochronology program
Alan P. Dickin
Centro Nacional de Investigación sobre la Evolución
School of Geography and Earth Sciences
Humana (CENIEH)
McMaster University
Paseo de Atapuerca s/n
1280 Main St. W.
09002 Burgos
Hamilton, ON
España
Canada L8S 4K1
mathieu.duval@cenieh.es
dickin@mcmaster.ca
Howell G. M. Edwards
Mathew Domeier Chemical and Forensic Sciences
Centre for Earth Evolution and Dynamics (CEED) School of Life Sciences
University of Oslo University of Bradford
Oslo BD7 1DP Bradford
Norway UK
mathew.domeier@fys.uio.no h.g.m.edwards@bradford.ac.uk
 CONTRIBUTORS xv

Phillip Endicott Andrew J. W. Gleadow

Départment Hommes, Natures, Sociétés School of Earth Sciences
Musée de l’Homme University of Melbourne
75116 Paris McCoy Bldy
France Melbourne, VIC 3010
phillip.endicott@gmail.com Australia
gleadow@unimelb.edu.au
Stephanie A. Ewing
Department of Land Resources and Environmental
Sciences John Gosse
Montana State University Department of Earth Sciences
817 Leon Johnson Hall Dalhousie University
Bozeman, MT 59717-3120 3006 LSC, 1459 Oxford St
USA Halifax, NS
stephanie.ewing@montana.edu Canada B3H 4R2
john.gosse@dal.ca
Morteza Fattahi
Institute of Geophysics Roi Granot
University of Tehran Department of Geological and Environmental Sciences
End of North Karegar Ave. Ben Gurion University of the Negev
1435944411 Tehran 84105 Beer Sheva
Iran Israel
morteza.fattahi@ouce.ox.ac.uk rgranot@bgu.ac.il
James Feathers
Department of Anthropology Guillaume Guérin
University of Washington Centre for Nuclear Technologies
Seattle, WA Technical University of Denmark
USA DTU Risø Campus
jimf@u.washington.edu 4000 Roskilde
Denmark
Alexander J. Finlay gugu@dtu.dk
Origin Analytical LTD
Welshpool, Powys
UK Karl B. Haase
alexfinlay@originanalytical.com CFC Laboratory
U.S. Geological Survey
Christina Gallup MS 432, 12201 Sunrise Valley Drive
Department of Earth and Environmental Sciences Reston, VA 20192
University of Minnesota Duluth USA
229 Heller Hall, 1114 Kirby Drive khaase@usgs.gov
Duluth, MN 55812
USA Chris Hall
cgallup@d.umn.edu Department of Earth and Environmental Sciences
University of Michigan
Miguel Garces 4534b C.C. Little Building, 1100 North University Ave
University of Barcelona Ann Arbor, MI 48109-1005
08028 Barcelona USA
Spain cmhall@umich.edu
mgarces@ub.edu

John I. Garver Galen P. Halverson

Department of Geology Department of Earth and Planetary Sciences/Geotop
Union College McGill University
807 Union St., Olin Building 3450 University Street
Schenectady, NY Montréal, QC
USA Canada H3A 0E8
garverj@union.edu galen.halverson@mcgill.ca
xvi CONTRIBUTORS

John M. Hanchar Arjun M. Heimsath

Department of Earth Sciences School of Earth and Space Exploration
Memorial University of Newfoundland Arizona State University
St. John’s, NL 781 Terrace Rd.
Canada A1B 3X5 Tempe, AZ 85287
jhanchar@mun.ca USA
arjun.heimsath@asu.edu
Ron Hancock
23 Oswald Crescent Robert Hendricks
Toronto, ON School Geography and Earth Sciences
Canada McMaster University
ronhancock@ca.inter.net Hamilton, ON
Canada
Judith Hannah hendricks.robert.r@gmail.com
AIRIE Program
Colorado State University
Fort Collins, CO Simon Y. W. Ho
USA School of Biological Sciences
and University of Sydney
Centre for Earth Evolution and Dynamics (CEED) Edgeworth David A11
University of Oslo Sydney, NSW 2006
Oslo Australia
Norway simon.ho@sydney.edu.au
judith.hannah@colostate.edu
Gregory W. L. Hodgins
Peter Harries Accelerator Mass Spectrometry Laboratory
School of Geosciences Department of Physics, School of Anthropology
University of South Florida University of Arizona
4202 E. Fowler Avenue, CPR107 1118 E. 4th Street
Tampa, FL 33620 Tucson, AZ 85721-0081
USA USA
harries@usf.edu ghodgins@physics.arizona.edu

Christine Hatté Alex Hodson

Laboratoire des Sciences du Climat et de l’Environnement School Geography and Earth Sciences
UMR8212 CEA-CNRS-UVSQ, Domaine du CNRS McMaster University
L.S.C.E. 12 avenue de la Terrasse Hamilton, ON
91198 Gif-sur-Yvette Canada
France hodsona@mcmaster.ca
christine.hatte@lsce.ipsl.fr

Larry M. Heaman Dirk Hoffmann

Department of Earth and Atmospheric Sciences Department of Human Evolution
University of Alberta Max Planck Institute for Evolutionary Anthropology
Edmonton, AB Deutscher Platz 6
Canada T6G 2E3 04103 Leipzig
larry.heaman@ualberta.ca Germany
dirk.hoffmann@eva.mpg.de
Peter J. Heaney
Department of Geosciences Quan Hua
Pennsylvania State University Australian Nuclear Science and Technology Organisation
Penn State 540 Deike Building Locked Bag 2001
University Park, PA 16802 Kirrawee DC, NSW 2232
USA Australia
pjheaney@psu.edu qhx@ansto.gov.au
 CONTRIBUTORS xvii

Lucie Hubert-Théou Darrell Kaufman

Department of Earth and Planetary Sciences/Geotop School of Earth Sciences and Environmental
McGill University Sustainability
3450 University Street Northern Arizona University
Montréal, QC 625 South Knoles Drive
Canada H3A 0E8 Flagstaff, AZ 86011-4099
lucie.hubert-theou@mail.mcgill.ca USA
darrell.kaufman@nau.edu
Konrad A. Hughen
Department of Marine Chemistry and Geochemistry Amanda Keen-Zebert
Woods Hole Oceanographic Institution Division of Earth and Ecosystem Sciences
266 Woods Hole Rd. MS#25 Desert Research Institute
Woods Hole, MA 02543-1050 2215 Raggio Parkway
USA Reno, NV 89512-1095
khughen@whoi.edu USA
keenzebert@gmail.com
Gary R. Huss
Hawai’i Institute of Geophysics and Planetology Simon Kelley
University of Hawai’i at Mānoa Department of Earth and Environmental Sciences
1680 East–west Road The Open University
Honolulu, HI 96822 Milton Keynes MK7 6AA
USA UK
ghuss@higp.hawaii.edu simon.kelley@open.ac.uk

William H. Kimbel
Trevor R. Ireland
Institute of Human Origins
Research School of Earth Sciences
School of Human Evolution and Social Change
The Australian National University
Arizona State University
Jaeger 5, Room 6, Building 61, Mills Road
Tempe, AZ 85028
Canberra, ACT 0200
USA
Australia
wkimbel.iho@asu.edu
trevor.ireland@anu.edu.au
Jeff Klein
Zenobia Jacobs Department of Physics and Astronomy
Centre for Archaeological Science University of Pennsylvania
School of Earth and Environmental Sciences Philadelphia, PA 19104
University of Wollongong USA
Wollongong, NSW 2522 klein.jef@gmail.com
Australia
zenobia@uow.edu.au L. Paul Knauth
School of Earth and Space Exploration
Mayank Jain Arizona State University
Center for Nuclear Technologies Tempe, AZ 85287-1404
Technical University of Denmark USA
DTU Risø Campus Knauth@asu.edu
Frederiksborgvej 399, Building 201
4000 Roskilde Matthew Kosnik
Denmark Department of Biological Sciences
maja@dtu.dk Macquarie University
Sydney, NSW 2109
A. J. Timothy Jull Australia
NSF-Arizona AMS Laboratory mkosnik@alumni.uchicago.edu
Physics Building
University of Arizona Vadim A. Kravchinsky
1118 East Fourth St University of Alberta
85721 Tucson, AZ Edmonton, AB
USA Canada T6G 2E1
jull@u.arizona.edu vadim@ualberta.ca
xviii CONTRIBUTORS

Wout Krijgsman Caroline Leland

Department of Earth Sciences Tree-Ring Laboratory
University of Utrecht Lamont-Doherty Earth Observatory of Columbia
Budapestlaan 17 University
3584 CD Utrecht Palisades, NY
The Netherlands USA
w.krijgsman@uu.nl cleland@ldeo.columbia.edu

Ioannis Liritzis
Michel Lamothe Laboratory of Archaeometry
Université du Québec à Montréal Department of Mediterranean Studies
C.P. 8888, Succ. Centre-Ville University of the Aegean
Montréal, Québec 85100 Rhodes
Canada H3C 3P8 Greece
lamothe.michel@uqam.ca liritzis@rhodes.aegean.gr

Robert Lanfear Gloria I. López

Ecology, Evolution, and Genetics Luminescence Laboratory
The Australian National University Centro Nacional de Investigación sobre la Evolución
116 Daley Road Humana CENIEH
Canberra, ACT 0200 Paseo Sierra de Atapuerca, 3
Australia Burgos 09002
rob.lanfear@gmail.com Spain
and
Leon Recanati Institute for Maritime Studies
University of Haifa
Andreas Lang Mt. Carmel
School of Environmental Sciences Haifa 31905
University of Liverpool Israel
4 Brownlow Street lopezgi.phd@gmail.com
Liverpool L69 3GP
UK
lang@liverpool.ac.uk David J. Lowe
School of Science, Faculty of Science and Engineering,
Earth and Ocean Sciences
Tobias Lauer The University of Waikato
Leibniz Institute for Applied Geophysics Hillcrest Rd
Stilleweg 2 3240 Hamilton
30655 Hannover New Zealand
Germany d.lowe@waikato.ac.nz
tobias.lauer@liag-hannover.de
Marco Martini
Dipartimento di Scienza dei Materiali and Sezione INFN
James K. W. Lee Universita’ degli Studi di Milano Bicocca
Department of Geological Sciences and Geological Via Cozzi 55
Engineering 20125 Milan
Queen’s University Italy
Miller Hall marco.martini@mater.unimib.it
Kingston, ON
Canada K7L 3N6
and István Márton
Department of Earth and Planetary Sciences Stockwork GeoConsulting Ltd
Macquarie University str. Protopop Aurel Munteanu, nr. 1
Sydney, NSW 2109 405400 Huedin, Cluj
Australia Romania
jim.lee@queensu.ca istvan.marton@stockwork.ro
 CONTRIBUTORS xix

James M. Mattinson Colin V. Murray-Wallace

Department of Earth Science School of Earth and Environmental Sciences
University of California University of Wollongong
Santa Barbara 1006 Webb hall Wollongong, NSW 2522
Santa Barbara, CA 93106-9630 Australia
USA cwallace@uow.edu.au
mattinson@geol.ucsb.edu
Amy Myrbo
Barbara Mauz LacCore, Department of Earth Sciences
School of Environmental Sciences University of Minnesota
University of Liverpool 7000 East Ave, L-397
4 Brownlow Street Minneapolis, MN
Liverpool USA
UK amyrbo@umn.edu
mauz@liverpool.ac.uk
Shu’nichi Nakai
Ian McDougall Earthquake Research Institute
Research School of Earth Sciences The University of Tokyo
The Australian National University Yayoi 1-1-1
Canberra, ACT 2601 Bunkyo-ku, Tokyo 113-0032
Australia Japan
Ian.McDougall@anu.edu.au snakai@eri.u-tokyo.ac.jp

Kyoungwon Min Stephen E. Nash

Department of Geological Sciences Denver Museum of Nature & Science
University of Florida 2001 Colorado Boulevard
241 Williamson Hall Denver, CO 80205
Gainesville, FL 32611 USA
USA stephen.nash@dmns.org
kmin@ufl.edu
Oliver Nebel
Research School of Earth Sciences
Martyna Molak The Australian National University
School of Biological Sciences Mills Road, Bldg. 61
University of Sydney Acton, ACT 0200
Sydney Australia
Australia oliver.nebel@anu.edu.au
martyna.molak@sydney.edu.au
Alexander Nemchin
Leah E. Morgan Department of Applied Geology
Scottish Universities Environmental Research Centre Curtin University of Technology
Rankine Avenue Perth, WA 6845
East Kilbride G75 0QF Australia
UK a.nemchin@curtin.edu.au
leah.morgan@glasgow.ac.uk
Leonid Neymark
Robert Morrison US Geological Survey
56-2773 Lahuiki Place MS 963
Hawi, HI 96719 Denver, CO 80225
USA USA
RobertForensics@aol.com lneymark@usgs.gov

Roland Mundil Perach Nuriel

Berkeley Geochronology Center Department of Geological and Environmental Sciences
2455 Ridge Road Stanford University
Berkeley, CA 94709 Stanford 94305-2115 CA
USA USA
rmundil@bgc.org perach@stanford.edu
xx CONTRIBUTORS

Allen P. Nutman Alistair W. G. Pike

School of Earth & Environmental Sciences Department of Archaeology
University of Wollongong University of Southampton
Wollongong, NSW Highfield Lane
Australia SO17 1BF Southampton
anutman@uow.edu.au UK
a.w.pike@soton.ac.uk

Jonathan O’Neil
Department of Earth Sciences Don Porcelli
University of Ottawa Department of Earth Sciences
140 Loui-Pasteur University of Oxford
Ottawa, ON South Parks Road
Canada K1N 6N5 OX1 3AN Oxford
jonathan.oneil@uottawa.ca UK
don.porcelli@earth.ox.ac.uk

James B. Paces
Geosciences and Environmental Change Science Center Ulrich Radtke
U.S. Geological Survey University of Duisburg-Essen
Denver Federal Center Universitätsstr. 2
Denver, CO 80225-004 Essen
USA Germany
jbpaces@usgs.gov rektor@uni-due.de

Randall Parrish E. Troy Rasbury

Department of Geology Department of Geosciences
University of Leicester and NERC Isotope Geosciences Stony Brook University
Laboratory, British Geological Survey Keyworth Room 254 ESS Building
Notts NG12 5GG Stony Brook, NY 11794
UK USA
rrp@nigl.nerc.ac.uk troy.rasbury@sunysb.edu

David Phillips Jesse Reimink

School of Earth Sciences Department of Earth and Atmospheric Sciences
The University of Melbourne University of Alberta
Parkville, VIC 3010 Edmonton, AB
Australia Canada T6G 2E3
dphillip@unimelb.edu.au reiminkjesse@gmail.com

Aaron J. Pietruszka Wolf Uwe Reimold

Denver Federal Center Museum für Naturkunde Berlin and
U.S. Geological Survey Humboldt-Universität zu Berlin
Denver, CO 80225 Berlin
USA Germany
apietruszka@usgs.gov uwe.reimold@mfn-berlin.de

Jeffrey S. Pigati David A. Richards

Denver Federal Center, U.S. Geological Survey School of Geographical Sciences
MS-980 University of Bristol
Denver, CO 80225 Bristol BS8 1SS
USA UK
jpigati@usgs.gov david.richards@bristol.ac.uk
 CONTRIBUTORS xxi

Daniel Richter Daniela Rubatto

Department of Human Evolution Research School of Earth Sciences
Max Planck Institute for Evolutionary Anthropology The Australian National University
Deutscher Platz 6 Mills Road, Bld.61
04103 Leipzig Canberra, ACT 0200
Germany Australia
and daniela.rubatto@anu.edu.au
Geomorphologie
University of Bayreuth
Bayreuth Joerg M. Schaefer
Germany Lamont-Doherty Earth Observatory
drichter@eva.mpg.de Columbia University
409 Comer 61 Route 9W
Palisades, NY 10964-8000
W. Jack Rink USA
School of Geography and Earth Sciences schaefer@ldeo.columbia.edu
McMaster University
1280 Main St. W.
Hamilton, ON Gerhard Schellmann
Canada L8S 4K1 Department of Physical Geography & Landscape Studies
rinkwj@mcmaster.ca University of Bamberg
Am Kranen 1
Bamberg
Helen M. Roberts Germany
Department of Geography and Earth Sciences gerhard.schellmann@uni-bamberg.de
Aberystwyth University
Llandinam Building, Penglais Campus
Aberystwyth SY23 3DB Bernd R. Schöne
UK Institute of Geosciences
hmr@aber.ac.uk University of Mainz
Mainz
Germany
schoeneb@uni-mainz.de
Richard G. Roberts
Centre for Archaeological Science
School of Earth and Environmental Sciences
University of Wollongong Derek W. G. Sears
Wollongong, NSW 2522 NASA Ames Research Center
Australia Mountain View, CA 94035
rgrob@uow.edu.au USA
derek.sears@nasa.gov

Rolf L. Romer Christian Seiler

Inorganic and Isotope Geochemistry School of Earth Sciences
GFZ German Research Centre for Geosciences University of Melbourne
Telegrafenberg McCoy Bldy
14473 Potsdam Melbourne, VIC 3010
Germany Australia
romer@gfz-potsdam.de seiler@unimelb.edu.au

David Selby
Alan D. Rooney Department of Earth Sciences
Department of Earth and Planetary Sciences University of Durham
Harvard University Cambridge Science Labs
Cambridge, MA Durham DH1 3LE
USA UK
alan.d.rooney@gmail.com david.selby@durham.ac.uk
xxii CONTRIBUTORS

Warren D. Sharp Vicente Soler Javaloyes

Berkeley Geochronology Center Estación Volcanológica de Canarias, IPNA-CSIC
2455 Ridge Road Avda. Astrofísico Francisco Snchez, n 3
Berkeley, CA 94709 38206 La Laguna
USA Spain
wsharp@bgc.org vsoler@ipna.csic.es

Holly Stein
Thibaud Simon-Labric AIRIE Program
Institute of Earth Surface Dynamics (IDYST) Colorado State University
Université de Lausanne (UNIL) Fort Collins, CO
Geopolis USA
1015 Lausanne and
Switzerland Centre for Earth Evolution and Dynamics (CEED)
thibaud.simon-labric@unil.ch University of Oslo
Oslo
Norway
Keith Sircombe holly.stein@colostate.edu
Geoscience Australia
Jerrabomberra Avenue, Symonston ACT 2617
Canberra, ACT 2601 Mordechai Stein
Australia Geological Survey of Israel
keith.sircombe@ga.gov.au 30 Malkhe Israel Street
95501 Jerusalem
Israel
motistein@gsi.gov.il
Anne Skinner
Chemistry Department
Williams College Gary E. Stinchcomb
47 Lab Campus Drive Department of Geosciences
Williamstown, MA 01267 Pennsylvania State University
USA 0302 Hosler Building
anne.r.skinner@williams.edu University Park, PA 16802
USA
ges130@psu.edu
Joseph M. Smoak
Department of Environmental Science, Policy and Claudine H. Stirling
Geography Department of Chemistry
University of South Florida University of Otago
140 7th Ave. South Dunedin
St. Petersburg, FL 33701 New Zealand
USA cstirling@chemistry.otago.ac.nz
smoak@mail.usf.edu
Markus Stoffel
Institute for Geological Sciences, University of Bern
Reza Sohbati
Baltzerstrasse 1 + 3
Department of Geoscience
3012 Bern
Aarhus University
Switzerland
Aarhus
markus.stoffel@dendrolab.ch
Denmark
and
Center for Nuclear Technologies Donna M. Surge
Technical University of Denmark Department of Geological Sciences
Building 201, DTU Risø University of North Carolina
Frederiksborgvej 399 104 South Road Mitchell Hall Campus
Roskilde Chapel Hill, NC 27599-3315
Denmark USA
resih@dtu.dk donna64@unc.edu
 CONTRIBUTORS xxiii

Kazuhiro Suzuki Jessica A. Thomas

The Center for Chronological Research Department of Biology
Nagoya University University of York
464-8602 Nagoya, Chikusa-ku, Aichi York
Japan UK
suzuki@nendai.nagoya-u.ac.jp jessica.thomas@york.ac.uk

Anders Svensson Jeroen W. Thompson

Niels Bohr Institute, Ice and Climate Research Department of Medical Physics and Applied Radiation
University of Copenhagen Sciences
Copenhagen McMaster University
Denmark 1280 Main St West
as@gfy.ku.dk Hamilton, ON
Canada L8S 4K1
thompjw@mcmaster.ca
Peter W. Swarzenski
US Geological Survey, Pacific Coastal and Marine Kristina Jørkov Thomsen
Science Center Centre for Nuclear Technologies
400 Natural Bridges Dr. Technical University of Denmark
Santa Cruz, CA 95060 Frederiksborgvej 399, Building 201, room S28
USA 4000 Roskilde
pswarzen@usgs.gov Denmark
krth@dtu.dk
Timothy Swindle
Lunar and Planetary Laboratory Helene Tissoux
University of Arizona DGR/GAT
1629 E. University Boulevard Bureau de Recherches Géologiques et Minières (BRGM)
Tucson, AZ 85721-0092 3 avenue Claude Guillemin
USA BP 36009
tswindle@email.arizona.edu 45060 Orléans
France
h.tissoux@brgm.fr
Paul Sylvester
Department of Geosciences
Texas Tech University Stephen Tooth
Lubbock, TX Department of Geography and Earth Sciences
USA Aberystwyth University
pjsylvester@gmail.com Llandinam Building, Penglais Campus
Aberystwyth SY23 3DB Ceredigion
UK
Larry Taylor set@aber.ac.uk
Department of Earth and Planetary Science
University of Tennessee Trond H. Torsvik
1412 Circle Drive Centre for Earth Evolution and Dynamics (CEED)
Knoxville, TN 37996-1410 University of Oslo
USA Oslo
lataylor@utk.edu Norway
t.h.torsvik@geo.uio.no
Matt Telfer
School of Geography, Earth and Environmental Science Sumiko Tsukamoto
University of Plymouth Leibniz Institute for Applied Geophysics (LIAG)
Room A513 Portland Square Stilleweg 2
Plymouth PL4 8AA 30655 Hannover
UK Germany
matt.telfer@plymouth.ac.uk sumiko.tsukamoto@liag-hannover.de
xxiv CONTRIBUTORS

Gillian Turner Tod Waight

School of Chemical and Physical Sciences Department of Geosciences and Natural Resource
Victoria University of Wellington Management, Geology Section
Wellington 6140 University of Copenhagen
New Zealand Øster Voldgade 10
gillian.turner@vuw.ac.nz 1350 Copenhagen
Denmark
todw@ign.ku.dk
Pieter van der Beek
Institute des Sciences de la Terre (ISTerre), CNRS Richard J. Walker
Université Joseph Fourier, BP 53 Isotope Geochemistry Laboratory
38041 Grenoble Department of Geology
France University of Maryland
pvdbeek@ujf-grenoble.fr College Park, MA 20742
USA
rjwalker@umd.edu
Ben van der Pluijm
Department of Earth and Environmental Sciences Jakob Wallinga
University of Michigan Soil Geography and Landscape group
4534b C.C. Little Building, 1100 North University Ave Wageningen University
Ann Arbor, MI 48109-1005 6700 AA Wageningen
USA The Netherlands
vdpluijm@umich.edu jakob.wallinga@wur.nl

Rachel Warnock
Jorge A. Vazquez School of Earth Sciences
SHRIMP-RG Laboratory University of Bristol
United States Geological Survey Wills Memorial Building Queen’s Road
345 Middlefield Rd Bristol
Menlo Park, CA 94025 UK
USA rachel.warnock@bristol.ac.uk
jvazquez@usgs.gov

Clare Warren
Jeff Vervoort Department of Earth and Environmental Sciences
School of the Environment The Open University
Washington State University MK7 6AA Milton Keynes
Pullman, WA 99164 UK
USA c.warren@open.ac.uk
vervoort@wsu.edu
John F. Wehmiller
Department of Geological Sciences
David Wacey University of Delaware
Centre for Microscopy Characterisation and Analysis Newark, DE 19716
University of Western Australia USA
Crawley, Perth jwehm@udel.edu
Australia
david.wacey@uwa.edu.au
Dominique Weis
Department of Earth, Ocean, and Atmospheric Sciences
Günther A. Wagner Pacific Centre for Isotopic and Geochemical Research
Geographisches Institut der Universität Heidelberg The University of British Columbia
Im Neuenheimer Feld 348 6339 Stores Road
69120 Heidelberg Vancouver, BC
Germany Canada V6T 1Z4
gawag-wagner@web.de dweis@eos.ubc.ca
 CONTRIBUTORS xxv

Kees Welten Peter K. Zeitler

Space Sciences Laboratory Department of Earth and Environmental Sciences
University of California Lehigh University
#7450 at Berkeley 7 Gauss Way 1 West Packer Avenue
Berkeley, CA 94720 Bethlehem, PA 18015-3001
USA USA
kcwelten@ssl.berkeley.edu peter.zeitler@lehigh.edu

Stephanie C. Werner
Centre for Earth Evolution and Dynamics Susan Zimmerman
University of Oslo Center for Accelerator Mass Spectrometry
Oslo Lawrence Livermore National Laboratory
Norway 7000 East Ave, L-397
stephanie.werner@geo.uio.no Livermore, CA 94551-0808
USA
John A. Westgate zimmerman17@llnl.gov
Department of Earth Sciences
University of Toronto
22 Russell St. Ludwig Zöller
Toronto, ON Geographisches Institut
Canada M5S 3B1 Universität Bayreuth
westgate@es.utoronto.ca Universitätsstr. 30
95447 Bayreuth
Jan R. Wijbrans Germany
Department of Earth Sciences, Faculty of Earth and Life ludwig.zoeller@uni-bayreuth.de
Sciences
VU University
De Boelelaan 1085 Horst Zwingmann
Amsterdam CSIRO Earth Science and Resource Engineering
The Netherlands Australian Resources Research Centre
j.r.wijbrans@vu.nl Technology Park 26 Dick Perry Avenue, Kensington
Perth, WA 6151
Simon A. Wilde Australia
School of Applied Geology and
Curtin University School of Earth and Environment
GPO Box U1987 The University of Western Australia
Perth Crawley, WA 6009
Australia Australia
s.wilde@curtin.edu.au and
Department of Applied Geology
Camilla Wilkinson Curtin University
Norges Geologiske Undersøkelse (NGU) Bentley, WA 6845
Leiv Eirikssons vei 39 Australia
Trondheim 7040 and
Norway Institut de Recherche sur les Archéomatériaux, UMR
camilla.wilkinson@ngu.no 5060 CNRS – Université de Bordeaux, Centre de
Recherche en Physique Appliquée à l’Archéologie
Jane Willenbring (CRP2A), Maison de l’archéologie
Department of Earth and Environmental Sciences 33607 Pessac
University of Pennsylvania France
Philadelphia, PA 19104-6313 horst.zwingmann@csiro.au
USA
jane.willenbring@sas.upenn.edu
 Preface
Scientific dating methods provide the timing, sequence,
and rates of geological, archaeological, and biological
events and processes. It is no exaggeration to say that
quantitative measurements of age (geochronology) pro-
vide the foundations for scientific understanding for many
disciplines within the earth and archaeological sciences.
The field of geochronology began with early applica-
tions of biostratigraphy and a basic understanding of
sedimentary processes. The ages and sequences fossils
provided the foundation for the geological time scale. At
the turn of the twentieth century, our understanding of
radioactivity revolutionized geochronology. This led to
the flourishing of numerous radiogenic isotopic dating
methods. In addition, of particular importance, was the
development of radiocarbon dating in 1949, which has
since yielded hundreds of thousands of age estimates for
earth scientists and archaeologists. Radiation exposure
methods, which utilize the effects of background radiation
on defects in minerals and biological materials, were
developed through the 1960s and the 1990s. More
recently, the development of molecular clock techniques
has resulted in a new approach to determine ages of events
in the history of biological evolution.
This volume is a comprehensive synthesis of the appli-
cations and physical basis for scientific dating methods in
use in the earth sciences, archaeology, and biology. All
widely-accepted scientific dating techniques – physical,
chemical, and biological – have been included, as well
as the most important materials which are amenable to
the application of scientific dating methods.
We trust that this volume will be of use to researchers
and students in the earth sciences and archaeology, who
wish to understand the scientific basis that underlies our
understanding of geological and archaeological chronol-
ogy. In addition, this volume may be useful to geologists
involved in exploration and exploitation of natural
resources, natural resource managers, and environmental
and archaeological consultants.
Each of the major dating methods is described in a main
entry that provides an in-depth review of the underlying
scientific principles of that method, including methods,
applications, uncertainties, applications, and limitations.
If appropriate, the most recent development in each field
is discussed. Each of these main entries was authored by
a leading expert in that field.
The majority of the entries in this volume are focused
on applications of scientific dating methods, and are
usually titled according to the material to be dated (e.g.,
“Carbonates, marine”), with the method in parenthesis.
We have attempted to provide comprehensive coverage
of organic and inorganic materials, including minerals,
rocks, archaeological materials, biominerals, plants, art
objects, water, and many more. Some entries focus on
rates of geological processes, such as sedimentation, fluid
flow, tectonics, cooling rates, and many more. By organiz-
ing entries by the application, rather than the methodol-
ogy, we hope that readers will be able to quickly locate
information most relevant for their interests and specific
needs.
Finally, this volume includes shorter, mini-entries with
key definitions, important materials, or notes on
instrumentation.
This volume was only possible through the extensive
contributions of the three associate editors and the large
editorial board who worked together to establish the range
of authors from 18 different countries who agreed to con-
tribute. The online version, which can be updated by the
authors as new information becomes available, provides
a dynamic dimension in the rapidly changing field of
geo-, bio-, and archaeo-chronology.
The field of dating methods continues to grow rapidly
through research scientists thinking of new ways to apply
the methods. Though it cannot be said that every single
application of dating methods is included, this volume sig-
nificantly expands the availability of knowledge through
its broad scope in each area in the field of dating.
 Acknowledgments
First of all we want to thank all of the authors for their long
hours composing the text and graphics of the entries, and
for answering our correspondence. Without their contribu-
tions, the encyclopedia could not have become a reality.
There are a large number of individuals who contrib-
uted to both the design and fulfillment of the aims of this
volume. Our section editors made great contributions in
this area and in finalizing the author list. The person who
made the biggest impact is Larry Heaman who stepped
in to handle much of the creation and editing of the radio-
genic isotope side of the volume. He has given of his time
extensively as decisions were made in all of the stages of
development, and we are both very grateful for this contri-
bution. Jim Paces also shouldered a large amount of the
creation and editing of the encyclopedia, and we thank
him for his contribution, especially in regard to the crea-
tion of the uranium series dating articles. Tim Jull also cre-
ated an entire section of the encyclopedia with a large
range of articles on radiocarbon applications, and without
him, we would not have achieved the great depth of the
entries we have. Many thanks to you Tim, from both of us.
The editorial board was seminal in creating the author-
ship of the encyclopedia. Our great thanks to Robert
A. Creaser, Don Davis, Geoff Duller, John Gosse, Simon
Ho, James K.W. Lee, Josep M. Pares, Peter W. Reiners,
and John F. Wehmiller, who were seminal in the early
stages in establishing the breadth of this volume. The
editorial board made decisions regarding how to “boil
down” which applications articles were the most relevant
to current developments in the fields of geochronology,
and made suggestions that led to the creation of synthe-
sized articles which encompassed more than one of the
original topics. Each of you made a great contribution,
and we are indebted to you for helping us condense our
early ideas into the most relevant entries that could be
created.
Our most sincere thanks go to the Springer personnel
who worked painstakingly throughout the process to
smooth our path. Petra van Steenbergen provided inspira-
tion, encouragement, and opportunity, while Sylvia
Blago, Simone Giesler, and Hermine Vloemans gave
countless hours regarding the correspondence and organi-
zation of entries and scheduling.
WJR would like to thank Jeroen Thompson, who
agreed to help me create the encyclopedia in its earliest
stages. WJR could not have gone forward without his
encouragement, but also his collaboration in creating the
original proposal and in creating the documents necessary
for its acceptance as a viable work of science. JWT would
like to gratefully acknowledge Jack Rink for shouldering
additional responsibilities during the project.
WJR would also like to acknowledge his graduate
students Kathleen Rodrigues, Robert Rombuck
Hendricks, and Alex Hodson, who helped him to make it
through this long-term project, at the cost of giving them
less time than he would have normally given them. To
Sarah, Katherine, and Rebecca, JWT expresses his love
and gratitude for their support. Finally, WJR dedicates
the volume to Carol; thank you dearest for willingly and
lovingly living with less of my presence.
W. Jack Rink
Jeroen W. Thompson
 A
ACASTA GNEISS COMPLEX
Samuel Bowring1, Tom Chacko2, Larry M. Heaman2 and
Jesse Reimink2
1
Department of Earth, Atmospheric and Planetary
Sciences, Massachusetts Institute of Technology,
Cambridge, MA, USA
2
Department of Earth and Atmospheric Sciences,
University of Alberta, Edmonton, AB, Canada
Synonyms
Acasta gneisses
Definition
The Acasta gneiss complex in northwestern Canada con-
tains some of the oldest dated rocks on earth and the oldest
dated using U-Pb in zircon. They therefore potentially
preserve clues to the nature and history of earth’s oldest
crust. The geology of the gneisses reflects a complex
multistage history of magmatism, metamorphism, and
deformation over more than 400 million years, and the
entire complex should be viewed as a small fragment
of Hadean to Archean continental crust. The
gneisses are compositionally diverse and range from
granite and tonalite to gabbro and serpentinized
ultramafic rocks. The oldest published crystallization
age is 4.031 +/
0.003 Ga. There are isotopic
and inherited zircon data consistent with the involvement
of crust as old as 4.2 Ga. There is no single age
of the Acasta gneiss complex (Bowring and
Williams, 1999).
Introduction
Components of the Acasta gneiss complex are among the
oldest known rocks on earth. They are exposed in north-
western Canada (65 100 N and 115 300 W) along the
J.W. Rink, J.W. Thompson (eds.), Encyclopedia of Scientific Dating Methods, DOI 10.1007/978-94-007-6304-3,
© Springer Science+Business Media Dordrecht 2015
western margin of the Archean Slave craton (>2.58 Ga),
in the core of a north-trending fold in the foreland of the
Wopmay orogen, a 2.02–1.84 Ga-old orogenic belt. The
Acasta gneisses range in age from 4.03 to ca 3.6 Ga
(Bowring and Williams, 1999; Stern and Bleeker, 1998;
Iizuka et al., 2006, 2007). All age groups are composition-
ally diverse and range from granite to quartz diorite to
tonalite. The deformational history is complex with sev-
eral episodes resulting in well-developed foliations. In
low-strain domains, crosscutting relationships can be
observed (Figure 1), but in general, deformation has
obscured primary relationships between different rock
types. Lens-shaped boudins of serpentinized ultramafic
rocks occur throughout the gneisses; some are 100’s of
meters in long dimension. No ca. 4.0–3.6 Ga
metasedimentary rocks have been discovered although
sparse outcrops of quartzite, iron formation, and pelite
are found locally tightly folded into older gneisses.
A wide variety of weakly deformed granitic dikes that
intrude older gneisses are ca 3.6, 3.4, 2.9, and 2.6 Ga
(Bleeker and Stern, 1997; Stern and Bleeker, 1998). Dur-
ing 1.88 Ga Calderian orogeny to the west, thrust sheets of
1.9–2.5 Ga rocks were thrust over the western edge of
Slave craton resulting in a set of north-trending folds and
metamorphism of underlying Archean rocks. Ar-Ar bio-
tite and U-Pb apatite dates record complex reheating and
cooling during this event and are as young as ca 1.77 Ga
(Sano et al., 1999)
Age constraints
U-Pb geochronological studies indicate that the Acasta
gneisses range in age from 4.03 to ca. 3.6 Ga with distinct
groupings at 4.03–3.94, 3.74–3.72, and 3.66–3.58 Ga,
although this may be an artifact due to a relatively small
number of published analyses. U-Pb zircon dates indicate
that the oldest igneous crystallization ages for Acasta
gneiss protoliths that range from granite to tonalite/diorite
2 ACASTA GNEISS COMPLEX
Acasta Gneiss Complex, Figure 1 Outcrop photos of ancient
gneissic rocks of the Acasta gneiss complex, Acasta River area,
N.W.T. (a) Layered multicomponent gneiss consisting of tonalite
(dark grey), granodiorite (grey), and granite (pink) layers. (b)
Layered tonalite-granodiorite gneiss crosscut by multiple
generations of younger granite.
are 4.012  0.006 to 4.030  0.003 Ga. Many zircons
from all rock types contain older cores with the oldest at
4.065  0.016 and 4.2 Ga (Iizuka et al., 2006), consistent
with the involvement of even older crust in their genera-
tion by partial melting or assimilation (Bowring and
Williams, 1999). In general, the geochemistry of the
Acasta gneisses is not distinctive from other Archean
and younger rocks; they are on average enriched in light
rare-earth elements with variable depletion in heavy rare-
earth elements thought to reflect the involvement of garnet
in the source area. Radiogenic isotope systematics in
whole rocks (Sm-Nd) and zircon (Lu-Hf) are also consis-
tent with the involvement of older continental crust. Many
of the rocks have zircons with thin overgrowths likely
related to episodes of metamorphism that range from 3.8
to 3.4 Ga.
Concluding remarks
The formation and preservation of continental crust early
in earth history is of broad interest to earth and planetary
scientists as the oldest continental crust records the only
history of magma formation and the role of water in gen-
erating granitic magmas more than 4 billion years ago.
With estimates for the age of the earth-moon system
between 4.5 and 4.4 Ga and the oldest detrital zircon from
Australia, 4.374  0.0006 Ga (Valley et al., 2014), the
Acasta gneiss complex with igneous rocks as old as
4.030 Ga with evidence for involvement of even older
crust offers important insights into early crustal evolution.
Finally, many of the rocks in the Acasta gneiss complex
are very similar in composition to those formed much later
in earth history by plate tectonic processes, and there is no
compelling evidence of the late heavy bombardment
(Mojzsis et al., 2014) preserved in the Acasta gneisses.
Bibliography
Bleeker, W., and Stern, R. A., 1997. The Acasta Gneisses: an imper-
fect sample of Earth’s oldest crust. In Cook, F., and Erdmer,
P. (eds.), Slave-Northern Cordillera Lithospheric Evolution
(SNORCLE) Transect and Cordillera Tectonics Workshop Meet-
ing, Lithoprobe Report 56. pp. 32–35.
Bowring, S. A., and Housh, T. B., 1995. The Earth’s early evolu-
tion. Science, 269, 1535–1540.
Bowring, S. A., and Williams, I. S., 1999. Priscian (4.00–4.03 Ga)
orthogneisses from northwestern Canada. Contributions to Min-
eralogy and Petrology, 134, 3–16.
Bowring, S. A., Housh, T. B., and Isachsen, C. E., 1990. The Acasta
Gneisses: Remnant of Earth’s Early Crust. Origin of the Earth.
New York: Oxford University Press.
Iizuka, T., Horie, K., Komiya, T., Maruyama, S., Hirata, T., Hidaka,
T., and Windley, B. F., 2006. 4.2 Ga zircon xenocryst in an
Acasta gneiss from northwestern Canada: evidence for early
continental crust. Geology, 34, 245–248.
Iizuka, T., Komiya, T., Ueno, Y., Katayama, I., Uehara, Y.,
Matuyama, S., Hirata, T., Johnson, S. P., and Dunkley, D. J.,
2007. Geology and zircon geochronology of the Acasta Gneiss
Complex, northwestern Canada: new constraints on its
tectonothermal history. Precambrian Research, 153, 179–208.
Mojzsis, S. J., Cates, N. L., Caro, G., Trail, D., Abramov, O.,
Guitreau, M., Blichert-Toft, J., Hopkins, M. D., and Bleeker,
W., 2014. Component geochronology in the polyphase ca
3920 Ma Acasta Gneiss. Geochimica et Cosmochimica Acta,
133, 68–96.
Sano, Y., Terada, K., Hidaka, H., Yokoyama, K., and Nutman, A. P.,
1999. Palaeoproterozoic thermal events recorded in the 4.0 Ga
Acasta gneiss, Canada: evidence from SHRIMP U-Pb dating of
apatite and zircon. Geochimica et Cosmochimica Acta, 63,
889–905.
Stern, R., and Bleeker, W., 1998. Age of the world’s oldest rocks
refined using Canada’s SHRIMP: the Acasta Gneiss Complex,
Northwest Territories, Canada. Geoscience Canada, 25, 27–31.
Valley, J. W., Cavosie, A. J., Ushikubo, T., Reinhard, D. A., Law-
rence, D. F., Larson, D. J., Clifton, P. H., Kelly, T. F., Wilde,
S. A., Moser, D. E., and Spicuzza, M. J., 2014. Hadean age for
a post-magma-ocean zircon confirmed by atom-probe tomogra-
phy. Nature Geoscience, 7, 219–223.
 ACCELERATOR MASS SPECTROMETRY
ACCELERATOR MASS SPECTROMETRY
A. J. Timothy Jull1 and George S. Burr2,3
1
NSF-Arizona AMS Laboratory, Physics Building,
University of Arizona, Tucson, AZ, USA
2
NSF Arizona Accelerator Mass Spectrometry
Laboratory, Department of Geosciences and Physics,
University of Arizona, Tucson, AZ, USA
3
Department of Geosciences, National Taiwan University,
Taipei, Taiwan
Synonyms
Accelerator dating; AMS; Atom counting
Definition
Accelerator mass spectrometry is a technique that combines
a particle accelerator with a mass spectrometer in order to
measure very low levels (10
16) of cosmogenic and anthro-
pogenic radionuclides employed for dating purposes.
Introduction
Accelerator mass spectrometry (AMS) is widely used to
measure rare isotope ratios of cosmogenic and anthropo-
genic nuclides. Cosmogenic isotopes are produced through
the interaction of cosmic rays with atmospheric molecules,
rocks at the earth’s surface (Dunai, 2010) and in extraterres-
trial settings. AMS is the analytical tool of choice for a range
of isotopes used for dating purposes, especially radiocarbon
dating and surface exposure dating. Table 1 lists a number of
these isotopes, hereafter referred to as AMS isotopes. These
can potentially be used to date samples from years to tens
of millions of years old. The chief advantage of AMS over
standard mass spectrometry is that it eliminates molecular
interferences using particles with 100 s of keV to 10s of
MeV energies. In recent years the technological develop-
ment of AMS instruments has focused on the production
of smaller (<1 MV) machines, which are less expensive to
fabricate, require less space, and are easier to maintain than
larger machines. This trend has seen an expansion of AMS
facilities around the world. Currently there are about
100 active AMS machines worldwide (Kutschera, 2013).
History and development of AMS
The concept and implementation of AMS date to the late
1970s, when Muller (1977) proposed the idea based on
a cyclotron experiment conducted in the 1930s to measure
3
H and 4He (Alvarez and Cornog, 1939). Muller showed that
a particle accelerator could be used as a mass spectrometer to
drastically reduce sample sizes and counting times and
improve precision and background levels encountered in
conventional decay counting of 14C (radiocarbon) and 10Be.
Within months, Nelson et al. (1977) and Bennett
et al. (1977) implemented the idea using linear particle
accelerators. These measurements were made using
3
Accelerator Mass Spectrometry, Table 1 Isotopes measured
with AMS for dating studies and their half-lives
Isotope Half-life (yr) Reference
10
Be 1.387  0.012  106 Korschinek et al. (2010)
32
Si 1.44  0.11  102 Fifield and Morgenstern (2009)
14
C 5.700  0.030  103 Kutschera (2013)
41
Ca 9.94  0.15  104 Jörg et al. (2012)
81
Kr 2.29  0.11  105 Baglin (2008)
36
Cl 3.013  0.015  105 Nica et al. (2012)
26
Al 7.18  0.18  105 Kutschera (2013)
53
Mn 3.7  0.4  106 Honda and Imamura (1971)
129
I 1.57  0.04  107 Tendow (1996)
relatively large accelerators operating with terminal volt-
ages of 7 and 8 MV and particles ranging from 35 to
40 MeV. Over the next several years, smaller (3 MV),
purpose-built AMS instruments began to appear. Since that
time, smaller instruments have been introduced and the
trend toward smaller and smaller machines continues.
Nowadays, especially for 14C, systems are being designed
with accelerator voltages as low as 200 kV. Such systems,
designed at ETH Zürich, have already been installed in sev-
eral laboratories (Suter, 2010). While further developments
may reduce this voltage further, some nuclides such as 10Be
and 36Cl still require higher energies to eliminate isobaric
interferences and maintain high transmission.
AMS design
Accelerator mass spectrometers all share the same basic
components: (1) ion source, (2) injection (low-energy)
magnet with an electrostatic field switching device
(bouncer), (3) particle accelerator and stripper, (4) high-
energy particle filters for magnetic and electrostatic separa-
tion, and (5) a particle detector (Figure 1). All designs of
AMS use negative ions that are produced in an ion source
and are then accelerated through a potential of up to
50 kV, focused into a particle beam and sent through an
injection magnet. The injection magnet performs an initial
separation of beams of different masses. However, molecu-
lar ions are still present at these lower energies. After the
ions pass through the injection magnet, they are further
accelerated by the voltage on the AMS terminal. The termi-
nal contains a “stripper” which can be a gas canal or thin
film. When the high-energy ions pass through the stripper,
molecules are dissociated and the injected particles lose
electrons to become positive. In tandem accelerators the
resulting positive particles continue to be accelerated in
the same direction, back to ground potential. One design,
the simplified “single-stage AMS,” places the ion source
at the accelerating potential and the detectors are at ground
potential. In this device the high-energy ions are analyzed
on their way to the detector using a variety of magnetic
and electrostatic filters. The ion beam is focused along the
path from the ion source to the detector, using different
4 ACCELERATOR MASS SPECTROMETRY
Accelerator Mass Spectrometry, Figure 1 Typical AMS design (From Synal, 2013).
combinations of Einzel, quadrupole, and gridded lenses,
accompanied by collimating slits and electrostatic steering
devices. The choice of lenses and their placement depend
on the design and model of each AMS and vary consider-
ably. The detection system can consist of solid-state
devices (surface-barrier detectors) or gas-filled detectors,
where further analysis of ion energy and energy loss in
a gas can be measured. AMS detectors generally can identify
ions by both mass and atomic number, allowing for complete
separation of isobaric interferences (Litherland et al., 2010).
Dating applications
AMS dating applications are primarily concerned with the
measurement of cosmogenic radionuclides, but also
include applications using anthropogenic isotopes. Radio-
carbon was the first cosmogenic nuclide to be widely
applied as a dating tool using AMS (Elmore and Phillips,
1987). Most radiocarbon is produced in the atmosphere as
a result of cosmic ray interactions with atmospheric mole-
cules. These reactions supply thermal neutrons that pro-
duce radiocarbon through the 14N(n,p)14C reaction.
Atmospheric radiocarbon rapidly spreads into terrestrial
and marine environments around the globe along
a variety of organic and inorganic pathways. When
a pathway is cut off, for example, when an organism dies,
its radiocarbon content begins to drop as it decays away at
a known rate. An AMS radiocarbon date can be deter-
mined on such material by comparison between the mea-
sured radiocarbon content in the sample and in
a standard representing the atmospheric value. The limit
of the AMS radiocarbon dating technique is about
55 kyr. Anthropogenic (non-cosmogenic) radiocarbon
can also be used for dating. Bomb carbon introduced into
the atmosphere through aboveground nuclear weapon
testing in the late 1950s and early 1960s produced
a spike which has been widely exploited for dating sam-
ples from the last 60 years (see entry “Bomb Carbon” by
Burr in this volume). In this case excess radiocarbon,
above the expected atmospheric value, is determined with
AMS. A third production pathway for radiocarbon is
through spallation reactions within rocks and meteorites.
This radiocarbon is well suited to in situ dating
(discussed next) and to determine the terrestrial ages of
meteorites (Jull, 2001).
In situ dating is concerned with the buildup of cosmo-
genic nuclides that occurs when a geologic surface is
exposed to cosmic rays (Gosse and Phillips, 2001). Some
examples of applications are to determine the age of a lava
flow, or the age of a glacial deposit marking the retreat of
the glacier. AMS is a truly pioneering method in the case
of in situ studies because the sample size requirements,
 ACCELERATOR MASS SPECTROMETRY
sensitivity, and precision of the AMS technique made in
situ dating possible. Most in situ dating studies measure
10
Be, 14C, 26Al, and 36Cl. In situ dating has advanced rap-
idly since the introduction of AMS and has seen a steady
expansion of applications. Examples include (1) surface
exposure dating on the earth and moon, (2) determination
of erosion and weathering rates, (3) burial dating, and
(4) determination of the terrestrial ages of meteorites.
The goal of surface exposure dating is to date fresh rock
or soil surfaces that have been continuously exposed to
cosmic rays since they formed and ideally have experi-
enced only minor erosion (Granger et al., 2013). In situ
isotopes can also be used to infer erosion and weathering
rates when the age of a surface is known independently.
In this case, the amount of material removed by both phys-
ical and chemical erosion can be assessed. When two cos-
mogenic isotopes are measured in the same rock, it can
also be possible to determine both surface exposure age
and erosion rate (Granger et al., 2013). A recent addition
to this group of in situ isotopes is 53Mn, which has been
measured in several laboratories and has been used to
study both terrestrial and extraterrestrial samples.
Another AMS isotope commonly used for dating appli-
cations is 129I. Like radiocarbon, both cosmogenic and
anthropogenic sources of 129I can be used for dating. 129I
is particularly useful for dating brines and evaporite
deposits (Fabryka-Martin et al., 1984) because iodine is
very soluble and becomes concentrated in these materials.
129
I has a very long half-life (1.57  107 years) and has
been used to date brine samples as far back as the Miocene
(Tomaru et al., 2009). Iodine is a near-conservative ele-
ment in groundwater and has showed promise as
a method suitable for dating young water (Schwehr
et al., 2005).
The remaining AMS isotopes given in Table 1 (32Si,
41
Ca, 81Kr) are less widely studied but have potential as
isotopic chronometers. 32Si has long been recognized as
a potentially useful radioisotope for dating oceanographic
samples (Lal et al., 1960), but poses significant analytical
challenges, in part due to its low abundance. Fifield and
Morgenstern (2009) have demonstrated the feasibility of
dating glacial ice with AMS measurements of 32Si. The
potential for 41Ca has also been long recognized (Paul
et al., 1985), but is a challenging isotope to measure.
41
Ca has been measured to determine cosmic ray exposure
ages of meteorites (Herzog et al., 2011). The strictly con-
servative nature of krypton in groundwater makes 81Kr
a valuable isotope in determining the residence time of
ancient groundwater. Collon et al. (2000) developed
a specialized sample processing technique to extract kryp-
ton from groundwaters of the Great Artesian Basin in Aus-
tralia and determined ages from about 2 to 4  105 years.
Bibliography
Alvarez, L., and Cornog, R., 1939. Helium and hydrogen of mass
3. Physical Review, 56, 613.
Baglin, C. M., 2008. Nuclear data sheets for A ¼ 81. Nuclear Data
Sheets, 109, 2257–2437.
5
Bennett, C. L., Beukens, R. P., Clover, M. R., Gove, H. E., Liebert,
R. B., Litherland, A. E., Purser, K. H., and Sondheim, W. E.,
1977. Radiocarbon dating using electrostatic accelerators: nega-
tive ions provide the key. Science, 198, 508–510.
Collon, P., Kutschera, W., Loosli, H. H., Lehmann, B. E., Purtschert,
R., Love, A., Sampson, L., Anthony, D., Cole, D., Davids, B.,
Morrissey, D. J., Sherrill, B. M., Steiner, M., Pardo, R. C., and
Paul, M., 2000. 81Kr in the Great Artesian Basin, Australia:
a new method for dating very old groundwater. Earth and Plan-
etary Science Letters, 182, 103–113.
Dunai, T., 2010. Cosmogenic Nuclides: Principles, Concepts, and
Applications in the Earth Surface Sciences. New York:
Cambridge University Press. 187 p.
Elmore, D., and Phillips, F. M., 1987. Accelerator mass spectrometry for
measurement of long-lived radioisotopes. Science, 236, 543–550.
Fabryka-Martin, J., Bentley, H., Elmore, D., and Airey, P. L., 1984.
Natural iodine-129 as an environmental tracer. Geochimica et
Cosmochimica Acta, 49, 337–347.
Fifield, L. K., and Morgenstern, U., 2009. Silicon-32 as a tool for
dating the recent past. Quaternary Geochronology, 4, 400–405.
Gosse, J. C., and Phillips, F. M., 2001. Terrestrial in situ cosmogenic
nuclides: theory and application. Quaternary Science Reviews,
20, 1475–1560.
Granger, D. E., Lifton, N. A., and Willenbring, J. K., 2013.
A cosmic trip: 25 years of cosmogenic nuclides in geology.
GSA Bulletin, 125(9/10), 1379–1402.
Hellborg, R., and Skog, G., 2008. Accelerator mass spectrometry.
Mass Spectrometry Reviews, 27, 398–427.
Herzog, G. F., Albrecht, A., Mai, P., Fink, D., Klein, J., Middleton,
R., Bogard, D. D., Nyquist, L. E., Shih, C.-Y., Garrison, D. H.,
Reese, Y., Masarik, J., Reedy, R. C., Rugel, G., Faestermann,
T., and Korschinek, G., 2011. Cosmic-ray exposure history of
the Norton County enstatite achondrite. Meteoritics & Planetary
Science, 46(2), 284–310.
Honda, M., and Imamura, M., 1971. Half-life of 53Mn. Physical
Review C, 4, 1182–1188.
Jörg, G., Amelin, Y., Kossert, K., and Gostomski, L. v., 2012. Pre-
cise and direct determination of the half-life of 41Ca. Geochimica
et Cosmochimica Acta, 88, 51–65.
Jull, A. J. T., 2001. Terrestrial ages of meteorites. In Peuker-
Ehrenbrink, B., and Schmitz, B. (eds.), Accretion of Extraterres-
trial Matter Throughout Earth’s History. New York: Kluwer/
Plenum, pp. 241–266.
Korschinek, G., Bergmaier, A., Faestermann, T., Gerstmann, U. C.,
Knie, K., Rugel, G., Wallner, A., Dillmann, I., Dollinger, G., von
Gostomski, L., Kossert, K., Maiti, M., Poutivtsev, M., and
Remmert, A., 2010. A new value for the half-life of 10Be by
heavy-ion elastic recoil detection and liquid scintillation
counting. Nuclear Instruments and Methods in Physics Research
B, 268, 187–191.
Kutschera, W., 2013. Applications of accelerator mass spectrome-
try. International Journal of Mass Spectrometry, 349–350,
203–218.
Lal, D., Goldberg, E. D., and Koide, M., 1960. Cosmic-ray-
produced silicon-32 in nature. Science, 131, 332–337.
Litherland, T., Zhao, X.-L., and Kieser, W. E., 2010. Mass spec-
trometry with accelerators. Mass Spectrometry Reviews, 30,
1037–1072.
Muller, R. A., 1977. Radioisotope dating with a cyclotron. Science,
196, 489–494.
Nelson, D. E., Korteling, R. G., and Stott, W. R., 1977. Carbon-14:
direct detection at natural concentrations. Science, 198, 507–508.
Nica, N., Cameron, J., and Singh, B., 2012. Nuclear data sheets for
A ¼ 36. Nuclear Data Sheets, 113, 1–155.
Paul, M., Fink, D., Meirav, O., Theis, S., and Englert, P., 1985.
Determination of Ca-41 production for meteorite studies by
accelerator mass-spectrometry. Meteoritics, 20(4), 726–727.
6 AGE OF THE EARTH
Schwehr, K. A., Santschi, P. H., Moran, J. E., and Elmore, D., 2005.
Near-conservative behavior of 129I in the Orange County aquifer
system, California. Applied Geochemistry, 20, 1461–1472.
Suter, M., 2010. Challenging developments in three decades of
accelerator mass spectrometry at ETH: from large particle accel-
erators to table size instruments. European Journal of Mass
Spectrometry, 16(3), 471–478, doi:10.1255/ejms.1078.
Synal, H.-A., 2013. Developments in accelerator mass spectrometry.
International Journal of Mass Spectrometry, 349–350, 192–202.
Tendow, Y., 1996. Nuclear data sheets for A ¼ 129. Nuclear Data
Sheets, 77, 631–770.
Tomaru, H., Lu, Z., Fehn, U., and Muramatsu, Y., 2009. Origin of
hydrocarbons in the Green Tuff region of Japan: 129I results from
oil field brines and hot springs in the Akita and Niigata Basins.
Chemical Geology, 264, 221–231.
AGE OF THE EARTH
Peter Barry and Larry Taylor
Department of Earth and Planetary Science, University of
Tennessee, Knoxville, TN, USA
Synonyms
Accretion time of Earth; Earth formation age; Planetary
differentiation; Timing of core formation
Definition
Current constraints on the age of the Earth suggest accre-
tion began approximately 4.54  0.05 b.y.a., shortly after
the formation of the solar system (4.567 b.y.a.). How-
ever, the exact accretion time of Earth is difficult to deter-
mine, with predictions from different accretion models
ranging from a few million up to about 100 m.y. after the
formation of the solar system. Not surprisingly, our under-
standing of the age of the Earth is deeply rooted in our
overall understanding of planetary science, particularly
in our ability to accurately date meteorites using various
radiogenic chronometers (e.g., U–Pb). The present-day
best estimate for the age of the Earth is not decidedly dif-
ferent than the initial estimate of 4.55  0.07 b.y.a. made
by Patterson in his pioneering 1956 work. However, the
resolution of the U–Pb isotope system is still not high
enough to distinguish between very small age variations
that may potentially separate the Earth, planets, and plan-
etary bodies (e.g., meteorites). As a result, the exact timing
of Earth formation is still a hotly debated issue, due in
large part to the fact that the first 500 m.y. of Earth
history are not well preserved (Halliday, 2004) in the
terrestrial rock record. The oldest known materials on
Earth are zircons from Western Australia, dated at
4.404  0.008 m.y. (Wilde et al., 2001); however, terres-
trial data of this age are limited and as a result this time
period is shrouded in mystery.
In order to circumvent the lack of data available for
early Earth history, geochronologists rely heavily on theo-
retical modeling and limited geochemical constraints from
short-lived radionuclides to decipher the absolute timing
of Earth formation. Xenon (129Xe and 136Xe) excesses in
Earth’s atmosphere suggest identifiable contributions
from the decay of the short-lived radionuclides 129I and
244
Pu, but in quantities considerably lower than predicted
from abundances observed in meteorites. This implies that
Earth did not achieve isotopic closure of the I–Pu–Xe sys-
tem until 100 m.y. after the formation of the solar system
(Pepin and Porcelli, 2006), an observation in agreement
with 235U/238U–207Pb/206Pb chronological methods
(Allegre et al., 1995).
A slightly different approach is to use the onset of dif-
ferentiation and core formation to define the exact age of
the Earth. In this regard, the U–Pb and 182Hf–182W iso-
tope systems are the most useful as they can potentially
reveal information about the timing of core formation.
However, these chronometers are unable to differentiate
between early accretion coupled with late core formation
and late accretion coupled with simultaneous core
formation. Nonetheless, the 182Hf–182W system unambig-
uously dates core formation between 10 and 15 m.y. after
the formation of the solar system (Yin et al., 2002;
Halliday, 2004) and, as a result, indirectly constrains
accretion timescales.
Accurately dating the Earth’s age is a fundamental
objective of Earth science and only with continued work
will the opposing models be fully reconciled with one
another. However, at present, the best age of Earth remains
4.54 b.y.
Bibliography
Allegre, C. J., Manhes, G., and Gopel, C., 1995. The age of the
earth. GCA, 59, 1445–1456.
Halliday, A. N., 2004. The origin and earliest history of the earth.
Treatise on Geochemistry, 1, 509–557.
Patterson, C., 1956. Age of meteorites and the earth. GCA, 10,
230–237.
Pepin, R. O., and Porcelli, D. R., 2006. Xenon isotope systematics,
giant impacts, and mantle degassing on the early earth. EPSL,
250, 470–485.
Wilde, S. A., Valley, J. W., Peck, W. H., and Graham, C. M.,
2001. Evidence from detrital zircons for the existence of conti-
nental crust and oceans on the earth 4.4 Gyr ago. Nature, 409,
175–178.
Yin, Q. Z., Jacobsen, S. B., Yamashita, K., Blichert-Toft, J., Telouk,
P., and Albarede, F., 2002. A short timescale for terrestrial planet
formation from Hf-W chronometry of meteorites. Nature, 418,
949–952.
Cross-references
Extraterrestrial Materials (K–Ar/Ar–Ar)
Jack Hills Zircon
Meteorites (Lu–Hf)
Meteorites (U–Pb)
Meteorites, Rubidium–Strontium, and Samarium–Neodymium
Chronology
Uranium–Lead, Extraterrestrial, Early Solar System
Uranium–Lead, Extraterrestrial, Planetary Materials
 ALPINE TERRANES (K–Ar/Ar–Ar)
ALPHA SPECTROSCOPY
Jeroen W. Thompson
Department of Medical Physics and Applied Radiation
Sciences, McMaster University, Hamilton, ON, Canada
Synonyms
Alpha counting; Thick source alpha counting
Definition
Measurement of U and Th concentrations in environmen-
tal samples through detection of alpha particles emitted by
isotopes in the U and Th decay chains.
Alpha particles emitted during radioactive decay are
extremely short range (in minerals, micrometers) but
deposit a significant quantity of energy. Therefore, the iso-
topes responsible for alpha emissions may be a large con-
tributor to the overall absorbed radiation dose for a datable
sample.
In order to measure the U and Th concentrations in
a mineral, the sample is ground to a powder and deposited
on a scintillator screen (mounted to a photomultiplier
tube) or on to a semiconductor detector. Alpha particles
deposit significantly more energy than beta particles or
gamma rays, so unwanted counts are rejected through
electronic processing of the signals recorded by the
detector.
Although thin samples would provide better energy res-
olution (essentially no loss of energy due to alpha particle
interactions within the sample), thick samples are pre-
ferred in order to achieve acceptable count rates. The sam-
ple thickness is therefore greater than the maximum range
of the naturally emitted alpha particles (“thick source
alpha counting”). Alpha dose rates are determined from
the count rate, sample density, and average alpha particle
range of all alpha emissions in a given decay chain
(Aitken, 1985).
U and Th concentrations may be determined through
other techniques, such as instrumental neutron activation
analysis (INAA) or x-ray fluorescence (XRF). However,
thick source alpha counting and high-resolution gamma
spectroscopy permit the identification of disequilibria in
the U and/or Th decay chains (Sjostrand and Prescott,
2002; Michael et al., 2008). This can result in more accu-
rate alpha dose rate measurements.
Bibliography
Aitken, M. J., 1985. Thermoluminescence Dating. London:
Academic.
Michael, C. T., Hein, A., Zacharias, N., and Kritidis, P., 2008. Dis-
equilibrium estimations in the U and Th series by using thick
source alpha-particle spectroscopy. Radiation Measurements,
43, 1149–1153.
Sjostrand, H., and Prescott, J. R., 2002. Thick source
alpha counting: the measurement of thorium. Ancient TL, 20,
7–10.
7
Cross-references
Electron Spin Resonance (ESR) Dating, General Principles
Radiation and Radioactivity
Radiation Dose Rate
ALPINE TERRANES (K–Ar/Ar–Ar)
Jan R. Wijbrans
Department of Earth Sciences, Faculty of Earth and Life
Sciences, VU University, Amsterdam, The Netherlands
Synonyms
Tertiary metamorphic terranes; Young metamorphic
terranes
Definition
Alpine terranes in the strictest sense can be defined geo-
graphically as terranes belonging to the region in south–
central Europe from the Maritime Alps and Ligurian Alps
along the Mediterranean coast between Genova and Mar-
seille in the west and the Vienna basin in the east where the
Tauern window ends in the northern belt and the Julian
Alps of Slovenia in the southern belt.
In a geological sense, much of southern Europe from
the Betics in southern Spain and the Rif in northern
Morocco to the Hellenides in Greece much of southern
Europe forms part of the Alpine orogeny. In fact, when
viewed as resulting from the collision of the Arabian–
African continent in the south and Eurasia in the north,
the Alpine orogeny can be traced much further to the
southeast to include Turkey and metamorphic complexes
along the Persian Gulf to end in Oman in the far southeast
(Figure 1).
The term Alpine terranes is sometimes used to include
young Tertiary mountain belts worldwide.
When approached from the perspective of metamorphic
processes, the Alpine terranes include the crystalline cores
of orogens formed by metamorphic processes since the
Cretaceous. Alpine terranes thus form a subset of the
metamorphic terranes.
Introduction
For the purpose of the present summary, it is useful to
propose a mixed definition based mainly on timing of meta-
morphic processes but including geographical consider-
ations. Alpine metamorphic terranes contain the evidence
for processes that deformed the mid- to deep levels of the
continental crust during the collision between Eurasia and
Arabia–Africa since the Cretaceous (Figure 1). Although
tectonic processes developed broadly synchronously, it is
appropriate to exclude the Himalayan orogeny here because
it resulted from the collision of India with Eurasia, and on
closer inspection, the tectonics and the time lines of oro-
genic processes are quite different. In its broadest sense,
the term “Alpine terrane” is used by some as a generic term
8 ALPINE TERRANES (K–Ar/Ar–Ar)
Alpine Terranes (K–Ar/Ar–Ar), Figure 1 The Alpine fold belt in southern Europe and the Middle East (Published with permission of
the author, H.W. Goede).
for all Tertiary metamorphic belts worldwide. Here we will
follow the more common definition: the mountain belts sit-
uated in south–central Europe extending in the southeast as
far as the Persian Gulf, with an occasional aside to pro-
cesses occurring in insights developed for mountain belts
elsewhere.
Much of the central and southern Eurasian continental
crust shows evidence for severe alteration during the
Variscan orogeny, roughly spanning the time period of
ca. 380–300 Ma. In the southern Arabian–African conti-
nent, the crystalline basement consists of rocks of
ca. 500–700 Ma, which belong to the Pan-African event
in the east, and Maghrebian rocks of Variscan age in the
west. The Rif mountains in northern Morocco form
the southern extent of an Alpine orogenic arc bordering
the Alboran Sea.
Between the main northern and southern blocks, during
the Mesozoic, there were smaller blocks of continental
crust that were exposed to an extensional regime in Trias-
sic and Jurassic times (e.g., Manatschal, 2004). Extension
at that time was accompanied by mafic underplating and
severe alteration of the continental lower and middle crust
between ca. 280–200 Ma. Evidence for this period of
pre-Alpine evolution in this domain can be found in the
Austro–Alpine thrust sheets in Eastern Switzerland and
Austria, in the bordering regions of northern Italy, in the
Pyrenees, and in the Catalan coast ranges. Rocks of this
age in the Southern Alps, south of the Insubric line sepa-
rating the Internal Alps in the north from the Southern
Alps, are interpreted as forming the thinned margin of
the Adria micro-continent (Bertotti et al., 1999). In con-
clusion, although in quite a substantial part of the Alpine
orogen metamorphic rocks of Triassic and Jurassic age
are exposed, tectonic analysis suggests strongly that these
rocks were formed during a phase of rifting that preceded
the phase of continental collision that formed the Alps.
The collisional phase that initiated the Alpine orogeny
was accompanied by deep subduction of the continental
crust, and much current research is focused on finding
out when these phases of deep subduction occurred. In
much of the Alpine domain, Paleocene–Eocene ages are
found. In the west, in the Betic Cordillera, subduction-
related metamorphism is possibly as young as the Oligo-
cene, but in parts of eastern Europe, in the Eastern Alps,
in the Julian Alps of Slovenia, and in the internal
Hellenides around the Ossa and Olympos ranges, Creta-
ceous ages of 90–100 Ma are associated with the
blueschist- and eclogite-forming events.
In several sectors of Alpine Europe, the phase of deep
subduction was followed by a phase of medium- to high-
temperature metamorphism at mid-crustal depths during
the Miocene; examples can be found in the Betics, the
Lepontine domain of the Swiss Alps, the Tauern window
in Austria, and the Greek Cyclades. Granitoid magmatism
and associated volcanics are found in the peri-Adriatic
domain of the southern Alps in northeast Italy, with ages
of ca. 45 Ma for the Adamello (Schoene et al., 2013),
32–24 Ma for the Bergell (Blanckenburg, 1992), and
 ALPINE TERRANES (K–Ar/Ar–Ar)
ca. 21 Ma for the Pohorje granites of Slovenia (Fodor
et al., 2008). In the Greek Cyclades, granites intruded dur-
ing the Miocene (Altherr and Siebel, 2002).
Active volcanism related to subduction can be found in
the Alboran domain in the west, in the Calabrian arc north
of Sicily, and in the Hellenic arc, where Santorini is the
most prominent volcano. Late Miocene to recent potassic
and ultrapotassic volcanism is found in the Betics in
southern Spain and in the Roman province in central Italy.
Historical perspectives
While the early efforts of dating the crystalline rocks of the
Alps proved quite effective in outlining the broad age
groups for the rocks found in the Alps, the general com-
plexity of the Alps as an orogen, and the fact that initially
quite large samples needed to be used for isotopic analy-
sis, prevented the early workers from obtaining a true pic-
ture of the Alpine geochronology. However, new insights
from mineralogy and petrology on the one hand, and tec-
tonics on the other, helped in refining the image that we
have of Alpine processes. The early division in
Eo-Alpine, Meso-Alpine, and Neo-Alpine periods for
rocks yielding Mesozoic, Paleogene, and Neogene ages
has largely been abandoned in the light of a better under-
standing of Alpine tectonics.
Probably the most important result of the early efforts
of understanding Alpine geochronology was the observa-
tion that minerals when dated with the K/Ar method
yielded systematically younger ages when compared with
their Rb/Sr ages (e.g., Jäger, 1979). This observation,
combined with the subsequent realization that minerals
that were formed during the ca. 320–350 Ma Variscan
yielded Alpine ages in a section with increasing
overprinting temperatures during the Alpine orogeny, led
to the understanding that the isotope systems in minerals
such as muscovite and biotite displayed open- or closed-
system behavior with respect to the radiogenic isotopes
formed from radioactive decay in these minerals. The
Berne geochronology group was in an ideal situation to
make this observation because Swiss petrologists using
the results from experimental petrology had come quite
far with assigning temperature estimates to the prograde
temperature sequence in the Swiss Alps from the external
units in the north and northwest to the internal zones in the
south (e.g., Frey et al., 1974). As a result of this early
work, the transition from closed to open system in the
K/Ar and Rb/Sr systems of common rock forming min-
erals such as muscovite and biotite could be assigned a
quantitatively determined temperature, for which the
Berne group used the term “blocking temperature”
(Jäger, 1979). As the blocking temperatures of minerals
appeared to be intrinsic properties, different temperatures
could be assigned to the age of minerals; it became possi-
ble to determine quantitative cooling rates for rocks
cooling after metamorphism.
At the time that the Berne team was quite far advanced
with calibrating the temperatures for the transition of
9
closed- to open-system behavior by comparison with tem-
perature estimates from thermobarometry, other
researchers focused on experimental determination of dif-
fusion rates for radiogenic isotopes such as isotopes of Sr
and Ar in minerals (e.g., Mussett, 1969; Giletti, 1974a;
Giletti, 1974b). The mathematical equations describing
diffusion in objects with simple shapes such as a sphere
or a cylinder can be derived under the assumption that dif-
fusion of particles and heat obey the same physical laws
and hence can be described using similar mathematical
equations. The diffusion equation allows calculation of
loss of, for example, radiogenic daughter isotopes from
minerals with time at a range of temperatures but can also
be used with some success to describe the diffusion of cat-
ions necessary to build new minerals as metamorphic
reactions proceed. As an example, time spans required to
achieve a threshold value of 90 % loss of radiogenic
daughter isotope are very long at low temperatures and
become gradually shorter at higher temperatures, and full
resetting can be expected when the time for 90 % loss of
radiogenic daughter isotopes is in the same order of magni-
tude as the time span of the metamorphic overprinting. The
diffusion equation also predicts a grain size effect, with
larger grains taking longer to lose all radiogenic daughter
isotopes. Finally, the mathematics of diffusion requires that
once the diffusing isotope has diffused out of the crystal, it
is effectively taken away and will not increase the concen-
tration of isotopes at the surface of the crystal occupying
the intergranular pore space. Some of the complexities in
the interpretation of age results in young mountain belts
can be attributed to the fact that not all of these boundary
conditions are always fulfilled. In the next section examples
will be given of such complexities in the interpretation of
age data in young mountain belts.
Complexities in interpretation
In Alpine metamorphic terranes, isotope geochronology
can be particularly powerful because in this age range
(generally processes take place during the period younger
than 66 Ma) the resolving power of the isotopic dating
techniques is high, and it can be expected that different
phases of metamorphic overprinting can be resolved. In
areas that have experienced high temperatures and where
recrystallization has gone to completion, results of isoto-
pic dating can be interpreted as recording cooling, and
the ages that are thus obtained are recording closure tem-
peratures (see entry “Metamorphic Terranes (K–
Ar/40Ar/39Ar)”). Dodson (1973) defined the closure tem-
perature in a cooling system as the temperature of the sys-
tem at the time defined by its measured isotopic age. Using
the mathematics of diffusion, Dodson demonstrated that a
closure temperature can be calculated when an activation
energy, Ea, and diffusion constant, Do, for a mineral phase
are known. A closure temperature is defined for mineral
phases that initially do not contain any radiogenic daugh-
ter isotope, and the actual value depends on the radius of
diffusion and rate of cooling of the rock. In an ideal
10 ALPINE TERRANES (K–Ar/Ar–Ar)
crystal, the radius of diffusion is likely to be equal to the
grain radius, but in deformed or twinned crystals, the dif-
fusion radius is more likely to be the radius of the
subgrains formed during twinning or deformation and
can be substantially less than the original grain radius.
Further, while based on the mathematical equations for
diffusion, Dodson’s closure temperatures are implicitly
defined only for the situation where the concentration of
radiogenic daughter isotope at the grain boundary is zero.
In reality, finite concentrations of radiogenic daughter
may occur especially in cases where the mobility in the
intergranular space is decreased. Examples of the simple
case of rocks cooling from high temperatures (>ca.
600  C) can be found in every Alpine metamorphic terrane
that reached high temperatures during metamorphic
overprinting. Examples of cases where the boundary con-
ditions for Dodson’s closure temperature definition are
violated can be found in most mountain belts in the
regions where metamorphic temperatures were lower
(<ca. 500  C), when minerals formed at a temperature
below their closure temperature, or where minerals were
not fully reset and still contain radiogenic daughter isotope
from the period preceding the studied metamorphic event.
The boundary conditions for calculating a closure temper-
ature may also be violated during high-pressure and
ultrahigh-pressure metamorphism when rock porosity is
severely decreased, thus effectively preventing transporta-
tion of radiogenic daughter isotope out of the system once
the first step of diffusion out of the crystal lattice has
occurred.
Examples
In a study of the metamorphic evolution of the Mulhacen
complex in the Betic Cordillera in southern Spain by De
Jong et al. (1992), muscovites yielded uniform ages in
the range of 11–12 Ma, whereas observation of the struc-
tures in the rocks suggested a fairly complex tectonic evo-
lution. The uniformity in K/Ar and 40Ar/39Ar ages
suggested that much of this tectonic complexity predated
the high-temperature event. In order to shed additional
light on this history that could not be measured from anal-
ysis of normal matrix muscovite, a megacryst of ca. 5 mm
diameter, i.e., some 20 times the size of the matrix, was
measured using a spot heating laser technique. Relying
on the prediction of diffusion theory that diffusion loss is
in part controlled by the grain radius, it was expected that
a signal predating the pervasive 11–12 Ma event could be
found. From spot analysis using an argon ion laser, a con-
centric pattern of ages increasing towards the center of the
grain was found, indicating that the megacryst was at least
27 Ma old. This can be taken as confirmation that grain
size and diffusion radius define isotopic closure: the
smaller grains record the younger cooling age, whereas a
larger grain yielded a much older cooling age.
An important observation was made by Baxter
et al. (2002), who worked in high-temperature metamor-
phic rocks in the Simplon area, part of the internal zone
of the Swiss Alps. Here, it appeared that there was a sys-
tematic bias in 40Ar/39Ar ages of biotites that were mea-
sured from mica schists and biotites in meta-basalts. As
the rocks were interlayered in a relatively small area, it
could be safely assumed that their metamorphic histories
were the same. Also the grain sizes were in the same
range. Yet the ages obtained from the minerals from the
meta-basalt were systematically higher than those found
from the mica schists. The interpretation was that the dif-
fusivity of radiogenic daughter isotope was less in the
meta-basalt and higher in the mica schist. A lower diffu-
sivity in the intercrystalline pore space may imply that a
substantial concentration of radiogenic daughter isotope
could be maintained in the pore space during the peak tem-
peratures of metamorphic overprinting. Isotopic closure is
firmly based on diffusion theory that assumes that the con-
centration of radiogenic isotope in the pore space remains
zero at all times, i.e., when radiogenic daughter isotope is
removed from the crystal by volume diffusion, it is
also removed from the system. When due to lower diffu-
sivity a higher concentration of radiogenic daughter
isotope can be maintained, this would result in the preser-
vation of systematically higher isotopic ages.
At metamorphic temperatures that are lower
(i.e., < ca. 600  C) and at metamorphic pressures that
are higher, the complexity of the preserved isotope signal
is expected to increase; at lower metamorphic tempera-
tures, isotopic resetting may not go to completion as at
these temperatures the system remains below the closure
temperatures of various minerals (Warren et al., 2012).
At increased burial depth, the porosity of the rock will
decrease, and with the concomitant decrease in diffusivity,
the concentration of radiogenic isotopes at the grain
boundaries will increase. As a result of these elevated con-
centrations of radiogenic isotopes, full resetting of
geochronometers may not be achieved.
As a result accumulation of radiogenic daughter isotope
may start at different times in different crystals of the same
mineral phase. Complex crystallization histories poten-
tially can be resolved, with a combination of detailed
microstructural characterization and access to single crys-
tal or spot dating techniques as demonstrated by Beltrando
et al. (2008). Another level of complexity is added in the
situation where a preexisting mineral assemblage is
overprinted. In an early study of this situation, Wijbrans
and McDougall (1986) demonstrated that phengites
formed during early high-pressure metamorphism at or
before ca. 45–50 Ma in Naxos, Greece, were progres-
sively overprinted by younger more muscovitic phengites
and true muscovites during a younger metamorphic phase
that ranged in peak temperature from >450  C to
ca. 700  C in the period 15–11 Ma. It could be demon-
strated in this case that as long as early phase phengites
remained present in the rock, an elevated age signal was
found. This observation underlined the conclusion that
recrystallization is the faster, more effective mechanism
to reset the isotope system. In addition, it was argued that
given the observation that full resetting of the early
 ALPINE TERRANES (K–Ar/Ar–Ar)
phengites was not observed, it should follow from diffu-
sion theory that the duration of the subsequent thermal
pulse that caused the high-temperature mineral assem-
blages may have been rather short.
In a study of the blueschist – greenschist metamor-
phism of the internal Hellenides, Lips et al. (1998) by care-
ful selection could isolate and date various phases of
blueschist phengites and younger greenschist muscovites
in an age range between 90 and 40 Ma, including an exam-
ple of a Variscan orthogneiss that exhibited ca. 320 Ma
protolith muscovites. They also found overprinting of
Alpines phengites with ages of ca. 65 Ma. At no time since
the Cretaceous were metamorphic temperatures high
enough to fully reset the isotope systems of the white
micas; complete ranges in ages from Variscan protolith
ages to late stage Alpine greenschist facies ages were
preserved.
In the internal Western Alps during Alpine collision,
high- and locally ultrahigh-pressure metamorphic condi-
tions were reached during the Paleocene–Eocene, as con-
firmed by U/Pb zircon dating, Sm/Nd dating, and Lu/Hf
dating of garnet-bearing mineral assemblages (Rubatto
and Hermann (2003), Lapen et al. (2003)). The K/Ar and
40
Ar/39Ar records of phengites and muscovites in these
areas yielded a complex picture with ages in the range of
40 Ma to 120–200 Ma, with outliers as high as
300 Ma (e.g., Scaillet, 2002). The simplest interpretation
would be to accept the ca. 40 Ma ages as consistent with
the U/Pb and Sm/Nd and Lu/Hf data, and everything older
is an artifact caused by excess argon. However, the
protolith age of the orthogneiss Variscan granite with ages
up to the range of ca. 300 Ma would be consistent with a
Variscan protolith history and might suggest inheritance
of an older argon signal that is preserved despite Alpine
tectonism and metamorphism in the Eocene.
Much of the complexity in the interpretation of K/Ar
and 40Ar/39Ar age data is caused by the mobility of 40Ar
in natural rock systems. Formed by the decay of 40 K,
40
Ar is a noble gas. As a neutral atom, argon is not a con-
stituent part of a crystal lattice of the mineral in which it
formed. The mobility of argon in minerals is best
described by a volume diffusion mechanism, for which a
pre-exponential diffusion constant Do and activation
energy Ea can be determined experimentally. Although
volume diffusion in pristine, undeformed crystals is likely
to be a very slow process, diffusion in a compact aggregate
of minerals along defects and grain boundaries is consid-
ered to be a faster process as the permeability along the
grain boundaries is higher than within the crystal itself
(Lee, 1995). Porosity may vary from very low to high, as
pore space may range from very low volume without
any interconnection to full interconnection allowing
advection of a fluid through the rock. In order to release
the radiogenic argon trapped within the crystals, the
slowest mechanism is via volume diffusion, whereas other
processes such as grain boundary diffusion and recrystal-
lization are likely to be much faster, as the bonds between
the ions forming the crystal lattice are broken in the latter
11
process, and trapped argon can be released via grain
boundaries (Wijbrans and McDougall, 1986).
The interpretation of dating results from rocks from
Alpine terranes thus may be complicated by the degree
of mobility of argon during metamorphic overprinting.
When ages are found with K/Ar and 40Ar/39Ar dating that
are older than expected, the system contains extraneous
argon. For practical purposes, two common forms of
extraneous argon can be defined: excess argon that is
released into the system when potassium-bearing minerals
break down and inherited argon, which is radiogenic Ar
that was incorporated in the lattice of a preexisting min-
eral. In the case of inherited argon, traces of a prehistory
may not have been fully erased during overprinting and
can be used in favorable cases to further constrain the ther-
mal evolution of the rock.
Summary and conclusion
In the Alpine orogenic belt in central and southern Europe,
rocks were exposed to elevated temperatures and pres-
sures caused by tectonic processes of subduction and col-
lision of continental crustal blocks during closure of
oceanic basin such as the Tethys and Paratethys basins.
These processes occurred predominantly during the
Paleogene and Neogene and in some sectors as early as
the Late Cretaceous. Earlier processes recorded in the
rocks appear to be related to the processes predating the
Alpine subduction and collision orogen. Evidence for ear-
lier processes and artifacts caused by the mobility or in
cases the lack of mobility of radiogenic daughter isotopes
add to the complexity of interpreting isotopic age data in
the low-temperature and high-pressure domains in moun-
tain belts.
Because metamorphic processes in the Alps occurred
during a relatively young period of Earth’s history, much
of the tectonic evolution causing metamorphism can be
resolved in time. As a result many of the mechanisms of
behavior of isotope systems were first described in
the Alps.
Bibliography
Altherr, R., and Siebel, W., 2002. I-type plutonism in a continental
back-arc setting: Miocene granitoids and monzonites from the
central Aegean Sea, Greece. Contributions to Mineralogy and
Petrology, 143, 397–415.
Baxter, E. F., DePaolo, D. J., and Renne, P. R., 2002. Spatially cor-
related anomalous 40Ar/39Ar “age” variations in biotites about a
lithologic contact near Simplon Pass, Switzerland: a mechanistic
explanation for excess Ar. Geochimica et Cosmochimica Acta,
66, 1067–1083.
Beltrando, M., Lister, G. S., Forster, M., Dunlap, W. J., Fraser, G.,
and Hermann, J., 2008. Dating microstructures by the 40Ar/39Ar
step-heating technique: deformation-pressure-temperature-time
history of the Penninic units of the Western Alps. Lithos, 113,
801–819.
Bertotti, G., Seward, D., Wijbrans, J., Ter Voorde, M., and Hurford,
A. J., 1999. Crustal thermal regime prior to, during, and after
rifting: a geochronological and modeling study of the Mesozoic
South Alpine rifted margin. Tectonics, 18, 185–200.
12 AMINO ACID
Blanckenburg, F., 1992. Combined high-precision chronometry and
geochemical tracing using accessory minerals – Applied to the
Central-Alpine Bergell intrusion, Central Europe. Chemical
Geology, 100, 19–40.
Brooker, R. A., Du, Z., Blundy, J. D., Kelley, S. P., Allan, N. L.,
Wood, B. J., Chamorro, E. M., Wartho, J. A., and Purton, J. A.,
2003. The ‘zero charge’ partitioning behaviour of noble gases
during mantle melting. Nature, 423, 738–741.
de Jong, K., Wijbrans, J. R., and Féraud, G., 1992. Repeated thermal
resetting of phengites during Miocene extension and wrenching
in the Betic Cordilleras – evidence from 40Ar/39Ar stepheating
and single grain laser probe dating in the Mulhacen complex.
Earth and Planetary Science Letters, 110, 173–191.
Dodson, M. H., 1973. Closure temperature in cooling geochrono-
logical and petrological systems. Contributions to Mineralogy
and Petrology, 40, 259–274.
Fodor, L. I., Gerdes, A., Dunkl, I., Koroknai, B., Pecskay, Z.,
Trajanova, M., Horvath, P., Vrabec, M., Jelen, B., Balogh, K.,
and Frisch, W., 2008. Miocene emplacement and rapid cooling
of the Pohorje pluton at the Alpine-Pannonian-Dinaridic junc-
tion, Slovenia. Swiss Journal of Geosciences, 101, S255–S271.
Frey, M., Hunziker, J. C., Franck, W., Bocquet, J., Dal Piaz, G. V.,
Jager, E., and Niggli, E., 1974. Alpine metamorphism of the
Alps. A review. Schweizerische Mineralogische und
Petrographische Mitteilingen, 54, 247–291.
Giletti, B. J., 1974a. Diffusion related to geochronology. In Hof-
mann, A. W., Giletti, B. J., Yoder, H. S., Jr., and Yund, R. A.
(eds.), Geochemical Transport and Kinetics. Washington: Car-
negie Institution. Carnegie Institution of Washington
634, pp. 61–76.
Giletti, B. J., 1974b. Studies in diffusion I: Argon in phlogopite
mica. In Hofmann, A. W., Giletti, B. J., Yoder, H. S., and
Yund, R. A. (eds.), Geochemical Transport and Kinetics. Wash-
ington: Carnegie Institution. Carnegie Institution of Washington
634, pp. 107–116.
Jäger, E., 1979. Introduction to geochronology. In Jäger, E., and
Hunziker, J. C. (eds.), Lectures in Isotope Geology. Berlin:
Springer, pp. 1–12.
Lapen, T. J., Johnson, C. M., Baumgartner, L. P., Mahlen, N. J.,
Beard, B. L., and Amato, J. M., 2003. Burial rates during pro-
grade metamorphism of an ultra-high-pressure terrane: an exam-
ple from Lago di Cignana, western Alps, Italy. Earth and
Planetary Science Letters, 215, 57–72.
Lee, J. K. W., 1995. Multipath diffusion in geochronology. Contri-
butions to Mineralogy and Petrology, 120, 60–82.
Lips, A. L. W., White, S. H., and Wijbrans, J. R., 1998. 40Ar/39Ar
laserprobe direct dating of discrete deformational events: contin-
uous record of Early Alpine tectonics in the Pelagonian Zone,
NW Aegean area, Greece. Tectonophysics, 298, 133–153.
Manatschal, G., 2004. New models for evolution of magma-poor
rifted margins based on a review of data and concepts from West
Iberia and the Alps. International Journal of Earth Sciences, 93,
432–466.
Mussett, A. E., 1969. Diffusion measurements and potassium-argon
method of dating. Geophysical Journal of the Royal Astronomi-
cal Society, 18, 257–303.
Rubatto, D., and Hermann, J., 2003. Zircon formation during fluid
circulation in eclogites (Monviso, Western Alps): implications
for Zr and Hf budget in subduction zones. Geochimica et
Cosmochimica Acta, 67, 2173–2187.
Scaillet, S., 2002. Excess 40Ar-transport scale and mechanism in
high-pressure phengites: a case study from an eclogitized
metabasite of the Dora-Maira nappe, western Alps. Geochimica
et Cosmochimica Acta, 60, 1075–1090.
Schoene, B., Schaltegger, U., Brack, P., Latkoczy, C., Stracke, A.,
and Gunther, D., 2013. Rates of magma differentiation
and emplacement in a ballooning pluton recorded by U-Pb
TIMS-TEA, Adamello batholith, Italy. Earth and Planetary Sci-
ence Letters, 355, 162–173.
Warren, C. J., Hanke, F., and Kelley, S. P., 2012. When can musco-
vite 40Ar/39Ar dating constrain the timing of metamorphic exhu-
mation? Chemical Geology, 291, 79–86.
Wijbrans, J. R., and McDougall, I., 1986. 40Ar/39Ar dating of white
micas from an Alpine high-pressure belt on Naxos (Greece):
resetting of the argon isotopic system. Contributions to Mineral-
ogy and Petrology, 93, 187–194.
Cross-references
Ar–Ar and K–Ar Dating
Metamorphic Terranes (K–Ar/40Ar/39Ar)
Minerals (40Ar–39Ar)
Potassium–Argon (Argon–Argon), Structural Fabrics
Sm–Nd Dating
AMINO ACID
John F. Wehmiller
Department of Geological Sciences, University of
Delaware, Newark, DE, USA
Definition
An organic compound containing an amino group (NH2),
a carboxylic acid group (COOH), and any of various side
groups, having the basic formula NH2CHRCOOH.
Amino acids are organic compounds composed of carbon,
oxygen, hydrogen, and nitrogen in a specific structure
where a central carbon atom is surrounded by a carboxyl
group, an amino group, and a side chain that may be
a single hydrogen or a more complex chain with multiple
carbon-bearing entities. The structure common to all
amino acids is shown in Figure 1. Approximately 20 com-
mon amino acids are found in nature, and they combine in
sequences of varying lengths to form peptides and pro-
teins that have specific biochemical functions. Amino
acids are found in a variety of geological samples and play
a role in understanding chemical evolution, the formation
and alteration of organic material in sediments, and
geochronology (Goodfriend et al., 2000).
The role of amino acids in geochronology is based on
the phenomenon of isomerism. Amino acids can exist in
two isomeric forms, commonly referred to as “left
handed” and “right handed.” These two forms are possible
because the central carbon to which the four different side
chains are bonded is asymmetric: two mirror-image con-
figurations around this central carbon are possible, and
these two are not geometrically equivalent.
Amino acids are found in sediments, soils, and in many
common biominerals, as constituents of the structural pro-
teins used by organisms such as mollusks, corals, forami-
nifera, etc., to synthesize the mineral phase. Amino acids
are also found in bone, teeth, and wood. In most cases,
 AMINO ACID RACEMIZATION DATING
Amino Acid, Figure 1 Mirror-image relation between the two
configurations of amino acid molecules. The four side chains are
bonded to the central carbon atom at the corners of
a tetrahedron. The two forms are geometrically distinct; hence
the conversion of one form to the other requires breaking
a bond and reassembling the amino acid as the other form.
H ¼ hydrogen; COOH ¼ carboxyl group; NH2 ¼ amino group;
R ¼ one of many possible side chains that define the different
amino acids.
the amino acids incorporated into the biomineral during
growth are of the 100 % left-handed configuration. Upon
death of the organism, these amino acids begin the process
of racemization, or conversion from the 100 % left-handed
form into an equilibrium mixture of 50 % of each of the
two forms. The rate at which this reaction occurs is
influenced by many factors, including temperature, sam-
ple type, alteration of the sample during its burial history,
and specific rates of breakdown of the proteins that
are preserved within the mineral matrix of the analyzed
samples. Common amino acid racemization (AAR) stud-
ies include the use of foraminifera from deep-sea sedi-
ments (Amino Acid Racemization, Marine Sediments),
marine mollusks from coastal deposits (Biostratigraphy;
Amino Acid Racemization, Coastal Sediments), and
snails, eggshells, or ostracodes from terrestrial environ-
ments (Amino Acid Racemization, Fluvial and Lacustrine
Sediments (AAR)). The timescale of these studies ranges
from a few hundred years to over a million years,
depending on the average temperatures to which
samples have been exposed. These biominerals function
to isolate the amino acids from the surrounding environ-
ment. Wood, soils, and bulk sediments also have been
used in some AAR geochronological studies, with mixed
success because of the open-system nature of these
samples.
Radiocarbon (14C) dating (see entry “Radiocarbon
Dating”) usually involves extracting carbon-bearing mate-
rial (either organic carbon or carbonate carbon) from sam-
ples that are to be dated. In some cases, it is preferable to
extract the amino acid component from these samples as
a test of the integrity of the 14C result (Hatte et al. 2012).
Comparison of 14C dates obtained on a bulk sample and
the amino acid fraction of that same sample provide an
indication of the integrity of the sample, as the amino acids
(bound within the biomineral) would be expected to be
less vulnerable to contamination from external sources
(Marom et al., 2012).
13
Bibliography
Goodfriend, G. A., Collins, M. J., Fogel, M. L., Macko, S. A., and
Wehmiller, J. F. (eds.), 2000. Perspectives in Amino Acid and
Protein Geochemistry. Oxford/New York: Oxford University
Press. xvi + 366 pp.
Hatte, C., Hodgins, G., Jull, A. J. T., Bishop, B., and Tesson, B.,
2008. Marine chronology based on 14C dating on diatom pro-
teins. Marine Chemistry, 109, 143–151.
Marom, A., McCullagh, J. S. O., Higham, T. F. G., Sinitsyn, A. A.,
and Hedges, R. E. M., 2012. Single amino acid radiocarbon dat-
ing of Upper Paleolithic modern humans. Proceedings of the
National Academy of Sciences, 109, 6878–6881.
Cross-references
Amino Acid Racemization, Biostratigraphy
Amino Acid Racemization, Coastal Sediments
Amino Acid Racemization Dating
Amino Acid Racemization, Eolianites
Amino Acid Racemization, Fluvial and Lacustrine Sediments (AAR)
Amino Acid Racemization, Marine Sediments
Amino Acid Racemization, Paleoclimate
Radiocarbon Dating
AMINO ACID RACEMIZATION DATING
Beatrice Demarchi and Matthew Collins
BioArCh, Department of Archaeology, University of
York, York, UK
Synonyms
AAR, Protein diagenesis geochronology
Definition
A method for estimating the relative age since death by
assessing the extent of postmortem conversion of biologi-
cal chiral forms of amino acids (L-enantiomers) to their
nonbiological counterparts (D-enantiomers).
Introduction
Amino acid racemization (AAR) dating is
a geochronological technique with a very long history.
Over the past 60 years, many researchers and laboratories
around the world have been involved with the develop-
ment of the method and its application to diverse environ-
ments. Its time depth and applicability to a wide range of
substrates are the main strengths of this method. Its main
weakness is the fact that it is a molecular- rather than an
atomic-scale reaction (cf. radionuclide decay), and as
a consequence the rate is sensitive to temperature. Useful
review articles for understanding the principles of AAR
dating are those of Miller and Brigham-Grette (1989),
Mitterer (1993), Rutter and Blackwell (1995), Johnson
and Miller (1997), and Wehmiller and Miller (2000). In
addition, two classic volumes on amino acid racemization
were produced: Biogeochemistry of Amino Acids (1980),
edited by Hare, Hoering, and King, and Perspectives in
14 AMINO ACID RACEMIZATION DATING
Amino Acid Racemization Dating, Figure 1 A short peptide (Val-Ser-Phe-Leu-Gln) to illustrate side chains (–R groups) and the
backbone structure formed by the condensation between the amino and carboxyl group of two amino acids (peptide bond). Proteins
are typically composed of 50–1,000s amino acid residues.
Amino Acid and Protein Geochemistry (2000), edited by
Goodfriend and colleagues. Following a session on AAR
dating that took place at the INQUA 2011 conference
(Bern, Switzerland), a special issue of the journal Quater-
nary Geochronology (vol. 16, pages 1–198, April 2013,
edited by Penkman and Kaufman) was published in
2013. This summarizes the current state of the art in
AAR research. We refer the interested reader to these pub-
lications for gaining a more in-depth understanding of the
vast field of protein diagenesis and its applications to
geochronology.
Here we provide a basic toolkit for understanding the
principles of AAR, beginning with the chemical mecha-
nisms of protein diagenesis, summarizing briefly the his-
tory of the development of the technique, and then
focusing in more detail on one of the methodologies of
AAR dating that has been developed in more recent years:
the intracrystalline protein diagenesis (IcPD) approach.
While the method does not supersede more traditional
approaches, it is an important step towards the integration
of AAR and biomineralization studies. The two are deeply
intertwined, as biomineral growth is controlled by proteins
which are then trapped in the mineral framework and
undergo postmortem degradation. Therefore, understand-
ing the complexities of protein diagenesis should be
viewed in conjunction with an understanding of biomin-
eral formation which, after death, will represent the pro-
tein’s burial environment.
We refer to the following entries in this encyclopedia
for additional information:
• Amino Acid Racemization, Biostratigraphy
• Amino Acid Racemization, Coastal Sediments
• Amino Acid Racemization, Eolianites
• Amino Acid Racemization, Fluvial and Lacustrine
Sediments (AAR)
• Amino Acid Racemization, Marine Sediments
What is an amino acid?
Amino acids, the building blocks of proteins, share
a common chemical character, each containing
a hydrogen atom (H), an amino group (NH2), a carboxyl
group (COOH), and an (R) group, or side-chain, bound
to a central carbon atom (C). Amino acids have both amine
and carboxylic acid functional groups and therefore can
act both as an acid and as a base. Most proteins are built
from sequences of 20 different amino acids; their chemical
characteristics are determined by the –R group and their
sequence in the protein by the genetic code (Figure 1).
What is amino acid racemization?
Chirality
Consider your hands; left and right (in Latin, left is laevo,
and right is dextro). Your hands have chiral symmetry:
they are mirror images of each other and cannot be
superimposed. As with your hands, so with all but the sim-
plest molecules: the configuration of their atoms in space
means that they too can exist in more than one form, which
may be mirror images but cannot be superimposed. All of
the amino acids that occur in proteins, except for glycine,
which is the simplest amino acid, have at least one asym-
metric carbon atom. They can therefore exist in two
nonsuperimposable stereoisomers (enantiomers), termed
(from the Latin) L- and D-forms. Some amino acids have
two asymmetric carbon atoms and can exist in four
 AMINO ACID RACEMIZATION DATING
different forms, known as diastereomers. Two of these
amino acids, isoleucine and threonine, are commonly
found in most proteins. L-isoleucine racemizes (or more
accurately in the case of two asymmetric carbon atoms,
epimerizes) almost exclusively to one form, called
D-alloisoleucine. Because diastereomers have different
chemical and physical properties, they are the easiest to
measure and were therefore the first to be used for
geochronology.
Dating
The basis of amino acid racemization (AAR) as a dating
technique is the measurement of the relative proportions
of the L- and D-chiral forms. In almost all living proteins,
amino acids are present in the L-form. The most com-
monly encountered exceptions are bacterial cell wall bio-
polymers, collectively termed peptidoglycan. The
presence of D-amino acids in peptidoglycan is a potential
confounding factor, although it is useful to estimate micro-
bial reworking.
Upon death L-amino acids will slowly and spontane-
ously convert to D-amino acids, which themselves sponta-
neously revert back to the L-form. Eventually, a situation
of dynamic equilibrium is reached, where there is a 1:1
(so-called racemic) mixture of L- and D-amino acids.
A simple mnemonic is that L-forms are found in “living”
organisms and D-forms accumulate in “dead” ones. This
interconversion reaction is called racemization.
Racemization analysis has been conducted on wood
(Engel et al., 1977), on textiles (Moini et al., 2011), and in
the biosciences on a range of tissues (skin, arteries, cornea)
(Ritz-Timme and Collins, 2002), but for geochronology
biominerals are the preferred target. Amino acids have been
reported in pyrite and have been isolated from soils and
rocks, siliceous sediments, and bioapatites (enamel) (e.g.,
Kimber et al., 1994; Nardi et al., 1994; Hearty and
Kaufman, 2000; Murray-Wallace et al., 2001; Harada
et al., 2002; Griffin et al., 2009), but most studies focus on
carbonate biominerals, principally aragonite and calcite:
mollusk shells, eggshell, ostracods, foraminifera, brachio-
pods, and corals (a selection of studies on carbonate
biominerals: Bada et al., 1970; Wehmiller and Hare, 1971;
Wehmiller, 1976, 1977; Masters and Bada, 1978; Andrews
et al., 1979; Wehmiller, 1980, 1982; Davies, 1983; Miller
et al., 1983; Müller, 1984; Hearty et al., 1986; Brooks
et al., 1990; Goodfriend, 1992; Bates, 1993; Sejrup and
Haugen, 1994; Wehmiller et al., 1995; Goodfriend
et al., 1996, 1997; Torres et al., 2000; Kaufman, 2003; Bar-
bour Wood et al., 2006; Penkman et al., 2007; Kaufman
et al., 2008; Hendy et al., 2012; Ortiz et al., 2013; Kaufman
et al., 2013; Penkman et al., 2013).
Racemization proceeds at different rates according to
(a) the chemical characteristics of the amino acid itself,
(b) the position of the amino acid within the protein, and
(c) burial temperature and other factors (pH, water avail-
ability, presence of cations and anions, microbial degrada-
tion). As we will see, factors operating in the burial
environment can be better controlled if a closed-system
15
fraction of amino acids (“intracrystalline”) is used for dat-
ing purposes. If all of these variables are known
(or comparable between a set of samples), then the ratio
between the D- and L-forms of a specific amino acid
(also called D/L value) will reflect the age since death of
the organism. This ratio will increase over time from
a value of 0 (when the organism is alive, and no D-amino
acids are present) to 1 (racemic equilibrium, where
a 50–50 % mixture of L- and D-forms is present). There-
fore, for a series of samples of increasing age, the D/L
value will also increase: this is the basis for building rela-
tive chronological frameworks with AAR. These frame-
works can then be calibrated, if the D/L value is
measured on samples of known age – for example,
a series of shells that have been dated by radiocarbon
(Kosnik et al., 2008). This allows building calibrated chro-
nological frameworks, which can then be used to date
samples independently. However, many studies focus on
Quaternary deposits, which lie well beyond the radiocar-
bon range: in this case, calibration must rely on other inde-
pendent chronologies, for example, U-series, OSL,
strontium isotopic stratigraphy, paleomagnetism, biostra-
tigraphy, and sedimentary evidence (e.g., Wehmiller and
Belknap, 1982; Hearty et al., 1986; Penkman
et al., 2011; Wehmiller et al., 2012). Calibration is espe-
cially important because pre-Holocene samples have
experienced a number of glacial-interglacial cycles and
have therefore been exposed to different burial tempera-
tures, which prevent Holocene-based calibrations to be
extended back to the Pleistocene (see also “Amino Acid
Racemization, Paleoclimate”).
A brief history of the technique
The first report of the presence of amino acids preserved in
geological samples was in 1954 when Philip Abelson
from the Carnegie Institute in Washington DC undertook
chromatographic analysis of a number of fossil samples,
including fossil fish more than 300 million years of age
(from the Devonian Period) (Abelson, 1954). Once the
group switched to ion exchange chromatography, they
noticed the presence of the nonbiological amino acid
alloisoleucine (aIle) in some fossil samples and speculated
that this was perhaps the result of an age-dependent chem-
ical reaction. Ed Hare from the same institute understood
that the epimerization of Ile into aIle was a potential tool
for geochronology and that as epimerization is
a chemical reaction governed by time and temperature,
the levels of aIle would increase in progressively older
samples. This was demonstrated by the experiments of
Hare and Mitterer (1969), who measured the rate of race-
mization of L-isoleucine to D-alloisoleucine in modern
shell fragments heated at high temperatures; they then
extrapolated these data to lower temperatures in order to
estimate the rate of racemization of L-isoleucine in fossil
shells to obtain what they believed to be an approximate
age for these fossil shells.
16 AMINO ACID RACEMIZATION DATING
This work blossomed in the 1960s with Ed Hare, Tom
Hoering, John Wehmiller, Kenneth King Jr., Richard
Mitterer, Sidney Fox, and Kaouru Harada, among others
(see the Mercenaria Scientific Family Tree that was
presented to Ed Hare in occasion of his retirement in
1998) rapidly expanding their range of analyses and sub-
strates. The focus was primarily on understanding the pro-
cesses and mechanisms of degradation: a range of studies
on marine sediments, for example, exploited the stable
thermal environment at the bottom of the sea in order to
decouple the effects of age and temperature on the pro-
cesses of racemization and therefore elucidate the kinetics
of isoleucine epimerization (Wehmiller and Hare, 1971;
Bada and Schroeder, 1972; Kvenvolden et al., 1973).
Almost in parallel at UC San Diego (SCRIPPS), Jeff
Bada identified that aspartic acid was racemizing much
more rapidly than isoleucine was epimerizing, offering
an opportunity to date much younger fossils. Bada and
coworkers applied this method to the dating of fossil
bones and to the determination of past temperatures, by
measuring the extent of racemization in several
radiocarbon-dated bones. Hare worried that bone lost pro-
teins too rapidly to be reliable, but so (to a lesser extent)
did shell (Wehmiller, 1977; Masters and Bada, 1977,
1978; Miller and Hare, 1980; Hare, 1988). High profile
studies, while the method was still under development
(the sites of Boxgrove (Bowen and Sykes, 1994); and
del Mar (Bada et al., 1974, 1984)), resulted in controver-
sial age estimates which led to a loss of support to the tech-
nique in parts of the community (Marshall, 1990).
Studies on the mechanisms of racemization and
pioneering applications of the method to archaeological
and geological contexts (including meteorites, e.g., Engel
and Nagy, 1982) continued through the 1980s, the 1990s,
and the 2000s. The application of AAR dating was mainly
as a relative dating tool to identify the age of different geo-
logical deposits: aminostratigraphy (Miller et al., 1979).
Research in this area continued, driven by a small number
of groups, mainly in the USA, for example, Giff Miller,
Darrel Kaufmann, and Glen Goodfriend working in the high
Arctic and elsewhere (Miller et al., 1983; Goodfriend, 1991,
1992; Kaufman and Brigham-Grette, 1993; Kaufman and
Sejrup, 1995; Goodfriend et al., 1996) and John Wehmiller
dating shell horizons along the east and west coasts of the
USA (Wehmiller et al., 1977; Wehmiller, 1982; Kennedy
et al., 1982; Rockwell, 1992; Wehmiller et al., 1995;
Wehmiller, 2013) and South America (Hsu et al., 1989;
Rutter et al., 1989). Datasets were also accumulated for
China (Oches and McCoy, 2001), Australia (Murray-
Wallace and Kimber, 1987; Murray-Wallace et al., 1993;
Miller et al., 1997, 2005), Africa (Brooks et al., 1990;
Miller et al., 1999), the Mediterranean (e.g., Hearty
et al., 1986), Great Britain and the North Sea (Bowen
et al., 1985, 1988; Bates, 1993; Meijer and Cleveringa,
2009). Many of these datasets are publicly available through
the NOAA website, where additional citations can be found,
as well as maps showing the scale of some of these AAR
investigations. During the same years, scientists such as
Ritz-Timme and Ohtani were using the method as
a forensic technique to estimate the age of tooth dentine
(Ohtani and Yamamoto, 1991; Ritz-Timme et al., 2000).
The next turning point in research would then occur
when two fields of research, which had been proceeding
somewhat in parallel throughout the twentieth century,
finally met: studies on protein-induced mineralization of
hard tissues and studies on protein diagenesis. This led
to the rationalization of the key concepts of intercrystalline
versus intracrystalline proteins in biominerals.
Key concepts
In considering the fate of amino acids in biominerals, it is
helpful to operationally define two pools of constituent
proteins, inter- and intracrystalline.
Intercrystalline
The intercrystalline fraction is defined as the components
which are destroyed by chemical treatment, occasionally
reduction with hydrazine (N2H4), usually oxidation with
sodium hypochlorite, NaOCl (Sykes et al., 1995). As the
latter is far more common, this will form the basis of subse-
quent discussion. The ability to destroy the organic matter
implies that it occupies interstices in the biomineral between
major contiguous blocks, such that they form a connected
network (Suzuki et al., 2011). In biominerals with micro-
crystals (e.g., bone), all proteins are intercrystalline, as the
individual mineral elements are too small to encapsulate
proteins. In the case of bone, proteins may be entrapped in
mineral aggregates (Salamon et al., 2005), but this has not
yet been the focus of geochronological studies.
Intracrystalline
The intracrystalline fraction is that which survives
prolonged oxidation/reduction. Since the definition is
operational, it leads to a number of issues: the concentra-
tion of the oxidant, the duration of the treatment, and the
optimal particle size need to be tested for each
biominerals. Prolonged exposure to bleach causes a slow
but steady decline in amino acid concentrations and an
increase in racemization, indicating that the system is not
perfectly isolated. The operational definition usually
refers to crushed biomineral (around 500 mm particle size)
exposed to 12 % wt/vol NaOCl for 48 h (Penkman
et al., 2008). The effect of oxidation with NaOCl
(bleaching) is shown schematically in Figure 2: the bleach
removes the intercrystalline proteins, more prone to con-
tamination and interaction/exchange with the environ-
ment. The protected intracrystalline fraction contains
both the original proteins and their degradation products.
This enables the quantification of the extent of protein
breakdown occurring via different reactions: peptide bond
hydrolysis and amino acid racemization and decomposi-
tion (e.g., dehydration of the amino acid Ser to Ala).
Within the intercrystalline fraction, the extent of reten-
tion of amino acids will be a function of the degree of pro-
tein breakdown, the rate of loss from the sample, and any
 AMINO ACID RACEMIZATION DATING 17

Amino Acid Racemization Dating, Figure 2 A strong oxidation treatment is able to remove the intercrystalline protein from
biominerals, leaving behind a fraction of the original proteins, degraded in situ (effectively, in a closed-system environment).

contribution from the burial environment itself. Therefore,
there will be a broad correlation between the age of the
sample and the extent of racemization, and local burial
conditions will dictate the overall value. The
intracrystalline fraction should behave in a more predict-
able fashion (Sykes et al., 1995). In the burial environ-
ment, the intercrystalline proteins are often lost quickly,
and therefore, many past studies that were analyzing
whole-shell proteins were effectively analyzing the
intracrystalline fraction. Therefore, the isolation of this
fraction through chemical pretreatment is more important
for samples with a short burial history. The added advan-
tage of bleaching is the removal of external contaminants,
e.g., exogenous amino acids. However, the effectiveness
of bleaching needs to be tested for each biomineral: for
example, the isolation of the intracrystalline proteins from
the chitin-rich ostracod valves has been shown to be not
straightforward (Bright and Kaufman, 2011). Nonethe-
less, the conceptualization of the intracrystalline fraction
has added a further piece to the jigsaw, along with other
experimental evidence on the behavior of proteins in
biominerals (high-temperature experiments, intergeneric
comparisons, and diagenetic models, e.g., Wehmiller,
1980; Mitterer and Kriausakul, 1984). Subsequent discus-
sion will refer to racemization analysis of the
intracrystalline fraction, unless otherwise explicitly stated.
The life and death of a protein: from
biomineralization to breakdown
Although the presence of water trapped in mollusk shell
carbonates had already been noted (Hudson, 1967), for
many years, the idea of the intracrystalline proteins
seemed to make little sense, even when the first detection
of organic material and water trapped within the crystals
was reported: a so-called “frothy” structure in single crys-
tals of Mytilus nacre (mother-of-pearl) was observed by
18 AMINO ACID RACEMIZATION DATING
Amino Acid Racemization Dating, Table 1 Gly-Asn repeats are found in the sequence of intracrystalline proteins isolated from
Pinctada
Start position Sequence
80 FYYMCCYTDDD-NGNGDGNGNGFNYLKSLYGGYGNGNG
80 FYYMCCYTDDD-NGNGNGNGNGFNYLKSLYGGYGNGNG
80 FYSLCCYTDDD-NGNGNGNGNGFNYLKSLYGGYGNGNG
80 FYSLCCYTDDD-NGNGNGNGNGNGLNYLKSLYGGYGNGNG
80 FYSLCCYTDDD-NGNGNGNGNGNGLNYLKSLYGGYGNGNG
80 FYSLCCYTDDD-NGNGNGNGNGNGLNYLKSLYGGYGNGNG
81 YYTLCCYTEDD-NGNGNGNGNGYGNGNGNGNGNNYLKYLFGGNGNGNG
81 YYTLCCYTDDD-NGNGNGNGNGYGNGNGNGNGNNYLKYLFGGNGNGNG
transmission electron microscopy (TEM; Towe and
Thompson, 1972).
Why are proteins trapped inside crystals? Decades of
research in biomineralization is starting to unveil the pro-
cess of entrapment of proteins within biominerals. Fast-
forming systems, such as eggshell, retain higher concen-
trations of amino acids in the intracrystalline fraction than
slow-forming systems, for example, mollusk shells, and
therefore in part incorporation may be kinetically driven.
Ostrich eggshell retains 30 nmol amino acids per mg
mineral after bleaching; limpet shells (Patella vulgata)
a mere 4 nmol per mg. However, both biominerals con-
tain  50 nmol amino acids per mg of unbleached (i.e.,
intracrystalline + intercrystalline) biomineral (Crisp
et al., 2013; Demarchi et al., 2013a). Importantly, the
intracrystalline proteins and, generally, proteins in
biominerals display a wide range of amino acid sequences
and three-dimensional structures (these can be found in
the NCBI database).
Proteins promote and regulate biomineralization. The
model proposed for the chicken eggshell protein
ovocleidin-17 (OC-17) (Freeman et al., 2010) highlights
the role of Arg-rich domains of the protein and their action
as “clamps” that bind calcium and promote the transfor-
mation from amorphous calcium carbonate (ACC) to crys-
talline mineral. If these crystallites aggregate to form
mesocrystals (Cölfen and Antonietti, 2005), it is easy to
envisage how the protein becomes surrounded and
entrapped. However, even in situations in which the pro-
tein does not act as a catalyst, other authors have proposed
models where remnant proteins are overgrown into the
mineral (Li et al., 2011; Okumura et al., 2013). The role
of these in modifying the structural properties of the min-
eral is now indeed an active area of research (e.g.,
Okumura et al., 2013).
Whole-genome sequencing and targeted analysis, such
as by immunohistochemistry, are beginning to reveal the
differences in the character of proteins found in
biominerals, and the new sequences and new tools for
analysis are rapidly accelerating knowledge of these sys-
tems. One example is the intriguing set of glycine-
asparagine (Gly-Asn or, in one-letter code, GN) repeats
End position Protein name
116 MA162_PINFU
116 MA165_PINFU
116 I7GQ94_PINFU
118 L8B660_PINFU
118 MA161_PINFU
118 MA163_PINFU
127 MAPE_PINMG
127 MA14_PINMA
found in the sequence of the protein N16.1 isolated from
the intracrystalline fraction of the mollusk Pinctada fucata
(see Table 1).
The Gly-Asn repeat is highly unstable and degradation
rates depend upon environmental temperature; therefore,
this might suggest that variants of this organism which
are biomineralizing at different conditions (e.g., water
temperatures) might have evolved a system whereby they
increase the number of the GN repeats to enhance the sta-
bility of the protein. From this brief review, it is evident
that we are only beginning to understand the role played
by the intracrystalline fraction; some may be accidental
overgrowth, others may be trapped catalysts, while still
others may be playing a functional role, perhaps even act-
ing as remote triggers controlling later-phase processes.
What we are beginning to understand is that the entrapped
intracrystalline proteins are not homogeneous and it is
therefore not surprising that their rates of decomposition
(and thus their rate of racemization) will vary from species
to species and ultrastructure to ultrastructure, the so-called
species effect (Lajoie et al., 1980; Wehmiller, 1980),
which is widely observed in AAR. Indeed, many whole-
shell AAR studies recognized the presence of different
peptide fractions, some more refractory than others (e.g.,
Kaufman and Sejrup, 1995), and this was used as a tool
for taxonomic identification (Andrews et al., 1985;
Wehmiller, 1989; Kaufman et al., 1992). What is more
surprising is perhaps the fact that the species effect is pre-
served within the intracrystalline fraction and a common
“template” cannot be found, although these proteins are
likely to be directly involved in the interaction with the
mineral phase.
Death and diagenesis
After deposition, organic remains are in thermodynamic
disequilibrium with their environment and they will
attempt to reach equilibrium with the surrounding envi-
ronment. The sequence of intermediate metastable condi-
tions is defined as a series of diagenetic pathways or
diagenetic trajectories (Wilson and Pollard, 2002). Several
difficulties arise in defining these for protein diagenesis,
 AMINO ACID RACEMIZATION DATING
Amino Acid Racemization Dating, Figure 3 Mechanism of dipeptide hydrolysis.
even when looking at the intracrystalline fraction, because
(a) the actual amino acid sequence (and composition) for
most proteins is unknown and (b) the temperature sensitiv-
ities of each of the reactions within the network of reac-
tions driving the overall diagenesis are difficult to assess.
Therefore, a “black box” approach is used in protein dia-
genesis studies, where three main reactions, occurring
simultaneously, are tracked, as they can provide informa-
tion on the time elapsed since degradation started, i.e.,
the time of death of an organism.
Hydrolysis
Peptide bonds are inherently unstable and will undergo
hydrolysis, the rate depending upon a number of factors:
the pH of the solution, the presence of catalysts, the avail-
ability of water (which is consumed in the reaction), and
the steric constraints upon the bond. One advantage of
the intracrystalline fraction is that all of the above factors
should be equivalent in the tissues of the same taxon.
Hydrolysis breaks the peptide bonds, yielding progres-
sively shorter peptides until only the basic building
blocks, the free (as opposed to peptide-bound) amino acid
residues, are present. Hydrolysis requires the presence of
chemically available water (Figure 3), which must be pre-
sent in the intracrystalline fraction, since (a) we observe
the rapid release of amino acids from the peptide-bound
fraction into the free amino acid pool due to peptide bond
hydrolysis and (b) it has been directly observed within
crystallites (Hudson, 1967; Gaffey, 1988). Furthermore,
diagenetic decomposition of hydroxy amino acids, Ser,
Thr, and Glu (Bada et al., 1978; Walton, 1998), and carbo-
hydrates (Collins et al., 1992) would increase the amount
of water available for hydrolysis of the intracrystalline
proteins. However, a residual un-hydrolyzed fraction is
usually found within the intracrystalline fraction of differ-
ent biominerals; Collins and Riley (2000) speculated that
this could be due either to a reduction in the availability
of water or to the presence of a recalcitrant pool of amino
acids.
Each protein possesses a unique primary structure and
contains a wide variety of peptide bonds of different
strengths. However, hydrolysis exhibits a certain degree
of specificity, and in the natural environment, under mild
pH and temperature conditions, three main mechanisms
of hydrolysis are likely to occur (Bada, 1991):
1. Cleavage of internal peptide bonds, which is rapid for
Ser, Ala, Gly, and Thr and slowest for Val and Ile.
19
2. Internal aminolysis at the N-terminus, yielding
diketopiperazine (cyclic dipeptides); this is more rapid
than internal cleavage for small peptides made up of
hydrophobic amino acids and at neutral pH. The
importance of this reaction therefore increases with
increasing extent of diagenesis.
3. Hydrolysis of an amino acid at the C-terminus, which
is acid/base catalyzed but is independent of pH
between pH 5 and 9 (Kahne and Still, 1988). The
importance of this reaction also increases with
proceeding diagenesis.
The overall extent of hydrolysis in fossil biominerals is
a useful proxy of the age of the sample and can be easily
quantified as the percentage of free amino acids (FAA)
over the total hydrolyzable amino acids (THAA): %FAA
¼ [FAA]/[THAA]  100.
The temperature sensitivity of peptide bond hydrolysis
is dependent on the relative stability of the bonds between
pairs of amino acids; overall, studies into the extent of
hydrolysis at high temperature (during laboratory simula-
tions of diagenesis or kinetic experiments) have shown
that, generally, the apparent activation energy for hydroly-
sis is lower than for racemization (Crisp et al., 2013;
Demarchi et al., 2013b; Tomiak et al., 2013). This implies
that at lower (burial) temperatures, hydrolysis will be
favored over racemization and that therefore hydrolysis
represents the rate-limiting pathway for overall diagene-
sis. Indeed, hydrolysis is a key reaction step in racemiza-
tion as it generates N-terminal residues, and these are the
most prone to undergo racemization (Figure 4).
Racemization
An amino acid can exist in one of four states: (a) entirely
bound within a peptide chain, (b) at the N-terminus,
(c) at the C-terminus, or (d) as a free amino acid. This
influences its ability to undergo racemization.
Free amino acids
The main focus of attention has traditionally been on the
mechanisms of racemization of free amino acids in aque-
ous solution; Neuberger (1948) proposed that abstraction
of the hydrogen linked to the central carbon leads to the
collapse of the tetrahedral geometry of the amino acid
and the formation of a planar carbanion. This is equally
likely to be attached by a proton from either direction, with
the possibility of umbrella-like inversion of the tetrahe-
dron, thus regenerating the L-form or producing the
20 AMINO ACID RACEMIZATION DATING
D-enantiomers (Figure 5). The rate-limiting step of the
reaction is therefore the abstraction of the proton. Increas-
ing the electron-withdrawing and resonance-stabilizing
capacities of the substituents attached to the central carbon
of an amino acid causes a corresponding increase in the
racemization rate (Bada and Schroeder, 1975). The mech-
anism is applicable not only to free amino acids but also to
amino acids bound in peptides and proteins: if the amino
group is protonated (as it would be at an N-terminal posi-
tion), then it is able to participate in resonance, facilitating
the loss of the hydrogen.
Different amino acids display different relative racemi-
zation rates: since racemization is enhanced by species
which are able to stabilize the carbanion intermediate,
a major role is played by the –R group. For example,
amino acids that have –R groups which are highly elec-
tron withdrawing (e.g., Ser) should have the fastest race-
mization rates at neutral pH, while those which have
substituents that are electron donating (e.g., Val) should
have the slowest rates (Bada, 1985). Since the –R group
is amino acid specific, each amino acid is characterized
by a different value of enthalpy and entropy, which reflect
Amino Acid Racemization Dating, Figure 4 Schematic
mechanism of peptide bond hydrolysis at the N-terminus,
promoting fast racemization of amino acids (AA) that are
released in the free pool.
Amino Acid Racemization Dating, Figure 5 Proposed mechanism of racemization of free amino acids, via a carbanion intermediate.
the electronic and steric factors determining racemization
rates for free amino acids in aqueous solution. Although
different factors enhance or retard the racemization rates,
racemization of free amino acids in aqueous solution has
generally been found to obey first-order reversible kinetics
(Bada and Schroeder, 1975):
ln½ð1 þ D=LÞ=ð1
D=LÞ
constant
¼ 2kt ðfor amino acids with one chiral centreÞ
ln½ð1 þ A=IÞ=ð1
K0 ðA=IÞÞ
constant
¼ ð1 þ K0 ÞkIle t ðfor IleÞ
where t is time and the constant term is different from zero
as it represents the small amount of D-enantiomer pro-
duced during sample preparation; K0 ¼ 1/Keq and Keq is
the equilibrium alloisoleucine/isoleucine (A/I) ratio,
which is 1.3 to 1.4 (Bada and Schroeder, 1975).
These equations can be used for deriving the racemiza-
tion rate of a given free amino acid in aqueous solution at
a known temperature. Typically, kinetic experiments are
conducted at three (or more) different temperatures and
the increase in D/L value monitored over time. When the
experimentally derived ln [(1 + D/L)/(1
D/L)] is plotted
against the time of isothermal heating, if the reaction is
first order, then a linear relationship is found, of slope
2k. The intercept represents the constant value, i.e., the
preparation-induced racemization. The racemization rates
obtained at the three (or more) temperature points can then
be used for extrapolating the racemization rate at any tem-
perature. This is possible because the reaction rate and the
temperature are directly related in the Arrhenius
relationship:
k ¼ Aeð
Ea=RTÞ
where k is the rate constant, A is the frequency factor
(year
1), Ea is the activation energy (kJ/mol or kcal/
mol), R is the gas constant (0.001987 kcal/mol or
0.008314 kJ/mol), and T is the absolute temperature
(degrees Kelvin, K). A plot of the natural logarithm of
k versus 1/T yields a straight line of slope Ea.
In the real world the situation is more complex, because
unlike free amino acids, the extent of racemization is
a consequence of a number of reactions, notably hydroly-
sis and racemization, with different temperature depen-
dencies, which means that net racemization rates are
unlikely to display a perfect linear relationship with
 AMINO ACID RACEMIZATION DATING
temperature. By estimating the activation energy and the
frequency factor, the racemization rate at the burial tem-
perature for fossil samples of unknown age can be approx-
imated. The temperature value used for fossil samples is
the “effective diagenetic temperature” (EDT; see “Amino
Acid Racemization, Paleoclimate”) representing the inte-
grated effect of all temperatures to which the samples at
a given depth have been exposed (Wehmiller, 1977).
Bound amino acids
Racemization of peptide-bound amino acids occurs via
complex pathways, which have not yet been fully clari-
fied. The body of knowledge acquired in the past decades
by experimenting on both synthetic peptides and
biominerals has allowed building a general model
(Wehmiller, 1980; Mitterer and Kriausakul, 1984)
whereby it is postulated that peptide-bound amino acids
are unable to racemize until progressive hydrolysis drives
them to a terminal position (Figure 4). However, excep-
tions to this general rule are the fast in-chain racemization
of Asn and Asp (Geiger and Clarke, 1987) and Ser
(Demarchi et al., 2013c) and the accelerated rate of race-
mization in diketopiperazines (Steinberg and Bada,
1981) at the peptide chain C-terminus. In general, the rate
of amino acid racemization within a protein retained in
a closed system depends on (a) the temperature of the
environment (or of the experiment), (b) the availability
of water for peptide bond hydrolysis, (c) the position of
the amino acid within the protein sequence throughout
diagenesis, and (d) the three-dimensional structure of the
protein.
Therefore, kinetic relationships derived from high-
temperature experiments are unable to capture the under-
lying chemical rationale of the mechanisms involved in
peptide-bound amino acid racemization. Mathematical
expressions have been developed that enable the calcula-
tion of apparent racemization rates and the assessment
of the temperature dependence of the reactions and, ulti-
mately, the extrapolation of an absolute age. These equa-
tions (first order, parabolic, power transformations;
Clarke and Murray-Wallace, 2006) are usually derived
from high-temperature experiments on biocarbonates,
and these can be further constrained by the measurement
of racemization in samples of well-defined age, for exam-
ple, on radiocarbon-dated shells. This then allows building
a calibrated age equation (e.g., Kosnik et al., 2008, 2013)
that can be used effectively to derive age since death for
subfossil biominerals. This is important, as studies on
the intracrystalline fractions of avian eggshell, mollusk
shells and corals have shown clearly that the pathways
of diagenesis at high temperature may not necessarily be
the same as those at low temperatures (Crisp et al., 2013;
Demarchi et al., 2013b; Tomiak et al., 2013); therefore,
the age-racemization relationship must be calibrated for
samples of known age and where the effective diagenetic
temperature can be reconstructed (e.g., foraminifera from
deep-sea cores). This is further confirmed by the recent
21
study of Allen et al. (2013), who fitted a variety of differ-
ent functions to racemization data and observed that high-
temperature and field data behave differently. In part, this
is because the relatively high activation energies
(120 kJ/mol) mean that unless the temperature range is
very narrow, data usually covers different stages within
the decomposition pathway. Because of this, an alternative
approach, which estimates rate differences simply by
relative scaling, has been proposed (the so-called
“model-free” approach; Crisp et al., 2013; Demarchi
et al., 2013b; Tomiak et al., 2013).
The temperature sensitivity of the reaction is especially
important when considering the calibration of samples
over Pleistocene timescales: temperature fluctuations
affect the rates of racemization, and therefore, the succes-
sion of glacial and interglacial stages needs to be consid-
ered (a recent example of combining kinetic models and
age calibration (by 87Sr/86Sr) over the Quaternary is the
study by Wehmiller and colleagues (2012) in the US
mid-Atlantic Coastal Plain).
Methodology
Sample types
In order to use a substrate for dating, the researcher has to
be confident that the material will prove suitable, i.e., that
the biominerals will have retained its original proteins.
These would have been degrading at a certain rate,
depending upon the proteins themselves (and therefore
the species effect; Lajoie et al., 1980), the pH of the envi-
ronment, and the burial temperature. Biominerals, such as
eggshell, mollusk shells, foraminifera, brachiopods, ostra-
cods, corals, and tooth enamel, are all good substrates for
AAR dating.
When looking at intracrystalline proteins, the most
common problem associated with AAR dating is that the
system is partially reset if the biomineral undergoes diage-
netic alteration, resulting in the opening up of a previously
closed system. Due to the difficulty of identifying diage-
netic transformation from aragonite to calcite, the latter
biomineral is favored. A comparison of the aragonitic
shells and calcitic opercula of Bithynia tentaculata (and
Bithynia troschelii) highlighted the limitations with the
former: in the case of the calcitic opercula, some samples
displayed evidence of distortion and recrystallization,
and these too displayed atypical patterns of racemization
(Penkman et al., 2007, 2013). However, this does not
mean that aragonitic shells should not be used: the pres-
ence and extent of diagenetic alterations can be character-
ized by microscopical observations and techniques such as
X-ray diffraction, and the use of multiple samples of dif-
ferent taxa is important for recognizing aberrant results.
Sample collection
The best way to collect samples for AAR is to collect as
many as reasonable, to note all information available
and, whenever possible, to define the sampling strategy
with the scientists in charge of the excavation of the site.
22 AMINO ACID RACEMIZATION DATING
Details on sample collection strategies can be found in
many studies and are summarized in Wehmiller and Miller
(2000). If collecting in the field, note the depositional
environment and the position of the specimens, especially
when dealing with high-energy depositional environ-
ments, with high likelihood of age mixing (e.g., Wehmiller
et al., 1995; Krause et al., 2010).
If collecting from a museum, the information on sample
collection and storage may not always be available. This is
important because one of the major issues to be taken into
account is whether the samples have been exposed to con-
ditions that have impacted on their thermal history,
increasing the levels of racemization: for example, drying
in an oven or exposure to heat during shallow burial. If
possible, avoid sampling from the surface of an outcrop,
but target instead samples that are buried under at least
1 m of sediments (or at a depth where the seasonal temper-
ature variations are dampened to less than 6  C (Wehmiller
and Miller, 2000)). Burning and heating, especially for
edible mollusks or eggshells, should also be taken into
account when collecting samples from archaeological
sites, e.g., shell middens, or cave deposits in proximity
to hearths. Some useful “burning indicators” can be iden-
tified on the basis of the bulk amino acid composition and
the covariance of D/L values of different amino acids (e.g.,
Brooks et al., 1991; Miller et al., 1999; Demarchi
et al., 2011; Crisp, 2013).
Due to the intergeneric differences in racemization
rates, aminostratigraphies can only be built for
monogeneric samples; therefore, whenever possible sam-
ple multiple taxa from each geological or archaeological
unit. At least five individual samples should be taken from
each depositional horizon (in both geological and archae-
ological settings) and taxon, in order to assess the natural
variability (Miller and Brigham-Grette, 1989) within the
taxon and the site.
How can amino acid racemization be measured?
Any mixture of analytes needs to be separated into its indi-
vidual components in order to achieve quantification. The
classic separation technique for complex mixtures is chro-
matography. Analysis of chiral amino acids is challenging,
because the two molecules (L- and D-enantiomers) are only
subtly different. Broadly, three different strategies have
been used. Originally, diastereomers were measured using
ion exchange chromatography, which could be followed
by a variety of different (some very sensitive) detection
methods. The method had the advantage of being robust,
sufficiently so for Ed Hare and his team to develop
a system that could be taken to the field. However, it had
one major disadvantage: it could only discriminate
between the two diastereomers of isoleucine, one of the
slowest-reacting amino acids.
A new method of gas chromatography (GC), first
developed at NASA-Ames, became routinely used during
the 1970s (see the first interlaboratory comparison study
led by Wehmiller in 1984). This guaranteed very good
separation of the constituent amino acids separated on
a chiral stationary phase (column), and although more
challenging than others, its extremely good separation of
multiple amino acids has meant that GC methods are still
used, notably by some forensic laboratories who demand
the very best separation and resolution, for example, when
assessing very low levels of racemization in young
(forensic) samples (Ohtani and Yamamoto, 1991; Ritz-
Timme and Collins, 2002).
In 1998 Kaufman and Manley published a key paper in
which they used a chiral derivative (based upon chemistry
previously developed by Brückner et al. (1991)) to sepa-
rate nine enantiomeric pairs on a conventional HPLC
(high-pressure liquid chromatography) reverse phase col-
umn (Kaufman and Manley, 1998). This RP-HPLC
method is very sensitive, paving the way for high-
throughput applications of the technique; scientific AAR
studies now commonly report hundreds of analyses. This
is now the most widely applied method, although alterna-
tives (e.g., UHPLC, ultrahigh-performance liquid chroma-
tography) are being considered to enhance the sensitivity
and the speed of the analyses, in order to further reduce
sample size and increase throughput. Published
interlaboratory comparison studies (Wehmiller, 1984,
2013; Powell et al., 2013) detail the different characteris-
tics of each method.
Present investigations and future directions
If suitable samples and suitable separation techniques are
used, AAR can be a very powerful dating tool. The most
successful application of the technique is perhaps in
assessing the age of microorganisms (e.g., foraminifera
tests) in deep-sea cores, where the temperature is stable
and the variability reduced (e.g., Müller, 1984; Kaufman
et al., 2008). In the Quaternary record, applications
include building calibrated geochronological frameworks
for the UK (e.g., Penkman et al., 2013) and sequences in
the USA (e.g., Wehmiller, 2013) and in the Mediterranean
(e.g., Hearty et al., 1986). Great attention has been given
to establishing long and precise sequences for the Holo-
cene, e.g., in Australia (e.g., Kosnik et al., 2013). Calibra-
tion of the D/L values is obtained through radiocarbon, but
advances in luminescence dating and cosmogenic
nuclides dating are likely to be in the spotlight of research
in the next decade.
Currently, AAR cannot be applied as a numerical
method per se, because of a) uncertainties associated with
the temperature history of each sample (especially for the
Quaternary record at midlatitudes) and b) difficulties in
assessing the kinetic parameters of each diagenesis reac-
tion. Paleotemperature models are however becoming
more and more sophisticated (see “Amino Acid Racemi-
zation, Paleoclimate”) and are likely to offer the unprece-
dented opportunity of refining our understanding of the
temperature- and age-driven diagenesis reactions. This,
combined with growing information on protein sequence
and structure, will likely transform the way AAR scientists
 AMINO ACID RACEMIZATION DATING
think and apply the technique. A greater emphasis on
interlaboratory trials (Wehmiller, 1984; Powell
et al., 2013; Wehmiller, 2013) is further evidence of the
growing maturity of the discipline.
Acknowledgments
We are truly indebted to two anonymous reviewers for
their comments on this manuscript. Their insight in the
earlier phases of development of the technique, which
they were willing to share with us, has been of invaluable
help in compiling this review. The responsibility for any
errors or inaccuracies still present is entirely ours.
Kirsty Penkman and the AAR group in York are
thanked for their help, support, and discussion over the
years.
Bibliography
Abelson, P., 1954. Paleobiochemistry: organic constituents of fos-
sils. Washington DC: Carnegie Institution of Washington, Year-
book, 53, pp. 97–101.
Allen, A. P., Kosnik, M. A., and Kaufman, D. S., 2013. Character-
izing the dynamics of amino acid racemization using time-
dependent reaction kinetics: a Bayesian approach to fitting
age-calibration models. Quaternary Geochronology, 18, 63–77.
Andrews, J. T., Bowen, D. Q., and Kidson, C., 1979. Amino acid
ratios and the correlation of raised beach deposits in south-west
England and Wales. Nature, 281, 556–558.
Andrews, J. T., Miller, G. H., Davies, D. C., and Davies, K. H.,
1985. Generic identification of fragmentary Quaternary mol-
luscs by amino acid chromatography: a tool for Quaternary and
palaeontological research. Geological Journal, 20, 1–20.
Bada, J., 1985. Racemization of amino acids. In Barrett, G.C. (ed.),
Chemistry and Biochemistry of the Amino Acids. London: Chap-
man and Hall, pp. 399–414.
Bada, J. L., 1991. Amino acid cosmogeochemistry. Philosophical
Transactions of the Royal Society of London. Series B, Biologi-
cal Sciences, 333, 349–358.
Bada, J. L., and Schroeder, R. A., 1972. Racemization of isoleucine
in calcareous marine sediments: kinetics and mechanism. Earth
and Planetary Science Letters, 15, 1–11.
Bada, J. L., and Schroeder, R. A., 1975. Amino acid racemization
reactions and their geochemical implications. Naturwis-
senschaften, 62, 71–79.
Bada, J. L., Luyendyk, B. P., and Maynard, J. B., 1970. Marine sed-
iments: dating by the racemization of amino acids. Science, 170,
730–732.
Bada, J. L., Schroeder, R. A., and Carter, G. F., 1974. New evidence
for the antiquity of man in North America deduced from aspartic
acid racemization. Science, 184, 791–793.
Bada, J. L., Shou, M.-Y., Man, E. H., and Schroeder, R. A., 1978.
Decomposition of hydroxy amino acids in foraminiferal tests;
kinetics, mechanism and geochronological implications. Earth
and Planetary Science Letters, 41, 67–76.
Bada, J. L., Gillespie, R., Gowlett, J. A. J., and Hedges, R. E. M.,
1984. Accelerator mass spectrometry radiocarbon ages of amino
acid extracts from Californian palaeoindian skeletons. Nature,
312, 442–444.
Barbour Wood, S. L., Krause, R. A., Jr., Kowalewski, M.,
Wehmiller, J., and Simões, M. G., 2006. Aspartic acid racemiza-
tion dating of Holocene brachiopods and bivalves from the
southern Brazilian shelf, South Atlantic. Quaternary Research,
66, 323–331.
23
Bates, M. R., 1993. Quaternary aminostratigraphy in Northwestern
France. Quaternary Science Reviews, 12, 793–809.
Bowen, D. Q., and Sykes, G. A., 1994. How old is Boxgrove man?
Nature, 371, 751.
Bowen, D., Sykes, G., Miller, G., Andrews, J., Brew, J., and Hare,
P., 1985. Amino acid geochronology of raised beaches in south
west Britain. Quaternary Science Reviews, 4, 279–318.
Bowen, D., Sykes, G., and Turner, C., 1988. Correlation of marine
events and glaciations on the Northeast Atlantic margin [and dis-
cussion]. Philosophical Transactions of the Royal Society of
London. Series B, Biological Sciences, 318, 619–635.
Bright, J., and Kaufman, D. S., 2011. Amino acid racemization in
lacustrine ostracodes, part I: effect of oxidizing pre-treatments
on amino acid composition. Quaternary Geochronology, 6,
154–173.
Brooks, A. S., Hare, P. E., Kokis, J. E., Miller, G. H., Ernst, R., and
Wendorf, F., 1990. Dating Pleistocene archaeological sites by
protein diagenesis in ostrich eggshell. Science, 248, 60–64.
Brooks, A., Hare, P., Kokis, J., and Durana, K., 1991. A burning
question: differences between laboratory-induced and natural
diagenesis in Ostrich eggshell proteins. Carnegie Institute of
Washington Yearbook, 2250, 176–179.
Brückner, H., Wittner, R., and Godel, H., 1991. Fully automated
high-performance liquid chromatographic separation of
DL-amino acids derivatized with o-phthaldialdehyde together
with N-isobutyryl-cysteine. Application to food samples.
Chromatographia, 32, 383–388.
Clarke, S. J., and Murray-Wallace, C. V., 2006. Mathematical
expressions used in amino acid racemisation geochronology –
a review. Quaternary Geochronology, 1, 261–278.
Cölfen, H., and Antonietti, M., 2005. Mesocrystals: inorganic
superstructures made by highly parallel crystallization and con-
trolled alignment. Angewandte Chemie International Edition,
44, 5576–5591.
Collins, M. J., and Riley, M. S., 2000. Amino acid racemization in
biominerals: the impact of protein degradation and loss. In
Goodfriend, G. A., Collins M. J., Fogel M. L., Macko S .A.,
Wehmiller J. F. (eds.), Perspectives in Amino Acid and
Protein Geochemistry. Oxford: Oxford University Press,
pp. 120–142.
Collins, M. J., Westbroek, P., Muyzer, G., and de Leeuw, J. W.,
1992. Experimental evidence for condensation reactions
between sugars and proteins in carbonate skeletons. Geochimica
et Cosmochimica Acta, 56, 1539–1544.
Crisp, M. K., 2013. Amino Acid Racemization Dating: Method
Development Using African Ostrich (Struthio camelus) Egg-
shell. Unpublished PhD thesis, York, Department of Chemistry,
University of York.
Crisp, M., Demarchi, B., Collins, M., Morgan-Williams, M.,
Pilgrim, E., and Penkman, K., 2013. Isolation of the intra-
crystalline proteins and kinetic studies in Struthio camelus
(ostrich) eggshell for amino acid geochronology. Quaternary
Geochronology, 16, 110–128.
Davies, K. H., 1983. Amino acid analysis of Pleistocene marine
molluscs from the Gower Peninsula. Nature, 302, 137–139.
Demarchi, B., Williams, M. G., Milner, N., Russell, N., Bailey, G.,
and Penkman, K., 2011. Amino acid racemization dating of
marine shells: a mound of possibilities. Quaternary Interna-
tional, 239, 114–124.
Demarchi, B., Rogers, K., Fa, D. A., Finlayson, C. J., Milner, N.,
and Penkman, K. E. H., 2013a. Intra-crystalline protein diagen-
esis (IcPD) in Patella vulgata. Part I: Isolation and testing of
the closed system. Quaternary Geochronology, 16, 144–157.
Demarchi, B., Collins, M. J., Tomiak, P. J., Davies, B. J., and
Penkman, K. E. H., 2013b. Intra-crystalline protein diagenesis
(IcPD) in Patella vulgata. Part II: Breakdown and temperature
sensitivity. Quaternary Geochronology, 16, 158–172.
24 AMINO ACID RACEMIZATION DATING
Demarchi, B., Collins, M. J., Bergstrom, E., Dowle, A., Penkman,
K. E. H., Thomas-Oates, J., and Wilson, J., 2013c. New experi-
mental evidence for in-chain amino acid racemization of serine
in a model peptide. Analytical Chemistry, 85(12), 5835–5842.
Engel, M. H., and Nagy, B., 1982. Distribution and enantiomeric
composition of amino acids in the Murchison meteorite. Nature,
296, 837–840.
Engel, M. H., Zumberge, J. E., and Nagy, B., 1977. Kinetics of
amino acid racemization in Sequoiadendron giganteum heart-
wood. Analytical Biochemistry, 82, 415–422.
Freeman, C. L., Harding, J. H., Quigley, D., and Rodger, P. M.,
2010. Structural control of crystal nuclei by an eggshell protein.
Angewandte Chemie, International Edition, 49, 5135–5137.
Gaffey, S. J., 1988. Water in skeletal carbonates. Journal of Sedi-
mentary Research, 58, 397–414.
Geiger, T., and Clarke, S., 1987. Deamidation, isomerization, and
racemization at asparaginyl and aspartyl residues in
peptides – succinimide-linked reactions that contribute to
protein-degradation. Journal of Biological Chemistry, 262,
785–794.
Goodfriend, G. A., 1991. Patterns of racemization and
epimerization of amino acids in land snail shells over the course
of the Holocene. Geochimica et Cosmochimica Acta, 55,
293–302.
Goodfriend, G. A., 1992. Rapid racemization of aspartic acid in
mollusc shells and potential for dating over recent centuries.
Nature, 357, 399–401.
Goodfriend, G. A., Brigham-Grette, J., and Miller, G. H., 1996.
Enhanced age resolution of the marine Quaternary record in
the Arctic using aspartic acid racemization dating of bivalve
shells. Quaternary Research, 45, 176–187.
Goodfriend, G. A., Flessa, K. W., and Hare, P., 1997. Variation in
amino acid epimerization rates and amino acid composition
among shell layers in the bivalve Chione from the Gulf of Cali-
fornia. Geochimica et Cosmochimica Acta, 61, 1487–1493.
Goodfriend, G. A., Collins, M. J., Fogel, M. L., Macko, S. A., and
Wehmiller, J. F. (eds.), 2000. Perspectives in Amino Acid and
Protein Geochemistry. New York: Oxford University Press.
Griffin, R., Chamberlain, A., Hotz, G., Penkman, K., and Collins,
M., 2009. Age estimation of archaeological remains using amino
acid racemization in dental enamel: a comparison of morpholog-
ical, biochemical, and known ages-at-death. American Journal
of Physical Anthropology, 140, 244–252.
Harada, N., Kondo, T., Fukuma, K., Uchida, M., Nakamura, T.,
Iwai, M., Murayama, M., Sugawara, T., and Kusakabe, M.,
2002. Is amino acid chronology applicable to the estimation of
the geological age of siliceous sediments? Earth and Planetary
Science Letters, 198, 257–266.
Hare, P., 1988. Organic geochemistry of bone and its relation to the
survival of bone in the natural environment. In Behrensmeyer,
A., and Hill, A. P. (eds.), Fossils in the Making: Vertebrate
Taphonomy and Paleoecology. Chicago: Chicago University
Press, pp. 208–219.
Hare, P. E., and Mitterer, R. M., 1969. Laboratory simulation of
amino acid diagenesis in fossils. Carnegie Institute of Washing-
ton Yearbook, 67, 205–208.
Hare, P. E., Hoering, T. C., and King, K., Jr. (eds.), 1980. Biogeo-
chemistry of Amino Acids. New York: Wiley.
Hearty, P. J., and Kaufman, D. S., 2000. Whole-rock aminostra-
tigraphy and Quaternary sea-level history of the Bahamas.
Quaternary Research, 54, 163–173.
Hearty, P. J., Miller, G. H., Stearns, C. E., and Szabo, B. J., 1986.
Aminostratigraphy of Quaternary shorelines in the Mediterra-
nean basin. Geological Society of America Bulletin, 97,
850–858.
Hendy, E. J., Tomiak, P. J., Collins, M. J., Hellstrom, J., Tudhope,
A. W., Lough, J. M., and Penkman, K. E. H., 2012. Assessing
amino acid racemization variability in coral intra-crystalline pro-
tein for geochronological applications. Geochimica and
Cosmochimica Acta, 86, 338–353.
Hsu, J. T., Leonard, E. M., and Wehmiller, J. F., 1989. Aminostra-
tigraphy of Peruvian and Chilean Quaternary marine terraces.
Quaternary Science Reviews, 8, 255–262.
Hudson, J., 1967. The elemental composition of the organic frac-
tion, and the water content, of some recent and fossil mollusc
shells. Geochimica et Cosmochimica Acta, 31, 2361–2378.
Johnson, B. J., and Miller, G. H., 1997. Archaeological applications
of amino acid racemization. Archaeometry, 39, 265–287.
Kahne, D., and Still, W. C., 1988. Hydrolysis of a peptide bond in
neutral water. Journal of the American Chemical Society, 110,
7529–7534.
Kaufman, D. S., 2003. Amino acid paleothermometry of Quater-
nary ostracodes from the Bonneville Basin, Utah. Quaternary
Science Reviews, 22, 899–914.
Kaufman, D. S., and Brigham-Grette, J., 1993. Aminostratigraphic
correlations and paleotemperature implications, Pliocene-
Pleistocene high-sea-level deposits, northwestern Alaska. Qua-
ternary Science Reviews, 12, 21–33.
Kaufman, D. S., and Manley, W. F., 1998. A new procedure for
determining dl amino acid ratios in fossils using reverse phase
liquid chromatography. Quaternary Science Reviews, 17,
987–1000.
Kaufman, D. S., and Sejrup, H.-P., 1995. Isoleucine epimerization
in the high-molecular-weight fraction of pleistocene Arctica.
Quaternary Science Reviews, 14, 337–350.
Kaufman, D. S., Miller, G. H., and Andrews, J. T., 1992. Amino
acid composition as a taxonomic tool for molluscan fossils: an
example from Pliocene-Pleistocene Arctic marine deposits.
Geochimica et Cosmochimica Acta, 56, 2445–2453.
Kaufman, D.S., Polyak, L., Adler, R., Channell, J. E., and Xuan, C.,
2008. Dating late Quaternary planktonic foraminifer Neoglobo-
quadrina pachyderma from the Arctic Ocean using amino acid
racemization. Paleoceanography, 23, PA3224, doi:10.1029/
2008PA001618.
Kaufman, D. S., Cooper, K., Behl, R., Billups, K., Bright, J., Gard-
ner, K., Hearty, P., Jakobsson, M., Mendes, I., O'Leary, M.,
Polyak, L., Rasmussen, T., Rosa, F., and Schmidt, M., 2013.
Amino acid racemization in mono-specific foraminifera from
Quaternary deep-sea sediments. Quaternary Geochronology,
16, 50–61.
Kennedy, G. L., Lajoie, K. R., and Wehmiller, J. F., 1982.
Aminostratigraphy and faunal correlations of late Quaternary
marine terraces, Pacific Coast, USA. Nature, 299, 545–547.
Kimber, R., Kennedy, N., and Milnes, A., 1994. Amino acid racemi-
zation dating of a 140,000 year old tephra-loess-palaeosol
sequence on the Mamaku Plateau near Rotorua, New Zealand.
Australian Journal of Earth Sciences, 41, 19–26.
Kosnik, M. A., Kaufman, D. S., and Hua, Q., 2008. Identifying out-
liers and assessing the accuracy of amino acid racemization mea-
surements for geochronology: I. Age calibration curves.
Quaternary Geochronology, 3, 308–327.
Kosnik, M. A., Kaufman, D. S., and Hua, Q., 2013. Radiocarbon-
calibrated multiple amino acid geochronology of Holocene mol-
luscs from Bramble and Rib Reefs (Great Barrier Reef, Austra-
lia). Quaternary Geochronology, 16, 73–86.
Krause, R. A., Jr., Barbour, S. L., Kowalewski, M., Kaufman, D. S.,
Romanek, C. S., Simões, M. G., and Wehmiller, J. F., 2010.
Quantitative comparisons and models of time-averaging in
bivalve and brachiopod shell accumulations. Paleobiology, 36,
428–452.
Kvenvolden, K. A., Peterson, E., Wehmiller, J., and Hare, P., 1973.
Racemization of amino acids in marine sediments determined by
gas chromatography. Geochimica et Cosmochimica Acta, 37,
2215–2225.
 AMINO ACID RACEMIZATION DATING
Lajoie, K. R., Wehmiller, J. F., and Kennedy, G. L., 1980. Inter- and
intra-generic trends in apparent racemization kinetics of amino
acids in Quaternary mollusks. In Hare, P. E., Hoering, T., and
King, K. (eds.), Biogeochemistry of Amino Acids. New York:
Wiley, pp. 305–340.
Li, H., Xin, H. L., Kunitake, M. E., Keene, E. C., Muller, D. A., and
Estroff, L. A., 2011. Calcite prisms from Mollusk shells (Atrina
rigida): swiss-cheese-like organic–inorganic single-crystal com-
posites. Advanced Functional Materials, 21, 2028–2034.
Marshall, E., 1990. Racemization dating: great expectations.
Science, 247, 799.
Masters, P. M., and Bada, J. L., 1977. Racemization of isoleucine in
fossil molluscs from Indian middens and interglacial terraces in
southern California. Earth and Planetary Science Letters, 37,
173–183.
Masters, P. M., and Bada, J. L., 1978. Amino acid racemization dat-
ing of bone and shell. Advances in Chemistry Series, 171,
117–138.
Meijer, T., and Cleveringa, P., 2009. Aminostratigraphy of Middle
and Late Pleistocene deposits in The Netherlands and the south-
ern part of the North Sea Basin. Global and Planetary Change,
68, 326–345.
Miller, G. H., and Brigham-Grette, J., 1989. Amino acid geochro-
nology: resolution and precision in carbonate fossils. Quater-
nary International, 1, 111–128.
Miller, G. H., and Hare, P., 1980. Amino acid geochronology: integ-
rity of the carbonate matrix and potential of molluscan fossils. In
Hare, P. E., Hoering, T. C., and King, K., Jr. (eds.), Biogeochem-
istry of Amino Acids. New York: Wiley, pp. 415–443.
Miller, G. H., Hollin, J. T., and Andrews, J. T., 1979. Aminostra-
tigraphy of UK Pleistocene deposits. Nature, 281, 539–543.
Miller, G. H., Sejrup, H. P., Mangerud, J., and Andersen, B. G.,
1983. Amino acid ratios in Quaternary molluscs and foraminif-
era from western Norway: correlation, geochronology and
paleotemperature estimates. Boreas, 12, 107–124.
Miller, G. H., Magee, J. W., and Jull, A., 1997. Low-latitude glacial
cooling in the Southern Hemisphere from amino-acid racemiza-
tion in emu eggshells. Nature, 385, 241–244.
Miller, G. H., Beaumont, P. B., Deacon, H. J., Brooks, A. S., Hare,
P. E., and Jull, A., 1999. Earliest modern humans in southern
Africa dated by isoleucine epimerization in ostrich eggshell.
Quaternary Science Reviews, 18, 1537–1548.
Miller, G. H., Fogel, M. L., Magee, J. W., Gagan, M. K., Clarke,
S. J., and Johnson, B. J., 2005. Ecosystem collapse in Pleisto-
cene Australia and a human role in megafaunal extinction.
Science, 309, 287–290.
Mitterer, R. M., 1993. The diagenesis of proteins and amino acids in
fossil shells. In Engel, M. H., and Macko, S. A. (eds.), Organic
Geochemistry. New York: Springer, pp. 739–753.
Mitterer, R. M., and Kriausakul, N., 1984. Comparison of rates and
degrees of isoleucine epimerization in dipeptides and tripeptides.
Organic Geochemistry, 7, 91–98.
Moini, M., Klauenberg, K., and Ballard, M., 2011. Dating silk by
capillary electrophoresis mass spectrometry. Analytical Chemis-
try, 83, 7577–7581.
Müller, P., 1984. Isoleucine epimerization in Quaternary planktonic
foraminifera: effects of diagenetic hydrolysis and leaching, and
Atlantic–Pacific intercore correlations. Meteor Forschungser-
gebnisse Reihe C Geologie und Geophysik, 38, 25–47.
Murray-Wallace, C., and Kimber, R., 1987. Evaluation of the amino
acid racemization reaction in studies of Quaternary marine sedi-
ments in South Australia. Australian Journal of Earth Sciences,
34, 279–292.
Murray-Wallace, C., Belperio, A., Gostin, V., and Cann, J., 1993.
Amino acid racemization and radiocarbon dating of interstadial
marine strata (oxygen isotope stage 3), Gulf St. Vincent, South
Australia. Marine Geology, 110, 83–92.
25
Murray-Wallace, C., Brooke, B., Cann, J., Belperio, A., and
Bourman, R., 2001. Whole-rock aminostratigraphy of the
Coorong Coastal Plain, South Australia: towards a 1 million year
record of sea-level highstands. Journal of the Geological Soci-
ety, 158, 111–124.
Nardi, S., Binda, P. L., Baccelle, L. S., and Concheri, G., 1994.
Amino acids of Proterozoic and Ordovician sulphide-coated
grains from western Canada: record of biologically-mediated
pyrite precipitation. Chemical Geology, 111, 1–15.
Neuberger, A., 1948. Stereochemistry of amino acids. Advances in
Protein Chemistry, 4, 297–383.
Oches, E. A., and McCoy, W. D., 2001. Historical developments
and recent advances in amino acid geochronology applied to
loess research: examples from North America, Europe, and
China. Earth-Science Reviews, 54, 173–192.
Ohtani, S., and Yamamoto, K., 1991. Age estimation using the race-
mization of amino acid in human dentin. Journal of Forensic
Sciences, 36, 792–800.
Okumura, T., Suzuki, M., Nagasawa, H., and Kogure, T., 2013.
Microstructural control of calcite via incorporation of
intracrystalline organic molecules in shells. Journal of Crystal
Growth, 381, 114–120.
Ortiz, J. E., Torres, T., and Pérez-Gonzlez, A., 2013. Amino acid
racemization in four species of ostracodes: taxonomic, environ-
mental, and microstructural controls. Quaternary Geochronol-
ogy, 16, 129–143.
Penkman, K. E. H., Preece, R. C., Keen, D. H., Maddy, D., Schreve,
D. C., and Collins, M. J., 2007. Testing the aminostratigraphy of
fluvial archives: the evidence from intra-crystalline proteins
within freshwater shells. Quaternary Science Reviews, 26,
2958–2969.
Penkman, K. E. H., Kaufman, D. S., Maddy, D., and Collins, M. J.,
2008. Closed-system behaviour of the intra-crystalline fraction
of amino acids in mollusc shells. Quaternary Geochronology,
3, 2–25.
Penkman, K. E., Preece, R. C., Bridgland, D. R., Keen, D. H.,
Meijer, T., Parfitt, S. A., White, T. S., and Collins, M. J., 2011.
A chronological framework for the British Quaternary based
on Bithynia opercula. Nature, 476, 446–449.
Penkman, K. E. H., Preece, R. C., Bridgland, D. R., Keen, D. H.,
Meijer, T., Parfitt, S. A., White, T. S., and Collins, M. J.,
2013. An aminostratigraphy for the British Quaternary based
on Bithynia opercula. Quaternary Science Reviews, 61,
111–134.
Powell, J., Collins, M. J., Cussens, J., MacLeod, N., and Penkman,
K. E. H., 2013. Results from an amino acid racemization inter-
laboratory proficiency study; design and performance evalua-
tion. Quaternary Geochronology, 16, 183–197.
Ritz-Timme, S., and Collins, M. J., 2002. Racemization of aspartic
acid in human proteins. Ageing Research Reviews, 1, 43–59.
Ritz-Timme, S., Rochholz, G., Schütz, H., Collins, M., Waite, E.,
Cattaneo, C., and Kaatsch, H.-J., 2000. Quality assurance in
age estimation based on aspartic acid racemisation. International
Journal of Legal Medicine, 114, 83–86.
Rockwell, T. K., 1992. Ages and deformation of marine terraces
between point conception and Gaviota: Western Transverse
Ranges, California. In Fletcher, C. H. I., and Wehmiller, J. F.
(eds.), Quaternary Coasts of the United States, Marine and
Lacustrine Systems. Tulsa: SEPM, pp. 333–341.
Rutter, N. W., and Blackwell, B., 1995. Amino acid racemization
dating. In Rutter, N. W., and Catto, N. R. (eds.), Dating Methods
for Quaternary Deposits. St. John’s: Geological Association of
Canada, pp. 125–167.
Rutter, N., Schnack, E. J., Rio, J. D., Fasano, J. L., Isla, F. I., and
Radtke, U., 1989. Correlation and dating of Quaternary littoral
zones along the Patagonian coast, Argentina. Quaternary Sci-
ence Reviews, 8, 213–234.
26 AMINO ACID RACEMIZATION, BIOSTRATIGRAPHY
Salamon, M., Tuross, N., Arensburg, B., and Weiner, S., 2005. Rel-
atively well preserved DNA is present in the crystal aggregates
of fossil bones. Proceedings of the National Academy of Sci-
ences of the United States of America, 102, 13783–13788.
Sejrup, H. P., and Haugen, J.-E., 1994. Amino acid diagenesis in the
marine bivalve Arctica islandica Linné from northwest Euro-
pean sites: only time and temperature? Journal of Quaternary
Science, 9, 301–309.
Steinberg, S., and Bada, J. L., 1981. Diketopiperazine formation
during investigations of amino acid racemization in dipeptides.
Science, 213, 544–545.
Suzuki, M., Okumura, T., Nagasawa, H., and Kogure, T., 2011.
Localization of intracrystalline organic macromolecules in mol-
lusk shells. Journal of Crystal Growth, 337, 24–29.
Sykes, G. A., Collins, M. J., and Walton, D. I., 1995. The signifi-
cance of a geochemically isolated intracrystalline organic
fraction within biominerals. Organic Geochemistry, 23,
1059–1065.
Tomiak, P. J., Penkman, K. E. H., Hendy, E. J., Demarchi, B.,
Murrells, S., Davis, S. A., McCullagh, P., and Collins, M. J.,
2013. Testing the limitations of artificial protein degradation
kinetics using known-age massive Porites coral skeletons.
Quaternary Geochronology, 16, 87–109.
Torres, T., Llamas, J., Canoira, L., Coello, F., García-Alonso, P., and
Ortiz, J., 2000. Aminostratigraphy of two Pleistocene marine
sequences from the Mediterranean coast of Spain: Cabo de
Huertas (Alicante) and Garrucha (Almería). In Goodfriend,
G. A., Collins, M. J., Fogel, M. L., Macko, S. A., and Wehmiller,
J. F. (eds.), Perspectives in Amino Acids and Protein Geochem-
istry. Oxford: Oxford University Press, pp. 263–278.
Towe, K. M., and Thompson, G. R., 1972. The structure of some
bivalve shell carbonates prepared by ion-beam thinning. Calci-
fied Tissue Research, 10, 38–48.
Walton, D., 1998. Degradation of intracrystalline proteins and
amino acids in fossil brachiopods. Organic Geochemistry, 28,
389–410.
Wehmiller, J., 1976. Amino acids in fossil corals: racemization
(epimerization) reactions and their implications for diagenetic
models and geochronological studies. Geochimica and
Cosmochimica Acta, 40, 763–776.
Wehmiller, J. F., 1977. Amino acid studies of the Del Mar, Califor-
nia, midden site: apparent rate constants, ground temperature
models, and chronological implications. Earth and Planetary
Science Letters, 37, 184–196.
Wehmiller, J. F., 1980. Intergeneric differences in apparent racemi-
zation kinetics in mollusks and foraminifera: implications for
models of diagenetic racemization. In Hare, P. E., Hoering, T.,
and King, K. (eds.), Biogeochemistry of Amino Acids. New
York: Wiley, pp. 341–345.
Wehmiller, J. F., 1982. A review of amino acid racemization studies
in Quaternary mollusks: Stratigraphic and chronologic applica-
tions in coastal and interglacial sites, pacific and Atlantic coasts,
United States, United Kingdom, Baffin Island, and tropical
islands. Quaternary Science Reviews, 1, 83–120.
Wehmiller, J. F., 1984. Interlaboratory comparison of amino acid
enantiomeric ratios in fossil Pleistocene mollusks. Quaternary
Research, 22, 109–120.
Wehmiller, J. F., 1989. Amino acid racemization: applications in
chemical taxonomy and chronostratigraphy of Quaternary fos-
sils. Short Courses in Geology, 5, 287–313.
Wehmiller, J. F., 2013. Interlaboratory comparison of amino acid
enantiomeric ratios in Pleistocene fossils. Quaternary Geochro-
nology, 16, 173–182.
Wehmiller, J., and Belknap, D., 1982. Amino acid age estimates,
Quaternary Atlantic coastal plain: comparison with U-series
dates, biostratigraphy, and paleomagnetic control. Quaternary
Research, 18, 311–336.
Wehmiller, J., and Hare, P., 1971. Racemization of amino acids in
marine sediments. Science, 173, 907–911.
Wehmiller, J. F., and Miller, G. H., 2000. Aminostratigraphic dating
methods in Quaternary Geology. In Noller, J. S., Sowers, J. M.,
and Lettis, W. R. (eds.), Quaternary Geochronology: Methods
and Applications. Washington DC: American Geophysical
Union, pp. 187–222.
Wehmiller, J. F., Lajoie, K. R., Kvenvolden, K. A., Peterson, E.,
Belknap, D. F., Kennedy, G. L., Addicott, W. O., Vedder, J. G.,
and Wright, R. W., 1977. Correlation and chronology of Pacific
Coast marine terrace deposits of continental United States by
fossil Amino Acid Stereochemistry – technique, evaluation, rel-
ative ages, kinetic model ages, and geological implications,
U.S. Geological Survey. Open-file report, pp. 77–680.
Wehmiller, J. F., York, L. L., and Bart, M. L., 1995. Amino acid
racemization geochronology of reworked Quaternary mollusks
on U.S. Atlantic coast beaches: implications for chronostra-
tigraphy, taphonomy, and coastal sediment transport. Marine
Geology, 124, 303–337.
Wehmiller, J. F., Harris, W. B., Boutin, B. S., and Farrell, K. M.,
2012. Calibration of amino acid racemization (AAR) kinetics
in United States mid-Atlantic Coastal Plain Quaternary mollusks
using 87Sr/86Sr analyses: evaluation of kinetic models and esti-
mation of regional Late Pleistocene temperature history. Quater-
nary Geochronology, 7, 21–36.
Wilson, L., and Pollard, A. M., 2002. Here today, gone tomorrow?
Integrated experimentation and geochemical modeling in studies
of archaeological diagenetic change. Accounts of Chemical
Research, 35, 644–651.
York, L. L., and Wehmiller, J. F., 1992. Aminostratigraphic results
from Cape Lookout, NC, and their relation to the preserved
Quaternary marine record of SE North Carolina. Sedimentary
Geology, 80, 279–291.
Cross-references
Amino Acid Racemization, Biostratigraphy
Amino Acid Racemization, Coastal Sediments
Amino Acid Racemization, Eolianites
Amino Acid Racemization, Fluvial and Lacustrine Sediments
(AAR)
Amino Acid Racemization, Marine Sediments
Amino Acid Racemization, Paleoclimate
Biostratigraphy
Chromatography
Luminescence Dating
Magnetostratigraphic Dating
Radiocarbon Dating
Terrestrial Cosmogenic Nuclide Dating
U-Series Dating
AMINO ACID RACEMIZATION, BIOSTRATIGRAPHY
Matthew Kosnik
Department of Biological Sciences, Macquarie
University, Sydney, NSW, Australia
Definition
Amino acid racemization biostratigraphy refers to using
the ratio of right-handed (D) to left-handed (L) amino
acids (D/L value) preserved in sediment or fossils to deter-
mine the age of sedimentary deposits including the
 AMINO ACID RACEMIZATION, BIOSTRATIGRAPHY
duration of time recorded in sedimentary deposits. This
can be done by direct comparison of D/L values to deter-
mine relative ages or by using calibrated D/L values to
estimate ages in calendar years.
Characteristics
The age of sedimentary deposits is of fundamental impor-
tance to many topics in stratigraphy, sedimentology, evo-
lutionary biology, and paleobiology. Amino acid
racemization (AAR) provides a cost-effective way to
directly date large numbers of fossils, and it has revolu-
tionized the way paleontologists have been able to study
the age of fossil deposits. AAR has made three key contri-
butions to biostratigraphy. (1) AAR D/L values are one of
many methods used to determine the relative age of sedi-
mentary deposits (older deposits have higher D/L values).
(2) AAR data are one of a few methods used to quantify
the extent of mixing in sedimentary deposits. (3) Cali-
brated AAR data have enabled cost-effective studies of
time-averaging (the amount of time recorded in a single
deposit).
The use of AAR in biostratigraphy dates back to the
1960s (e.g., Hare and Mitterer, 1967) and took hold in
the 1970s (e.g., Bada et al., 1970; Bada and Protsch,
1973). Because the rate of racemization increases with
increasing temperature, the timespan over which AAR is
useful depends on climactic conditions. AAR is useful
over longer timespans in colder climates, stretching back
over a million years in areas such as New Zealand (Bowen
et al., 1998, and see Amino Acid Racemization, Coastal
Sediments).
Post-deposition mixing and other factors impose funda-
mental limits on the resolution of fossil records. While
D/L ratios can be used to demonstrate that two sediment
layers have distinct age distributions (i.e., that one deposit
is older than another), calibrated AAR enables the amount
of time within a deposit and between deposits to be quan-
tified. Calibrated AAR has been widely and primarily
used to study sediment mixing processes in Holocene-
aged sediments (e.g., Meldahl et al., 1997; Kowalewski
et al., 1998, 2000; Carroll et al., 2003; Kidwell et al.,
2005; Barbour-Wood et al., 2006; Kosnik et al., 2007,
2009; Krause et al., 2010; Kosnik et al., 2013; Simonson
et al., 2013). The use of AAR as a cost-effective means
of dating large numbers of shells has enabled these studies
to date multiple shells per sedimentary deposit and quan-
tify the amount of time recorded rather than relying on
a single age estimate. These datasets have enabled others
to test mechanistic models of sediment mixing and fossil
assemblage formation (e.g., Olszewski, 2004).
In other cases, AAR has been used for recognition of
taphonomic processes, such as shell preservation and
transport (Murray-Wallace and Belperio, 1994; Wehmiller
et al., 1995; Martin et al., 1996). In these latter studies, it is
often observed that samples of one or more Pleistocene-
aged specimens can be found mixed into Holocene shell
accumulations and that the source units for these reworked
27
Pleistocene samples can also be identified in the region
using AAR (Wehmiller et al., 1995; Martin et al., 1996).
A key advantage of AAR when dating fossil deposits,
relative to other methods such as Pb-210 or OSL, is that
AAR directly dates the fossils rather than the surrounding
sediments. Directly dating the fossils avoids any biases
associated with differential sediment mixing.
Bibliography
Bada, J. L., and Protsch, R., 1973. Racemization reaction of aspartic
acid and its use in dating fossil bones. Proceedings of the
National Academy of Science, 70, 1331–1334.
Bada, J. L., Luyendyk, B. P., and Maynard, J. B., 1970. Dating by
the racemization of amino acids. Science, 170, 730–732.
Barbour-Wood, S. L., Krause, R. A., Jr., Kowalewski, M.,
Wehmiller, J. F., and Simoes, M. G., 2006. Aspartic acid racemi-
zation dating of Holocene brachiopods and bivalves from the
southern Brazilian shelf, South Atlantic. Quaternary Research,
66, 323–331.
Bowen, D. Q., Pillans, B., Sykes, G. A., Beu, A. G., Edwards, A. R.,
Kamp, P. J. J., and Hull, A. G., 1998. Amino acid geochronology
of pleistocene marine sediments in the Wanganui basin; a New
Zealand framework for correlation and dating. Journal of the
Geological Society of London, 155, Part 3: 439-446.
Carroll, M., Kowalewski, M., Simões, M. G., and Goodfriend,
G. A., 2003. Quantitative estimates of time averaging in
terebratulid brachiopod shell accumulations from a modern trop-
ical shelf. Paleobiology, 29, 381–402.
Hare, P. E., and Mitterer, R. M., 1967. Nonprotein amino acids in
fossil shells. Yearbook Carnegie Institution of Washington, 65,
362–364.
Kidwell, S. M., Best, M. M. R., and Kaufman, D. S., 2005. Tapho-
nomic trade-offs in tropical marine death assemblages: differen-
tial time averaging, shell loss, and probable bias in siliciclastic
vs. carbonate facies. Geology, 33, 729–732.
Kosnik, M. A., Hua, Q., Jacobsen, G. E., Kaufman, D. S., and Wüst,
R. A., 2007. Sediment mixing and stratigraphic disorder
revealed by the age-structure of Tellina shells in Great Barrier
Reef sediment. Geology, 35, 811–814.
Kosnik, M. A., Hua, Q., Kaufman, D. S., and Wüst, R. A., 2009.
Taphonomic bias and time- averaging in tropical molluscan
death assemblages: differential shell half-lives in Great Barrier
Reef sediment. Paleobiology, 34(4), 565–586.
Kosnik, M. A., Kaufman, D. S., and Hua, Q., 2013. Radiocarbon-
calibrated multiple amino acid geochronology of Holocene mol-
luscs from Bramble and Rib Reefs (Great Barrier Reef, Austra-
lia). Quaternary Geochronology, 16, 73–86.
Kowalewski, M., 1996. Time-averaging, overcompleteness, and the
fossil record. Journal of Geology, 104, 317–326.
Kowalewski, M., Goodfriend, G. A., and Flessa, K. W., 1998. The
high-resolution estimates of temporal mixing in shell beds: the
evils and virtues of time averaging. Paleobiology, 24, 287–304.
Kowalewski, M., Serrano, G. E. A., Flessa, K. W., and Goodfriend,
G. A., 2000. Dead delta’s former productivity: two trillion shells
at the mouth of the Colorado River. Geology, 28, 1059–1062.
Krause, R. A., Barbour, S. L., Kowalewski, M., Kaufman, D. S.,
Romanek, C. S., Simões, M. G., and Wehmiller, J. F., 2010.
Quantitative comparisons and models of time-averaging in
bivalve and brachiopod shell accumulations. Paleobiology,
36(3), 428–452.
Martin, R. E., Wehmiller, J. F., Harris, M. S., and Liddell, W. D.,
1996. Comparative taphonomy of bivalves and foraminifera
from Holocene tidal flat sediments, Bahia La Choya, Sonora,
Mexico (northern Gulf of California): taphonomic grades and
temporal resolution. Paleobiology, 22, 80–90.
28 AMINO ACID RACEMIZATION, COASTAL SEDIMENTS
Meldahl, K. H., Flessa, K. W., and Cutler, A. H., 1997. Time aver-
aging and postmortem skeletal survival in benthic fossil assem-
blages: quantitative comparisons among Holocene
environments. Paleobiology, 23, 207–229.
Murray-Wallace, C., and Belperio, A. P., 1994. Identification of
remanie fossils using amino acid racemisation. Alcheringa, 18,
219–227.
Olszewski, T. D., 2004. Modeling the influence of taphonomic
destruction, reworking, and burial on time-averaging in fossil
accumulations. Palaios, 19, 39–50.
Simonson, A. E., Lockwood, R., and Wehmiller, J. F., 2013. Three
approaches to radiocarbon calibration of amino acid racemiza-
tion in Mulinia lateralis from the Holocene of the Chesapeake
Bay, USA. Quaternary Geochronology, 16, 62–72.
Wehmiller, J. F., York, L. L., and Bart, M. L., 1995. Amino acid
racemization geochronology of reworked Quaternary mollusks
on U.S. Atlantic coast beaches; implications for chronostra-
tigraphy, taphonomy, and coastal sediment transport. Marine
Geology, 124, 303–337.
Cross-references
Amino Acid Racemization Dating
Amino Acid Racemization, Coastal Sediments
AMINO ACID RACEMIZATION, COASTAL
SEDIMENTS
John F. Wehmiller
Department of Geological Sciences, University of
Delaware, Newark, DE, USA
Definitions
Amino acid racemization: The phenomenon of conversion
of “left-handed” (L or “levo”) amino acids to their “right-
handed” (D or “dextro”) form. In most living systems,
100 % of the amino acids are of the L form, and the
conversion results in an equal mixture of D and L when
the racemization reaction is complete.
Coastal sediments: Sedimentary material that accumulates
and is preserved in coastal environments such as estuaries,
beaches, shallow continental shelves, back-barrier bays
and lagoons, and dunes. These sediments can be domi-
nated by biological material (shells) and those dominated
by inorganic material that may form by cementation or
evaporation, such as carbonates or evaporates and mud
or clay. Sediments range in age (time since deposition)
from days to millions of years.
Introduction
Amino acid racemization (AAR) dating has been applied
to a large variety of coastal sediments representing
a range of geological ages and depositional environments
(Figure 1). In most cases, samples analyzed for the extent
of racemization are of biogenic origin (mollusks, forami-
nifera, mixed carbonate sediments, or corals). AAR has
been applied to coastal sediments as young as a few hun-
dred years and as old as several million years. The goals
of these studies include as follows: (1) correlation and rel-
ative age assignments for Pleistocene coastal or shallow
marine units, (2) assessing Quaternary tectonic deforma-
tion rates and/or relative sea-level histories,
(3) deciphering the chronology of Pleistocene glacial
advances in coastal regions, and (4) understanding Holo-
cene sedimentary processes including age mixing. In most
cases, multiple samples of different genera have been ana-
lyzed from each collection site, and relative ages are
inferred from apparent clusters of D/L values
(aminozones).
Topics reviewed here include the following categories,
with representative references: (1) emergent deposits,
active continental margins, such as the Mediterranean
(Hearty et al., 1986, 2007) and the Pacific coasts of South
America (Hsu et al., 1989) and North America
(Wehmiller, 2013); (2) emergent deposits and passive con-
tinental margins, such as Australia (Murray-Wallace,
1995, 2000) and the Atlantic coasts of South America
(Aguirre et al., 1995) and North America (Wehmiller,
2013); (3) offshore deposits and passive margins
(Murray-Wallace et al., 2005; Meijer and Cleveringa,
2009; Wehmiller et al., 2010); (4) tropical islands and car-
bonate reefs (Hearty et al., 2007); (5) coastal regions
affected by glaciations (Miller and Mangerud, 1985;
Kaufman and Brigham-Grette, 1993; Refsnider et al.,
2013); and (6) age-mixing studies in coastal regions
(Wehmiller et al., 1995; Kosnik et al., 2009). The latter
three topics are discussed in more detail in other entries
(see Amino Acid Racemization, Eolianites). AAR results
are used to determine relative ages within a study region,
but if associated with independent age calibrations
(derived from U–Th, 14C, or other numerical dating
method), then AAR can be used to provide semiquantita-
tive ages using appropriate model(s) for the overall kinet-
ics of racemization.
AAR studies of active continental margins
Along the Pacific coast of North America, south of Cape
Mendocino, strike-slip relative motion is related to the
northwest motion of the Pacific Plate. Coastal units and
marine terraces in this region usually record relatively
low uplift rates (approx. <0.2 m/ka years) with a few nota-
ble exceptions in localized areas. North of Cape
Mendocino, where plate convergence related to spreading
from the Juan de Fuca Ridge dominates the region, ages
for uplifted marine terraces suggest more variable uplift
rates, occasionally as high as 1 m/ka years (Kennedy et al.,
1982; Dupré et al., 1991; Muhs et al., 1992a, 2004).
Uranium-thorium dating of solitary corals has been
applied to numerous marine terrace sites along the North
American Pacific coast (Muhs et al., 2004). These U–Th
results provide rigorous calibration for the regional
aminostratigraphy, allowing age estimates for AAR
results from other sites with no independent age control.
Two fundamental principles regarding racemization were
tested and evaluated using samples from Pacific coast
 AMINO ACID RACEMIZATION, COASTAL SEDIMENTS 29

Amino Acid Racemization, Coastal Sediments, Figure 1 World map showing coastal regions where AAR studies have been
conducted. Letters on map denote the following: A, active margins; P, passive margins; G, coastal areas influenced by glaciations; X,
offshore studies. Solid circles on map identify tropical islands for which AAR data exist.

terraces (Wehmiller et al., 1977; Kennedy et al., 1982):
(1) D/L values in multiple taxa increase with increasing
terrace elevation for vertical sequences of terraces, and
(2) D/L values increase with decreasing latitude (increas-
ing temperature) in samples of known equal age. Mollus-
can faunal assemblages indicative of local water
temperatures at the time of formation of late Pleistocene
terrace deposits also are age indicative because of their
association with the marine climate record and associated
calibrated AAR results (Kennedy et al., 1982). The origi-
nal aminostratigraphic framework (Kennedy et al., 1982)
was extended with results from uplifted terraces on the
Baja California Peninsula (Keenan et al., 1987) and the
Gulf of California (Ortlieb, 1991). Regionally specific
studies of AAR in uplifted terraces in Southern California
include Santa Barbara County (Rockwell et al., 1992), the
Palos Verdes Hills (Muhs et al., 1992b), and San Diego
County (Kern and Rockwell, 1992). Localized zones of
unusually rapid uplift rates (up to 5 m/ka) are inferred
from the ages (AAR, C14, and U–Th) of terraces found
along the northern margin of the Santa Barbara Channel
(Wehmiller, 1992, 2013).
The marine terraces along the South American Pacific
coast between 15 and 30 S are influenced by the sub-
duction of the Nazca Plate beneath the South American
Plate. In the northern area (Peru) six distinct aminozones,
estimated to represent ages up to 900 ka, are recognized,
with independent calibration using electron spin reso-
nance (ESR) dating (Hsu et al., 1989). In the more south-
ern region (northern and central Chile), five aminozones
were recognized, the oldest estimated to represent an age
of 700 ka (Hsu et al., 1989). In several cases, multiple
aminozones were identified within a single localized ter-
race sequence. The highest uplift rates in Peru (0.5 m/
ka) are found where the Nazca Ridge is being subducted
under that South American Plate (Hsu et al., 1989; Hsu,
1992; Goy et al., 1992; Saillard et al., 2011). Uplifted ter-
races are highest above the southern flank of the Nazca
Ridge, being lower both to the north and south of the local
region of ridge subduction. To the south, terrace uplift
rates are generally less than 0.2 m/ka (Hsu, 1992; Ortlieb
et al., 1996). In some cases these low uplift rates have
resulted in terrace reoccupation, with deposits of at least
two interglacial ages being found within what appears to
be a single mapped morphologic unit (Leonard and
Wehmiller, 1992).
A broad regional study of AAR data in mollusks
(Glycimeris and Arca) from uplifted marine deposits in
the western and central Mediterranean spans a latitude
range of 10 and a longitude range of 25 (Hearty
et al., 1986). The AAR data cluster into five distinct
aminozones, the second youngest being associated with
four U–Th coral ages that cluster around 126  7 ka,
equivalent to marine isotope substage 5e (Hearty et al.,
1986). More detailed studies of selected sites in the west-
ern Mediterranean (Torres et al., 2000, 2013) using mod-
ern analytical methods (see Chromatography), while
consistent with the fivefold aminostratigraphy, indicate
that variability in D/L values within an aminozone can
occur because of age mixing and/or shell alteration.
Coastal Quaternary units are preserved intermittently
around the North and South Islands of New Zealand,
where active subduction and strike-slip faulting occur
associated with the convergence of the Australian and
30 AMINO ACID RACEMIZATION, COASTAL SEDIMENTS
Pacific Plates. AAR data from these sites were occasion-
ally obtained on wood (Pillans, 1990) but more often
using marine mollusks (Ota et al., 1996; Bowen et al.,
1998). Along the south shore of the North Island of New
Zealand, up to four kilometers of Plio-Pleistocene sedi-
ments are exposed in the Wanganui Basin with at least five
aminozones representing deposition over the past one mil-
lion years (Bowen et al., 1998). Supporting chronologic
data include magnetostratigraphy and fission-track dating
of interbedded volcanic ashes. At least seven nearby
uplifted marine terraces can be correlated with portions
of the Wanganui section using associated AAR data
(Bowen et al., 1998), and similar correlation can be made
with aminozones on the South Island of New Zealand (Ota
et al., 1996).
AAR studies of passive continental margins
The Atlantic margin of the United States ( 45 N to
24 N) includes sites of some of the earliest AAR studies
of Quaternary marine mollusks (Mitterer, 1974; Muhs
et al., 2004; Wehmiller, 2013). Exposed Quaternary
coastal units in this region occur within 10 m of sea
level, reflecting minimal tectonic deformation. Neverthe-
less, there are underlying structures that influence the pat-
tern of Quaternary sedimentation, the Albemarle
Embayment (eastern North Carolina) being one major
example. The Quaternary section in the Albemarle
Embayment is up to 100 m thick, thinning to no more than
10 m to the north and south, over the Norfolk and Cape
Fear arches, respectively. Combined U–Th, AAR, and Sr
dating indicates that at least eight separate phases
(aminozones) of sedimentation between 2 Ma and
100 ka are preserved in the Albemarle section
(Wehmiller et al., 2010, 2012). To the south of the
Albemarle, multiple coast parallel paleoshorelines are rec-
ognized, but fewer AAR aminozones (3–5) are recog-
nized, most likely because the record is thin and less
complete. To the north of the Albemarle, a record of
paleoshorelines and fluvial incision during multiple gla-
cial-interglacial cycles within the Chesapeake Bay drain-
age system is preserved (Groot et al., 1990; Genau et al.,
1994; Mirecki et al., 1995; Wehmiller, 2013). Units dated
by AAR in this region are found within approx. 5 m of
present sea level. AAR data from the Delmarva Peninsula,
combined with stratigraphic and geomorphic analysis,
indicate that the Peninsula evolved in a southerly direction
during the Pleistocene, responding to multiple cycles of
fluvial incision and valley filling during glacial-
interglacial sea-level cycles. AAR data for beach shells
in the US mid-Atlantic region demonstrate age-mixing
processes associated with coastal dynamics and subsur-
face stratigraphy (Wehmiller et al., 1995).
AAR and associated radiocarbon, U–Th, or lumines-
cence data exist for Quaternary coastal units on the west-
ern, southern, and eastern coasts of Australia, including
some offshore marine cores (Murray-Wallace, 1995,
2000; Murray-Wallace et al., 2001, 2005; Hearty and
O’Leary, 2008). AAR data for the southern coast can be
grouped into at least four aminozones representing deposi-
tion during most of the Pleistocene. Whole-rock
(eolianite) AAR age estimates combined with lumines-
cence dating from the Coorong Coastal Plain and the
mouth of the River Murray, South Australia, show that
a series of coastal barriers have developed in this region
over the past 500 ka in response to glacial-interglacial
sea-level cycles (Murray-Wallace et al., 2001, 2010).
Extensive AAR data exist for coastal units on the west-
ern coast of Australia, primarily based on the analysis of
eolianites, samples composed of nearly 100 % carbonate
fragments (Hearty, 2003; Hearty and O’Leary, 2008;
O’Leary et al., 2008). Six aminozones are recognized,
spanning an interval from the Holocene (<10 ka) to over
500 ka. A global synthesis and discussion of the geochro-
nology of the last interglacial (MIS 5e) shoreline (Hearty
et al., 2007) include AAR data from multiple Australian
sites and also sites from the central Pacific, western Atlan-
tic, and the Mediterranean.
Hearty and Aharon (1988) and Kosnik et al. (2007,
2009, 2013) reported AAR data for samples from the
Great Barrier Reef, offshore eastern Australia. These stud-
ies have focused on Holocene specimens, often using age-
calibrated samples to establish the kinetics of racemization
in different sample types. Kosnik’s studies also emphasize
the use of AAR in evaluating the magnitude of age mixing
within Holocene assemblages.
AAR data for the coastal plain of Argentina between
35 S and 55 S for multiple taxa identify both Holo-
cene and Pleistocene aminozones (Rutter et al., 1989,
1990; Aguirre et al., 1995). Two Pleistocene aminozones
are recognized, one representing at least a portion of the
last interglacial high sea stand. Electron spin resonance
(ESR) provides limited independent calibration. Last
interglacial shoreline deposits are usually found between
10 and 15 m above sea level. AAR data indicate that some
deposits, at elevations up to 40 m, are older than the last
interglacial. Holocene coastal/marine deposits are also
found in this region, at elevations up to 5 m, so there is
ample potential for age mixing.
Carr et al. (2010) reported AAR data for marine mol-
lusks and terrestrial gastropods from sites along the coast
of South Africa. Optically stimulated luminescence
(OSL) results provided the primary age control, indicating
formation of beach deposits and associated dunes between
138  7 ka and 113  6 ka, at elevations up to 8 m above
sea level. AAR data, while limited, are consistent with
these age estimates.
AAR studies of submerged deposits, continental
margins
AAR data for shallow submerged portions of the continen-
tal margins of Europe, Australia, North America, and Ant-
arctica also exist. In many cases, these studies compare
offshore data with results from nearby onshore sites.
Along the US Atlantic margin, aminozones representing
 AMINO ACID RACEMIZATION, COASTAL SEDIMENTS
multiple intervals of deposition during the Pleistocene are
encountered from offshore New Jersey, Maryland, Dela-
ware, and North Carolina (Toscano and York, 1992; Groot
et al., 1995; Wehmiller et al., 2010; Miller et al., 2013). In
most cases there are associated lithostratigraphic, paleon-
tological, and geophysical data to aid in the environmental
analysis of the AAR data. Sr-isotope age estimates provide
some qualitative calibration for some of the older (early
Pleistocene) aminozones (Wehmiller et al., 2012; Miller
et al., 2013). These continental margin records usually
contain evidence of multiple transgressive-regressive
cycles associated with sea-level changes during the Pleis-
tocene, as in the case of southeastern Australia (Murray-
Wallace et al., 2005), the North Sea (Meijer and
Cleveringa, 2009), and the Black Sea (Nicholas et al.,
2011). AAR results from deeper water environments off
northeastern Australia (Hearty et al., 2004) demonstrate
the kinetic effects of different bottom water temperatures
(see Amino Acid Racemization, Marine Sediments).
Low-latitude carbonate islands and reefs
The stratigraphy and geochronology of deposits on
numerous low-latitude islands from the Atlantic and
Pacific Oceans have been studied using AAR, using either
land snails, whole-rock analyses of carbonate eolianites
(see Amino Acid Racemization, Eolianites), or marine
mollusks. In many cases the results contribute to the chro-
nology of local eolianite (dune) formation, usually associ-
ated with local relative sea-level histories. Studies of
Pleistocene units from Atlantic, Mediterranean, or Carib-
bean islands include Bermuda (Hearty et al., 1992,
2007), the Bahamas (Hearty and Kaufman, 2000; Hearty,
2010), Jamaica (Goodfriend and Mitterer, 1993), Barba-
dos (corals) (Wehmiller et al., 1976), Madeira
(Goodfriend et al., 1996a), the Canary Islands (Zazo
et al., 2002), and the Mediterranean region (Hearty et al.,
1986; Hearty, 1987; Rose et al., 1999). Pacific Island stud-
ies include Lord Howe Island (Brooke et al., 2003;
Woodroffe et al., 2006), New Guinea (Hearty and Aharon
1988), and Hawaii (Hearty et al., 2000, 2007). AAR stud-
ies of Holocene corals from assorted localities establish
the kinetics of racemization during the earliest phases of
diagenesis (Hearty and Aharon, 1988; Goodfriend et al.,
1992; Hendy et al., 2012; Tomiak et al., 2013). The tropi-
cal temperatures at which these corals are found enhance
the overall rate of racemization, allowing high-resolution
kinetic studies.
Coastal regions influenced by glaciations
Discussions of these regions will be found in separate
entries (Glacial Landscape (Cosmogenic Nuclide)) but
selected references for these areas are reviewed here (see
Figure 1).
High-latitude coastal regions that have been affected by
glaciations reveal a complex record of Quaternary deposi-
tion. AAR-dated samples are often transported from their
original site of deposition, but results from these samples
31
can constrain histories of ice dynamics and local relative
sea-level history. Along the Arctic margins of Pleistocene
ice sheets, ice margin and sea-level histories are
documented in northeastern Siberia, northwest Alaska,
Baffin Island, Greenland, and Spitzbergen (Miller et al.,
1992; Kaufman, 1992; Kaufman and Brigham-Grette,
1993; Funder et al., 1994; Brigham-Grette and Hopkins,
1995; Goodfriend et al., 1996b; Kelly et al., 1999;
Brigham-Grette et al., 2001; Gualtieri et al., 2003;
Mangerud et al., 2008; Refsnider et al., 2013).
Along the southern margin of the Laurentide Ice Sheet,
clusters of D/L values in mollusks identified multiple
advances of ice in the Hudson Bay Lowland, including
events that were recognizable only in transported samples
found in glacial sediments (Andrews et al., 1983;
Thorleifson et al., 1992). AAR data for in-place estuarine
mollusks identified both pre- and post-last glacial fossilif-
erous deposits in the St. Lawrence valley (southeastern
Canada) using AAR analyses of marine mollusks
(Occhietti et al., 1996).
A long history of AAR study of sites along the southern
margin of the Scandinavian Ice Sheet and associated sites
in the United Kingdom exists. Miller and Mangerud
(1985) and Sejrup et al. (2000) analyzed AAR in mollusks
found in marine and glacial deposits of the North Sea,
Norway, Denmark, Germany, and the Netherlands. Meijer
and Cleveringa (2009) summarized AAR data for the
region, generated by several laboratories over three
decades. Bowen (2000) and Penkman et al. (2011, 2013)
used AAR data to interpret the glacial, marine, and
nonmarine records of the United Kingdom. Almost all of
these studies demonstrate that multiple phases of deposi-
tion, either shallow marine or fluvial, are apparent in the
regional record, extended back through the entire
Pleistocene.
Application of AAR to high-latitude Southern hemi-
sphere sites includes southern Chile (Clapperton et al.,
1995) and Antarctica (Hart and Webb, 1998; Forsberg
et al., 2003; Gardner et al., 2006). Each of these studies
includes AAR and other independent chronological infor-
mation to reconstruct local histories of glacial advance and
retreat. As with other high-latitude studies, the low tem-
peratures of these regions result in very slow racemization
kinetics, meaning that low D/L values are observed even
in samples that are perhaps as old as 4 Ma.
Summary and conclusions
The use of racemization in the study of Quaternary coastal
environments and processes has a history almost as long
as the history of the AAR dating method itself, as some
of the earliest studies involved fossils from coastal sites
along the US Atlantic coast (see Wehmiller, 2013, for
review). Regions of study range from polar to equatorial
latitudes, with potential age ranges for application being
determined by the effective temperatures in these climates.
Many coastal studies, in particular, benefit from associ-
ated U–Th dating of associated corals (at low- or
32 AMINO ACID RACEMIZATION, COASTAL SEDIMENTS
mid-latitude sites), the preferred sample type for U-series
dating. In addition, AAR data from coastal studies have
yielded important insights into geochemical and taxo-
nomic factors that influence racemization, particularly in
situations where long relative age sequences of samples
are available. Extensive AAR study of major active conti-
nental margins includes marine terraces on the Pacific
coasts of North and South America and the Mediterra-
nean; passive continental margins include the Atlantic
coasts of North and South America and much of the coast
of Australia, including several long chronologic sections
spanning the entire Quaternary. Studies of sites in high
northern latitude regions in North America (Baffin Island)
and Europe (United Kingdom, Scandinavia, and the
southern North Sea) have been historically important for
the development of AAR methods and have contributed
substantial insights into the Quaternary glacial and climate
history of northern hemisphere ice sheets. Data from
lower-latitude sites provide a broad perspective for com-
parison with mid- and high-latitude results, and a large
dataset exists for a variety of types of carbonate (marine
and eolian) deposits on tropical islands from the Atlantic,
Pacific, and Mediterranean. These latter results come from
a variety of sample types not usually sampled from higher-
latitude sites, and the rapid racemization at low-latitude
temperatures permits high-resolution kinetic and age-
mixing studies.
Bibliography
Aguirre, M. L., Bowen, D. Q., Sykes, S. A., and Whatley, R. C.,
1995. A provisional aminostratigraphical framework for late
Quaternary marine deposits in Buenos Aires province, Argen-
tina. Marine Geology, 128, 85–104.
Andrews, J. T., Shilts, W. W., and Miller, G. H., 1983. Multiple
deglaciations of the Hudson Bay Lowlands, Canada, since depo-
sition of the Missinaibi (last-interglacial?) formation. Quater-
nary Research, 19, 18–37.
Bowen, D. Q., 2000. Revised aminostratigraphy for land-sea corre-
lations from the northeastern North Atlantic region. In
Goodfriend, G. A., Collins, M. J., Fogel, M. L., Macko, S. A.,
and Wehmiller, J. F. (eds.), Perspectives in Amino Acid and Pro-
tein Geochemistry. Oxford/New York: Oxford University Press,
pp. 253–262.
Bowen, D. Q., Pillans, B., Sykes, G. A., Beu, A. G., Edwards, A. R.,
Kamp, P. J. J., and Hull, A. G., 1998. Amino acid geochronology
of Pleistocene marine sediments in the Wanganui basin; a New
Zealand framework for correlation and dating. Journal of the
Geological Society of London, 155(Part 3), 439–446.
Brigham-Grette, J., and Carter, L. D., 1992. Pliocene marine trans-
gressions of northern Alaska: circumarctic Correlations and
Paleoclimate. Arctic, 43, 74–89.
Brigham-Grette, J., and Hopkins, D. M., 1995. Emergent marine
record and paleoclimate of the last interglaciation along the
northwest Alaskan coast. Quaternary Research, 43, 159–173.
Brigham-Grette, J., Hopkins, D. M., Ivanov, V. F., Basilyan, A.,
Benson, S. L., Heiser, P., and Pushkar, V., 2001. Last interglacial
(Isotope Stage 5) glacial and sea level history of coastal
Chukotka Peninsula and St. Lawrence Island, western Beringia.
Quaternary Science Reviews, 20, 419–436.
Brooke, B. P., Murray-Wallace, C., Woodroffe, C. D., and Heijnis,
H., 2003. Quaternary aminostratigraphy of eolianite on Lord
Howe Island, Southwest Pacific Ocean. Quaternary Science
Reviews, 22, 387–406.
Carr, A. S., Bateman, M. D., Roberts, D. L., Murray-Wallace, C.,
Jacobs, Z., and Holmes, P. J., 2010. The last interglacial sea-level
high stand on the southern Cape coastline of South Africa.
Quaternary Research, 73, 351–363.
Clapperton, C. M., Sugden, D. E., Kaufman, D. S., and McCulloch,
R. D., 1995. The last glaciations in Central Magellan Strait,
southernmost Chile. Quaternary Research, 44, 133–144.
Dupré, W. R., Morrison, R. B., Clifton, H. E., Lajoie, K. R., Ponti,
D. J., Powell, C. L. I. I., Mathieson, S. A., Sarna-Wojcicki,
A. M., Leithold, E. L., Lettis, W. R., McDowell, P. F., Rockwell,
T. K., Unruh, J. R., and Yeats, R. S., 1991. Quaternary geology of
the Pacific margin. In Morrison, R. B. (ed.), Quaternary Non-
glacial Geology, Conterminous US. Boulder, CO: Geological
Society of America. The Geology of North America, Vol. K-2,
pp. 141–214.
Forsberg, C. F., Lovlie, R., Jansen, E., Solheim, A., Sejrup, H. P.,
and Lie, H. E., 2003. A 1.3-myr palaeoceanographic record from
the continental margin off Dronning Maud Land, Antarctica.
Palaeogeography, Palaeoclimatology, Palaeoecology, 198,
223–235.
Funder, S., Hjort, C., and Landvik, J. Y., 1994. The last glacial
cycles in East Greenland, an overview. Boreas, 23, 283–293.
Gardner, N., Hall, B., and Wehmiller, J. F., 2006. Pre-Holocene
raised beaches at Cape Ross, Southern Victoria Land, Antarctica.
Marine Geology, 229, 273–284.
Genau, R. B., Madsen, J. A., McGeary, S., and Wehmiller, J. F.,
1994. Seismic reflection identification of Susquehanna River
paleochannels on the mid-Atlantic coastal plain. Quaternary
Research, 42, 166–175.
Goodfriend, G. A., and Mitterer, R. M., 1993. A 45,000-yr record of
a tropical lowland biota; the land snail fauna from cave sedi-
ments at Coco Ree, Jamaica. Geological Society of America
Bulletin, 105, 18–29.
Goodfriend, G. A., Hare, P. E., and Druffel, E. R. M., 1992. Aspartic
acid racemization and protein diagenesis in corals over the last
350 years. Geochimica et Cosmochimica Acta, 56, 3847–3850.
Goodfriend, G. A., Cameron, R. A. D., Cook, L. M., Courty, M.-A.,
Fedoroff, N., Livett, E., and Tallis, J., 1996a. The Quaternary
eolian sequence of Madeira; stratigraphy, chronology, and
paleoenvironmental interpretation. Palaeogeography,
Palaeoclimatology, Palaeoecology, 120, 195–234.
Goodfriend, G. A., Brigham-Grette, J., and Miller, G. H., 1996b.
Enhanced age resolution of the Marine Quaternary Record in
the Arctic using aspartic acid racemization dating of bivalve
shells. Quaternary Research, 45, 176–187.
Goy, J. L., Machare, J., Ortlieb, L., and Zazo, C., 1992. Quaternary
shorelines in southern Peru: a record of global sea-level fluctua-
tions and tectonic uplift in Chala Bay. Quaternary International,
15(16), 99–112.
Groot, J. J., Ramsey, K. W., and Wehmiller, J. F., 1990. Ages of the
Bethany, Beaverdam, and Omar Formations of Southern Dela-
ware. Newark, NJ: Delaware Geological Survey. Delaware Geo-
logical Survey report of investigations number 47, pp. 1–19.
Groot, J. J., Benson, R. N., and Wehmiller, J. F., 1995. Palynologi-
cal, foraminiferal and aminostratigraphic studies of quaternary
sediments from the U. S. middle Atlantic upper continental
slope, continental shelf and coastal plain. Quaternary Science
Reviews, 14, 17–49.
Gualtieri, L., Vartanyan, S., Brigham-Grette, J., and Anderson,
P. M., 2003. Pleistocene raised marine deposits on Wrangel
Island, northeast Siberia and implications for the presence of
an east Siberian ice sheet. Quaternary Research, 59, 399–410.
Hart, C. P., and Webb, P. N., 1998. Amino acid based geochronol-
ogy of Unit 3.1 (Quaternary), CRP-1, Victoria Land Basin,
Antarctica. Terra Antartica, 5, 697–701.
 AMINO ACID RACEMIZATION, COASTAL SEDIMENTS
Hearty, P. J., 1987. New data on the Pleistocene of Mallorca. Qua-
ternary Science Reviews, 6, 245–257.
Hearty, P. J., 2003. Stratigraphy and timing of eolianite deposition
on Rottnest Island, western Australia. Quaternary Research,
60, 211–222.
Hearty, P. J., 2010. Chronostratigraphy and morphological changes
in Cerion land snail shells over the past 130 ka on Long Island,
Bahamas. Quaternary Geochronology, 5, 50–64.
Hearty, P. J., and Aharon, P., 1988. Amino acid chronostratigraphy
of late quaternary coral reefs: Huon Peninsula, New Guinea,
and the Great Barrier Reef, Australia. Geology, 16, 579–583.
Hearty, P. J., and Kaufman, D. S., 2000. Whole-rock aminostra-
tigraphy and quaternary sea-level history of the Bahamas. Qua-
ternary Research, 54, 163–173.
Hearty, P. J., and O’Leary, M. J., 2008. Carbonate eolianites, quartz
sands, sea-level cycles, western Australia. Quaternary Geochro-
nology, 3, 26–55.
Hearty, P. J., Miller, G. H., Stearns, C. E., and Szabo, B. J., 1986.
Aminostratigraphy of quaternary shorelines in the Mediterranean
Basin. Geological Society of America Bulletin, 97, 850–858.
Hearty, P. J., Vacher, H. L., and Mitterer, R. M., 1992. Aminostra-
tigraphy and ages of Pleistocene limestones of Bermuda; with
suppl. data 92–07. Geological Society of America Bulletin,
104, 471–480.
Hearty, P. J., Kaufman, D. S., Olson, S. L., and James, H. F., 2000.
Stratigraphy and whole-rock amino acid geochronology of key
Holocene and last interglacial carbonate deposits in the Hawai-
ian Islands. Pacific Science, 54, 423–442.
Hearty, P. J., O'Leary, M. J., Kaufman, D. S., Page, M. C., and
Bright, J., 2004. Amino acid geochronology of individual fora-
minifer (Pulleniatina obliquiloculata) tests, north Queensland
margin, Australia; a new approach to correlating and dating Qua-
ternary tropical marine sediment cores. Paleoceanography, 19,
14–14.
Hearty, P. J., Hollin, J. T., Neumann, A. C., O’Leary, M. J., and
McCulloch, M., 2007. Global sea-level fluctuations during the
last interglaciation (MIS5e). Quaternary Science Reviews, 26,
2090–2112.
Hendy, E. J., Tomiak, P. J., Collins, M. J., Hellstrom, J., Tudhope,
A. W., Lough, J. M., and Penkman, K. E. H., 2012. Assessing
amino acid racemization variability in coral intra-crystalline pro-
tein for geochronological applications. Geochimica et
Cosmochimica Acta, 86, 338–353.
Hsu, J. T., 1992. Quaternary uplift of the Peruvian coast related to
the subduction of the Nazca Ridge; 13.5 to 15.6 degrees south
latitude. Quaternary International, 15–16, 87–97.
Hsu, J. T., Leonard, E. M., and Wehmiller, J. F., 1989. Aminostra-
tigraphy of Peruvian and Chilean Quaternary marine terraces.
Quaternary Science Reviews, 8, 255–262.
Kaufman, D., 1992. Aminostratigraphy of Pliocene-Pleistocene high-
sea-level deposits, Nome coastal plain and adjacent nearshore
area, Alaska. Geological Society of America Bulletin, 104, 40–52.
Kaufman, D. S., and Brigham-Grette, J., 1993. Aminostratigraphic
correlations and paleotemperature implications, Pliocene-
Pleistocene high-sea-level deposits, northwestern Alaska.
Quaternary Science Reviews, 12, 21–33.
Keenan, E. M., Ortlieb, L., and Wehmiller, J. F., 1987. Amino acid
dating of quaternary marine terraces, Bahia Asuncion, Baja Cal-
ifornia Sur, Mexico. Journal of Coastal Research, 3, 297–305.
Kelly, M., Funder, S., Houmark-Nielsen, M., Knudsen, K. L.,
Kronborg, C., Landvik, J. Y., and Sorby, L., 1999. Quaternary
glacial and marine environmental history of northwest Green-
land; a review and reappraisal. Quaternary Science Reviews,
18, 373–392.
Kennedy, G. L., Lajoie, K. R., and Wehmiller, J. F., 1982.
Aminostratigraphy and faunal correlations of late quaternary
marine terraces, Pacific coast, USA. Nature, 299, 545–547.
33
Kern, J. P., and Rockwell, T. K., 1992. Chronology and deformation
of marine shorelines, San Diego County, California. In Fletcher,
C. H., III, and Wehmiller, J. F. (eds.), Quaternary Coasts of the
United States: Marine and Lacustrine Systems. Tulsa, OK:
SEPM (Society for Sedimentary Geology). Special publication
number 48, pp. 377–382.
Kosnik, M. A., Hua, Q., Jacobsen, G.E., Kaufman, D. S., and Wust,
R. A., 2007. Sediment mixing and stratigraphic disorder
revealed by the age-structure of Tellina shells in Great Barrier
Reef sediment. Geology, 35(9), 811–814.
Kosnik, M. A., Hua, Q., Kaufman, D. S., and Wüst, R. A., 2009.
Taphonomic bias and time-averaging in tropical molluscan death
assemblages: differential shell half-lives in Great Barrier Reef
sediment. Paleobiology, 35, 565–586.
Kosnik, M. A., Kaufman, D. S., and Hua, Q., 2013. Radiocarbon-
dated multiple amino acid geochronology of Holocene molluscs
from Bramble and Rib Reefs (Great Barrier Reef, Australia).
Quaternary Geochronology, 16, 73–86.
Leonard, E. M., and Wehmiller, J. F., 1992. Low uplift rates and ter-
race reoccupation inferred from mollusk aminostratigraphy,
Coquimbo Bay area, Chile. Quaternary Research, 38, 246–259.
Mangerud, J., Kaufman, D., Hansen, J., and Svendsen, J. I., 2008.
Ice-free conditions in novaya zemlya 35 000–30 000 cal years
B.P., as indicated by radiocarbon ages and amino acid racemiza-
tion evidence from marine molluscs. Polar Research, 27,
187–208.
Meijer, T., and Cleveringa, P., 2009. Aminostratigraphy of middle
and late Pleistocene deposits in the Netherlands and the southern
part of the North Sea basin. Global and Planetary Change, 68,
326–345.
Miller, G. H., and Mangerud, J., 1985. Aminostratigraphy of Euro-
pean marine interglacial deposits. Quaternary Science Reviews,
4, 215–278.
Miller, G. H., Funder, S., de Vernal, A., and Andrews, J. T., 1992.
Timing and character of the last interglacial-glacial transition in
the eastern Canadian Arctic and northwest Greenland. In Clark,
P. U., and Lea, P. D. (eds.), The Last Interglacial-Glacial Transi-
tion in North America. Boulder, CO: Geological Society of
America. Special Paper 270, pp. 223–231.
Miller, K. G., Sugarman, P. J., Browning, J. V., Sheridan, R. E.,
Kulhanek, D. K., Monteverde, D. H., Wehmiller, J. F., Lombardi,
C., and Feigenson, M. D., 2013. Pleistocene sequence stratigra-
phy of the shallow continental shelf, offshore New Jersey: con-
straints of Integrated Ocean Drilling Program Leg 313 core
holes. Geosphere, 9, 74–95.
Mirecki, J. E., Wehmiller, J. F., and Skinner, A., 1995. Geochronol-
ogy of quaternary coastal units, southeastern Virginia. Journal of
Coastal Research, 11, 1135–1144.
Mitterer, R. M., 1974. Pleistocene stratigraphy in southern Florida
based on amino acid diagenesis in fossil Mercenaria. Geology,
2, 425–428.
Muhs, D. R., Rockwell, T. K., and Kennedy, G. L., 1992a. Late
Quaternary uplift rates of marine terraces on the Pacific Coast
of North America, southern Oregon to Baja California Sur. Qua-
ternary International, 15–16, 121–133.
Muhs, D. R., Miller, G. H., Whelan, J. F., and Kennedy, G. L., 1992b.
Aminostratigraphy and oxygen isotope stratigraphy of marine ter-
race deposits, Palos Verdes Hills and San Pedro areas, Los
Angeles County, California. In Fletcher, C. H., III, and Wehmiller,
J. F. (eds.), Quaternary Coasts of the United States: Marine and
Lacustrine Systems. Tulsa, OK: SEPM (Society for Sedimentary
Geology). Special publication number 48, pp. 363–376.
Muhs, D. R., Wehmiller, J. F., Simmons, K. R., and York, L. L.,
2004. Quaternary sea level history of the United States. In Gilles-
pie, A. R., Porter, S. C., and Atwater, B. F. (eds.), The Quater-
nary Period in the United States. Amsterdam: Elsevier,
pp. 147–183.
34 AMINO ACID RACEMIZATION, COASTAL SEDIMENTS
Murray-Wallace, C., 1995. Aminostratigraphy of Quaternary
coastal sequences in southern Australia; an overview. Quater-
nary International, 26, 69–86.
Murray-Wallace, C. V., 2000. Quaternary coastal aminostra-
tigraphy: Australian data in a global context. In Goodfriend,
G. A., Collins, M. J., Fogel, M. L., Macko, S. A., and Wehmiller,
J. F. (eds.), Perspectives in Amino Acid and Protein Geochemis-
try. Oxford/New York: Oxford University Press, pp. 279–300.
Murray-Wallace, C., Brooke, B. P., Cann, J. H., Belperio, A. P., and
Bourman, R. P., 2001. Whole-rock aminostratigraphy of the
Coorong coastal plain, South Australia; towards a 1 million year
record of sea-level highstands. Journal of the Geological Society
of London, 158(Part 1), 111–124.
Murray-Wallace, C., Ferland, M. A., and Roy, P. S., 2005. Further
amino acid racemisation evidence for glacial age, multiple
lowstand deposition on the New South Wales outer continental
shelf, southeastern Australia. Marine Geology, 214, 235–250.
Murray-Wallace, C., Bourman, R. P., Prescott, J. R., Williams, F.,
Price, D. M., and Belperio, A. P., 2010. Aminostratigraphy and
thermoluminescence dating of coastal aeolianites and the later
Quaternary history of a failed delta: the River Murray mouth
region, South Australia. Quaternary Geochronology, 5, 28–49.
Nicholas, W. A., Chivas, A. R., Murray-Wallace, C., and Fink, D.,
2011. Prompt transgression and gradual salinisation of the Black
Sea during the early Holocene constrained by amino acid race-
mization and radiocarbon dating. Quaternary Science Reviews,
30, 3769–3790.
O’Leary, M. J., Hearty, P. J., and McCulloch, M., 2008. Geomorphic
evidence of major sea-level fluctuations during marine isotope
substage-5e. Cape Cuvier, Western Australia, Geomorphology,
102, 595–602.
Occhietti, S., Balescu, S., Lamothe, M., Clet, M., Cronin, T.,
Ferland, P., and Pichet, P., 1996. Late stage 5 glacio-isostatic
sea in the St. Lawrence Valley, Canada and United States. Qua-
ternary Research, 45, 128–137.
Ortlieb, L., 1991. Quaternary vertical movements along the coasts
of Baja California and Sonora. In Dauphin, J. P., and Simoneit,
B. R. T. (eds.), The Gulf and Peninsular Province of the Califor-
nia. Tulsa, OK: American Association of Petroleum Geologists,
Memoir 47, pp. 447–480.
Ortlieb, L., Zazo, C., Goy, J. L., Dabrio, C., and Machare, J., 1996.
Pampa del Palo; an anomalous composite marine terrace on the
uprising coast of southern Peru. Journal of South American
Earth Sciences, 9, 367–379.
Ota, Y., Pillans, B., Berryman, K., Beu, A., Fujimori, T., Miyauchi,
T., Berger, G., Beu, A. G., and Climo, F. M., 1996. Pleistocene
coastal terraces of Kaikoura Peninsula and the Marlborough
coast, South Island, New Zealand, New Zealand. Journal of
Geology and Geophysics, 39, 51–73.
Penkman, K. E. H., Preece, R. C., Bridgland, D. R., Keen, D. H.,
Meijer, T., Parfitt, S. A., White, T. S., and Collins, M. J., 2011.
A chronological framework for the British Quaternary based
on Bithynia opercula. Nature, 476, 446–449.
Penkman, K. E. H., Preece, R. C., Bridgland, D. R., Keen, D. H.,
Meijer, T., Parfitt, S. A., White, T. S., and Collins, M. J., 2013.
An aminostratigraphy for the British Quaternary based on
Bithynia opercula. Quaternary Science Reviews, 61, 111–134.
Pillans, B., 1990. Pleistocene marine terraces in New Zealand:
a review. New Zealand Journal of Geology and Geophysics,
33, 219–231.
Refsnider, K. A., Miller, G. H., Frechette, B., and Rood, D. H.,
2013. A chronological framework for the Clyde Foreland For-
mation, Eastern Canadian Arctic, derived from amino acid race-
mization and cosmogenic radionuclides. Quaternary
Geochronology, 16, 21–34.
Rockwell, T. K., Nolan, J. M., Johnson, D. L., and Patterson, R. H.,
1992. Ages and deformation of marine terraces between Point
Conception and Gaviota, western transverse ranges, California.
In Fletcher, C. H., III, and Wehmiller, J. F. (eds.), Quaternary
Coasts of the United States: Marine and Lacustrine Systems.
Tulsa, OK: SEPM (Society for Sedimentary Geology). Special
publication number 48, pp. 333–341.
Rose, J., Meng, X., and Watson, C., 1999. Palaeoclimate and
palaeoenvironmental responses in the western mediterranean
over the last 140 ka; evidence from Mallorca, Spain. Journal of
the Geological Society of London, 156(Part 2), 435–448.
Rutter, N., Schnack, E. J., del Rio, J., Fasano, J. L., Isla, F. I., and
Radtke, U., 1989. Correlation and dating of Quaternary littoral
zones along the Patagonian coast, Argentina. Quaternary Sci-
ence Reviews, 8, 213–234.
Rutter, N., Radtke, U., and Schnack, E. J., 1990. Comparison of
ESR and amino acid data in correlating and dating Quaternary
shorelines along the Patagonian coast, Argentina. Journal of
Coastal Research, 6, 391–411.
Saillard, M., Hall, S. R., Audin, L., Farber, D. L., Regard, V., and
Herail, G., 2011. Andean coastal uplift and active tectonics in
southern Peru: 10Be surface exposure dating of differentially
uplifted marine terrace sequences (San Juan de Marcona, 15.4
degree S). Geomorphology, 128, 178–190.
Sejrup, H. P., Larsen, E., Landvik, J., King, E. L., Haflidason, H.,
and Nesje, A., 2000. Quaternary glaciations in southern
Fennoscandia; evidence from southwestern Norway and the
northern North Sea region. Quaternary Science Reviews, 19,
667–685.
Thorleifson, L. H., Wyatt, P. H., Shilts, W. W., and Nielsen, E.,
1992. Hudson Bay lowland Quaternary stratigraphy: evidence
for early Wisconsinan glaciation centered in Quebec. In Clark,
P. U., and Lea, P. D. (eds.), The Last Interglacial-Glacial Transi-
tion in North America. Geological Society: Boulder. Geological
Society of America Special Paper, 270, pp. 207–221.
Tomiak, P. J., Penkman, K. E. H., Hendy, E. J., Demarchi, B.,
Murrells, S., Davis, S. A., McCullagh, P., and Collins, M. J.,
2013. Testing the limitations of artificial protein degradation
kinetics using known-age massive Porites coral skeletons.
Quaternary Geochronology, 16, 87–109.
Torres, T., Llamas, J. F., Canoira, L., Coello, F. J., Garcia-Alonso,
P., and Ortiz, J. E., 2000. Aminostratigraphy of two Pleistocene
marine sequences from the Mediterranean coast of Spain:
Cabo de Huertas (Alicante) and Garrucha (Almeria). In
Goodfriend, G. A., Collins, M. J., Fogel, M. L., Macko, S. A.,
and Wehmiller, J. F. (eds.), Perspectives in Amino Acid and Pro-
tein Geochemistry. Oxford/New York: Oxford University Press,
pp. 263–278.
Torres, T., Ortiz, J. E., and Aribas, I., 2013. Variations in racemiza-
tion/epimerization ratios and amino acid content of Glycymeris
shells in raised marine deposits in the Mediterranean. Quater-
nary Geochronology, 16, 35–49.
Toscano, M. A., and York, L. L., 1992. Quaternary stratigraphy and
sea-level history of the US Middle Atlantic Coastal Plain.
Quaternary Science Reviews, 11, 301–328.
Wehmiller, J. F., 1992. Aminostratigraphy of southern California
Quaternary marine terraces. In Fletcher, C. H., III, and
Wehmiller, J. F. (eds.), Quaternary Coasts of the United States:
Marine and Lacustrine Systems. Tulsa, OK: Society of Eco-
nomic Paleontologists and Mineralogists. Special publication
number 48, pp. 317–321.
Wehmiller, J. F., 2013. United States Quaternary coastal sequences
and molluscan racemization geochronology – what have they
meant for each other over the past 45 years? Quaternary Geo-
chronology, 16, 3–20.
Wehmiller, J. F., Hare, P. E., and Kujala, G. A., 1976. Amino acids
in fossil corals; racemization (epimerization) reactions and their
implications for diagenetic models and geochronological stud-
ies. Geochimica et Cosmochimica Acta, 40, 763–776.
 AMINO ACID RACEMIZATION, EOLIANITES
Wehmiller, J. F., Lajoie, K. R., Kvenvolden, K. A., Peterson, E. A.,
Belknap, D. F., Kennedy, G. L., Addicott, W. O., Vedder, J. G.,
and Wright, R. W., 1977. Correlation and chronology of Pacific
coast marine terraces of continental United States by amino acid
stereochemistry: technique evaluation, relative ages, kinetic
model ages, and geologic implications. U. S. Geological Survey
Open-File Report 77–680, 196 p. Available at http://dspace.
udel.edu:8080/dspace/handle/19716/10498
Wehmiller, J. F., York, L. L., and Bart, M. L., 1995. Amino acid
racemization geochronology of reworked Quaternary mollusks
on U.S. Atlantic coast beaches; implications for chronostra-
tigraphy, taphonomy, and coastal sediment transport. Marine
Geology, 124, 303–337.
Wehmiller, J. F., Thieler, E. R., Miller, D., Pellerito, V., Bakeman
Keeney, V., Riggs, S. R., Culver, S., Mallinson, D., Farrell,
K. M., York, L. L., Pierson, J., and Parham, P. R., 2010.
Aminostratigraphy of surface and subsurface Quaternary sedi-
ments, North Carolina Coastal Plain, USA. Quaternary
Geochronology, 5, 459–492.
Wehmiller, J. F., Harris, W. B., Boutin, B. S., and Farrell, K. M.,
2012. Calibration of amino acid racemization (AAR) kinetics
in United States mid-Atlantic coastal plain Quaternary mollusks
using 87Sr/86Sr analyses: evaluation of kinetic models and esti-
mation of regional late Pleistocene temperature history. Quater-
nary Geochronology, 7, 21–36.
Woodroffe, C. D., Kennedy, D. M., Brooke, B. P., and Dickson,
M. E., 2006. Geomorphological evolution of Lord Howe Island
and carbonate production at the latitudinal limit to reef growth.
Journal of Coastal Research, 22, 188–201.
Zazo, C., Goy, J. L., Hillaire-Marcel, C., Gillot, P., Soler, V.,
Gonzalez, J. A., Dabrio, C. J., and Ghaleb, B., 2002. Raised
marine sequences of Lanzarote and Fuerteventura revisited;
a reappraisal of relative sea-level changes and vertical move-
ments in the eastern Canary Islands during the Quaternary.
Quaternary Science Reviews, 21, 2019–2046.
Cross-references
Amino Acid
Amino Acid Racemization Dating
Amino Acid Racemization, Biostratigraphy
Amino Acid Racemization, Eolianites
Amino Acid Racemization, Fluvial and Lacustrine Sediments
(AAR)
Amino Acid Racemization, Marine Sediments
Amino Acid Racemization, Paleoclimate
AMINO ACID RACEMIZATION, EOLIANITES
Colin V. Murray-Wallace
School of Earth and Environmental Sciences, University
of Wollongong, Wollongong, NSW, Australia
Definition
Occurring along many of the world’s coastlines, eolianites
are geographically extensive, wind-blown accumulations
of skeletal carbonate sands derived from the physical
breakdown of marine invertebrates from adjacent conti-
nental shelves. Eolianite successions have been of interest
as archives of long-term environmental change and their
35
potential for reconstructing Quaternary relative sea-level
histories. Amino acid racemization (AAR) has been
applied to the dating of eolianite successions (“whole-
rock” dating) and has involved assessments of relative
age, stratigraphical correlation, and, in more ideal situa-
tions, the derivation of numerical ages for eolianite units.
Fossil molluscs from interstratified shelly facies have also
been dated using AAR to constrain the timing of eolianite
deposition.
Introduction
Eolianites are stacked, sheet-like deposits of skeletal car-
bonate sand that crop out along many of the world’s coast-
lines (Figure 1). The bioclastic sediment comprising
eolianite is generated in highly productive inner continen-
tal shelf environments, and the sediment is entrained land-
wards by waves, shallow water currents, and eolian
processes. Eolianite sediment typically comprises sand-
sized (0.063–2 mm) fragments of skeletal carbonate
derived from the physical breakdown of marine inverte-
brates such as molluscs, echinoids, bryozoans, corals,
and coralline algae. Foraminifers and sponge spicules are
also commonly present. The term eolianite was introduced
by Sayles (1931) in his report on the carbonate dune suc-
cessions of Bermuda. Non-carbonate minerals may
include quartz and heavy minerals. Eolianites have also
been described using the lithological term calcarenite to
avoid the genetic definition, as many facies within these
successions were deposited by subaqueous processes.
Although by definition, calcarenites contain  50 % cal-
cium carbonate as framework grains and cement, some
eolianites have been noted to have as little as 10 % by
mass of calcium carbonate (Murray-Wallace et al.,
2001). Eolianite successions represent shore-parallel,
composite ancient dune forms which flank many of the
world’s coastlines. They have been described from coast-
lines in tropical and temperate regions, and particularly
well-described successions occur in the Bahamas, Ber-
muda, Hawaii, western India, Mediterranean, South
Africa, and southern Australia (McKee and Ward, 1983;
Fairbridge, 1995; Vacher and Quinn, 1997; Brooke,
2001; Bateman et al., 2004; Hearty and O’Leary, 2008).
Australian eolianite successions extend from Shark Bay,
Western Australia, to western Victoria and represent the
world’s largest eolianite province.
The origin of eolianites has been of global controversy
in view of conflicting observations of the field relation-
ships of these deposits (Fairbridge, 1995). Along some
coastlines eolianites have clearly formed at times of
interglacial high sea levels analogous to the present
Holocene interglacial such as the Coorong Coastal Plain
in southern Australia (Sprigg, 1952; Murray-Wallace
et al., 1998). This has been confirmed by dating using
a variety of geochronological methods, the results of
which correlate with sea-level highstands of the past
300,000 years for many of the world’s coastlines
(Vacher and Quinn, 1997). On the Coorong Coastal Plain
36 AMINO ACID RACEMIZATION, EOLIANITES

Amino Acid Racemization, Eolianites, Figure 1 Eolianite of Robe Range near the town of Robe, in southern South Australia.
Robe Range is a composite barrier structure comprising skeletal carbonate sand units of the late Pleistocene warm interstadials MIS
5c and MIS 5a and capped by unconsolidated but vegetated Holocene dune sand. At this site the tabular cross-bedded sands are
separated by weakly developed paleosol. The uppermost unit is capped by a pervasive and strongly indurated calcrete. A person
stands as scale at the crest of the structure. The cliff is approximately 20 m high.

in southern Australia, the eolianite record extends back
in time over 1 Ma (Murray-Wallace et al., 2001). Other
coastlines, however, indicate eolianite formation at times
of lower sea level during glacials sensu lato in addition to
deposition during interglacials such as parts of the
Hawaiian Islands (Stearns, 1985) and the Swan Coastal
Plain, near Perth in Western Australia (Price
et al., 2001). During such intervals, global ice volumes
were larger and sea levels were lower than present. In
these situations, eolianites formed in response to stronger
onshore winds carrying sand-sized skeletal carbonate
fragments landwards from the partially, or completely,
subaerially exposed continental shelves. The calcium
carbonate was formed at times of higher sea level, and
the subsequent landward migration of the carbonate sed-
iment occurred by eolian processes following a fall in sea
level (forced regression) or wave, shallow water currents,
and eolian activity accompanying a relative rise in sea
level (transgressive-related sedimentation). The outer
surfaces of the dune accumulations were then consoli-
dated by vegetation cover and soil development. Further
evidence for the glacial age (sensu lato) of some eolianite
successions is that the basal units extend below present
sea level implying a lower sea level at the time of their
formation and scuba diving has identified erosional
shoreline notches within dune limestones that also
formed at times of lower sea level (Stearns, 1985).
Numerous sedimentary structures are preserved within
eolianites from which inferences about relative sea level
may be derived (McKee and Ward, 1983). Eolian trough
cross-bedding and other sedimentary structures (e.g.,
topset, foreset, bottomset, grain avalanche structures, trace
fossils such as animal footprints) represent relational
sea-level indicators, having clearly formed above tidal
datum (Figure 1). In settings where active eolian dunes
have migrated into standing bodies of water such as
coastal lagoons, however, it may be possible to infer
relative sea level, albeit crudely, based on the presence
of distinctive fossils and interfingering relationships with
lagoonal muds.
Eolianite “whole-rock” dating
Despite advances in geochronological methods, it remains
a difficult task to establish the age of many Quaternary
sedimentary successions and landforms. In many field
contexts, requisite mineral phases may be absent (e.g., fos-
sil corals for U-series dating), or the relevant assumptions
of geochronological methods are not upheld. AAR has
been applied in the dating of eolianites, particularly in
contexts where quartz sand is not available for lumines-
cence dating or where entire fossils and other mineralogi-
cal constituent are absent that would otherwise have been
suitable for dating by other methods (Milnes et al., 1987;
Hearty et al., 1992; Kimber et al., 1994; Kindler and
Hearty, 1997; Hearty, 1998; Murray-Wallace et al., 2001;
Brooke et al., 2003; Murray-Wallace et al., 2010). AAR
has been applied to dating eolianite successions as the
 AMINO ACID RACEMIZATION, EOLIANITES
method is applicable to a substantial portion of the Quater-
nary timescale. Technological advances in the detection
limits of amino acid residues in fossils (e.g., reverse-
phase, high-performance liquid chromatography) have
now made it possible to analyze individual microfossil
components from eolianites (e.g., foraminifers such as
Elphidium sp.).
As with all AAR-based chronologies, the age range of
the whole-rock method is a function of diagenetic temper-
ature. Lower latitude sites, characterized by higher current
mean annual temperatures, will have experienced higher
diagenetic temperatures. Accordingly, the age range of
the technique is spatially variable and will be less in lower
latitudes than for higher-latitude regions with lower diage-
netic temperature histories. Based on a regional study of
isoleucine epimerization of eolianite in the Bahamas
(current mean annual temperature 23–25  C), Hearty
and Kaufman (2000) concluded that the method is appli-
cable to the past 450–500 ka (i.e., back to about MIS
13). Whole-rock analyses from the Coorong Coastal Plain
in southern Australia (CMAT 14.7  C) have yielded a leu-
cine D/L value of 0.843 for eolianite from the East
Naracoorte Range, which based on magnetostratigraphy
is regarded as older than 780 ka, and therefore indicating
that in this region the practical limit of the AAR method
is approximately 1 Ma (Murray-Wallace et al., 2001).
Whole-rock dating is based on the notion that sedimen-
tary particles, if sufficiently well preserved, may be
directly dated by AAR and reliably indicates the age of
the sedimentary deposit. This is potentially the most com-
plicated application of AAR as considerable background
information is required for each field site. Similarly, the
results from one field area cannot be assumed to directly
apply to another, even within a restricted geographical
area. For example, the residence time of sedimentary par-
ticles (i.e., time lag between skeletal carbonate formation
and their final incorporation within a geological deposit)
needs to be quantified for each sedimentary environment.
This commonly involves the analyses of modern beach
sand, sampled from beach surfaces (e.g., upper 2 cm of
sand). Quantification of the residence time may be
achieved through calibration by an independent dating
method such as radiocarbon dating on a subset of frame-
work grains, but the value determined remains unique to
that environment and local field setting.
Dating eolianites: analytical methods
Coarse-grained skeletal carbonate sand (bioclasts) from
eolianite is most suited to the whole-rock method
(Hearty, 1998; Murray-Wallace et al., 2001) as the
pretreatment strategies are easier with larger bioclasts
(e.g., HCl digestion of outer carbonate grain surfaces).
Typically, such sedimentary particles are represented by
comminuted molluscs, foraminifers, bryozoans, echi-
noids, and corals.
An important requirement for whole-rock dating is that
the sediment has been rapidly buried and not subjected to
37
prolonged intervals of subaerial exposure, thus signifi-
cantly reducing uncertainties about racemization rates
and diagenetic temperatures. Shallowly buried sediment
(< 1 m) will be characterized by disproportionately high
D/L values, a function of the exponential effect of higher,
summer temperatures on diagenetic racemization.
Accordingly, rapid, deep burial largely removes sediment
from the influences of weathering, pedogenic processes,
erosion, and subaerial exposure, ensuring that the sedi-
ment sampled has remained buried since deposition and
is more likely to be better preserved. Collection of
whole-rock samples from depths greater than 1 m will
ensure that the diagenetic temperature history reflects tem-
perature changes associated with long-term climate
changes, rather than seasonal temperature variations. The
extent of racemization in whole-rock samples will accord-
ingly represent the integrated kinetic effect of racemiza-
tion in each of the fossil fragments. Commonly, the most
reliable approach for sediment sampling is to auger into
the face of an outcrop (natural rock face or quarry expo-
sure) ensuring that the location of the sediment sample is
at least 1 m below the natural ground surface. It is also crit-
ical from the perspective of diagenetic temperature histo-
ries to note the relation of the sediment sample to any
subaerial exposure surfaces such as calcrete horizons
within the sedimentary succession. Settings subjected to
regular alternate wetting and drying should be avoided
such as outcrops within the intertidal zone as they are
likely to have been subjected to prolonged leaching and
diffusive loss of amino acids, particularly in terms of
amino acids bound within lower molecular weight
peptides and free amino acids. Such contexts also com-
monly show the effects of reprecipitation of calcium
carbonate.
Eolianite sample preparation initially involves the
physical disaggregation of sediment to remove any car-
bonate cement adhering to sediment grain surfaces.
Removal of cement involves cleaning in distilled water
in an ultrasonic bath followed by an H2O2 rinse to remove
organic compounds and followed by a dilute acid (HCl)
etch to remove any additional cement and the outer por-
tion of sediment grains. The mineralogical integrity of
the framework grains can be examined in part through
microscopy and X-ray diffraction analysis of an aliquot
of the whole-rock sample. Commonly, a subset of sedi-
ment grains for AAR analysis are selected through sieving
sediment (e.g., particle sizes 250–850 mm; Hearty, 1998;
63–500 mm; Murray-Wallace et al., 2001). Analyses have
commonly been undertaken on the total hydrolyzable
amino acids of the intercrystalline fraction, and it is advis-
able to analyze free amino acids from the same sample
before undertaking acid hydrolysis, as an additional
means of evaluating integrity of results. Given the com-
plexities of whole-rock dating, analyses of the
intracrystalline fraction would also be advisable following
the analytical protocol set out in Penckman et al. (2008).
More recently the analysis of single foraminifers such as
the genus Elphidium sp. has greatly enhanced the dating
38 AMINO ACID RACEMIZATION, EOLIANITES
of eolianite (Lachlan, 2012) using reverse-phase, high-
performance liquid chromatography (RP-HPLC)
(Kaufman and Manley, 1998).
Complexities in eolianite dating
The dating of eolianite is most effective in field settings in
which cyclic sedimentation has produced the successions,
such as successive interglacials. In these contexts,
sufficient time has elapsed between the deposition of dif-
ferent sedimentary units, for them to be statistically
resolved based on extent of AAR. A related advantage is
that in the most general sense, sedimentary successions
of different interglacials may have experienced similar
changes in diagenetic temperatures over multiple glacial
cycles.
AAR whole-rock sample preparation should also
involve the physical removal of strongly abraded and
etched grains from the sample population of grains before
HCl digestion. The reworking of older sedimentary parti-
cles into younger sediments potentially complicates the
application of AAR to whole-rock dating, particularly if
the proportion of reworked framework grains is not quan-
tified. Older grains will be characterized by higher D/L
values provided that the grains have not lost amino acids
through in situ leaching (diffusive loss of free amino acids
and amino acids bound in lower molecular weight pep-
tides). The magnitude of reworking of older skeletal car-
bonate grains into modern sediment will reflect the
sediment dynamics of individual study sites and will be
a function of local carbonate bioproductivity in the forma-
tion of modern carbonate grains and the rate of input of
older carbonate grains from the erosion and reworking of
Holocene or Pleistocene sedimentary units. In southern
Australia, modern beach sediments have yielded high
D/L values indicating a significant input of relict carbon-
ate grains from locally outcropping Pleistocene eolianite
successions (Murray-Wallace et al., 2001, 2010). At Par-
sons Beach on Fleurieu Peninsula, South Australia, for
example, modern beach sand collected from the upper-
most 2 cm of a beach face yielded D/L values of 0.253
for glutamic acid and 0.189 for valine (total hydrolyzable
amino acids measured by HPLC) compared with 0.614
and 0.545, respectively, from a last interglacial eolianite
succession at Knight Beach on Fleurieu Peninsula
(Murray-Wallace et al., 2010). Modern marine molluscs
(alive at the time of collection or recently dead) consis-
tently yield low D/L values in the range of 0.02–0.03 for
valine and glutamic acid, respectively, further highlighting
the magnitude of sediment reworking in the Parsons
Beach example. At Long Beach, Guichen Bay near Robe
in southern Australia, modern beach sediment yielded
a D/L value of 0.225 for leucine (total hydrolyzable amino
acids by gas chromatography) and 0.679 for free amino
acids (Murray-Wallace et al., 2001). A last interglacial
eolianite from a coastal barrier termed Woakwine
I (132  6 ka) yielded a mean leucine D/L value of
0.451  0.099.
Aminostratigraphy
AAR has most commonly been applied to the dating of
eolianites in the context of aminostratigraphy (Miller
and Hare, 1980). Aminostratigraphy involves the classifi-
cation of strata into chronostratigraphical units or
aminozones that correspond with intervals of time, termed
geochronological units (Salvador, 1994). Litho-, bio-, and
morphostratigraphical evidence is generally used to sup-
port these chronostratigraphically defined units, as well
as calibration by other geochronological methods. The
fundamental unit of aminostratigraphy, the aminozone,
refers to a single time-stratigraphical unit representing a
well-defined interval of time (e.g., single stages of the
marine oxygen isotope record) and is delineated by the
clustering of amino acid enantiomeric values on replicate
fossils from a single, mappable lithostratigraphical unit.
According to Miller and Hare (1980), this approach
involves the least ambiguous application of AAR data,
as it largely overcomes the uncertainties associated with
racemization kinetic models and diagenetic temperature
histories. Traditional practice has seen the application of
this tool in restricted geographical settings to overcome
regional uncertainties in diagenetic temperature histories.
By far the majority of applications of AAR have
involved stratigraphical correlation and assessments of
relative age. The earliest attempts at assessing age equiva-
lence of disjunct sedimentary successions tended only to
be undertaken within restricted geographical regions, so
that the assumption was upheld that different sedimentary
deposits of presumed equivalent age had experienced
similar diagenetic temperatures.
Selected case studies
One of the first examples of AAR “whole-rock” dating
was reported on the extent of racemization of aspartic acid,
alanine, leucine, and valine (total hydrolyzable amino
acids) in carbonate sediment from a 3 m profile at
Piednippie Hall in South Australia (Milnes et al., 1987).
Their results revealed a monotonic increase in the extent
of racemization down profile with the most promising
results for the fast racemizing acid, alanine. Since this
early reconnaissance study, whole-rock dating of eolianite
has represented a valuable application of AAR in studying
Quaternary coastal carbonate successions (Hearty, 1998;
Murray-Wallace et al., 2001). In a comprehensive study
of the geology of Eleuthera Island, Bahamas, Hearty
(1998) used the degree of isoleucine epimerization in skel-
etal carbonate sands as a method for discriminating
eolianites of contrasting age. Sediments from eolian
dune, beach, and wash-over facies revealed a generally
high level of concordance for replicate samples,
suggesting that the “whole-rock” method may yield reli-
able geochronological information for Quaternary
morphostratigraphical studies. For example, twelve ana-
lyses on a minor paleosol developed on beach and eolian
sediments yielded a mean D/L value for isoleucine of
0.379  0.027 (Hearty, 1998). The coefficient of variation
 AMINO ACID RACEMIZATION, EOLIANITES
for these data of 7.12 % compares favorably with results
typically obtained on fossil molluscs from a single deposit
(Murray-Wallace, 1995). Thus, although the age of
lithoskels comprising skeletal carbonate sands may be
potentially of contrasting ages, the overall averaging
effect in well-sorted sand may yield a high level of repro-
ducibility in replicate analyses. Using the whole-rock
method, Hearty (1998) was able to discriminate intergla-
cial deposits on Eleuthera Island correlating with MIS
1, 5a, 5c, 7, 9/11, and  13.
The whole-rock method has also been applied to
a succession of eolianite coastal barriers of the Coorong
Coastal Plain in southern Australia (Murray-Wallace
et al., 2001). Up to thirteen well-defined barriers occur
across the 90 km coastal plain associated with deposition
during interglacial sea-level highstands since the onset of
the middle Pleistocene (Murray-Wallace et al., 2001).
The ages of the barriers had previously been determined
on the basis of luminescence dating (Huntley
et al., 1993, 1994). Samples for whole-rock dating were
obtained from the same stratigraphical units studied by
Huntley et al. (1993, 1994). The extent of leucine racemi-
zation (total hydrolyzable amino acids and free amino
acids) in the Pleistocene skeletal carbonate sand
(63
500 mm) increases monotonically with age and is
consistently higher than for entire fossil molluscs from
the same stratigraphical units, reflecting the lengthy resi-
dence time for bioclasts within the study area (Murray-
Wallace et al., 2001). The extent of racemization in the
whole-rock samples conforms to a model of apparent par-
abolic kinetics, and the numerical ages determined on the
basis of degree of leucine racemization were generally in
accord with the luminescence ages reported by Huntley
et al. (1993, 1994). The oldest amino acid racemization-
derived age of 935  178 ka was for eolianite from the
East Naracoorte Range, which is known to predate the
Brunhes-Matuyama boundary (>780 ka).
Summary and conclusions
Amino acid racemization has been applied in establishing
geochronological frameworks for eolianite formation for
a range of Quaternary coastal successions from around
the world. The principal advantages of applying AAR to
the dating of eolianites include the ability to derive relative
ages significantly beyond the range of radiocarbon
methods (>50 ka) and, for some coastal regions, signifi-
cantly beyond the range of U/Th or traditional methods
of luminescence dating and accordingly resolve the ages
of several interglacials. With refinements in chromato-
graphic detection of amino acids such as reverse-phase,
high-performance liquid chromatography, it is now possi-
ble to analyze individual carbonate grains or foraminifers
and examine the dynamics of sedimentary environments
(e.g., quantify the magnitude of reworking of sedimentary
particles). The analysis of eolianites by AAR is a far from
trivial task and the “black box” approach to analysis
should be avoided.
39
Bibliography
Bateman, M. D., Holmes, P. J., Carr, A. S., Horton, B. P., and
Jaiswal, M. K., 2004. Aeolianite and barrier dune construction
spanning the last two glacial – interglacial cycles from the south-
ern Cape, South Africa. Quaternary Science Reviews, 23,
1681–1698.
Brooke, B. P., 2001. The distribution of carbonate eolianite. Earth-
Science Reviews, 55, 135–164.
Brooke, B. P., Murray-Wallace, C. V., Woodroffe, C. D., and
Heijnis, H. J., 2003. Quaternary aminostratigraphy of eolianite
on Lord Howe Island, southwestern Pacific Ocean. Quaternary
Science Reviews, 22, 387–406.
Fairbridge, R. W., 1995. Eolianite and eustasy: early concepts on
Darwin’s voyage of HMS Beagle. Carbonates and Evaporites,
10, 92–101.
Hearty, P. J., 1998. The geology of Eleuthera Island, Bahamas:
a rosetta stone of Quaternary stratigraphy and sea-level history.
Quaternary Science Reviews, 17, 333–355.
Hearty, P. J., and Kaufman, D. S., 2000. Whole-rock aminostra-
tigraphy and Quaternary sea-level history of the Bahamas.
Quaternary Research, 54, 163–173.
Hearty, P. J., and O’Leary, M. J., 2008. Carbonate eolianites, quartz
sands, and Quaternary sea-level cycles, Western Australia:
a chronostratigraphic approach. Quaternary Geochronology, 3,
26–55.
Hearty, P. J., Vacher, H. L., and Mitterer, R. M., 1992. Aminostra-
tigraphy and ages of the Pleistocene limestones of Bermuda.
Geological Society of America Bulletin, 104, 471–480.
Huntley, D. J., Hutton, J. T., and Prescott, J. R., 1993. The stranded
beach-dune sequence of south-east South Australia: a test of
thermoluminescence dating, 0–800 ka. Quaternary Science
Reviews, 12, 1–20.
Huntley, D. J., Hutton, J. T., and Prescott, J. R., 1994. Further
thermoluminescence dates from the dune sequence in
southeast of South Australia. Quaternary Science Reviews, 13,
201–207.
Kaufman, D. S., and Manley, W. F., 1998. A new procedure for
determining enantiomeric (D/L) amino acid ratios in fossils
using reverse phase liquid chromatography. Quaternary Science
Reviews, 17, 987–1000.
Kimber, R. W. L., Kennedy, N. M., and Milnes, A. R., 1994. Amino
acid racemization dating of a 140 000 year old tephra-loess-
palaeosol sequence on the Mamaku Plateau near Rotorua, New
Zealand. Australian Journal of Earth Sciences, 41, 19–26.
Kindler, P., and Hearty, P. J., 1997. Geology of the Bahamas: archi-
tecture of Bahamian Islands. In Vacher, H. L., and Quinn,
T. (eds.), Geology and Hydrogeology of Carbonate Islands.
Amsterdam: Elsevier. Developments in Sedimentology,
Vol. 54, pp. 141–160.
Lachlan, T., 2012. Aminostratigraphy and Luminescence Dating of
the Pleistocene Bridgewater Formation, Kangaroo Island,
South Australia: An Archive of Long Term Climate and
Sea-Level Change. PhD thesis, University of Wollongong.
McKee, E. D., and Ward, W. C., 1983. Eolian environment. In
Scholle, P. A., Bebout, D. G., and Moore, C. H. (eds.), Carbon-
ate Depositional Environments. Oklahoma: Tulsa. American
Association of Petroleum Geologists Memoir, Vol. 33,
pp. 131–170.
Miller, G. H., and Hare, P. E., 1980. Amino acid geochronology:
integrity of the carbonate matrix and potential of molluscan fos-
sils. In Hare, P. E., Hoering, T. C., and King, K. (eds.), Biogeo-
chemistry of Amino Acids. New York: Wiley, pp. 415–444.
Milnes, A. R., Kimber, R. W. L., and Phillips, S. E., 1987. Studies in
calcareous Aeolian landscapes of southern Australia. In
Tungsheng, L. (ed.), Aspects of Loess Research. Beijing: China
Ocean Press, pp. 130–139.
40 AMINO ACID RACEMIZATION, FLUVIAL AND LACUSTRINE SEDIMENTS (AAR)
Murray-Wallace, C. V., 1995. Aminostratigraphy of Quaternary
coastal sequences in southern Australia – an overview. Quater-
nary International, 26, 69–86.
Murray-Wallace, C. V., Belperio, A. P., and Cann, J. H., 1998. Qua-
ternary neotectonism and intra-plate volcanism: the Coorong to
Mount Gambier Coastal Plain, Southeastern Australia: A review.
In Stewart, I. S., and Vita-Finzi, (eds.), Coastal Tectonics. Lon-
don: Geological Society. Special publication, 146, pp. 255–267.
Murray-Wallace, C. V., Brooke, B. P., Cann, J. H., Belperio, A. P.,
and Bourman, R. P., 2001. Whole-rock aminostratigraphy of
the Coorong Coastal Plain, South Australia: towards
a 1 million year record of sea-level highstands. Journal of the
Geological Society (London), 158, 111–124.
Murray-Wallace, C. V., Bourman, R. P., Prescott, J. R., Williams, F.,
Price, D. M., and Belperio, A. P., 2010. Aminostratigraphy and
thermoluminescence dating of coastal aeolianites and the later
Quaternary history of a failed delta: the River Murray mouth
region, South Australia. Quaternary Geochronology, 5, 28–49.
Penckman, K. E. H., Kaufman, D. S., Maddy, D., and Collins, M. J.,
2008. Closed-system behaviour of the intra-crystalline fraction
of amino acids in mollusc shells. Quaternary Geochronology,
3, 2–25.
Price, D. M., Brooke, B. P., and Woodroffe, C. D., 2001. Thermolu-
minescence dating of aeolianites from Lord Howe Island and
South-West Western Australia. Quaternary Science Reviews,
20, 841–846.
Salvador, A. S., 1994. International Stratigraphic Guide, 2nd edn.
Boulder: Geological Society of America, 214 pp.
Sayles, R. W., 1931. Bermuda during the ice age. Proceedings of the
American Academy of Arts and Sciences, 66, 381–467.
Sprigg, R. C., 1952. The Geology of the South-East Province, South
Australia, with Special Reference to Quaternary Coast-Line
Migrations and Modern Beach Developments. Adelaide:
K.M. Stevenson Government Printer. Geological Survey of
South Australia, Bulletin, Vol. 29, 120 pp.
Stearns, H. T., 1985. Geology of the State of Hawaii, 2nd edn. Palo
Alto: Pacific Books, 325 pp.
Vacher, H. L., and Quinn, T. M. (eds.), 1997. Geology and Hydroge-
ology of Carbonate Islands. Amsterdam: Elsevier. Develop-
ments in Sedimentology, Vol. 54, 948 pp.
Cross-references
Amino Acid
Amino Acid Racemization, Coastal Sediments
Amino Acid Racemization, Marine Sediments
Amino Acid Racemization, Paleoclimate
AMINO ACID RACEMIZATION, FLUVIAL AND
LACUSTRINE SEDIMENTS (AAR)
John F. Wehmiller
Department of Geological Sciences, University of
Delaware, Newark, DE, USA
Synonyms
Climate history; Fossils; Geochronology; Lakes; Rivers
Definition
Amino acid racemization. The phenomenon of conversion
of “left-handed” (L or “levo”) amino acids to their
“right-handed” (D or “dextro”) form. In most living sys-
tems, 100 % of the amino acids are of the L form, and
the conversion results in an equal mixture of D and
L when the racemization reaction is complete.
Fluvial sediments, lacustrine sediments. Sedimentary
material that accumulates and is preserved in river,
stream, or lake environments. These can include sedi-
ments dominated by biological material (shells) and
those dominated by inorganic material that may form
by cementation or evaporation, such as carbonates or
evaporates and mud or clay. Sediments range in age
(time since deposition) from days to millions of years
and are found in a variety of terrestrial sedimentary
environments.
Introduction
Amino acid racemization (AAR) studies of fluvial and
lacustrine deposits usually require the analysis of
a biogenic mineral (e.g., bivalve, gastropod, ostracode,
or egg shell) or organic material associated with these
environments. These terrestrial deposits include river ter-
race and delta sediments, glacial and pluvial lake sedi-
ments, and eolian sediments derived from combined
glacial and fluvial sources. In some cases fluvial deposits
merge with coastal deposits in areas influenced by sea
level change, providing an opportunity to link global ice
volume (sea level) history with local fluvial stratigraphy.
Archaeological studies that include AAR often involve
samples associated with one or more of these terrestrial
depositional processes (Johnson and Miller, 1997). In
many situations the geochronological study of these envi-
ronments provides an opportunity for comparison of mul-
tiple dating methods (Oches et al., 1996; Owen et al.,
2007; Bright et al., 2010). Additionally, when suitably cal-
ibrated, AAR results yield insights into local or regional
temperature histories for the late Pleistocene (Goodfriend
and Mitterer, 1993; Miller et al., 1997). Examples of
AAR applications in each of these depositional systems
are reviewed in the following sections.
Shallow (<2 m) ground temperatures can vary signifi-
cantly because of exposure and moisture conditions
(Wehmiller et al., 2000). Consequently, the interpretation
of AAR data from many fluvial, eolian, colluvial, or
archaeological deposits must account for possible differ-
ences in burial temperature histories. This issue is particu-
larly important for any sample that has experienced a large
portion of its history either exposed at the land surface,
heated by fire, or buried just a few cm below land surface.
Lacustrine and marine samples are less vulnerable to this
effect because their initial burial environment is usually
under water. Many AAR studies demonstrate the effect
of shallow burial heating (Wehmiller, 1977; Goodfriend,
1987; Miller and Brigham-Grette, 1989; Ellis et al.,
1996; Wehmiller et al., 2000; Owen et al., 2007), and
many others indicate the potential thermal effects on rela-
tive age interpretations (Miller and Brigham-Grette, 1989;
Johnson and Miller, 1997; Miller et al., 1997).
 AMINO ACID RACEMIZATION, FLUVIAL AND LACUSTRINE SEDIMENTS (AAR)
Fluvial and lacustrine deposits, Western
United States
During Pleistocene glacial intervals, a series of large lakes
existed in the Western United States where present-day
precipitation is quite low (Reheis et al., 2012 and refer-
ences therein). The modern remnants of these large lakes
include Great Salt Lake (Utah), and Pyramid and Walker
Lakes (Nevada). The hydrologic balance of these
lake regions was vastly different from the present, involv-
ing a major increase in the ratio of precipitation to evapo-
ration during glacial times. AAR is one of the many
chronological tools applied to the development of
a regional chronology for lake level fluctuation during
the Quaternary, using either deposits created in the deeper
portions of the lake (Oviatt et al., 1999) or shoreline
deposits created around the high stand lake margins
(McCoy, 1987).
Lake Bonneville is the name given to the large lake that
occupied what now is most of the state of Utah. AAR ana-
lyses of gastropods from various shallow-water deposits
of Lake Bonneville identified five aminozones, each
representing different phases of deposition during the
middle and late Pleistocene (McCoy, 1987; Oviatt et al.,
1987). Numerical ages of these deposits are estimated
using associated radiocarbon (C14) and U-Th data and
volcanic ash deposits that are found in association with
the oldest aminozone. Available geochronological data
suggest that Lake Bonneville reached peak volumes at
the end of marine isotope stages 4 and 2, roughly 60 and
12 ka, respectively (Kaufman et al., 2001), as well as at
earlier times that are less well constrained. The ages of
the shallow-water deposits constrain the times of maxi-
mum lake levels (deepest lake); samples from cores from
near the center of the Bonneville Basin provide informa-
tion on the transitional phases when lake levels may have
been at low or intermediate levels (Oviatt et al., 1999;
Kaufman, 2003a). AAR results from these cores, also
combined with volcanic ash chronology, demonstrate
a long and discontinuous record of lacustrine deposition
over the last 700 ka.
Paleoclimatologic information from calibrated
Lake Bonneville AAR data
AAR data can be used to infer effective temperatures if the
analyzed samples have been independently dated (see sec-
tion “Cross-Reference” list). AAR analyses of well-dated
ostracodes from late Pleistocene sediments in the Lake
Bonneville record imply that full glacial (MIS 2) tempera-
tures in the basin were approximately 10  C colder than
present (Kaufman, 2003b). Effective temperatures for
older units were not as cold (Kaufman, 2003b). Ostracode
AAR results from Summer Lake, a smaller pluvial lake in
Oregon, imply unrealistically cold effective temperatures
when compared with the Bonneville results, suggesting
that other factors besides temperature may influence
AAR kinetics in this sample type (Bright and Kaufman,
2011; Reichert et al., 2011).
41
Western North America river systems
The Pleistocene history of the Colorado River includes
phases when large lakes were created upstream of large
lava dams (see Kaufman et al., 2002). Deposits created
by these large lakes are quite rare, and the ages of some
potential lacustrine deposits in the Grand Canyon have
been debated. AAR and 14C results from mollusks and
ostracodes indicate that lacustrine deposits found in the
eastern Grand Canyon were formed <45 ka BP in small
travertine-dammed ponds rather than in the large deep
lakes that would have been formed in association with
the much older (>100 ka to 2 Ma) lava-dammed lakes
(Kaufman et al., 2002).
At the downstream end of the Colorado River, the riv-
er’s delta has changed dramatically during the past century
in response to upstream water diversions. Part of this
change has been documented using 14C-calibrated AAR
results for mollusks from beach ridges within the delta
system. Nearly all the analyzed shells formed prior to
1930 and since 1000 AD, indicating a decline in benthic
productivity in the delta region corresponding to the time
of diversion of water flow in the river (Kowalewski
et al., 2000).
Fluvial terraces of the Chama/Rio Grande River sys-
tem in the Espanola Basin of northern New Mexico
record the downcutting of the rivers during the middle
and late Pleistocene (Dethier and McCoy, 1993). AAR
data from gastropods found in slack-water deposits pre-
served on these terraces identify five aminozones with
increasing D/L values found at increasing elevations
above the modern river. Rapid deposition of shells and
limited surface exposure are inferred, thereby minimiz-
ing shallow thermal effects on the results. The oldest
of these zones is 620 ka based on volcanic ash chro-
nology (ash of the same age is used for calibration of
aminostratigraphy in the Lake Bonneville sequence –
see above), younger ones being estimated to represent
310, 170, 95, and <19 ka, respectively (Dethier and
McCoy, 1993).
Bright et al. (2010) used AAR data, along with
associated 14C and IRSL results, to revise the chronology
of a bison-bearing fluvial deposit in northern Mexico. Pre-
vious Ar40/Ar39 ages on associated lava flows suggested
an age of 440  130 ka for this deposit, far older than
expected for the presence of Bison in this region. The
revised interpretation of the combined geochronological
data indicates a late Pleistocene (<40 ka) age.
Fluvial and lacustrine deposits, Eastern United
States and Southeastern Canada
AAR data from this region are helpful in understanding
the geochronology of deposits formed during advances
or retreats of the Laurentide Ice Sheet, a major influence
on the climate and environment of eastern North America
during the Quaternary. Ostracodes, gastropods, and
bivalves are intermittently preserved in some of the classic
stratigraphic sections from this region, ranging from the
42 AMINO ACID RACEMIZATION, FLUVIAL AND LACUSTRINE SEDIMENTS (AAR)
Mississippi River Valley to the St. Lawrence Valley. AAR
data for gastropods from a broad range of sites, mostly in
loess, from the Mississippi River Valley (Clark et al.,
1989; Miller et al., 1993; Oches et al., 1996) define at least
four phases of eolian (loess) deposition over the past
750 ka, the most recent phase corresponding to the late
Pleistocene Wisconsinan (MIS 4-3-2) glacial advance into
the region. Elsewhere, portions of this regional aminostra-
tigraphic record are preserved at isolated sites both within
the Laurentide Ice Sheet margin (Karrow et al., 1997;
Karrow, 2004) and near the ice sheet margin in southeast-
ern Indiana (Miller et al., 1993) and in the St. Lawrence
River Valley (Occhietti et al., 1996).
Fluvial, lacustrine, and glacial deposits, United
Kingdom
An extensive and detailed record of glacial, fluvial, and
lacustrine sedimentation in the British Isles has provided
an opportunity for many geochronological studies involv-
ing AAR analysis of nonmarine (and some marine) sam-
ples (Bowen et al., 1989; Bowen, 2000; Penkman et al.,
2013). Many samples have been collected from intergla-
cial stratotypes with associated faunal, floral, or archaeo-
logical information, including many with independent
chronological data. The AAR studies of UK sites likely
represent the most intense coverage of AAR data of any
region of the world, as numerous investigations (most
within an area 3  3 ), involving many laboratories and
collections have occurred over nearly 40 years. The
advent of sensitive instrumentation and new strategies to
isolate and analyze the most robust amino acid component
now provides a regional aminostratigraphic framework for
the British Quaternary record, a complex of glacial, non-
glacial, fluvial, and marine sequences (Penkman et al.,
2011, 2013). Because of the relatively low effective tem-
peratures for the region and the slow racemization rate
during glacial intervals, the difference in D/L values for
samples from one interglacial and the next can be small,
but when these results are considered within lithostra-
tigraphic or fluvial terrace relative age sequences, mean-
ingful age estimates and correlations with the global
marine isotopic record can be established (Penkman
et al., 2011, 2013). Aminozones corresponding to intergla-
cial stages 11, 9, 7, and 5 are evident (Bowen, 2000;
Penkman et al., 2011, 2013). Aminozones representing
corresponding phases of deposition are recognized in the
coastal-marine records of the southern North Sea (Meijer
and Cleveringa, 2009).
The Nile River delta
Sediments of the eastern Nile River Delta (Egypt) contain
late Holocene marine bivalves and gastropods for which
radiocarbon-calibrated AAR results have been obtained
(Goodfriend and Stanley, 1996, 1999). As with other stud-
ies, the relative ease and inexpensive cost of AAR ana-
lyses allow for large numbers of analyses to be
conducted, far more than could be done using radiocarbon
alone. Calibrated AAR kinetics indicate the extent of
mixing of older Holocene samples into younger deposits
(Goodfriend and Stanley, 1996) and also indicate that this
portion of the Nile Delta underwent rapid strand-plain
accumulation in the ninth century A.D., likely leading to
relocation of major port cities (Goodfriend and Stanley,
1999).
Fluvial and travertine deposits, Iberian Peninsula
In Central Spain, tufa deposits are found within fluvial ter-
races in small tributaries of the Tagus River, a region of
karst terrain. AAR data for ostracodes from these terrace
deposits, combined with racemization kinetic modeling
for the region (Ortiz et al., 2004), suggest phases of tufa
deposition that correspond with most interglacial warm
phases since 400 ka (Torres et al., 2005; Ortiz et al.,
2009).
Lake Eyre Australia: lacustrine chronology and
paleoclimatology
As with the combined stratigraphic and paleoclimatic
studies of the pluvial lakes of the Western United States,
a similar record occurs in the Madigan Gulf region of Lake
Eyre, South Australia, a region influenced by monsoonal
precipitation (Magee et al., 1995). This playa lake has
been episodically flooded during the late Quaternary,
and a combination of radiocarbon, uranium-series, lumi-
nescence, and racemization data indicates that the lake
reached a maximum depth of 25 m during the early part
of MIS 5. A Holocene high lake interval did not reach this
same depth, and intervening phases of variable moisture
and deflation during dry lakes dominated the regional
history.
AAR studies of ostrich eggshell (OES) from the Lake
Eyre region yield insights into the magnitude of cooling
during the late Pleistocene (Miller et al., 1997). Combined
14
C, AAR, and luminescence data document the extent of
racemization for samples with ages from last interglacial
to the present, and the contrast in rates for samples
<20 ka compared with those between 60 and 20 ka
implies a glacial temperature reduction of approximately
8  C (Miller et al., 1997). See also the entry “Amino Acid
Racemization, Paleoclimate”.
Summary
AAR study of terrestrial (fluvial, lacustrine, eolian, or
archaeological) deposits has often been conducted in asso-
ciation with other dating methods that provide either cali-
bration for AAR or insights into other processes affecting
the physical record. Many AAR terrestrial studies focus
on issues of correlation of local records with regional ter-
restrial records or global ice volume records, in order to
understand the climatic links between local processes
and the global record. Many of these terrestrial records
demonstrate discontinuous or episodic deposition, in
 AMINO ACID RACEMIZATION, FLUVIAL AND LACUSTRINE SEDIMENTS (AAR)
phase with either interglacial or glacial stages, depending
on the physical process being dated. Volcanic ash or paleo-
magnetic geochronology establishes a calibration for the
older parts of many records. Advances in instrumental
technology have allowed large numbers of small samples
(ostracodes, land snails, etc.) to be analyzed, thereby
establishing a firm statistical basis for chronostratigraphic
interpretations. The use of calibrated AAR results to infer
late Pleistocene temperature histories is demonstrated by
examples from the Western United States, the United
Kingdom, Australia, and Jamaica.
Bibliography
Bowen, D. Q., 2000. Revised aminostratigraphy for land-sea corre-
lations from the northeastern North Atlantic margin. In
Goodfriend, G. A., Collins, M. J., Fogel, M. L., Macko, S. A.,
and Wehmiller, J. F. (eds.), Perspectives in Amino acid and Pro-
tein Geochemistry. Oxford/New York: Oxford University Press,
pp. 252–262.
Bowen, D. Q., Hughes, S., Sykes, G. A., and Miller, G. H., 1989.
Land-sea correlations in the pleistocene based on isoleucine
epimerization in non-marine molluscs. Nature, 340, 49–51.
Bright, J., and Kaufman, D. S., 2011. Amino acids in lacustrine
ostracodes, part III: effects of pH and taxonomy on racemization
and leaching. Quaternary Geochronology, 6, 574–597.
Bright, J., Kaufman, D. S., Forman, S. L., McIntosh, W. C., Mead,
J. I., and Baez, A., 2010. Comparative dating of a Bison-bearing
late Pleistocene deposit, Térapa, Sonora, Mexico. Quaternary
Geochronology, 5, 631–643.
Clark, P. U., Nelson, A. R., McCoy, W. D., Miller, B. B., and
Barnes, D. K., 1989. Quaternary aminostratigraphy of Missis-
sippi Valley loess. Bulletin of the Geological Society of America,
101, 918–926.
Dethier, D. P., and McCoy, W. D., 1993. Aminostratigraphic rela-
tions and age of Quaternary deposits, northern Espanola basin,
New Mexico. Quaternary Research, 39, 222–230.
Ellis, G. L., Goodfriend, G. A., Abbott, J. T., Hare, P. E., and von Endt,
D. W., 1996. Assessment of integrity and geochronology of archae-
ological sites using amino acid racemization in land snail shells:
examples from Central Texas. Geoarchaeology, 11, 189–213.
Goodfriend, G. A., 1987. Chronostratigraphic studies of sediments
in the Negev desert, using amino acid epimerization analysis of
land snail shells. Quaternary Research, 28, 374–392.
Goodfriend, G. A., and Mitterer, R. M., 1993. A 45,000-yr record of
a tropical lowland biota: the land snail fauna from cave sedi-
ments at Coco Ree, Jamaica. Bulletin of the Geological Society
of America, 105, 18–29.
Goodfriend, G. A., and Stanley, D. J., 1996. Reworking and discon-
tinuities in Holocene sedimentation in the Nile Delta; documen-
tation from amino acid racemization and stable isotopes in
mollusk shells. Marine Geology, 129, 271–283.
Goodfriend, G. A., and Stanley, D. J., 1999. Rapid strand-plain
accretion in the northeastern Nile Delta in the 9th century A.D.
and the demise of the port of Pelusium. Geology, 27, 147–150.
Johnson, B. J., and Miller, G. H., 1997. Archaeological applications
of amino acid racemization. Archaeometry, 39, 265–287.
Karrow, P. F., 2004. Late Quaternary stratigraphic comparisons in
south-central Ontario and western New York and the OIS 5E to
early 3 interval. Northeastern Geology and Environmental Sci-
ences, 26, 202–210.
Karrow, P. F., Seymour, K. L., Miller, B. B., and Mirecki, J. E.,
1997. Pre-late Wisconsinan Pleistocene biota from southeastern
43
Michigan, USA. Palaeogeography, Palaeoclimatology,
Palaeoecology, 133, 81–101.
Kaufman, D. S., 2003a. Dating deep-lake sediments by using amino
acid racemization in fossil ostracodes. Geology, 31, 1049–1052.
Kaufman, D. S., 2003b. Amino acid paleothermometry of Quater-
nary ostracodes from the Bonneville Basin, Utah. Quaternary
Science Reviews, 22, 899–914.
Kaufman, D. S., Forman, S. L., and Bright, J., 2001. Age of the Cut-
ler Dam Alloformation (late Pleistocene), Bonneville Basin,
Utah. Quaternary Research, 56, 322–334.
Kaufman, D. S., O’Brien, G., Mead, J. I., Bright, J., and Umhoefer,
P., 2002. Late Quaternary spring-fed deposits of the Grand Can-
yon and their implication for deep lava-dammed lakes. Quater-
nary Research, 58, 329–340.
Kowalewski, M., Serrano, G., Flessa, K. W., and Goodfriend, G. A.,
2000. Dead delta’s former productivity: two trillion shells at the
mouth of the Colorado river. Geology, 28, 1059–1062.
Magee, J. W., Bowler, J. M., Miller, G. H., and Williams, D., 1995.
Stratigraphy, sedimentology, chronology and palaeohydrology
of Quaternary lacustrine deposits at Madigan Gulf, Lake Eyre,
South Australia. Palaeogeography, Palaeoclimatology,
Palaeoecology, 113, 3–42.
McCoy, W. D., 1987. Quaternary aminostratigraphy of the
Bonneville Basin, western United States. Bulletin of the Geolog-
ical Society of America, 98, 99–112.
Meijer, T., and Cleveringa, P., 2009. Aminostratigraphy of middle
and late Pleistocene deposits in the Netherlands and the southern
part of the North Sea basin. Global and Planetary Change, 68,
326–345.
Miller, G. H., and Brigham-Grette, J., 1989. Amino acid geochro-
nology: resolution and precision in carbonate fossils. Quater-
nary International, 1, 111–128.
Miller, B. B., McCoy, W. D., Wayne, W. J., and Brockman, C. S.,
1992. Ages of the Whitewater and Fairhaven tills in
southwestern Ohio and southeastern Indiana. In Clark, P. U., and
Lea, P. D. (eds.), The Last Interglacial-Glacial transition in North
America. Boulder: Geological Society of America. Geological
Society of America Special Paper, Vol. 270, pp. 89–98.
Miller, B. B., Palmer, D. F., McCoy, W. D., Smith, A. J., and
Colburn, M. L., 1993. A pre-Illinoian Pleistocene fossil assem-
blage from near Connersville, southeastern Indiana. Quaternary
Research, 40, 254–261.
Miller, G. H., Magee, J. W., and Jull, A. J. T., 1997. Low-latitude
glacial cooling in the Southern Hemisphere from amino-acid
racemization in emu eggshells. Nature, 385, 241–244.
Occhietti, S., Balescu, S., Lamothe, M., Clet, M., Cronin, T.,
Ferland, P., and Pichet, P., 1996. Late stage 5 glacio-isostatic
sea in the St. Lawrence Valley, Canada and United States.
Quaternary Research, 45, 128–137.
Oches, E. A., McCoy, W. D., and Clark, P. U., 1996. Amino acid
estimates of latitudinal temperature gradients and geochronol-
ogy of loess deposition during the last glaciation, Mississippi
Valley, United States. Geological Society of America Bulletin,
108, 892–892.
Ortiz, J. E., Torres, T., Delgado, A., Julia`, R., Llamas, F. J., Soler,
V., and Delgado, J., 2004. Numerical dating algorithms of amino
acid racemization ratios analyzed in continental ostracodes of the
Iberian Peninsula (Spain). Application to Guadix-Baza Basin
(southern Spain). Quaternary Science Reviews, 23, 717–730.
Ortiz, J. E., Torres, T., Delgado, A., Reyes, E., and Díaz-Bautista,
A., 2009. A review of the Tagus river tufa deposits (central
Spain): age and paleoenvironmental record. Quaternary Science
Reviews, 28, 947–963.
Oviatt, C. G., McCoy, W. D., and Reider, R. G., 1987. Evidence for
a shallow early or middle Wisconsin-age lake in the Bonneville
Basin, Utah. Quaternary Research, 27, 248–262.
44 AMINO ACID RACEMIZATION, MARINE SEDIMENTS
Oviatt, C. G., Thompson, R. S., Kaufman, D. S., Bright, J. E.,
and Forester, R. M., 1999. Reinterpretation of the Burmester
core, Bonneville basin, Utah. Quaternary Research, 52,
180–184.
Owen, L. A., Bright, J., Finkel, R. C., Jaiswal, M. K., Kaufman,
D. S., Mahan, S., Radtke, U., Schneider, J. S., Sharp, W.,
Singhvi, A. K., and Waren, C. N., 2007. Numerical dating of
a late Quaternary spit-shoreline complex at the northern end of
Silver Lake, Mojave Desert, California: testing the applicability
of radiocarbon, luminescence, terrestrial cosmogenic nuclide,
electron spin resonance, U-series, and amino acid racemization.
Quaternary International, 166, 87–110.
Penkman, K. E. H., Preece, R. C., Bridgland, D. R., Keen, D. H.,
Meijer, T., Parfitt, S. A., White, T. S., and Collins, M. J., 2011.
A chronological framework for the British Quaternary based
on Bithynia opercula. Nature, 476, 446–449.
Penkman, K. E. H., Preece, R. C., Bridgland, D. R., Keen, D. H.,
Meijer, T., Parfitt, S. A., White, T. S., and Collins, M. J.,
2013. An aminostratigraphy for the British Quaternary based
on Bithynia opercula. Quaternary Science Reviews, 61,
111–134.
Reheis, M. C., Bright, J., Lund, S. P., Miller, D. M., Skipp, G., and
Fleck, R. J., 2012. A half-million-year record of paleoclimate
from the Lake Manix Core, Mojave Desert, California.
Palaeogeography, Palaeoclimatology, Palaeoecology,
365–366, 11–37.
Reichert, K. L., Licciardi, L., and Kaufman, D. S., 2011. Amino
acid racemization in lacustrine ostracodes, part II:
paleothermometry in Pleistocene sediments at Summer Lake,
Oregon. Quaternary Geochronology, 6, 174–185.
Torres, T., Ortiz, J. E., Garcia de la Morena, M. A., Llamas, F. J., and
Goodfriend, G. A., 2005. Ostracode-based aminostratigraphy
and aminochronology of a tufa system in central Spain. Quater-
nary International, 135, 21–33.
Wehmiller, J. F., 1977. Amino acid studies of the Del Mar, Califor-
nia midden site: apparent rate constants, ground temperature
models, and chronological implications. Earth and Planetary
Science Letters, 37, 184–196.
Wehmiller, J. F., Stecher, H. A., III, York, L. L., and Friedman, I.,
2000. The thermal environment of fossils: effective ground tem-
peratures at aminostratigraphic sites on the U. S. Atlantic coastal
plain. In Goodfriend, G. A., Collins, M. J., Fogel, M. L., Macko,
S. A., and Wehmiller, J. F. (eds.), Perspectives in Amino Acid
and Protein Geochemistry. Oxford/New York: Oxford Univer-
sity Press, pp. 219–250.
Cross-references
Amino Acid Racemization Dating
Amino Acid Racemization, Paleoclimate
Feldspar, Infrared-Stimulated Luminescence
Glacial Landscape (Cosmogenic Nuclide)
Lacustrine Environments (14C)
Luminescence Dating
Luminescence Dating, Loess
Luminescence, Fluvial Sediments
Luminescence, Glacial Sediments
Marine Isotope Stratigraphy
Marine Varves
Radiocarbon Dating
Radiocarbon Dating of Terrestrial Carbonates
Sediments, Terrestrial (Paleomagnetism)
Tephrochronology
Terrestrial Cosmogenic Nuclide Dating
AMINO ACID RACEMIZATION, MARINE SEDIMENTS
Darrell Kaufman
School of Earth Sciences and Environmental
Sustainability, Northern Arizona University, Flagstaff,
AZ, USA
Definition
Amino acid racemization. The phenomenon of conversion
of “left-handed” (L or “levo”) amino acids to their “right-
handed” (D or “dextro”) form. In most living systems,
100 % of the amino acids are of the L form, and the con-
version results in an equal mixture of D and L when the
racemization reaction is complete.
Marine Sediment. Material that accumulates in the
marine environment. This article focuses on material col-
lected in cores from deep-sea settings.
Among the wide range of applications of amino acid geo-
chronology, this technique is especially well suited for dat-
ing deep-sea sediments using foraminifera. Foraminifera
inhabit most of the World Ocean and they contain relatively
high concentrations of amino acids that are well retained by
their carbonate test. The stable thermal environment of deep-
sea sites minimizes the often-complicating effect of variable
temperature on the long-term rate of racemization. Some of
the earliest research on amino acid geochronology took
advantage of the long-term stability of deep-sea settings to
investigate the diagenesis of amino acids over geologic time,
including the rate of racemization in foraminifera
(Wehmiller and Hare, 1971; Bada and Schroeder, 1972;
King and Hare, 1972; Kvenvolden et al., 1973; Schroeder
and Bada, 1977; Bada et al., 1978; Bada and Man, 1980;
King, 1980; Belknap and Doyle, 1986; Stathopolos
et al., 1987; Robbins and Brew, 1990). The extent of amino
acid racemization in fossil foraminifera has also been used to
estimate the ages of Quaternary marine sediment (Müller,
1984; Macko and Aksu, 1986; Sejrup and Hughen, 1992;
Knudsen and Sejrup, 1993; Murray-Wallace and Belperio,
1994; Harada et al., 1996), and in a few cases, to reconstruct
paleotemperature histories (Bada and Man, 1980; Lehman
et al., 1988; Johnson et al., 1990).
Two general approaches can be used to convert the
extent of amino acid racemization to a numeric time scale:
In the first approach, the effects of time and temperature
on the extent of racemization are determined in modern
shells subjected to high-temperature laboratory experi-
ments (e.g., Kaufman, 2006). This relation, together with
a model of racemization kinetics, is used to calculate the
age of a sample if its temperature history is known.
Because accelerating the rate of racemization under high
temperature may not precisely mimic the net effect of
long-term diagenetic processes, most attempts to quantify
the temperature sensitivity of long-term racemization
extend their analyses to ambient temperatures and geo-
logic environments by including analyses of Holocene
 AMINO ACID RACEMIZATION, MARINE SEDIMENTS 45

Amino Acid Racemization, Marine Sediments, Figure 1 Extent of racemization (D/L) for aspartic acid (Asp) in independently
dated samples of all 179 mono-specific foraminifera samples from deep-sea cores in the Pacific, Atlantic, and Arctic oceans extending
back to 1.5 Ma (Kaufman et al., 2013). Colors indicate the current site temperature grouped into four categories. In addition, the
modeled rate of racemization based on laboratory-heated and radiocarbon-dated P. obliquiloculata for three temperatures is shown
for comparison (dashed lines; Kaufman, 2006). Data are plotted on logarithmic scales to improve data presentation.

samples whose ages are known from radiocarbon dating,
and whose temperature history can be reasonably inferred
from instrumental measurements. A more secure approach
that does not require assumptions about temperature his-
tory is to calibrate the rate of racemization by analyzing
securely dated samples of a particular taxon from a region
where temperature histories are uniform (Sejrup
et al., 1984; Harada and Handa, 1995; Hearty
et al., 2004; Kaufman et al., 2008, 2013). The calibrated
reaction rate is then used to date samples of the same taxon
of unknown age from the same environment where tem-
perature histories are similar.
Figure 1 shows the extent of aspartic acid racemization
(Asp D/L) in independently dated, mono-specific forami-
nifera samples from the Pacific, Atlantic, and Arctic
oceans (Kaufman et al., 2013). These data provide the
basis for developing a calibrated age equation to apply to
undated samples (the second approach described above).
Figure 1 also shows the modeled rate of racemization
based on laboratory-heated and radiocarbon-dated forami-
nifera for three temperatures (Kaufman, 2006). The ages
of samples can be determined by analyzing their D/L values
and assuming a reasonable temperature history (the first
approach).
Most studies of amino acid geochronology based on
foraminifera have used gas chromatography or
ion-exchange HPLC to separate the D- and L-amino acids,
both of which require hundreds of foraminifera tests for a
single analysis (several milligrams of fossils).
Consequently, these studies relied on samples of mixed
foraminiferal taxa. Sample size requirements using HPLC
reverse-phase chromatography (RPC) are at least two
orders of magnitude lower (Kaufman and Manley,
1998), allowing analysis of fewer, and even single speci-
mens in the case of a taxon with a large test (Hearty
et al., 2004). Subdividing a sample into small subsamples,
each analyzed individually, can improve the accuracy of
the results because collections of microfossils commonly
include a few individuals whose D/L values fall outside
the mean of the others. These outliers might result from
post-depositional reworking, post-collection contamina-
tion, or an aberrant diagenetic pathway. Regardless of
the cause, by routinely analyzing multiple subsamples
per level (10 or more), these aberrant results can be identi-
fied objectively and excluded from the data set. The
resulting mean D/L values are more accurate than those
based on just a few subsamples.
The rate of racemization differs among foraminifera
taxa. An early investigation of mono-specific samples
(King and Neville, 1977) showed that the extent of race-
mization (epimerization of isoleucine) in different taxa
from the same level within a sediment core could differ
by as much as a factor of 4 between species. More recent
analyses based on the same approach and augmented by
laboratory heating experiments do not show a difference
of this magnitude among mono-specific samples, either
as fossils or as laboratory-heated specimens (Kaufman
et al., 2013). Differences in racemization rates might be
46 AMINO ACID RACEMIZATION, MARINE SEDIMENTS
attributed to the effect of contamination of the type
described by Stathoplos and Hare (1993) and Sejrup and
Haugen (1992) and that can be recognized by the rela-
tively high serine content (Kaufman et al., 2013).
Previous studies recognized the potential of exogenous
amino acids to contaminate foraminifera and to skew the
results (Sejrup and Hughen, 1992; Stathoplos and Hare,
1993). Many studies have investigated the procedures
for pretreating samples to isolate the most reliably indige-
nous amino acid fraction (e.g., Penkman et al., 2008), and
for applying data-screening criteria to identify and exclude
outliers based on statistical tests of internal consistency
among multiple amino acids (e.g., Kosnik and Kaufman,
2008). In addition, the concentration of serine, a liable
amino acid that is present in only low concentrations in
fossils, has been used widely as an indicator of contamina-
tion by young amino acids (e.g., Kaufman et al., 2013).
Bibliography
Bada, J. L., and Man, E. H., 1980. Amino acid diagenesis in Deep
Sea Drilling Project cores; kinetics and mechanisms of some
reactions and their applications in geochronology and in
paleotemperature and heat flow determinations. Earth-Science
Reviews, 16, 21–55.
Bada, J. L., and Schroeder, R. A., 1972. Racemization of isoleucine
in calcareous marine sediments. Earth and Planetary Science
Letters, 15, 1–11.
Bada, J. L., Shou, M. Y., Man, E. H., and Schroeder, R. A., 1978.
Decomposition of hydroxy amino acids in foraminiferal tests;
kinetics, mechanism and geochronological implications. Earth
and Planetary Science Letters, 41, 67–76.
Belknap, D. F., and Doyle, L. J., 1986. Amino acid racemization
dating in foraminifera from U.S. Atlantic and Gulf Coast slope
cores. Abstracts – SEPM Midyear Meeting 3, 8.
Harada, N., and Handa, N., 1995. Amino acid chronology in the fos-
sil planktonic foraminifers, Pulleniatina obliquiloculata from
Pacific Ocean. Geophysical Research Letters, 22, 2353–2356.
Harada, N., Handa, N., Ito, M., Oba, T., and Matsumoto, E., 1996.
Chronology of marine sediments by the racemization reaction
of aspartic acid in planktonic foraminifera. Organic Geochemis-
try, 24, 921–930.
Hearty, P. J., O’Leary, M. J., Kaufman, D. S., Page, M., and Bright,
J., 2004. Amino acid geochronology of individual foraminifer
(Pulleniatina obliquiloculata) tests, north Queensland margin,
Australia: a new approach to correlating and dating Quaternary
tropical marine sediment cores. Paleoceanography, 19,
PA4022, doi:10.1029/2004PA001059.
Johnson, B. J., Lehman, S. J., and Miller, G. H., 1990. Arrhenius
parameters for the foram species Globorotalia menardii and
deduction of bottom water temperature changes during the
Pleistocene-Holocene transition. Abstracts with Programs –
Geological Society of America, 22(7), 146.
Kaufman, D. S., 2006. Temperature sensitivity of aspartic and
glutamic acid racemization the foraminifera Pulleniatina. Qua-
ternary Geochronology, 1, 188–207.
Kaufman, D. S., and Manley, W. F., 1998. A new procedure for
determining enantiomeric (D/L) amino acid ratios in fossils
using reverse phase liquid chromatography. Quaternary Science
Reviews (Quaternary Geochronology), 17, 987–1000.
Kaufman, D. S., Polyak, L., Adler, R., Channell, J. E. T., and Xuan,
C., 2008. Dating late Quaternary planktonic foraminifer
Neogloboquadrina pachyderma from the Arctic Ocean using
amino acid racemization. Paleoceanography, 23, PA3224.
Kaufman, D. S., Cooper, K., Behl, R., Billups, K., Bright, J., Gard-
ner, K., Hearty, P., Jokobbson, M., Mendes, I., O’Leary, M.,
Polyak, L., Rasmussen, T., Rosa, F., and Schmidt, M., 2013.
Amino acid racemization in mono-specific foraminifera from
Quaternary deep-sea sediments. Quaternary Geochronology,
16, 50–61.
King, K. J., 1980. Applications of amino acid biogeochemistry for
marine sediments. In Hare, P. E., Hoering, T. C., and King, K.,
Jr. (eds.), Biogeochemistry of Amino Acids. Wiley, pp. 377–391.
King, K., Jr., and Hare, P. E., 1972. Amino acid composition of the
test as a taxonomic character for living and fossil planktonic
foraminifera. Micropaleontology, 18, 285–293.
King, K., Jr., and Neville, C., 1977. Isoleucine epimerization for
dating marine sediments: importance of analyzing monospecific
foramiferal samples. Science, 195, 1333–1335.
Knudsen, K. L., and Sejrup, H. P., 1993. Pleistocene stratigraphy in
the Devils Hole area, central North Sea; foraminiferal and
amino-acid evidence. Journal of Quaternary Science, 8, 1–14.
Kosnik, M. A., and Kaufman, D. S., 2008. Identifying outliers and
assessing the accuracy of amino acid racemization measure-
ments for geochronology: II. Data screening. Quaternary Geo-
chronology, 3(4), 328–341.
Kvenvolden, K. A., Peterson, E., Wehmiller, J., and Hare, P. E.,
1973. Racemization of amino acids in marine sediments deter-
mined by gas chromatography. Geochimica et Cosmochimica
Acta, 37, 2215–2225.
Lehman, S. J., Miller, G. H., and Jones, G. A., 1988. Glacial Holo-
cene bottom water temperature changes deduced from
epimerization rate changes in foraminifers. Eos, Transactions
American Geophysical Union, 69(44), 1229.
Macko, S. A., and Aksu, A. E., 1986. Amino acid epimerization in
planktonic foraminifera suggests slow sedimentation rates for
Alpha Ridge, Arctic Ocean. Nature, 322, 730–732.
Müller, P. J., 1984. Isoleucine epimerization in Quaternary plank-
tonic foraminifera; effects of diagenetic hydrolysis and leaching,
and Atlantic-Pacific intercore correlations. Meteor-
Forschungsergebnisse Reihe C. Geologie und Geophysik, 38,
25–47.
Murray-Wallace, C. V., and Belperio, A. P., 1994. Identification of
remanié fossils using amino acid racemization. Alcheringa, 18,
219–227.
Penkman, K. E. H., Kaufman, D. S., Maddy, D., and Collins, M. J.,
2008. Closed-system behaviour of the intra-crystalline fraction
of amino acids in mollusk shells. Quaternary Geochronology,
3, 2–25.
Robbins, L. L., and Brew, K., 1990. Proteins from the organic
matrix of core-top and fossil planktonic foraminifera.
Geochimica et Cosmochimica Acta, 54, 2285–2292.
Schroeder, R. A., and Bada, J. L., 1977. Kinetics and mechanism of
the epimerization and decomposition of threonine in fossil fora-
minifera. Geochimica et Cosmochimica Acta, 41, 1087–1095.
Sejrup, H. P., and Haugen, J. E., 1992. Foraminiferal amino acid
stratigraphy of the Nordic seas; geological data and pyrolysis
experiments. Deep-Sea Research, Part A. Oceanographic
Research Papers, 39(Supplement 2A), 603–623.
Sejrup, H. P., Rokoengen, K., and Miller, G. H., 1984. Isoleucine
epimerization in Quaternary benthonic foraminifera from the
Norwegian continental shelf: a pilot study. Marine Geology,
56, 227–239.
Stathoplos, L., and Hare, P. E., 1993. Bleach removes labile amino
acids from deep sea foraminiferal shells. Journal of Foraminif-
eral Research, 23, 102–107.
Stathopolos, L., Hare, P. E., and Kennett, J. P., 1987. Modeling
amino acid diagenesis in fossilforaminiferal tests. Eos, Transac-
tions American Geophysical Union, 68(44), 1333–1334.
Wehmiller, J., and Hare, P. E., 1971. Racemization of amino acids in
marine sediments. Science, 173, 907–911.
 AMINO ACID RACEMIZATION, PALEOCLIMATE
Cross-references
Amino Acid
Amino Acid Racemization Dating
AMINO ACID RACEMIZATION, PALEOCLIMATE
Matthew Collins and Beatrice Demarchi
BioArCh, Department of Archaeology, University of
York, York, UK
Definition
Amino acid racemization (AAR): changes in the chirality
of amino acids, the building blocks of proteins, resulting
from the time-and-temperature-driven breakdown of pro-
teins. AAR is used in the geosciences as a relative dating
technique, typically comparing the ratio of D/L amino
acid enantiomers in biominerals.
Paleoclimate: the complex interrelated changes brought
upon by temperature shifts caused by orbital forcing of the
geosphere, hydrosphere, atmosphere, biosphere, and
anthroposphere throughout Earth’s history.
Introduction
The level of racemization in any one sample is in effect
a measure of the integrated time/temperature history (ther-
mal age) that accounts for the observed degree of protein
diagenesis. Therefore the contribution of amino acid race-
mization (AAR) to understanding past climate changes is
twofold: it can be used either as a geochronological tool,
to reconstruct the timing of major climatic events, or (less
commonly) as a paleothermometer in the case of samples
of known age, because the higher the observed rate of
racemization (k, see Arrhenius plot in Figure 1), the
warmer the burial history.
Dating past climate cycles
AAR geochronology can help to clarify the timing of past
climatic cycles and the extent of the environmental
responses to these changes, from a local to a global scale.
One of the most successful applications is the identifica-
tion of interglacial episodes in the patchy terrestrial sedi-
mentary record, which can thus be linked to the global
temperature fluctuations recorded in ocean and ice cores
(Oxygen Isotope Stages or Marine Isotope Stages). This
is particularly effective at high latitudes, where cyclic fluc-
tuations in global temperature have resulted in the accu-
mulation of sediments typically during warm ice-free
periods; here AAR can help to assign these deposits to
specific climate cycles (Penkman et al., 2013). In these
and in other midlatitude regions, models have been devel-
oped that link the acceleration of the racemization rates to
interglacial temperature increase and sea-level highstands
(e.g., Hearty et al., 1986).
47
Paleothermometry
The rates of protein decomposition (diagenesis)
increase exponentially with temperature. Effective
burial temperature (Teff, also known as the effective dia-
genetic temperature or EDT (Wehmiller, 1977)) is the
temperature inferred from rate constants determined in
dated samples (Figure 1). This is influenced by (i) sea-
sonal temperature fluctuations (which increase with dis-
tance from the equator but are dampened with burial
depth and shading) and (ii) long-term changes in global
temperature.
Paleothermometry and AAR are therefore inextricably
entwined. Accurate age estimates can be obtained on the
basis of D/L values, when the temperature history (Teff)
of a sample is known; conversely, when the absolute ages
of the sample are known, Teff can be calculated. In both
cases the kinetic parameters of the reaction must be
established (usually estimated from higher temperatures
using an Arrhenius plot; see Figure 1). Age estimations
are more likely to be more accurate where the extent of
seasonal fluctuation is minimized (e.g., sediment cores),
since near-surface sediments (especially deposits of
anthropogenic origin, e.g., shell middens) may experience
short-lived periods of extreme temperature.
The accuracy of the estimates of the kinetic parame-
ters affects these calculations (McCoy, 1987); kinetic
models based on high-temperature (artificial) diagenesis
(Figure 1, Arrhenius plot) can only be constrained to
the “real” burial temperatures using samples that can
accurately be dated by independent means. This is often
done using radiocarbon, hence covering the Holocene
and Marine Isotope Stages 2 and 3. The disparity
between Holocene Teff and that of the sample can be
used to estimate changes in paleotemperature (DT) at
the site, for example, to quantify the extent of cooling
during the Last Glacial (e.g., Miller et al., 1987, 1997).
If multiple samples of known age are available, it is then
possible to impute a paleotemperature record, but only
relatively few attempts have been made to do this
due to the need for long, high-resolution chronologies
(Kaufman, 2003).
Conclusions
Individual biominerals can be considered as temperature
loggers which report a single (integrated) thermal age.
Improvements in our ability to model paleotemperature
mean that in the future it should be possible to predict
the thermal history of a specific site at a specific burial
depth; an example where this is being attempted is the
http://thermal-age.eu website. When combined with
good kinetic estimates of protein decomposition, it
may be possible to use such models to estimate the
absolute age of a sample from AAR (or indeed other
temperature dependent data, such as hydration of
silicates).
48 ANCIENT INKS: A FORENSIC ART HISTORICAL PERSPECTIVE
Amino Acid Racemization, Paleoclimate, Figure 1 Increase in the extent of racemization as an effect of time and temperature, e.g.,
throughout the last glacial/interglacial cycle.
Bibliography
Hearty, P. J., Miller, G. H., Stearns, C. E., and Szabo, B. J., 1986.
Aminostratigraphy of Quaternary shorelines in the Mediterra-
nean basin. Geological Society of America Bulletin, 97,
850–858.
Kaufman, D. S., 2003. Amino acid paleothermometry of Quater-
nary ostracodes from the Bonneville Basin, Utah. Quaternary
Science Reviews, 22, 899–914.
McCoy, W. D., 1987. The precision of amino acid geochronology
and paleothermometry. Quaternary Science Reviews, 6,
43–54.
Miller, G., Jull, A., Linick, T., Sutherland, D., Sejrup, H., Brigham,
J., Bowen, D., and Mangerud, J., 1987. Racemization-derived
late Devensian temperature reduction in Scotland. Nature, 326,
593–595.
Miller, G. H., Magee, J. W., and Jull, A., 1997. Low-latitude glacial
cooling in the Southern Hemisphere from amino-acid racemiza-
tion in emu eggshells. Nature, 385, 241–244.
Penkman, K. E. H., Preece, R. C., Bridgland, D. R., Keen, D. H.,
Meijer, T., Parfitt, S. A., White, T. S., and Collins, M. J.,
2013. An aminostratigraphy for the British Quaternary based
on Bithynia opercula. Quaternary Science Reviews, 61,
111–134.
Wehmiller, J. F., 1977. Amino acid studies of the Del Mar, Califor-
nia, midden site: apparent rate constants, ground temperature
models, and chronological implications. Earth and Planetary
Science Letters, 37, 184–196.
ANCIENT INKS: A FORENSIC ART HISTORICAL
PERSPECTIVE
Howell G. M. Edwards
Chemical and Forensic Sciences, School of Life Sciences,
University of Bradford, Bradford, UK
Definition
Ink: A colored, usually liquid, material used for writing
which is applied with a brush, stylus or pen to a variety
of surfaces such as paper, papyrus, parchment, vellum or
fabrics.
Inks have been used to record events and data in docu-
mentation and for the decorative arts since 2700 BCE;
a description of the different types of ink used historically
for nearly 5,000 years will be followed by a survey of ana-
lytical information available based on their chemical com-
position which can provide a chronological assessment. In
conclusion, analytical case studies are presented briefly of
two important historical manuscripts, namely, the Vinland
Map, which purportedly was drawn in the early fifteenth
century and shows the northeastern coast of North
 ANCIENT INKS: A FORENSIC ART HISTORICAL PERSPECTIVE
America some 60 years before the voyage of Columbus,
and the Voynich Manuscript, a mysterious historiated
fifteenth-century herbal which has thus far defied transla-
tion and which has been termed the most enigmatic
ancient manuscript in existence.
Introduction
The earliest form of writing, cuneiform, involved the
inscribing of characters into wet clay which was then fired
to produce ceramic tablets; the adoption of cut reed pens
and fluid inks applied to paper was perfected in China
around 2700 BCE and attributed to the Chinese philoso-
pher Tien-Lcheu. Hence, ink has existed for over 4,700
years of human history – its name derives from the Latin
incaustum and the early French encre; a major problem
in the dating of inked artifacts is the rather limited range
of experimental compositions of historical inks, which
can reliably be described as being either early carbon-
based or later iron gallotannate inks (Thompson, 1996;
Brunelli and Crawford, 2003; Cradock, 2009).
The earliest ink, formulated in China from carbon black
suspended in an aqueous solution of water and gum ara-
bic, is now known as India ink because the best carbon
supplies at that time for this purpose were sourced in India;
the carbon was synthesized by burning wood such as pine,
in which the partially combusted resin helped to bind the
sooty deposits, in a limited amount of air under an
upturned iron or ceramic dish. Other carbons were sourced
from the calcination of animal bone and ivory, which pro-
duced a deeper black color – these contained residues of
calcium phosphate derived from the hydroxyapatite com-
ponent and provide an analytical signature of bone black
or ivory black.
The discovery of natural mineral oils and petroleums
afforded another opportunity for the production of
a deep black sooty residue upon combustion in a limited
supply of oxygen. A hierarchical basis for recipes for the
production of carbonaceous soot existed in which certain
botanical materials were highly prized for combustion to
form the blackest inks, such as peach stones, almond
shells, and vine twigs. There was much empiricism in
the formulation of early inks as evidenced by the universal
adoption of gum arabic as a binding agent; in addition to
assisting the suspension of the insoluble carbon particles
in the aqueous ink medium, the water-soluble gum arabic
modified the viscosity of the ink, assisting in the writing
flow when applied with reed pens, quills, and brushes
and also improving the adhesion of the ink to the substrate.
However, the addition of too much gum arabic resulted in
a brittleness of the dried ink applied and a tendency for the
writing to flake off – hence, the debris found between the
leaves of ancient manuscripts frequently contains particles
of ink from the associated script which can provide a rich
source of analytical information without involving the
destruction of the manuscript text. Different names have
been recorded for carbon black inks through the ages,
49
dependent upon formulation or source, such as bistre
(Seaver, 2004), an extract from sooty fires which pos-
sessed a warm brown color, and sepia, a dark, semitrans-
parent ink from cuttlefish much used by the Romans.
Despite prior historical usage for a period of some
3,800 years, from about 1130 ACE iron gall ink replaced
carbon black ink as the favored medium of writing; iron
gall ink, more correctly described as an iron gallotannate,
was the first water-based ink to be made from a chemical
reaction between aqueous solutions of iron (II) sulfate
and extracts of oak galls with the addition of gum
arabic. Oak galls are spherical nutlike protuberances
resulting from the egg laying of wasps on oak trees – the
best galls were those fully developed from which the
emerging wasp larvae had hatched. As encountered with
the carbon-based inks, there are many recipes in existence
for the manufacture of iron gall inks; indeed, the chronol-
ogy for the first appearance of iron gall inks must predate
the mediaeval since Pliny in the first century ACE
describes in detail the preparation of aqueous gall solu-
tions which blacken in the presence of copperas, an iron
sulfate ore, which can be used for writing. An example
of the complexity of an iron gall ink composition is pro-
vided by the following recipe for an ink “of the highest
quality” in the sixteenth century: 8 oz powdered Aleppo
galls, 4 oz logwood chips, 4 oz iron sulfate, 3 oz powdered
gum arabic, 1 oz copper sulfate or verdigris (basic copper
acetate), and 1 oz sugar – all heated and triturated in 12 lb
of water followed by filtration and the addition of alum,
ammonia, beer, lemon juice, oil of cloves, ground walnuts,
lavender, wine, boiled oil, and extract of amber or shellac
in brandy to minimize the growth of mould.
Iron gallotannate inks quickly became the medium of
choice for mediaeval scribes because, unlike the carbon-
based inks, they interacted chemically with cellulose sub-
strates, conferring better adhesion and permanence on the
script (Thompson, 1956). However, this improvement in
the better writing permanence of iron gall inks caused
severe corrosion problems for paper manuscripts in partic-
ular. In some cases, this process resulted in the formation
of holes in the manuscript (lacunae) in place of the writing
(Figure 1); many manuscripts have suffered irreversibly in
this way and pose problems for conservation and the pres-
ervation of their integrity (Reissland and de Groot, 1999;
Reissland, 2002). It has been found that arresting the
decay can be achieved by the application of calcium bicar-
bonate, lime, magnesite, and calcium phytate, but gener-
ally irreparable damage has resulted for the original
script and text (Barrow, 1972; Delage et al., 1990;
Rouchon-Quillet et al., 2004).
The corrosive effect of iron gallotannate inks upon cel-
lulosic substrates can be related to the iron-catalyzed
breakdown of cellulose in an acidic environment (Proost
et al., 2004); in this process the Fe2+ react with acidic
decomposition products of the cellulose to form hydroxyl
and oxygenyl radicals, from which the subsequent
50 ANCIENT INKS: A FORENSIC ART HISTORICAL PERSPECTIVE
Ancient Inks: A Forensic Art Historical Perspective, Figure 1 A
historical manuscript showing evidence of serious corrosive
deterioration from iron gall ink on paper.
creation of hydrogen peroxide destroys the cellulose sub-
strate and oxidizes the iron to Fe3+ (Bulska and Wagner,
2004). Hence, it has been suggested (Proost et al., 2004)
that measurement of the Fe2+/Fe3+ ratio in an ancient iron
gall ink could provide a means of assessing its age,
although clearly the actual rate of degradation of the ink
would be dependent upon several environmental factors,
not the least of which would be the recipe and formulation
of the original ink. The elemental migration of iron atoms
into the substrate also can provide a measure of the age of
the script, as determined from Auger spectroscopy, but pit-
falls can be encountered in the form of hair follicles in vel-
lum substrates (de Pas and Flieder, 1976).
The next development in ink manufacture occurred in
the mid-nineteenth century, when a new range of organic
dye-based colored inks became available following upon
Perkin’s synthesis of mauveine; at the London Society of
Arts in 1858, the advent of this new range of brightly col-
ored dyestuffs was advocated for ink manufacture with the
additional adoption of graphite in place of soot for carbon-
based inks.
Other ancient inks
It is true that most ancient manuscript texts and writing
have been accomplished using black ink, but other colors
have been identified, although these are strictly limited;
examples include red inks (cinnabar, dragon’s blood,
Brazilwood extracts), purple (folium, caput mortuum,
purpurum extracts) and gold. The major problem that
was identified early on with the use of alternatives to car-
bon black or iron gall-based inks is that unless the pigment
was a mineral, such as cinnabar or in some rarer cases
hematite, the extracts from botanical or marine sources
were dyestuffs, which were badly affected by light and
handling – such was the case for folium (also known as
turnsole) and purpurum. The ancient scribes and artists
termed these dyes “fugitive” and thereby recognized their
impermanence (Ball, 2001; Lee, 2007).
The problem that faces analytical scientists and art his-
torians is the dating of inked manuscripts; unlike
historiated manuscripts such as Books of Hours and the
Lindisfarne Gospels, ancient texts and maps often do not
have any coloring pigments associated with them, so the
identification of the chemical composition of the ink was
an important analytical source of information. Radiocar-
bon dating procedures are usually not practicable for the
dating of ancient inks because of the quantities of speci-
men required in the destructive analysis, although of
course some useful ancillary information can result from
the radiocarbon dating of the manuscript substrates.
Dating of ancient inks
Traditional methods of characterization of ancient inks
(de Pas and Flieder, 1976; Cradock, 2009) involve the
application of techniques such as proton-induced X-ray
emission (PIXE) spectroscopy, thin layer chromatogra-
phy, ultraviolet–visible spectroscopy, infrared spectros-
copy, Raman microscopy, micro-X-ray fluorescence
spectroscopy, capillary electrophoresis, Rutherford back-
scattering spectroscopy, scanning electron microscopy,
scanning Auger spectroscopy, and mass spectrometry
(Senvaitiene et al., 2005; La Porte and Stephens, 2012).
The major problems facing the analysts arise from varia-
tion of ink composition over the centuries. The central
analytical discrimination for the differentiation between
iron gall inks and carbon-based inks is the recognition of
elemental carbon signals versus those of metallic iron.
Secondary issues relate to the analytical identification of
potential additives such as gum arabic, shellac, and cop-
per; in this respect, infrared spectroscopy and mass spec-
trometry provide valuable molecular information that
other techniques do not. Thin layer chromatography has
been found to be not particularly useful for the composi-
tional discrimination of ancient inks (Senvaitiene
et al., 2005). In contrast, the identification of elemental
iron by SEM/EDAXS, micro-XRF, and PIXE techniques
is often used to indicate the presence an iron gall ink in
preference to a carbon-based ink. Iron gall inks possess
an ultraviolet absorption with a sharp, strong band at
215 nm and a shoulder at 269 nm, characteristic of an iron
gallotannate complex, compared with corresponding
bands at 218 and 274 nm for the uncomplexed
gallotannic acid.
 ANCIENT INKS: A FORENSIC ART HISTORICAL PERSPECTIVE
Raman spectroscopy can be used to detect amorphous
and graphitic carbon and has been applied to the analysis
of ancient carbons and inks: a typical Raman spectrum
of an ancient carbon derived from a botanical source gives
two features, a broad band at 1,320 cm
1 ascribed to sp3
hybridized carbon atoms (the D band) and a sharper band
at 1,580 cm
1 attributed to sp2 hybridized carbon atoms
(the G band).
As an illustration of the information that can be
deduced from the ink analysis of ancient manuscripts,
two important manuscripts purporting to be from the fif-
teenth century will be described in more detail here –
one of these, the Vinland Map, possesses only black ink
on two sheets of vellum (Skelton et al., 1965; Seaver,
2004), whereas the other, the Voynich Manuscript, is poly-
chrome and comprises 240 sheets of vellum, but with
again only black ink script (Kennedy and Churchill,
2004; D’Imperio, 2012). Both manuscripts were discov-
ered in the early to mid-twentieth century with limited his-
torical provenances and have since been the subject of
some intense analytical eschatological and historical
research.
Case study 1: the Vinland map
The Vinland Map is a vellum map of the Old World, mea-
suring approximately 28  40 cm, that identifies an area in
the Western Atlantic, Vinilanda Insula, the so-called Vin-
land of Scandinavian Viking folklore, showing an area to
the northeast of North America, described thereon as “a
new and fertile land to the west.” It first appeared in
1957 bound together with a manuscript known as the Tar-
tar Relation (the Historia Tartarorum), which had an
unusual composition of vellum and paper and two leaves
of vellum alternating with six of paper, the latter
exhibiting bulls’ head watermarks which were identifiable
with the Basle Council of the 1430s. The cartographic
importance of the Vinland Map centered upon its
predating the voyage of Christopher Columbus and his
epic discovery of the New World in 1492 by approxi-
mately 60 years and fed the rumor that Columbus could
actually have used such a map based on earlier Viking sea-
faring exploits, notably those of Leif Ericson (Seaver,
1998).
In 1957, the Vinland Map and Tartar Relation were
examined by British Museum experts in ancient maps
and incunabula (Skelton et al., 1965). The philanthropist
Paul Mellon donated them to the Beinecke Rare Book
and Manuscript Library in Yale University, and in 1972
the Yale University Library commissioned an analysis of
the Vinland Map, thereby starting a controversy which
has raged for over 40 years and is still ongoing! The scien-
tific analysis of the Vinland Map opened with a detailed
polarized microscopic and X-ray diffraction (XRD) analy-
sis of the inked areas (McCrone and McCrone, 1974) from
which it was concluded that the presence of anatase,
a polymorphic form of titanium (II) oxide, in the ink dated
the map firmly to the twentieth century (Gettens and Stout,
51
1942) and indicating the Vinland Map to be a fake
(Baynes-Cope, 1974). Further analyses using scanning
electron microscopy and energy-dispersive X-ray (EDX)
(McCrone, 1988; McCrone, 1999) determined that the
ink was not an iron gallotannate as found on the associated
Tartar Relation but was carbon black. The ink and parch-
ment of the Vinland Map were reanalyzed (Cahill
et al., 1987) using proton-induced X-ray emission
(PIXE) spectroscopy and challenged the McCrone conclu-
sion (Delage et al., 1990). Radiocarbon dating of the vel-
lum (Donahue et al., 2002) gave a date of 1432 +/
11
ACE, which agreed to within one standard deviation with
the watermarked date on the associated paper in the Tartar
Relation.
In 2002, a definitive Raman microspectroscopic study
of the Vinland Map was undertaken (Brown and Clark,
2002) with 632.8 nm laser excitation. The inked areas
were apparently composed of two parts: a yellow line
which was strongly adherent to the parchment substrate
and an overlaid black line which showed evidence of
severe loss in parts due to the black pigment “flaking
off.” The analysis of the black ink gave the characteristic
D and G bands for amorphous carbon at 1,325 and
1,580 cm
1, respectively. The presence of anatase was
evident, with characteristic bands at 142 and 398 cm
1
and seems to be indicative of a clever forgery; however,
care must be taken as anatase has actually been identified
in genuine archaeological artifacts significantly predating
its established synthesis in the twentieth century (Edwards
et al., 2006).
Case study 2: the Voynich manuscript
In 1912, Wilfrid Voynich revealed the discovery of his
eponymous manuscript, which apparently dates to the
early fifteenth century: the 240 vellum pages with black
script and historiated polychrome figures, now labelled
the “world’s most mysterious manuscript,” resemble
a herbal compilation of botanical medicine and
alchemy. Believed to be written either in a secret code
or in a lost language, the major problem with this man-
uscript is that it has thus far defied all attempts at trans-
lation. The historical provenance of the Voynich
Manuscript, like that of the Vinland Map, is steeped in
international intrigue (Kennedy and Churchill, 2004;
D’Imperio, 2012). A recent comprehensive and elegant
analysis of the iron gall ink on the Voynich Manuscript
has been undertaken (Barabe, 2009) using polarized
light microscopy, SEM/EDAX spectrometry, micro-
XRD, and IR spectroscopy.
Unlike the Vinland Map, the Voynich Manuscript is
historiated and has the advantage of supporting analytical
information resulting from the interrogation of the associ-
ated pigments to assist in the chronology (Birren, 1965;
Feller, 1966; Harley, 1982; Roy, 1993; Fitzhugh, 1997;
Berrie, 2007; Eastaugh et al., 2004), which seems to be
well established in the mediaeval period.
52 ANCIENT INKS: A FORENSIC ART HISTORICAL PERSPECTIVE
Conclusion
The identification of and analytical discrimination
between ancient inks and the recognition of the corrosive
nature of iron gallotannate inks are well documented, but
the dating of specimen manuscripts is generally not feasi-
ble solely by examination of the ink composition alone;
the consideration of associated information from radiocar-
bon dating of the manuscript substrates, eschatology, lex-
icography, and chronological placement of historiated
pigments materially assists in the verification of the antiq-
uity of a manuscript.
Bibliography
Ball, P., 2001. Bright Earth: The Invention of Colour. London:
Viking/Penguin Books.
Barabe, J. G. 2009. Materials Analysis of the Vinland Map:
A Report to the Curator of Modern European Books and
Manuscripts, Beinecke Rare Book & Manuscripts Library, Yale
University, USA, pp. 1–6.
Barrow, W. J., 1972. Manuscripts and Documents: Their Deteriora-
tion and Restoration. Charlottesville: University of Virginia
Press.
Baynes-Cope, A. D., 1974. The scientific examination of the
Vinland Map in the research laboratory of the British Museum.
Geographical Journal, 140, 208–211.
Berrie, B. H. (ed.), 2007. Artists’ Pigments: A Handbook of Their
History and Characteristics. Oxford/New York/Washington:
National Gallery of Art/Oxford University Press, Vol. 4.
Birren, P., 1965. History of Colour in Painting. New York: Van
Nostrand/Reinhold.
Brown, K.L., and Clark, R.J.H., 2002. Analysis of Pigmentary
Materials on the Vinland Map and Tartar Relation by Raman
Microprobe Spectroscopy. Analytical Chemistry, 74,
3658–3661.
Brunelli, R. L., and Crawford, K. R., 2003. Advances in the Foren-
sic Analysis and Dating of Writing Ink. Springfield:
C.C. Thomas.
Bulska, E., and Wagner, B., 2004. The study of ancient manuscripts
exposed to iron-gall ink corrosion, Chapter 17. In Janssens, K.,
and van Grieken, R. (eds.), Non-Destructive Micro-Analysis of
Cultural Heritage Materials. Dordrecht: Elsevier.
Cahill, T. A., Schwab, R. N., Kusko, B. H., Eldred, R. A., Moeller,
G., Dutshke, D., and Wick, D. L., 1987. The Vinland Map
re-visited: new compositional evidence on its inks and parch-
ments. Analytical Chemistry, 59, 829–833.
Cradock, P., 2009. Scientific Investigation of Copies, Fakes and
Forgeries. Oxford: Elsevier Butterworth-Heinemann.
D’Imperio, M. E., 2012. The Voynich Manuscript: An Elegant
Enigma. Seattle, USA:CreateSpace Independent Publishing
Platform. Amazon.com.
de Pas, M., and Flieder, F., 1976. History and prospects for black
manuscript inks. In Broumelle, N., and Smith, P. (eds.), Conser-
vation and Restoration of Pictorial Art. London: Butterworths.
Delage, E., Grange, M., Kusko, B., and Menei, E., 1990. Apparition
de l’encre metallogallique en Egypte a partir de la Collection de
Papyrus du Louvre. Revue d’Egyptologie, 41, 213–217.
Donahue, D. H., Olin, J. S., and Harbottle, G., 2002. Determination
of the radiocarbon age of the parchment of the Vinland Map.
Radiocarbon, 44, 45–52.
Eastaugh, N., Walsh, V., Chaplin, T., and Siddall, R., 2004. Pigment
Compendium: A Dictionary of Historical Pigments. Oxford:
Elsevier Butterworth-Heinemann.
Edwards, H. G. M., Nik Hassan, N. F., and Middleton, P. S., 2006.
Anatase: A pigment in ancient artwork or a modern Usurper?
Analytical and Bioanalytical Chemistry, 384, 1356–1365.
Feller, R. L. (ed.), 1996. Artists’ Pigments: A Handbook of Their
History and Characteristics. Oxford/New York/Washington:
National Gallery of Art/Oxford University Press, Vol. 1.
Fitzhugh, E. W. (ed.), 1997. Artists’ Pigments: A Handbook of Their
History and Characteristics. Oxford/New York/Washington:
National Gallery of Art/Oxford University Press, Vol. 3.
Gettens, R. L., and Stout, G. L., 1942. Painting Materials: A Short
Encyclopaedia. New York: D. Van Nostrand.
Harley, R. L., 1982. Artists’ Pigments, 1600–1835. London:
Butterworths Scientific.
Kennedy, G., and Churchill, R., 2004. The Voynich Manuscript: The
Unsolved Riddle of an Extraordinary Book which has Defied
Interpretation for Centuries. London: Orion Books.
La Porte, G. M., and Stephens, J. C., 2012. Analytical techniques
used for the forensic examination of writing and printing inks.
In Kobilinskiy, L. (ed.), Forensic Chemistry Handbook.
Hoboken, NJ: Wiley, pp. 225–248.
Lee, D., 2007. Nature’s Palette: The Science of Plant Color.
Chicago: University of Chicago Press.
McCrone, W. C., 1988. The Vinland Map 1988. Analytical Chemis-
try, 60, 1009–1018.
McCrone, W. C., 1999. The Vinland Map 1999. Microscope, 47,
71–74.
McCrone, W. C., and McCrone, L. B., 1974. The Vinland Map ink.
Geographical Journal, 140, 212–214.
Proost, K., Janssens, K., Wagner, B., Bulska, E., and Schreiner, M.,
2004. Determination of localised Fe2+/Fe3+ ratios in inks of his-
toric documents by means of micro-XANES. Nuclear Instru-
ments and Methods in Physics Research B, 213, 723–728.
Reissland, B., 2002. Iron gall ink corrosion – progress in
visible degradation. In Mesh, J. A., and Tennent, N. H. (eds.),
Contributions to Conservation. London: James & James,
pp. 113–118.
Reissland, B., and de Groot, S., 1999. Ink corrosion: comparison of
currently used aqueous treatments for paper objects. In Preprint
from the 9th Ink Congress of IADA, ed. M.S. Koch, Royal Acad-
emy of Fine Arts School of Conservation, Copenhagen, Den-
mark, Nilsen Bogtryk, Copenhagen. pp. 121–129.
Rouchon-Quillet, V., Bernard, J., Wattiaux, A., and Fournes, L.,
2004. The impact of gallic acid on iron gall ink corrosion.
Applied Physics, 79, 389–392.
Roy, A. (ed.), 1993. Artists’ Pigments: A Handbook of Their History
and Characteristics. Oxford/New York/Washington: National
Gallery of Art/Oxford University Press, Vol. 2.
Seaver, K. A., 1998. The Frozen Echo: Greenland and the Explora-
tion of North America, ca. AD1000 – 1500. Stanford: Stanford
University Press.
Seaver, K. A., 2004. Maps, Myths and Men. Stanford: Stanford
University Press.
Senvaitiene, J., Beganskiene, A., Tautkus, S., Padarauskas, A., and
Kareiv, A., 2005. Characterization of historical writing inks by
different analytical techniques. Chemiya, 16, 34–38.
Skelton, R. A., Manston, T., and Painter, G. D., 1965. The Vinland
Map and the Tartar Relation. New Haven: Yale University Press.
Thompson, D. V., 1956. Materials and Techniques of Mediaeval
Painting. New York: Dover Publications.
Thompson, J. C., 1996. Manuscript Inks: A Personal Exploration of
Materials and Modes. Portland, OR: Caber Press.
Cross-references
Radiocarbon Dating
 APATITE
APATITE
W. Jack Rink
School of Geography and Earth Sciences, McMaster
University, Hamilton, ON, Canada
Synonyms
Chlorapatite (chlorine-rich variety); Dahllite (carbonate-
bearing hydroxyapatite); Fluorapatite (fluorine-rich vari-
ety); Francolite (carbonate-rich variety); Hydroxyapatite
(hydroxyl-rich variety)
Definition
Apatite is described herein and as a biomineral (biological
apatite). The mineral apatite has the formula
Ca5(PO4)3(F, Cl, OH) and is characterized (Figure 1) by
phosphorus tetrahedrons which share oxygen with nine-
coordinated calcium sites and singly charged anion lattice
sites hosting F, Cl, or OH that are surrounded by a planar
arrangement of three calcium atoms (Beevers and
McIntryre, 1946). It occurs widely as an accessory mineral
in all classes of rocks: igneous, metamorphic, and sedi-
mentary (Klein and Hurlbut, 1999). It is also found in peg-
matites and other veins, probably of hydrothermal origin.
Apatite, Figure 1 Apatite structure viewed along the
c-crystallographic axis: green ¼ calcium, red ¼ oxygen,
orange ¼ phosphorus, white ¼ fluorine, chlorine, or hydroxyl.
53
The source of its datability is the radioactive trace element
uranium that substitutes for calcium.
It is most commonly used as a dating material in
thermochronology of large-scale earth processes via three
different dating methods: (U-Th)/He, 3He/4He, and fission
track, recently covering topics such as exhumation studies
of mountain ranges (Cecil et al., 2006), transient erosion
of orogens (Rahl et al., 2007), glacial erosion (Shuster
et al., 2005; Ehlers et al., 2006), and erosion of foreland
basins (Cederbom et al., 2011) or exotic applications like
the age and temperature of shock metamorphism of mete-
orites (Min et al., 2004). Apatite’s versatility as a dating
material is attested through the wide array of other dating
methods that have been applied. Thomson et al. (2012)
have reported the successful application U-Pb dating in
apatite using laser ablation multicollector inductively
coupled plasma mass spectrometry. In addition, cosmo-
genic nuclide dating of apatite using 3He (Farley
et al., 2006), uranium-series dating of phosphorites
(Jarvis et al., 1994), and Lu-Hf and Pb step-leaching geo-
chronology in metamorphic apatites have also been
reported.
Bibliography
Barfod, G. H., Krogstad, E. J., Frei, R., and Albarède, F., 2005.
Lu-Hf and PbSL geochronology of apatites from Proterozoic ter-
ranes: a first look at Lu-Hf isotopic closure in metamorphic apa-
tite. Geochimica Cosmochimica Acta, 69, 1847–1859.
Beevers, C. A., and McIntyre, D. B., 1946. The atomic structure of
fluor-apatite and its relation to that of tooth and bone material.
Mineralogical Magazine, 27, 254–257.
Cecil, M. R., Ducea, M. N., Reiners, P. W., and Chase, C. G., 2006.
Cenozoic exhumation of the Northern Sierra Nevada, California,
from (U-Th)/He thermochronology. Geological Society of Amer-
ica Bulletin, 118, 1481–1488.
Cederbom, C. E., van der Beek, P., Schlunegger, F., Sinclair, H. D.,
and Oncken, O., 2011. Rapid extensive erosion of the Northern
Alpine Foreland Basin at 5–4 Ma. Basin Research, 23, 528–550.
Ehlers, T. A., Farley, K. A., Rusmore, M. E., and Woodsworth, G. J.,
2006. Apatite (U-Th)/He signal of large-magnitude accelerated
glacial erosion, southwest British Columbia. Geology, 34,
765–768.
Farley, K. A., Libarkin, J., Mukhopadhyay, S., and Amidon, W.,
2006. Cosmogenic and nucleogenic 3He in apatite, titanite and
zircon. Earth and Planetary Science Letters, 248, 451–461.
Jarvis, I., Burnett, W. C., Nathan, Y., Almbaydin, F. S. M., Attia,
A. K. M., Castro, L. N., Flicoteaux, R., Hilmy, M. E., Husain,
V., Qutawnah, A. A., Serjani, A., and Zanin, Y. N., 1994. Phos-
phorite geochemistry: state-of-the-art and environmental con-
cerns. Eclogae Geologicae Helveticae, 87, 643–700.
Klein, C., and Hurlbut, C. S., 1999. Manual of Mineralogy (after
James D. Dana). New York: Wiley. 681 p.
Min, K., Reiners, P. W., Nicolescu, S., and Greenwood, J. P., 2004.
Age and temperature of shock metamorphism of Martian mete-
orite Los Angeles from (U-Th)/He thermochronometry. Geol-
ogy, 32, 677–680.
Rahl, J. M., Ehlers, T. A., and van der Pluijm, B. A., 2007. Quanti-
fying transient erosion of orogens with detrital
thermochronology from syntectonic basin deposits. Earth and
Planetary Science Letters, 256, 147–161.
54 AQUIFER CHARACTERISTICS (U-SERIES)

Shuster, D. L., Ehlers, T. A., Rusmore, M. E., and Farley, K. A.,
2005. Rapid glacial erosion at 1.8 Ma revealed by 4He/3He
thermochronometry. Science, 310, 1668–1670.
Thomson, S. N., Gehres, G. E., Ruiz, J., and Buchwaldt, R., 2012.
Routine low-damage apatite U-Pb dating using laser ablation-
multicollector-ICPMS. Geochemistry, Geophysics, Geosystems,
13, Q0AA21, doi:10.1029/2011 GC003928.
Cross-references
Fission Track Dating and Thermochronology
Terrestrial Cosmogenic Nuclide Dating
Uranium–Lead Dating
U-Series Dating
AQUIFER CHARACTERISTICS (U-SERIES)
Don Porcelli
Department of Earth Sciences, University of Oxford,
Oxford, UK
the decay constant) of all the radionuclides in a decay
series are the same. When this is disrupted by an event
or process that changes the relative proportions of the iso-
topes within a decay series, the activity of each isotope
will evolve until it again reaches equilibrium with its par-
ent isotope, over a time of several half-lives of the isotope
itself. Within groundwaters, activities vary widely
between the radionuclides (which are therefore in disequi-
librium), as shown in Figure 2, due to the contrasting
chemical behavior of the elements involved. The highest
activities are for 222Rn, which, as a noble gas, does not
react with the aquifer solids. The actinide U, which is sol-
uble in oxidizing waters, is often present in intermediate
activities that are moderated by removal onto aquifer
rocks by adsorption and, in some cases, precipitation.
The alkaline earth element Ra and the actinide Th are
strongly depleted due to water/rock interactions. Both of
these elements have very short-lived as well as longer-
lived isotopes, and so their isotope compositions reflect

Definition
Groundwater concentrations of naturally occurring iso-
topes in the 238U, 235U, and 232Th decay series that are
used to identify and quantify a range of aquifer
characteristics.

Essential concepts
The radionuclides 238U, 235U, and 232Th decay in a series
of shorter-lived radionuclides, and these decay series
(Figure 1) provide an important tool for investigating var-
ious aquifer characteristics, including chemical parame-
ters, flow patterns, and transport of actinides. The
relationships between the radionuclides in groundwater
provide the potential for readily quantifying parameters
of water/rock interaction from direct measurements of
waters alone, with trace element source terms and bulk Aquifer Characteristics (U-Series), Figure 2 The ranges of
adsorption coefficients inferred from comparison between activity concentrations (in becquerels per liter or decays per
second per liter) measured in three aquifers are shown side by
the different radionuclides. side for select isotopes in the 238U decay series (After Porcelli,
The point of comparison for groundwater radionuclides 2008). There are wide variations for each isotope reflecting site
is the secular equilibrium generally occurring within min- conditions and between isotopes reflecting the different
erals, in which the activities (the number of atoms times chemistries of the elements.

Aquifer Characteristics (U-Series), Figure 1 The isotopes of the 238U, 235U, and 232Th decay series. The isotopes in bold are those
most commonly measured in groundwater studies. The dashed arrows indicate where isotopes have been left out that have very
short half-lives and so are measured only in rare cases. Radioactive half-lives are given below each isotope in units of billions of years
(Ga), thousands of years (ka), years (a), or days (d).
 AQUIFER CHARACTERISTICS (U-SERIES)
processes occurring over a range of timescales. The con-
centration of each isotope is also controlled by the rate at
which it is supplied. Ongoing processes that maintain a
steady-state supply of each isotope will maintain disequi-
librium between parent and daughter isotopes; however,
when there are changes in hydrodynamic conditions,
activity concentrations will adjust to the new steady-state
supply over a time of several half-lives of the daughter iso-
tope. Overall, since the isotopes in each decay series are
tied to one another by decay systematics, with daughter
nuclide productions and distributions dependent upon par-
ent distributions, combined studies of the elements in
these series can generate considerable information regard-
ing radionuclide water/rock interactions.
It is generally difficult to quantify aquifer parameters
that control trace element behavior by direct measure-
ments. Host rocks typically cannot be as readily sampled
as groundwater, and average values for rocks recovered
from boreholes are difficult to obtain. Some of this infor-
mation has been sought from aquifer tests using experi-
mentally introduced tracers and from laboratory
experiments on aquifer materials, but these studies are
labor intensive and necessarily limited in scale of both
time and space. Measurement of radioactive decay series
isotopes allows a number of aquifer characteristics to be
investigated including mixing of groundwater into marine
and surface water reservoirs, tracking of groundwater
flow, estimating radionuclide migration rates, determining
recent precipitation, locating redox boundaries within an
aquifer, and dating groundwater within an aquifer.
The supply of radon and other radionuclides
by recoil
As shown in Figure 1, some radionuclides are produced by
high-energy a decay, while some are produced by
low-energy b decay. In the former, decay is sufficiently
energetic that during decay, the daughter isotope is pro-
pelled 100 Å (10
8 m) through the mineral in which it
was produced, possibly resulting in expulsion from the min-
eral. It is a physical process, in contrast to the chemical pro-
cesses affecting the radionuclides involved. Short-lived
nuclides in groundwater are supplied primarily by this
mechanism as well as by decay of dissolved and adsorbed
parent isotopes. The recoil rate is therefore a key parameter.
For long-lived nuclides in the decay series, weathering of
U- and Th-bearing minerals may also be significant.
The radionuclide that is simplest to understand is 222Rn
(Figure 3). Since Rn is an unreactive noble gas, groundwa-
ter 222Rn is entirely dissolved, and this can be directly
measured in waters pumped to the surface. It is removed
only by decay and not by adsorption or precipitation.
The main supply is from recoil caused by a decay of
226
Ra that is contained in minerals with 238U. The recoil
rate is simply equal to the rate at which 222Rn is removed
by decay. Therefore, the recoil rate can be determined sim-
ply from the rate of decay of dissolved 222Rn, and this can
then be used to determine the recoil rate of other nuclides.
55
The fraction of the total 222Rn produced in the aquifer
solids that is supplied to groundwater through recoil is
dependent upon the size of the 226Ra-bearing grains, with
small grains having a larger proportion of the contained
226
Ra within recoil distance of grain surfaces. Measured
222
Rn groundwater activities have been obtained that cor-
respond to as much as 10 % of the total amount pro-
duced in the aquifer rock. Where the proportion of recoil
222
Rn in groundwater is this large, parental 226Ra is typi-
cally interpreted to reside in very small mineral grains or
in secondary surface coatings.
Adsorption of Ra, Th, and U
Another important parameter determining element behav-
ior in an aquifer is the bulk adsorption coefficient, which is
the ratio of atoms adsorbed onto surfaces to the atoms in
solution. This is approximately equal to the retardation
factor, which is the ratio of the rate of groundwater flow
to the rate of element migration. Therefore, elements with
a high partition coefficient, and so high retardation factor,
migrate correspondingly slower. This parameter can be
determined for each isotope from the rate of recoil and
its groundwater concentration and so can provide overall
rates of transport of Ra, Th, and U as well as of water/rock
interaction processes affecting these elements. This is
regardless of the actual mechanisms involved, and so the
information complements small-scale physical chemistry
studies that can identify the controlling processes.
The case for 234Th is shown in Figure 2b. When the
234
Th concentrations dissolved in groundwater and
adsorbed on mineral surfaces are in steady state, the input
from recoil is equal to the removal rate by decay in solu-
tion and on the surfaces. If the recoil rate is known from
the concentration of 222Rn, then the adsorption coefficient
of 234Th, and so of the element Th, can be determined.
Similarly, the adsorption coefficients of Ra, Po, Pb, and
U can also be determined. It should be noted that the sup-
ply of any daughter nuclide is dependent not only upon
recoil from parent atoms within primary minerals but also
by removal onto adsorption sites and incorporation into
secondary minerals. Precise calculations therefore are
somewhat more complex, since weathering, recoil,
adsorption, and precipitation have affected the distribution
of the parent isotope in the decay series, which in turn is
dependent upon the effects of parent isotopes further up
the decay chain. Mathematical treatments of simple aqui-
fer models can be found in Krishnaswami et al. (1982),
Ku et al. (1992), and Porcelli (2008).
Using these methods, the Ra partition coefficients in
various freshwater aquifers have been determined to be
in the range of 103–104. In saline aquifers, the adsorption
of Ra drops substantially due to competition for adsorp-
tion sites. In freshwater aquifers, Th is generally adsorbed
101–102 times more efficiently than Ra. Values for U in
the same aquifers of 103 have been found, though else-
where there is evidence for much lower adsorption (see
Porcelli, 2008). Values for Po vary as much as 101–106,
56 AQUIFER CHARACTERISTICS (U-SERIES)
Aquifer Characteristics (U-Series), Figure 3 Recoil rates of nuclides in the 238U decay series are determined from 222Rn
concentrations. Applied to 234Th, the extent of adsorption of Th can be calculated. The isotope 234U is supplied by both recoil and
weathering, which leads to 234U/238U activity ratios (designated as 234U/238U) in groundwater of greater than 1.
and Pb has been found to have values of 10
2–101 times
those of Po (Porcelli, 2014).
Interpreting U isotope compositions
The (234U/238U) ratio (the ratio of activities of the two iso-
topes, as denoted by the parentheses) can provide useful
information on aquifer characteristics and serve as
a groundwater tracer. While the concentrations of 238U
in groundwaters are controlled by weathering (like other,
stable trace elements), which generally releases U
with (234U/238U) ¼ 1, the concentration of 234U is con-
trolled by both weathering and recoil (Figure 3). The
decay of 238U produces 234Th, which can be recoiled out
of mineral grains and which then produces 234U by
low-energy b decay. The (234U/238U) ratio therefore
reflects the ratio of the rate of recoil to the rate of
weathering. The recoil flux is generally simply propor-
tional to the surface area and U concentration of U- bear-
ing grains. The weathering rate is controlled by the
composition of the host grains and groundwater chemis-
try. It is also proportional to the surface area and
U concentration.
A number of aquifer conditions will have predictable
effects on the (234U/238U) ratio and U concentration of
groundwater. Differences in the minerals that contain
U will affect both (234U/238U) ratios and U concentrations,
since it will change the rate at which U is released by
weathering. Similarly, changes in groundwater chemistry
can change the weathering rate of minerals. Differences
in the U precipitation rate will result in different
U concentrations, but not in different (234U/238U) ratios.
Differences in the adsorption coefficient of U, e.g., due
to changes in pH or the presence of adsorbing minerals,
will have no effect on the U isotope composition. Differ-
ences in the surface area of U-bearing minerals, e.g., by
differences in average grain size or in the abundances of
U-bearing minerals, will affect recoil and weathering of
U equally and so will not change the (234U/238U) ratio of
the water.
Applications
The systematics of isotopes within the U and Th decay
series provide a range of tools, useful over different time-
scales, which can be used to trace groundwater, identify
aquifer processes, and quantify water/rock interactions.
Examples include:
Determining marine groundwater discharge. There can be
significant discharge of groundwater offshore and into
shallow coastal waters, and this can be identified with
Ra isotopes. Saline coastal groundwaters are enriched
in the long-lived 226Ra and especially 228Ra compared
to seawater. The short-lived 223Ra and 224Ra isotopes
are also enriched in saline and fresh groundwaters and
highly depleted in seawater. Assuming that the concen-
trations in coastal waters are in steady state, from esti-
mates of the water residence time, the influx of
groundwater-derived Ra, and so groundwater fluxes,
can be determined. The (228Ra/226Ra) ratio can also be
used to identify groundwater sources, since younger
groundwaters, or aquifers recently flushed by saline
water intrusion, have higher ratios since it takes longer
for 226Ra concentrations to build up outside minerals.
Short-lived 222Rn can also be used to evaluate very
recent groundwater inputs. These methods have been
used to successfully map groundwater discharge across
large coastal areas (Charette et al., 2008).
Tracking groundwater flow. In freshwater systems,
U isotopes may be used to trace groundwater flow and
mixing, while Ra and Th are too strongly adsorbed to
be useful as water tracers. Aquifers with different char-
acteristics may generate different groundwater
(234U/238U) ratios that can serve as fingerprints and so
can be used in hydrologic studies to identify mixing
and flow patterns. For example, older groundwaters in
aquifers with low weathering rates may have higher
(234U/238U) ratios than those in shallow aquifers so that
the location and magnitude of fluxes into overlying
waters can be identified. Similarly, where there are
aquifers composed of very different rock types with
 AQUIFER CHARACTERISTICS (U-SERIES)
different weathering release rates of U, (234U/238U)
ratios can be used to trace groundwater mixing. This
can include identifying higher velocity flow paths
where waters enter an aquifer with distinctly different
characteristics (Roback et al, 2001).
Identifying groundwater inputs to surface waters. Simi-
larly, the contributions of waters from different ground-
waters into lakes and rivers can be identified using
(234U/238U) ratios, especially when coupled with other
tracers such as 87Sr/86Sr ratios. This can include the rel-
ative proportions of inputs from shallow and deep aqui-
fers or from aquifers of different rock types and
watersheds along a river (Riotte and Chabaux, 1999).
Estimating pollutant radionuclide migration rates. Deter-
mination of the behavior of naturally occurring U and
Th can be used to predict the mobility of released
anthropogenic actinides and so can be part of site risk
assessments for nuclear facilities. Naturally occurring
nuclides are present in very low concentrations, so that
the measured extent of adsorption cannot be directly
extrapolated to pollutants with much higher concentra-
tions and possibly accompanied by changes in aquifer
chemistry. However, they still provide a baseline for
establishing the relevant adsorption factors and may
remain relevant for lower levels of contamination
(Porcelli, 2008).
Determining “recent” mineral precipitation. The nature
of minerals precipitating within an aquifer reflects the
chemical conditions of the groundwaters. While rocks
within aquifers generally are composed of primary min-
erals having U and Th decay series nuclides within sec-
ular equilibrium, secondary minerals that incorporate
radionuclides from groundwater will initially show sub-
stantial secular disequilibrium. The resulting
230
Th-234U-238U compositions can be used to date the
hydrogenic minerals and so determine the timing and
history of precipitation from migrating solutions.
Locating redox boundaries. Under reducing conditions,
U forms relatively insoluble U4+ and precipitates as
UO2. As groundwater moves past a redox boundary,
groundwater U concentrations drop and the enhanced
recoil rates from fine-grained U oxides increase the (234-
U/238U) ratio, providing a way for mapping aquifer
redox fronts from U isotope distributions (Osmond
and Cowart, 2000).
Dating groundwaters. There have been various studies
that used U isotopes for dating groundwaters. If the
238
U concentration along a flow line is constant or
varies in a clearly defined way and the recoil input of
234
U is constant, then the (234U/238U) ratio evolves pri-
marily by decay and can be used to determine the age of
the groundwater (Ivanovich et al., 1991). However,
such conditions rarely are evident. For shorter time-
scales, Ra might be used. In highly saline groundwaters
where Ra is not adsorbed and where the recoil inputs of
Ra isotopes are constant, then the 228Ra activity will
increase over 20a until it reaches a steady state, and
so can be used to date young waters (Kiro et al.,
57
2013). In principle, the longer-lived isotope 226
Ra can
be used for waters with ages up to 5 ka.
Summary
Isotopes in the 238U, 235U, and 232Th decay series chains
provide numerous tools to study aquifer processes. With
a wide range of half-lives, these isotopes can be used to
investigate very rapid processes such as adsorption as well
as longer-term processes from precipitation to basin-wide
aquifer flow. The elements represented in the decay series
also have a range of chemistries, providing investigative
methods that are applicable to different aquifer aspects
and conditions. This includes the strongly adsorbed ele-
ments Th, Pb, and Po, the relatively mobile U, and Ra,
which is only strongly adsorbed in fresh waters. There
are some fundamental aspects of radionuclide behavior
that require further investigation, including the reversibil-
ity of adsorption, factors that may affect release from min-
erals by recoil or leaching, and the effect of colloids.
Progress in these fields, as well as new aquifer studies, will
undoubtedly lead to new methods for extracting informa-
tion on aquifer characteristics from the decay series.
Bibliography
Charette, M. A., Moore, W. S., and Burnett, W. C., 2008. Uranium-
and thorium-series nuclides as tracers of submarine groundwater
discharge. In Krishnaswami, S., and Cochran, J. K. (eds.), U-Th
Series Nuclides in Aquatic Systems. Amsterdam: Elsevier,
pp. 155–192.
Ivanovich, M., Frohlich, K., and Hendry, M. J., 1991. Uranium-
series radionuclides in fluids and solids, Milk River aquifer,
Alberta, Canada. Applied Geochemistry, 6, 405–418.
Kiro, Y., Weinstein, Y., Starinsky, A., and Yechieli, Y., 2013.
Groundwater ages and reaction rates during seawater circulation
in the Dead Sea aquifer. Geochimica et Cosmochimica Acta,
122, 17–35.
Krishnaswami, S., Graustein, W. C., Turekian, K. K., and Dowd, F.,
1982. Radium, thorium, and radioactive lead isotopes in ground-
waters: application to the in-situ determination of adsorption-
desorption rate constants and retardation factors. Water
Resources Research, 6, 1663–1675.
Ku, T.-L., Luo, S., Leslie, B. W., and Hammond, D. E., 1992.
Decay-series disequilibria applied to the study of rock-water
interaction and geothermal systems. In Ivanovich, M., and
Harmon, R. S. (eds.), Uranium-Series Disequilibrium. Oxford:
Clarendon, pp. 631–668.
Osmond, J. K., and Cowart, J. B., 2000. U-series nuclides as tracers
in groundwater hydrology. In Cook, P. G., and Herczeg, A. L.
(eds.), Environmental Tracers in Subsurface Hydrology. New
York: Springer, pp. 145–173.
Porcelli, D., 2008. Investigating groundwater processes using U-
and Th-series nuclides. In Krishnaswami, S., and Cochran,
J. K. (eds.), U/Th Series Radionuclides in Aquatic Systems.
Amsterdam: Elsevier. Radioactivity in the Environment, Vol.
13, pp. 105–153.
Porcelli, D., 2014. A method for determining the extent of bulk
210
Po and 210Pb adsorption and retardation in aquifers. Chemical
Geology, 382, 132–139.
Riotte, J., and Chabaux, F., 1999. (234U/238U) activity ratios in
freshwaters as tracers of hydrological processes: the Strengbach
watershed (Vosges, France). Geochimica et Cosmochimica Acta,
63, 1263–1275.
58 Ar–Ar AND K–Ar DATING
Roback, R. C., Johnson, T. M., McLing, T. L., Murrell, M. T., Luo,
S., and Ku, T.-L., 2001. Uranium isotopic evidence for ground-
water chemical evolution and flow patterns in the eastern Snake
River Plain aquifer, Idaho. Geological Society of America Bulle-
tin, 113, 1133–1141.
Cross-references
Chemical Weathering (U-Series)
Uranium Series, Opal
U-Series Dating
Ar–Ar AND K–Ar DATING
James K. W. Lee
Department of Geological Sciences and Geological
Engineering, Queen’s University, Kingston, ON, Canada
Department of Earth and Planetary Sciences, Macquarie
University, Sydney, NSW, Australia
Synonyms
(Argon–argon): Ar–Ar, 40Ar–39Ar, 40Ar/39Ar, 39Ar–40Ar,
39
Ar/40Ar, argon-40/argon-39; (Potassium–argon): K–Ar,
40
K–40Ar, 40Ar–40K
Definition
Potassium–argon dating. An absolute dating method
based on the natural radioactive decay of 40K to 40Ar used
to determine the ages of rocks and minerals on geological
time scales.
Argon–argon dating. A variant of the K–Ar dating method
fundamentally based on the natural radioactive decay of
40
K to 40Ar, but which uses an artificially generated iso-
tope of argon (39Ar) (produced through the neutron irradi-
ation of naturally occurring 39K) as a proxy for 40K.
K–Ar dating
The potassium–argon (K–Ar) geochronological method
is one of the oldest absolute dating methods and is based
upon the occurrence of a radioactive isotope of potassium
(40K), which naturally decays to a stable daughter isotope
of argon (radiogenic 40Ar, also known as 40Ar*). For this
reason, the K–Ar method is one of the few radiometric
dating techniques in which the parent (40K, a solid) is
a different phase from the daughter (40Ar, a gas). The
method was first suggested by Goodman and Evans
(1941) and one of the earliest K–Ar ages was published
by Smits and Gentner (1950). Because potassium is
a major or minor element in many minerals, the K–Ar dat-
ing technique has been used to date a diverse range of rock
types. A comprehensive and detailed overview of the
method can be found in Dalrymple and Lanphere (1969).
The conventional K–Ar method became widely used
soon after its development and can give reliable ages on
many rapidly cooled rocks (e.g., volcanics, dykes). There
are, however, a number of limitations with respect to the
interpretation of K–Ar ages (further discussed below) which
has led to a very limited use of the K–Ar method in
current studies. These limitations are largely overcome by
the Ar–Ar method, however, which has now superseded
the K–Ar method as the geochronological method of choice
in dating K-bearing minerals and is further described below.
Types of materials which can be dated
Any potassium-bearing solid material has the potential to
be dated by either the K–Ar or Ar–Ar methods. Because
many common rock-forming minerals contain K, the
Ar–Ar method is one of the most commonly employed
geochronological techniques in many geological studies.
The most common materials that are particularly well
suited to dating using the K–Ar or Ar–Ar technique
include:
K-feldspar Biotite Phlogopite Glasses (e.g., tektites)
Plagioclase Muscovite Nepheline Clays and slates
Hornblende Sericite Alunite Basalts
In general, it should be noted that pure, unaltered minerals
yield the best results.
Clays and materials composed of very small grains can
be problematic in Ar–Ar dating, depending on their grain
size. If the grain size is smaller than about 1 mm, an experi-
mental effect known as “39Ar recoil” associated with the
neutron irradiation of the sample can make the interpretation
of the results difficult; this is further discussed below. For
this reason, K–Ar dating may remain the preferred method
of dating clays and very fine-grained K-bearing minerals.
Natural isotopic abundances
Potassium has three naturally occurring isotopes with
the following natural abundances: 39K (93.2581 %),
40
K (0.01167 %), and 41K (6.7302 %) (Steiger and Jäger,
1977). Of these three isotopes, only 40K is radioactive.
Argon also has three naturally occurring isotopes with
the following natural abundances: 36Ar (0.337 %), 38Ar
(0.063 %), and 40Ar (99.6 %) (Nier, 1950; Steiger and
Jäger, 1977). Although all three isotopes are stable, 40Ar
is the only isotope that is also a radiogenic daughter prod-
uct (resulting from the decay of 40K). From these abun-
dances, the atmospheric ratio of 40Ar/36Ar [often
denoted as (40Ar/36Ar)a] is thus 295.5. This ratio is
required in the calculation of 40Ar–39Ar ages, as explained
below.
Radioactive decay scheme
The K–Ar method is unique among geochronological
schemes because the radioactive parent (40K) undergoes
a complex branched radioactive decay to both radiogenic
40
Ca and 40Ar (Figure 1).
The dominant mode of decay (89.5 % of the time) is the
emission of a b-particle leading to the formation of radio-
genic 40Ca. However, 40K can also decay via one of three
 Ar–Ar AND K–Ar DATING
electron 40K
capture
(10.32%)
40Ar
β− decay
electron (89.52%)
capture
(0.16%)
γ decay β+ and γ
decay
(~0.001%)
40
Ca
ground state
40Ar
ground state
Ar–Ar and K–Ar Dating, Figure 1 40K decay scheme (After
McDougall and Harrison, 1999), based on the decay constants of
Steiger and Jäger (1977).

other modes to radiogenic 40Ar. The most common mode
is via electron capture; this may result in the formation of
40
Ar in an excited state that subsequently drops to the
ground state with the additional release of a g-ray
(10.3 % of the time) or, less commonly, may directly form
at the ground state (0.16 % of the time). A very rare mode
of decay to 40Ar (only 0.001 % of the time) resulting from
the emission of a positron (b+) has also been hypothesized.
The most widely used set of decay constants for 40K is
published by Steiger and Jäger (1977): lb
¼ 4.962 
10
10 a
1 and le (all other branches combined) ¼ 0.581
 10
10 a
1 for a total decay constant of l ¼ 5.543 
10
10 a
1. Thus, the corresponding half-life of 40K is ln
2/l ¼ 1,250 Ma.
High-resolution geochronological studies integrating
data from different isotopic systems (e.g., the timing of
mass extinctions, the geological time scale) have gener-
ated widespread renewed interest throughout the geochro-
nological community in the accuracy and precision of the
40
K decay constants (l ¼ le + lb) due to the existence of
small but statistically significant age discrepancies of
same precisely dated events by the K–Ar and U–Pb sys-
tems. Renne et al. (1998) first noted an 2.1 % discrep-
ancy between the value of le used by geologists and that
used by nuclear physicists – a potential source of system-
atic error. Min et al. (2000) statistically evaluated the
nuclear physics database as well as estimated the total
decay constant l as calibrated with U–Pb and historical
dating; their findings led them to suggest tentative values
for l which differ from the currently used value by

1.4 % to
3.1 %, leading to an international call for an
59
improved set of all geochronological decay constants
(Begemann et al., 2001). After a careful intercalibration
of the widely used Fish Canyon sanidine (FCs) standard
with primary K–Ar standards, Jourdan and Renne
(2007) stated that there was an imminent need for the revi-
sion of the 40K decay constant l, particularly in the
electron-capture branch (i.e., to 40Ar). Kuiper et al. (2008)
published a unique study indirectly recalibrating l by dat-
ing sanidine from very precisely dated, astronomically
tuned ash beds; their revised estimate of l brings into syn-
chronicity the Ar–Ar and U–Pb ages of a number of glob-
ally significant geological events (e.g., K–T and P–Tr
boundaries). Recently, Renne et al. (2010) used a Monte
Carlo optimization methodology of all available
40
K activity data and integrated 238U–206Pb and
40
Ar–39Ar data pairs to estimate an optimal l and
corresponding uncertainties. Despite these advances,
however, a common set of updated K decay constants
has yet to be agreed upon and adopted by the global
scientific community.
K–Ar age equation
From first principles, it can be shown that the K–Ar age
equation is

40
1 l Ar
t ¼ ln 40 þ 1 ð1Þ
l le K
where t ¼ age, le ¼ ¼ 0.581  10
10 a
1, l ¼ 5.543 
10
10 a
1, and 40K and 40Ar* are the number of atoms
(e.g., moles) of each element, respectively. Note that the
asterisk (*) in 40Ar* is used by convention in the literature
to denote radiogenic 40Ar (i.e., derived from the
radioactive decay of 40K in the same sample), as opposed
to 40Ar from other sources, such as the atmosphere (often
denoted as 40Ara) or excess Ar from the Earth’s crust or
mantle (sometimes denoted as 40ArE).
Thus, in order to calculate a K–Ar age [which is
directly proportional to the 40Ar*–40K ratio according to
Eq. (1)], one must determine both the 40Ar* and 40K in
the sample.
Ar–Ar dating
Ar–Ar dating was introduced in the mid-1960s, with the
publication of a landmark paper by Merrihue and Turner
(1966). The most comprehensive and detailed overview
of all aspects of the method can be found in McDougall
and Harrison (1999) although some of the data interpreta-
tion sections tend to reflect author biases. Dickin (2005)
also provides a more concise summary of the technique.
The Ar–Ar method is fundamentally based on the
K–Ar method. However, instead of directly measuring
40
K, the sample is placed in a nuclear reactor and
bombarded with fast neutrons (having energies
>0.1 MeV) to transform a proportion of the potassium
(specifically, the fraction associated with 39K) to another
isotope of argon, 39Ar. The process results in the addition
60 Ar–Ar AND K–Ar DATING

of the neutron (n) to the 39K nucleus, resulting in the ejec- Common practice is to gather all of the constant terms
tion of a proton (p) and energy in a process that can be to define an irradiation parameter J, where
summarized as follows:
 Z
l 39 K
39
19 K þ n ! 39
18 Ar þ p þ energy ð2Þ J ¼ 40 D fðE ÞsðE ÞdE ð6Þ
le K ab
E
Equation (2) can also be written in the abbreviated form
39
K(n, p)39Ar. 39Ar produced in this way is commonly so that substitution into Eq. (5) yields
designated as 39ArK denoting the fact that is was derived (
 )
from the neutron irradiation of K. Because 39ArK is not 1 40
Ar

a naturally occurring argon isotope, it is directly propor- t s ¼ ln J 39 þ1 ð7Þ

l ArK s
tional to the amount of 39K (and hence 40K, because the
natural abundances of the K isotopes are known, i.e., the or rearranging,
39
K–40K ratio is known and constant, as stated above) in
the sample. Thus, 39Ar serves as an effective proxy for elts
1 eltu
1
40
K (and therefore, the 40Ar*–39ArK ratio is directly J ¼ 40  ¼   ð8Þ
Ar
=39 ArK s 40
Ar
=39 ArK u
proportional to the 40Ar*–40K ratio).
However, for a sample of unknown age (herein desig-
Ar–Ar age equation nated by the subscript u), we obtain from Eq. (7) an anal-
The 40Ar–39ArK ratio cannot be directly substituted for the ogous age equation for the unknown:
40
Ar*–40K ratio in Eq. (1) because the amount of 39Ar (
 )
generated from the neutron irradiation will clearly be 1 40
Ar

a function of the amount of 39K in the sample, the duration t u ¼ ln J 39 þ1 ð9Þ

l ArK u
of the irradiation, the neutron flux, and the proportion of
fast neutrons within the neutron flux, i.e., Thus, by substituting Eq. (8) into Eq. (9), we obtain the
Z Ar–Ar age equation:
39
ArK ¼ KD fðE ÞsðE ÞdE
39 (  )
1
40
Ar
=39 ArK u  lts 
E t u ¼ ln 40
39  e
1 þ1 ð10Þ

39  Z l Ar = ArK s
K
¼ 40 K 40 D fðE ÞsðE ÞdE ð3Þ
K ab This equation can be compared with the K–Ar age
E equation [Eq.(1)]. Significantly, this shows that an age
where D is the duration of the irradiation, f(E) is the neu- standard must always be irradiated with an unknown
tron flux at energy E, s(E) is the neutron capture cross sec- sample.
tion at energy E for the conversion reaction described in
Eq. (2), and (39K–40K)ab is the 39K–40K ratio from the Sample preparation
natural atomic abundances. Sample preparation for both K–Ar and Ar–Ar dating is
Substitution of Eq. (3) into Eq. (1) thus yields generally similar. In both cases, the most important crite-
0 2 3 1

 Z rion is the use of fresh, unaltered samples which are free
1 @ l 40 Ar
4 39 K
t ¼ ln D fðE ÞsðE ÞdE5 þ 1A of inclusions or other contaminants. Also, it is essential
l le 39 ArK 40
K ab to know the geological context of the minerals to be dated
E
so thin-section petrography of the sample to be dated
ð4Þ should be the first step. Where aliquots of a mineral are
The determination of the neutron irradiation parameters required, rock samples are usually crushed to the grain
in Eq. (4) is not trivial and will vary for each irradiation. size of the minerals to be dated. After sieving and ultra-
Therefore, the Ar–Ar age determination can be greatly sonic washing in acetone and distilled water, mineral sep-
simplified if an age standard (sometimes called arates are obtained through standard techniques, such as
a monitor mineral) is used. If the age of the standard is the use of magnetic separators or heavy liquids. Ultrapure
known (designated as ts), then Eq. (4) becomes mineral separates are required because even a small
(
amount of impurities or other minerals can contaminate
 an argon analysis. As a result, mineral separates are usu-
1 l 40 Ar

ts ¼ ln ally inspected carefully under a binocular microscope

l le 39 ArK s and only pristine grains (i.e., free from inclusions, defects,
2 3 9 ð5Þ

39  Z = or other mineral contaminants) are retained. In this way,
4 K mineral separates with >99.9 % purity can be obtained.
D fðE ÞsðE ÞdE 5 þ 1
40
K ab ; The application of lasers in Ar–Ar dating has greatly
E simplified the sample preparation process. Because much
 Ar–Ar AND K–Ar DATING
less sample is required, e.g., single crystals are often ana-
lyzed, mineral separation can be relatively quick. Further-
more, the use of lasers as age microprobes has allowed the
analysis of minerals in situ, typically in a polished thick
(100–200 mm) section.
In Ar–Ar dating, the samples to be dated (either single
crystals, aliquots of crystals, or a thick section) are com-
monly wrapped in Al foil or encapsulated in silica tubes,
along with the age standard, and then irradiated in
a nuclear reactor. The age of a suitable standard and the
optimum duration of irradiation should all be carefully
considered at this stage. More details can be found in
McDougall and Harrison (1999).
Analytical methods
The following is a general summary of the analytical
methods employed in both K–Ar and Ar–Ar dating.
Detailed descriptions of the methodology for each dating
technique can be found in Dalrymple and Lanphere
(1969) and McDougall and Harrison (1999), respectively.
Potassium
The analysis of potassium typically involves
wet-chemical or physical methods.
Most K–Ar studies in the early literature employed the
use of flame photometry to determine K concentration,
which is based on the fact that all elements emit character-
istic photons when excited electrons drop back to their
ground state. In general, the rock or mineral is dissolved
in acid and unwanted elements (such as Si) are removed
through a variety of chemical reactions. By heating the
potassium-enriched solution with a gas–air flame, mea-
surement of the distinct photon wavelengths emitted by
the excited electrons will distinguish the various elements,
and the measurement of the intensity of the radiation is
directly proportional to the amount of the element in the
sample. Unknown samples are calibrated by also analyz-
ing standard solutions of various (known) potassium con-
centrations and by adding an internal alkali metal standard
(typically Li) to the unknown solution. Uncertainties are
typically on the order of 1 %.
Other analytical methods that have been used less fre-
quently to measure K concentrations include atomic
absorption spectroscopy, based on the absorption of radi-
ant energy by potassium atoms, which produce an absorp-
tion spectrum characteristic of that element; X-ray
fluorescence, based on the X-ray emission spectrum of
potassium resulting from bombarding the sample with an
X-ray beam; neutron activation, in which an artificial
unstable isotope of K (42K) is created by irradiating the
sample with neutrons to convert 41K and its subsequent
g-decay is measured; and isotope dilution, in which
a known quantity of an isotopic tracer (known as
a “spike,” typically 41K) is added to the unknown solution
before measuring all of the K isotopes in a solid-source
mass spectrometer.
61
Argon
Argon in both K–Ar and Ar–Ar dating is measured in a
two-step process typically involving extraction and purifi-
cation from the mineral sample and then measurement and
calculation of the 40Ar*. Because argon is a gas, the easiest
way to extract argon is to fuse the sample at high temper-
ature. This is usually done in a high-vacuum system using
a radio-frequency (induction) or resistance furnace or
a laser (in Ar–Ar dating only). Because other gases, such
as N2, O2, or CO2, can also be released, the argon is puri-
fied through the use of liquid nitrogen traps, Zr-alloy get-
ters, and Ti sponges. The purified argon is then analyzed in
an ultrahigh vacuum system in a noble gas mass spectrom-
eter. Simply, the argon that enters the mass spectrometer is
first ionized in the ion source. The resultant beam of pos-
itive Ar ions is collimated and then accelerated through
both electric and magnetic fields that separate the beam
into a series of beams that follow a curved flight path, each
with a distinct charge/mass ratio. Older mass spectrome-
ters had only a single detector to measure the incoming
beam current (proportional to the number of Ar atoms),
but latest-generation mass spectrometers now have
multicollector capability.
In K–Ar dating, because 40Ar* and 40K must be mea-
sured separately, a method is required to determine the
absolute concentration of 40Ar* in the unknown sample.
The most widely used method is by isotope dilution. In
this method, a known quantity of an argon tracer (also
called a “spike”) is mixed with the argon from the
unknown sample. The distinguishing characteristic of
the spike is that it must have a known argon isotopic com-
position, i.e., the ratio of 36Ar–38Ar–40Ar, different to the
natural abundance of these isotopes. Since the isotopic
composition of the mixture can be measured in the mass
spectrometer and the argon isotopic composition and
amount of spike are known, it is then possible to calculate
the isotopic composition and amount of the argon from the
unknown. Argon tracers are typically enriched in 38Ar,
and all three naturally occurring argon isotopes (36Ar,
38
Ar, and 40Ar) are measured in the mass spectrometer.
Dalrymple and Lanphere (1969) describe the isotope dilu-
tion process in more detail.
In Ar–Ar dating, only the ratio of two argon isotopes
(40Ar* and 39ArK) is required in the age equation
[Eq. (10)], so the argon analysis is, in principle, much
simpler than in K–Ar dating because absolute concentra-
tions are not required. However, because of neutron-
induced interference reactions (described below), five
argon isotopes (36Ar, 37Ar, 38Ar, 39Ar, and 40Ar) must be
measured.
Data corrections
There are several important corrections that must be
applied to the raw data in order to calculate an
accurate age.
62 Ar–Ar AND K–Ar DATING
Blank correction
Although argon is measured in an ultrahigh vacuum envi-
ronment in both K–Ar and Ar–Ar dating, system blanks
from both the extraction line and mass spectrometer must
be subtracted from the total amount of each argon isotope
measured.
Atmospheric correction
This is the most fundamental correction in both K–Ar and
Ar–Ar dating. Because 40Ar is the most abundant isotope
of argon in the atmosphere, it can be adsorbed onto the sur-
face of mineral grains or analytical equipment or can be
incorporated into minerals during crystallization, so
a correction must be made to subtract this atmospheric
component (40Ara) from the total amount of 40Ar mea-
sured. Since it is impossible to distinguish 40Ara from
40
Ar*, another isotope of argon present in the atmosphere
(36Ara) is also measured in the mass spectrometer.
Since the atmospheric ratio (40Ar/36Ar)a is 295.5, the
atmospheric 40Ar component is easily calculated as
40
Ara ¼ 295.5  36Ara. Depending on the age of the sam-
ple or the amount of gas being measured, this correction
can be significant, e.g., for young samples or samples with
very low K.
Neutron-induced interferences
In Ar–Ar dating, neutron bombardment of most minerals
generates numerous reactions with elements other than
K that produce a variety of argon isotopes. These
neutron-induced interferences must therefore be corrected
in order to obtain accurate ages. The most important reac-
tions are [cf. Eq. (2)] 40Ca(n, na)36Ar, 40Ca(n, a)37Ar
(37Ar is radioactive and decays to 37Cl with a half-life of
35.1 days), 37Cl(n, g)38Cl ! 38Ar (38Cl is radioactive
and decays to 38Ar with a half-life of 37.3 min), 42Ca
(n,a)39Ar, 39K(n,p)39Ar, and 40K(n,p)40Ar. From these
reactions, it can be seen that neutron-induced interferences
will affect the measurement of five argon isotopes (36Ar,
37
Ar, 38Ar, 39Ar, and 40Ar) and are the reason why all of
these isotopes must be measured. McDougall and Harri-
son (1999) describe in detail the added complexities in
the data correction process resulting from neutron irradia-
tion in Ar–Ar dating.
Radioactive decay correction
In Ar–Ar dating, two of the artificial argon isotopes pro-
duced by neutron irradiation (37ArCa and 39ArK) are them-
selves radioactive with half-lives that are on a time scale of
days to years. 37ArCa has a half-life of about 35.1 days and
39
ArK has a half-life of 269 years (Stoenner et al., 1965),
so further corrections for the time interval between neu-
tron irradiation and analysis in the laboratory must also
be made.
Modes of analysis
Conventional (total fusion) age
The analysis of a sample in conventional K–Ar
dating produces a single age, since the sample must
be completely dissolved and fused in order to extract
the 40K and 40Ar, respectively. The analogous process
(total fusion) can be performed in the Ar–Ar analysis of
a sample to yield a single, total fusion age. Because of
the decreased sensitivity of early-generation mass spec-
trometers, several milligrams of purified mineral separates
often had to be analyzed to obtain enough argon to calcu-
late an age. Improved sensitivity of current-generation
mass spectrometers now permits the Ar–Ar analysis of
single crystals which can be fused with either a furnace
or a laser. In fact, lasers can be used as age microprobes
to obtain intragrain Ar–Ar age analyses, in which
a small amount of material can be ablated by the laser
within a mineral grain and yields a total fusion age for that
spot. Ar–Ar laser microprobes thus permit the direct deter-
mination of the 40Ar (age) distribution within single crys-
tals. For more information about single-crystal laser
dating, see the entry “Single-Crystal Laser Fusion.”
Ar–Ar step-heating
Because the Ar–Ar method only requires the extraction of
gaseous argon isotopes, Merrihue and Turner (1966) noted
that the sample did not have to be fused in a single step,
but could be incrementally heated in steps of increasing
temperature, and the argon isotopes extracted during each
temperature step could be analyzed to yield an age for that
step. This incremental-heating (or step-heating) technique
thus was a major advantage over conventional K–Ar dat-
ing because it provided a way to examine the distribution
of 40Ar (and hence, ages) within minerals. The fundamen-
tal principle is based on solid-state diffusion theory and is
illustrated in Figure 2.
Consider a homogeneous spherical mineral grain. If the
grain is heated at a relatively low temperature (T1) for
a finite period of time, the argon isotopes diffusing out
of the grain can only come from a thin region (i.e., shell)
along the boundary of the grain and will yield an age t1
associated with that region. If the grain is now heated at
a slightly higher temperature (e.g., T2 > T1) for the same
length of time, the argon isotopes will diffuse from
a spherical region that extends slightly deeper into the
grain and will yield an age t2. In this way, successively
higher steps of increasing T will sample regions deeper
and deeper into the grain until the highest temperature
steps (e.g., T9) will sample the argon from the grain core.
The age of each step is commonly referred to as an appar-
ent age because it may not necessarily be identical to the
overall age of the mineral grain.
By plotting the apparent ages of each of the steps (t1–t9)
in a diagram of apparent age versus the cumulative
fraction of 39ArK released, a so-called apparent age spec-
trum can be constructed. For example, consider the same
spherical grain in Figure 2. If 39K is homogeneous
 Ar–Ar AND K–Ar DATING
T1
T9
Ar–Ar and K–Ar Dating, Figure 2 Hypothetical spherical grain
subject to step-heating. By heating the grain in incremental
steps of increasing temperature, argon (and hence, the age
distribution within the grain) can be sampled progressively from
the grain boundary at the lowest temperature (T1 – dark blue) to
the grain core at the highest temperature (T9 – red).
throughout the grain (i.e., K is not chemically zoned), then
the 39ArK produced from the fast-neutron irradiation of the
39
K should also be uniformly distributed throughout the
grain from core to rim. If the 40Ar distribution is uniform
throughout the grain, then 40Ar–39ArK ratio (and thus,
apparent age) should be identical for every step (t1–t9)
and the resultant apparent age spectrum yields
a “plateau” age (Figure 3). The significance of a plateau
age is discussed in the following section.
Ar–Ar data interpretation
Because argon is a noble gas, its mobility within crystals is
strongly dependent on temperature. Thus, the interpreta-
tion of Ar–Ar age spectra is based primarily on solid-state
(volume) diffusion theory.
Plateau ages
There is no convention that is widely accepted by the geo-
chronology community as a standard definition of
a plateau in an Ar–Ar age spectrum. Thus, the definition
of a plateau should be explicitly stated where ambiguity
may result. Common elements in any definition of
a plateau generally include a series of contiguous steps
which overlap within error and which comprise
a significant fraction of the 39ArK released from the sam-
ple. For example, the Fleck et al. (1977) definition states
that a plateau is defined by three or more contiguous steps
63
Ar–Ar and K–Ar Dating, Figure 3 Apparent age spectrum for
a hypothetical grain with a uniform distribution of 40Ar and 39ArK
throughout the grain. The horizontal length of each step
corresponds to the fraction of 39ArK from the total 39ArK released
and the vertical width corresponds to the apparent age and
corresponding age uncertainty of that step (in this example,
2s). The color for each step corresponds to the region of the
mineral grain in Figure 2 from which the age of that step is
derived. The weighted mean age from all of the steps yields an
“integrated” age for the sample, which is equivalent to the age
that would have been obtained if the sample was fused in
a single step. If a significant fraction of the age spectrum
(typically >50 %) is composed of steps having identical ages
within error, then the weighted mean age of those steps is said
to define a plateau age.
comprising at least 50 % of the 39ArK released which over-
lap at the 2s level. The plateau age is calculated as the
weighted mean of the ages of all steps included in the pla-
teau, with the inverse of the variance of the age of each
step used as the weights. Examples of plateau age spectra
are shown in Figures 3, 4, and 5.
The existence of a plateau is, in general, interpreted to
indicate a uniform distribution of 40Ar* within that region
of the mineral grain(s) sampled by those steps, and thus,
a plateau age is often considered to be geologically mean-
ingful. Understanding the geological context of the dated
sample is critical in accurately interpreting a plateau age,
however, because of argon’s ability to diffuse under ele-
vated temperature conditions. For example, if the dated
sample comes from a rapidly cooled igneous dyke which
has undergone no metamorphism since intrusion, the pla-
teau age may represent the actual formation age of the
dyke. If the dated sample comes from a metamorphosed
rock, however, the plateau age may represent the age of
metamorphism rather than the protolith age, if the rock
has been reheated to a temperature high enough for any
accumulated argon to diffuse out of the mineral at the time
of metamorphism, effectively “resetting” the Ar–Ar age
of the mineral.
64 Ar–Ar AND K–Ar DATING
Ar–Ar and K–Ar Dating, Figure 4 Ar–Ar age spectrum
exhibiting an argon-loss profile as defined by steps 1–3. Note
that steps 4–9 still define a plateau with an age of ~23.6 Ma.
Ar–Ar and K–Ar Dating, Figure 5 Ar–Ar age spectrum
exhibiting an excess-argon profile as defined by steps 1–3. Note
that steps 4–9 still define a plateau with an age of ~23.6 Ma.
An added complication in interpreting plateau ages
arises in the step-heating of hydrous minerals such as
amphiboles and micas. Gaber et al. (1988), Lee
et al. (1991), Sletten and Onstott (1998), and Lo
et al. (2000) found that during an Ar–Ar experiment,
hydrous minerals may not remain stable at elevated tem-
peratures due to oxidation and dehydroxylation of the
crystal structure while being step-heated in vacuo. This
structural decomposition of the crystal can lead to the for-
mation of numerous extended defects throughout the
grains and result in the homogenization of argon gradients
which could lead to the generation of an apparent plateau
(i.e., a pseudo-plateau) which has no geological signifi-
cance. Thus, the plateau ages from such samples should
be interpreted with caution.
Ar loss
There are three main scenarios in which an age spectrum
showing apparent Ar loss can be generated. The first is
due to reheating. If a mineral has experienced
a moderate–high T geological event, such as proximity
to an igneous intrusion, contact metamorphism, hot fluid
flow, etc., argon can diffuse out of the mineral. If argon
diffusion is not sustained for protracted periods of time,
it is possible that the mineral may experience only partial
argon loss, such that argon is lost from a limited region
of the grain nearest the grain boundary and the original
argon is retained in the grain core. In this case, an apparent
age spectrum may look like that in Figure 4. A second sce-
nario involves slow regional cooling. If a mineral is sub-
ject to slow cooling resulting from regional-scale
metamorphism or exhumation, partial argon loss can
occur over long periods of time if the rate of cooling is
not rapid enough to retard Ar diffusion. This will also
yield an age spectrum similar to Figure 4. Finally, a third
scenario involves impurities (e.g., inclusions) in the ana-
lyzed crystals, which often degas in the lower T steps
and can cause an apparent Ar-loss profile. Fortunately, it
is often possible to determine if impurities are the cause
by examining other argon isotopes (see section “Ca/K
and Cl/K Spectra” below). Therefore, knowledge of the
sample and its local geological context is important in
interpreting such age spectra.
Excess 40Ar
Thermal activation can also result in the diffusion of 40Ar
from the external environment into mineral grains, if the
chemical potential of argon (often referred to as the argon
partial pressure) is much greater outside the crystal than
inside. Because this 40Ar is not associated with the
in situ decay of 40K (i.e., radiogenic 40Ar*), it is com-
monly referred to as excess Ar (40ArE). There are many
possible sources of excess Ar. One source is the host rock
itself; if there are K-bearing minerals, local thermal
reheating can release some of the 40Ar* to the surround-
ings, where it then becomes 40ArE. Another common
source is hot Ar-rich fluids which may carry 40ArE from
the lower crust or distal regions. Incorporation of 40ArE
from the external environment means that age spectra tend
to have anomalously old ages in the lower T steps
(Figure 5).
Sometimes, anomalously old ages are reflected in the
highest T steps. Although the reasons for this will vary
from sample to sample, one potential source is 40ArE
released from contaminant phases (e.g., pyroxene). Such
phases may only release excess argon at high T or the
phases may occur as internal inclusions which can only
degas when the host mineral itself decomposes at high T.
 Ar–Ar AND K–Ar DATING
Ar–Ar and K–Ar Dating, Figure 6 Ar–Ar age spectrum
exhibiting an excess-argon profile in the low and high T steps.
This is sometimes referred to as a “saddle-shaped” age
spectrum.
An example of an age spectrum displaying excess argon in
the low and high T steps is shown in Figure 6. An excellent
review of the role of excess Ar in Ar–Ar dating may be
found by Kelley (2002).
39
Ar recoil
When a fast neutron interacts with a 39K nucleus to form
39
Ar during neutron irradiation, there is a recoil energy
also associated with this process. Turner and Cadogan
(1974) calculated that the average recoil distance which
an 39Ar atom can move from the original position of the
39
K atom is 0.08 mm, which has been subsequently con-
firmed by direct experiment (Villa, 1997). Although the
39
Ar recoil distance is very short compared to the length
scale of most mineral grains (and therefore, will have
a negligible effect in such cases), it can have
a significant effect for very fine-grained materials (e.g.,
1 mm) or samples with mm-scale domains, where the
length scale of argon diffusion is of the same order of mag-
nitude. In particular, it has been shown that significant
39
Ar loss can occur in minerals such as clays since it can
be recoiled entirely out of the crystal structure (e.g.,
Halliday, 1978). This has led to the development of encap-
sulation techniques when performing 40Ar–39Ar dating on
such fine-grained minerals, in which the sample is placed
in a vacuum-evacuated, sealed capsule before irradiation
in order to capture the ejected 39Ar, which can then be
measured subsequently in the mass spectrometer by
breaking the capsule in vacuo and then reintegrating the
measured 39Ar trapped in the capsule into the step-heating
results (e.g., Dong et al., 1995, 1997).
Alternatively, the recoiled 39Ar may not be lost to the
surrounding “void” but may become tightly imbedded in
an adjacent K-poor phase. This is illustrated in Figure 7.
65
n 40Ar 39Ar
n
40Ar 39Ar
n
40Ar
n
39Ar
39Ar
n
Ar–Ar and K–Ar Dating, Figure 7 39Ar recoil. Recoil energy
from the fast-neutron conversion of 39K can either eject 39Ar
atoms or reimplant them in adjacent phases, where they can
be liberated at high temperature during a step-heating
experiment.
For example, Huneke and Smith (1974) and Lo and
Onstott (1989) observed this phenomenon occurring
between a K-rich phase and olivine and biotite and chlo-
rite, respectively. In such cases, the resultant age spectrum
can display anomalously high ages in the low T steps
(reflecting the loss of 39Ar recoiled near the grain bound-
ary) monotonically decreasing to anomalously low ages
in the high T steps as the 39Ar is degassed from the
K-poor phases. Such an age spectrum is shown in
Figure 8.
For this reason, Ar–Ar studies involving fine-grained
samples must interpret the results with care. In such spe-
cific situations, it may also be worthwhile to consider
using conventional K–Ar dating, which avoids such
difficulties.
Ca/K and Cl/K spectra
One useful aspect about the 40Ar–39Ar technique is that it
has the ability to yield significant information about the
chemical composition of the samples that are being dated.
The foundation of the 40Ar–39Ar technique (through the
39
K(n,p)39Ar reaction) already provides chemical infor-
mation on the K content in the sample. As noted in the sec-
tion on Data Corrections, however, there are also other
neutron-induced interferences, leading to the production
of 37Ar from 40Ca and 38Ar from 37Cl. Thus, an 37Ar–39Ar
ratio is directly related to a Ca/K ratio and an 38Ar–39Ar
ratio is directly related to a Cl/K ratio. Examples of Ca/K
and Cl/K spectra are shown in Figure 9.
These argon isotope ratios can provide useful informa-
tion on the degassing of phases during a step-heating run.
For example, consider Figures 4 and 9. These spectra were
all derived from the same sample, and a comparison of the
first three steps in all of the spectra shows that the apparent
argon-loss profile in Figure 4 actually corresponds to
argon released from a high Ca/K, low Cl/K reservoir, in
66 Ar–Ar AND K–Ar DATING

contrast to a reservoir which has a uniform Ca/K and Cl/K Isochron diagrams
over the steps corresponding to the plateau. This strongly The total amount of 40Ar measured by a mass spectrome-
suggests that the first three steps showing apparent argon- ter during the analysis of an unknown for both K–Ar
loss profile do not necessarily reflect a geologically mean- and Ar–Ar dating is a function of two main sources of
ingful event, but rather likely reflect argon released from 40
Ar – radiogenic 40Ar from the in situ decay of 40K in
a small amount of a contaminating phase, such as plagio- the sample (40Ar*) and any initial 40Ar from the atmo-
clase or epidote. Note that this also does not necessarily sphere (40Ari). In other words,
affect the validity of the plateau age, but rather helps to
discern the geological significance of various portions of 40
Artotal ¼ 40 Ari þ 40 Ar
ð11Þ
an apparent age spectrum. A very illustrative study dem-
onstrating the use of Ca/K spectra coupled with electron This is analogous to many other geochronological sys-
microprobe analyses to interpret a complex age spectrum tems, e.g., initial Sr in Rb–Sr and initial Pb in Pb–Pb. By
is by Hanes et al. (1985). substituting the K–Ar age equation of Eq. (1) into
Eq. (11), we find


le 40  lt 
40
Artotal ¼ 40 Ari þ K e
1 ð12Þ
l
The difficulty in using Eq. (12) is that it is impossible to
know exactly how much 40Ari is in the sample, so to get
around this, we divide Eq. (12) by stable,
non-radiogenic, naturally occurring Ar isotope (36Ar) to
get

40 
40 

Ar Ar le  lt  40 K
¼ þ e
1 36 ð13Þ
36
Ar total 36
Ar i l Ar
In a plot of (40Ar/36Ar) versus (40K/36Ar), Eq. (13) has
the form of a straight line y ¼ mx + b. Note that the slope
m ¼ (le/l)(elt–1) can therefore be used to determine the
age t, and the y-intercept b ¼ (40Ar/36Ar)i is often the
atmospheric ratio (40Ar/36Ar)a ¼ 295.5.
In K–Ar dating, several minerals which have varying
amounts of K (e.g., amphibole, biotite, and K-feldspar)
Ar–Ar and K–Ar Dating, Figure 8 Ar–Ar age spectrum from a rock sample are usually analyzed and the results
exhibiting the effects of 39Ar recoil artifacts. Anomalously high are plotted on such a diagram. If the data form a straight
ages are seen in the first few steps which gradually decline to line, this is termed a mineral isochron and the slope can
anomalously low ages in the final few steps. therefore be used to calculate the age of the sample.

Ar–Ar and K–Ar Dating, Figure 9 (a) Ca/K spectrum showing elevated Ca/K ratios in the first three steps in comparison to the
remaining steps. (b) Cl/K spectrum showing lower Ca/K ratios in the first three steps in comparison to the remaining steps. These
spectra correspond to the same sample analyzed in Figure 4. Such spectra can be useful in interpreting degassing patterns in Ar–Ar
step-heating experiments.
 Ar–Ar AND K–Ar DATING 67

An analogous procedure can be used to analyze whole- associated with measuring 36Ar is much larger in relative
rock samples across a region, in which case a resultant line terms and so this error correlation is significant.
would be termed a whole-rock isochron. The goodness-of- To reduce this large correlation of errors, the inverse
fit of such lines is usually indicated by the mean square of isochron (also known as isotope correlation) diagram
the weighted deviates (MSWD) (McIntyre et al., 1966; places the largest measured peak (40Ar) in the
York, 1969). denominator.
In Ar–Ar dating, we substitute Eq. (9) into Eq. (11) to Dividing Eq. (14) by 40Artotal
get  lt 
39 
 lt 
40
Ari e
1 Ar
e
1 39 1 ¼ 40 þ 40
ð16Þ
40
Artotal ¼ Ari þ
40
ArK ð14Þ Artotal J Ar total
J
If 40Ari ¼ Ari • u, where u is the initial (40Ar/36Ar)i
36
and dividing again by a stable, non-radiogenic, naturally ratio, then
occurring Ar isotope (36Ar), we find  lt 
39 

40 
40   lt  Ari u
36
e
1 Ar
Ar Ar e
1 39 ArK 1 ¼ 40 þ ð17Þ
36
¼ 36 þ 36 ð15Þ Artotal J 40
Ar total
Ar total Ar i J Ar
and dividing by u:
This is analogous to the K–Ar isochron equation  lt 
39 

36 
[Eq. (13)], so in a plot of (39Ar/36Ar) versus (40Ar/36Ar), Ar 1 e
1 Ar
Eq. (15) has the form of a straight line y ¼ mx + b. Note that ¼
ð18Þ
the slope m ¼ (elt–1)/J can therefore be used to determine
40
Ar total u Ju 40
Ar total
the age t, and the y-intercept b ¼ (40Ar–36Ar)i is often the In a plot of (39Ar/40Ar) versus (36Ar/40Ar), Eq. (18) has
atmospheric ratio (40Ar–36Ar)a ¼ 295.5. An example of the form of a straight line y ¼ mx + b. Note that in this case,
such an isochron diagram is shown in Figure 10. y-intercept b ¼ 1/u is often the inverse of the atmospheric
One feature of such isochron diagrams in Ar–Ar dating ratio (40Ar
36Ar)a ¼ 1/295.5 ¼ 3.384  10
3, and the
is that there is a strong correlation of uncertainties between x-intercept (when the 36Ar–40Ar equals zero) is simply
the abscissa (39Ar–36Ar) and ordinate (40Ar–36Ar), shown 39
Ar/40Ar* ¼ J/(elt–1) from Eq. (8) and therefore yields
by the rotated error ellipses which lie subparallel to the an intercept age t. An example of an inverse isochron
line in Figure 10. This is a result of the common denomi- diagram is shown in Figure 11.
nator (36Ar) in each; because 36Ar is such a small peak to The data which form a line on an isotope correlation
measure in relation to 40Ar, the relative uncertainty diagram can be interpreted to reflect mixing between two
Ar reservoirs (or components) in the sample. More
complex mixing patterns between different argon reser-
voirs within a crystal may also be discerned using
such diagrams (e.g., Phillips and Onstott, 1988). Some-
times, Ar–Ar isochron diagrams may be somewhat mis-
leading because nonlinear data may appear to have
a good linear correlation because of the strong correlation
of errors; however, Dalrymple et al. (1988) have
shown that there is no difference between isochron versus
inverse isochron plots as long as rigorous mathematics for
the least squares fitting of lines with correlated errors
is used.

Probability distribution diagrams

Ar–Ar and K–Ar Dating, Figure 10 Ar–Ar isochron diagram
using the data obtained from the sample shown in Figure 4.
Note that the first three temperature steps do not lie on a line
corresponding to the steps defining the plateau. A best-fit line
through these steps yields a slope corresponding to an Ar–Ar
isochron age of 23.6 Ma, which is identical to the plateau age.
Probability distribution diagrams (Figure 12) are generally
used to plot the cumulative results of a group of single-
grain total fusion analyses. Analogous to histograms, the
diagrams plot the age on the x-axis versus the relative
probability (i.e., the sum of the normal distributions of
the individual analyses of single-grain total fusion ages
based on their experimental uncertainties; Deino and
Potts, 1992). They are particularly useful in detrital prov-
enance studies where source regions of different ages
can be distinguished.
68 Ar–Ar AND K–Ar DATING
Ar–Ar and K–Ar Dating, Figure 11 Ar–Ar inverse isochron
diagram using the data obtained from the sample shown in
Figure 4. Note that the first three temperature steps do not lie on
a line corresponding to the steps defining the plateau. A best-fit
line through these steps yields an intercept age of 23.6 Ma,
which is identical to the plateau age in Figure 4 and the isochron
age in Figure 10.
Argon terminology
There are several possible sources of argon in rocks. The
following glossary describes these various sources:
Artificially induced Ar – argon (typically 36Ar, 37Ar, 38Ar,
39
Ar, or 40Ar) produced in a mineral during fast-neutron
irradiation in a nuclear reactor, due to neutron interac-
tions with Cl, Ca, and K.
Atmospheric Ar – argon from the atmosphere, may be
either 36Ara, 38Ara, or 40Ara. The present-day
40
Ara/36Ara is 295.5.
Cosmogenic Ar – any Ar isotope (typically 36Ar, 37Ar,
38
Ar, 39Ar, or 40Ar) produced from cosmic ray interac-
tions with various elements (including Ca, Ti, Fe) in
a mineral, mainly through spallation reactions, but also
through neutron capture, normally only a concern when
dealing with extraterrestrial samples.
Excess Ar – any Ar (typically 40Ar) which is incorporated
into minerals by processes other than the in situ radioac-
tive decay of 40K in the crystal, includes inherited Ar
but may also include Ar derived from physical contam-
ination by other minerals or Ar introduced into a crystal
from other sources after its formation, e.g., by metamor-
phic fluids.
Inherited Ar – argon (typically 40Ar) incorporated into the
crystal lattice of a mineral at the time of its formation
which is NOT associated with the radiogenic 40Ar* pro-
duced from the in situ decay of 40K in the crystal (see
excess Ar). This is one form of excess Ar.
Radiogenic Ar – 40Ar* formed from the in situ decay of
40
K in a mineral.
muscovite; n = 58
Relative Probability
850 900 950 1000 1050
Apparent 40Ar/ 39Ar Age (Ma)
Ar–Ar and K–Ar Dating, Figure 12 Probability distribution
diagram of the Ar–Ar ages from 58 single crystals of detrital
muscovite derived from a sandstone. The higher the peak, the
greater the number of grains of that age. In this example, the
muscovite grains appear to be derived from at least three
different sources or regions – a dominant source with an age of
880–890 Ma, another source with an age of ~950 Ma, and a third
(minor) source with an age of ~1,000 Ma.
Trapped Ar – argon that is trapped within a rock or min-
eral at the time of its formation or during a subsequent
event, can exist in the crystal lattice (¼ inherited Ar)
or as fluid inclusions (which would typically be
expected to contain Ar of atmospheric composition for
terrestrial samples).
Applications
K–Ar
At present, there are two main uses of the conventional
K–Ar method. First, it can be used to calibrate fundamen-
tal physical parameters such as the K decay constant, since
it is a method grounded on first principles, i.e., the method
is completely independent of other geochronological
methods. Second, it is well-suited for dating potassium-
bearing fine-grained materials (such as clays or
glauconites) where 39Ar recoil would be problematic.
In general, however, the method of choice for dating
most K-bearing materials in a variety of geological studies
is the Ar–Ar method.
Ar–Ar
The presence and abundance of K-bearing minerals in
a variety of rocks, coupled with the strong temperature
dependence of argon diffusion in minerals, have led to
the widespread application of Ar–Ar dating in a variety
of geological applications. There are a plethora of studies
in each of the general areas below which are far too
 Ar–Ar AND K–Ar DATING
numerous to list; some illustrative references are included,
and more can also be found in more specific entries cross-
referenced in this work.
In general, Ar–Ar dating has been applied most often in
the following areas of study which are described briefly
below.
Refinement of the geological time scale and
time-scale boundaries
Ar–Ar dating and U–Pb dating are currently the most
commonly used techniques in establishing the timing of
geological events that are used to mark the various tempo-
ral boundaries in the geological time scale. The accuracy
and precision of both of these methods remain unmatched
by any other geochronological technique. Ar–Ar dating
has been used to refine the age of many boundaries such
as the Cretaceous–Paleogene boundary by dating the age
of the Chicxulub crater off the coast of Mexico (e.g.,
Renne et al., 2013) or the Permian–Triassic boundary by
dating large-scale flood basalts (e.g., Reichow
et al., 2009).
Hominid evolution
The high precision of the Ar–Ar technique makes it an
excellent method in which to date fossil evidence of the
earliest origins of the human species. By precisely dating
minerals (such as sanidines) from volcanic ash beds which
lie above and below fossilized remains, Ar–Ar dating has
proven to be extremely useful in understanding the evolu-
tion of hominids. Examples of Ar–Ar studies include the
dating of hominid evolution in western Africa (e.g., Lea-
key et al., 1995, 1998), the earliest hominids in Java
(Swisher et al., 1994), and early hominids in Georgia
(Gabunia et al., 2000), which appear to shift the date of
hominid dispersal out of Africa at least half a million years
earlier than previously thought. Antón and Swisher (2004)
provide a general review of hominid evolution out of
Africa, and an overview of the method applied to paleoan-
thropology and archeology is provided by Deino
et al. (1998).
Thermochronology
Because Ar is a noble gas, it is particularly sensitive to
temperature. If ambient temperatures are high enough,
Ar will preferentially leave the mineral structure via diffu-
sive transfer and escape to the surroundings. Thus, rocks
which have been reheated after crystallization (e.g., due
to metamorphism) may partially or completely lose Ar
from their minerals, depending on the diffusion character-
istics of the mineral, the temperature, and the time
involved. The temperature dependence of Ar–Ar ages
has opened up an entire field of geochronology known
as thermochronology where the thermal evolution of rocks
and regions can be studied using multiple geochronologi-
cal techniques; of these, Ar–Ar technique is one of the
most versatile.
69
As mentioned in the section on Modes of Analysis, the
interpretation of argon data is fundamentally based on
solid-state diffusion theory. There is now a considerable
body of knowledge associated with the application of dif-
fusion theory in geochronology, starting with the land-
mark paper by Dodson (1973), who developed the
mathematics defining the concept of closure temperature
(Tc). This key concept, which is defined as the temperature
of a cooling system at the time defined by the age of
a mineral, has led to the development of field
thermochronology, in which the thermal (i.e., temperature
T, time t) histories of rocks and regions can be elucidated
with geochronological techniques. Because argon diffuses
at different rates in minerals (due to difference in chemical
composition and the structure of the crystal lattice), differ-
ent K-bearing minerals span a different range of Ar clo-
sure temperatures. Argon diffusion experiments have
determined that, in general, for K-bearing minerals having
the same length scale of diffusion (i.e., grain size) and
cooling rate, argon closure temperatures increase in the
following order: Tc (K-feldspar) < Tc (biotite) < Tc
(muscovite) < Tc (hornblende). As a result, studying rocks
having these minerals means that an Ar–Ar dating study
could elucidate thermal histories over a temperature range
of potentially 150–600  C; this temperature range covers
a significant portion of the thermal history of many rocks,
making Ar–Ar thermochronology a very powerful tool in
the study of the evolution of geological terranes. Compli-
cated numerical models attempting to deconvolute ther-
mal histories from the step-heating of single crystals of
K-feldspar have been developed (e.g., the multidiffusion
domain (MDD) model of Lovera et al., 1989 and subse-
quent papers thereafter), but these remain controversial
because of a lack of correspondence to actual physical fea-
tures which are observed in the analyzed minerals, both in
nature and during argon step-heating experiments (see
Parsons et al., 1999, 2010). The entry on
Thermochronology provides more details on the funda-
mental principles of this field.
Many thermochronological Ar–Ar studies have been
conducted throughout the world, spanning most of the
Earth’s history. One of the earliest Ar–Ar thermochro-
nological studies was by Berger and York (1981), who
analyzed hornblende, biotite, plagioclase, and K-feldspar
to determine the cooling history of the Haliburton High-
lands in the Grenville Province, Ontario. An overview of
many of these studies and a more detailed analysis of
thermochronological principles can be found in
McDougall and Harrison (1999).
Structural geology
Ar–Ar dating has found widespread application in solving
structural geology problems because it can date K-bearing
phases (especially micas) that form during deformational
events. Moreover, the development of the laser micro-
probe in Ar–Ar dating (see “Single-Crystal Laser
Fusion”) enables mineral grains to be dated in situ and,
70 Ar–Ar AND K–Ar DATING
thus, potentially allows multiple deformation events to be
dated in a single rock. Many recent studies have also
highlighted the importance of considering defect struc-
tures associated with deformation in controlling Ar mobil-
ity and therefore in interpreting Ar–Ar ages; the study of
white micas by Kramar et al. (2001) highlights the utility
of the 40Ar–39Ar laser microprobe in this regard.
Tectonics
The ability to document a significant portion of the low–
mid-temperature history of geological terranes has made
Ar–Ar dating the method of choice in many tectonic stud-
ies. In general, thermochronology is used to elucidate
regional thermal T–t histories by dating a variety of
K-bearing minerals that record different parts of the geo-
logical evolution of a terrane. Many Ar–Ar applications
in tectonics have been published. A novel application of
Ar–Ar thermochronology was presented by Camacho
et al. (2005), who used a variety of Ar–Ar techniques in
combination with diffusion theory to determine the dura-
tion of a complete orogenic cycle (subduction and exhu-
mation) in the Bergen Arcs of Norway.
Economic geology
Ar–Ar dating is well suited for the timing of mineraliza-
tion in ore deposits because K-bearing minerals are com-
monly associated either with the ore or with ore-forming
processes. Minerals such as biotite, K-feldspar, musco-
vite, and alunite are commonly dated in such studies.
Ar–Ar dating has been applied in the study of a variety
of ore deposits including porphyry Cu–Mo (e.g., Quang
et al., 2003), epithermal Au (e.g., Bissig et al., 2002), vein
Au (e.g., Kontak et al., 1998; Mark et al., 2010), Missis-
sippi Valley-type Pb (e.g., Sherlock et al., 2004),
unconformity-type U (e.g., Polito et al. 2004), and dia-
mondiferous kimberlitic (e.g., Phillips et al., 1999)
deposits.
Detrital provenance
Ar–Ar dating of single crystals of K-bearing minerals has
provided useful insights into the provenance of sediments
derived from adjacent source regions. Analogous to the
U–Pb dating of single crystals of zircon, the dating of
grains of detrital biotite and muscovite can be equally use-
ful if the thermal history of the rocks has not been substan-
tially disturbed. A recent overview of the use of Ar–Ar
dating in detrital provenance studies is given in Hodges
et al. (2005).
Paleomagnetism and the geomagnetic polarity
time scale
Ar–Ar dating has been particularly useful in dating paleo-
magnetic poles for plate reconstructions as well as in refin-
ing the geomagnetic polarity time scale. An excellent
summary is given in York (1984) with subsequent devel-
opments outlined in Spell and McDougall (1992).
Sedimentary basin evolution
Sedimentary basins have been the subject of intense study
because of the opportunity to study a wealth of crustal pro-
cesses as well as a rich source of ore and hydrocarbon
deposits. Ar–Ar dating is generally considered to record
higher-temperature processes in such basins and is also
used in the study of the provenance of sediments.
A recent overview of the use of Ar–Ar and other
thermochronometers in understanding the thermal histo-
ries of sedimentary basins is given in Armstrong (2005).
Cosmochronology
One of the earliest applications of 40Ar–39Ar dating was in
the study of meteorites as well as lunar samples brought
back from the US Apollo missions. In particular, there is
a vast body of literature on the 40Ar–39Ar dating of
whole-rock basalts and related rocks which led not only
to an increased understanding of the genesis of the moon
and solar system but also to many refinements in the
development of the technique. Many of these results are
summarized in McDougall and Harrison (1999).
Advantages and limitations
With a 40K half-life of 1,250 Ma, the K–Ar and Ar–Ar
methods are well-suited for dating most geological mate-
rials from the Paleocene to the age of the Earth and solar
system. Dating very young materials, e.g., < 1 Ma, can
be particularly challenging with the Ar–Ar method due
to very short irradiation times and larger uncertainties
resulting from variations in the neutron flux and propor-
tionally larger corrections from blanks, neutron interfer-
ences, and mass discrimination in the mass spectrometer.
However, extremely young rocks dating from historical
times (i.e., the eruption of Mt. Vesuvius in 79AD) have
been successfully dated using the Ar–Ar method (Renne
et al., 1997; Lanphere et al., 2007).
In addition, the K–Ar and Ar–Ar dating methods
occupy a unique position among all geochronological
methods because Ar is a noble gas. As K–Ar and Ar–Ar
ages are therefore particularly dependent on the tempera-
ture of geological systems, this has both its advantages
and disadvantages. In particular, one of the greatest
strengths is the ability to record the geological history of
a sample over a range of temperatures. However, in the
absence of additional geological and petrological informa-
tion, it may not be possible to know if a K–Ar or Ar–Ar
age represents a crystallization, metamorphic, or
cooling age.
K–Ar
At present, there are two main advantages of the conven-
tional K–Ar method. First, it is a method grounded on
first principles, i.e., the method is completely independent
of other geochronological methods. This is essential
where reevaluation of fundamental physical parameters
such as the K decay constant may be important (see
section “Radioactive Decay Constants: A Review”).
 Ar–Ar AND K–Ar DATING
Second, it is well suited for dating potassium-bearing fine-
grained materials (such as clays or glauconites) where
39
Ar recoil would be problematic.
A major disadvantage of the K–Ar method is the sensi-
tivity of a K–Ar age to the thermal history of the rocks.
While a K–Ar age may reflect the formation age or crys-
tallization age of rocks, this age can be readily disturbed
or even reset due to a variety of geological processes such
as slow cooling, thermal reheating, or regional metamor-
phism. For example, in many early studies, it was noted
that K–Ar dates did not agree with either (a) dates for
other potassium-bearing minerals from the same rock or
(b) ages using other geochronological methods (e.g.,
Rb–Sr, U–Pb). Thus, one of the most serious challenges
with the conventional K–Ar method is that, in isolation,
it is impossible to tell whether a K–Ar date reflects the
time of crystallization, time of a subsequent metamor-
phism, or some time in between. There are some other dis-
advantages with conventional K–Ar dating. In the K–Ar
method, absolute quantities of potassium and argon are
measured on separate aliquots of the sample in order to
calculate the 40Ar–40K ratio and, hence, the age of the
sample; consequently, K–Ar ages tend to have much
larger uncertainties than Ar–Ar ages. Moreover, if the
sample is inhomogeneous, this practice can significantly
reduce the accuracy of the final age. In addition, because
the sample is completely fused to measure K and Ar, only
a single date can be obtained per analysis. Finally, rela-
tively large quantities (milligrams) of pure sample are
required for a K–Ar analysis which means that there must
be significant quantities of sample available to be ana-
lyzed, also increasing sample preparation requirements.
Fortunately, all of these disadvantages are effectively
eliminated by using the 40Ar–39Ar technique.
Ar–Ar
There are numerous advantages in using the 40Ar–39Ar
method. It is more convenient in terms of analysis because
only isotopes of Ar need to be measured from a single ali-
quot of sample, unlike K–Ar dating in which different
techniques are required to measure K and Ar, respectively.
Problems with sample inhomogeneity and differing
machine calibrations are eliminated because all of the
measurements are performed on the same sample with
the same equipment. Because only ratios are needed
(i.e., 40Ar/39Ar), Ar–Ar ages will inherently be more pre-
cise than K–Ar ages and can typically achieve routine ana-
lytical uncertainties less than 0.5 %. The high sensitivity
of today’s noble gas mass spectrometers means that very
little sample is required for analysis; single-grain dating
is routine. A variety of techniques (furnace/laser step-
heating, total fusion, laser spot-fusion) can be employed
to suit the goals of the project. The method can yield sig-
nificant information on the thermal history of the sample.
Finally, other argon isotopes can be used to infer some
information about the chemical composition (K, Ca, Cl)
of the sample.
71
Ar–Ar dating does have some disadvantages. A nuclear
reactor with a suitable fast-neutron fluence is required.
This also means that irradiated samples are radioactive
and must be appropriately stored; safe handling and sam-
ple preparation procedures must also be developed. Neu-
tron irradiation results in the formation of artificial Ar
isotopes from K, Ca, and Cl which must be corrected in
calculating an Ar–Ar age, leading to more complicated
data processing. The irradiation also results in 39Ar recoil,
which can negatively affect the ages of very fine-grained
samples or samples with mm-scale domains. Finally, all
of these factors also mean that the method is generally
more time consuming and costly in comparison to K–Ar
dating.
Summary
The Ar–Ar method has largely superseded conventional
K–Ar dating because of its numerous advantages. The
presence of K-bearing minerals in a variety of rocks has
led to the widespread application of Ar–Ar dating to solve
a variety of geological problems involving the geological
time scale, hominid evolution, structural geology and tec-
tonics, economic geology, detrital provenance, paleomag-
netism, the evolution of sedimentary basins, and the origin
of extraterrestrial materials. Moreover, the strong relation-
ship between argon mobility and temperature makes the
Ar–Ar method an ideal thermochronometer. To apply it
successfully, however, knowledge of the microstructural,
mineralogical, petrological, and geological context of the
dated samples must be known in order to obtain meaning-
ful interpretations of the age data. With this information,
the real power of the Ar–Ar method lies in its unique abil-
ity to solve diverse problems in the earth sciences and to
elucidate a significant portion of the temperature–time
histories of geological terranes.
Bibliography
Antón, S. A., and Swisher, C. C., III, 2004. Early dispersal of homo
from Africa. Annual Reviews in Anthropology, 33, 271–296.
Armstrong, P. A., 2005. Thermochronometers in sedimentary
basins. Reviews in Mineralogy and Geochemistry, 58, 499–525.
Begemann, F., Ludwig, K. R., Lugmair, G. W., Min, K., Nyquist,
L. E., Patchett, P. J., Renne, P. R., Shih, C.-Y., Villa, I. M., and
Walker, R. J., 2001. Call for an improved set of decay constants
for geochronological use. Geochimica et Cosmochimica Acta,
65, 111–121.
Berger, G. W., and York, D., 1981. Geothermometry from 40Ar/39Ar
dating experiments. Geochimica et Cosmochimica Acta, 45,
795–811.
Bissig, T., Clark, A. H., Lee, J. K. W., and Hodgson, C. J., 2002.
Miocene landscape evolution in the Chilean flat-slab transect:
uplift history and geomorphologic influences on epithermal pro-
cesses in the El Indio-Pascua Au (–Ag, Cu) belt. Economic
Geology, 97, 971–996.
Camacho, A., Lee, J. K. W., Hensen, B. J., and Braun, J., 2005.
Short-lived orogenic cycles and the eclogitization of cold crust
by spasmodic hot fluids. Nature, 435, 1191–1996.
Dalrymple, G. B., and Lanphere, M. A., 1969. Potassium-Argon
dating: principles, techniques, and applications to geochronol-
ogy. San Francisco: WH Freeman. 258 p.
72 Ar–Ar AND K–Ar DATING
Dalrymple, G. B., Lanphere, M. A., and Pringle, M. S., 1988. Cor-
relation diagrams in 40Ar/39Ar dating: is there a correct choice?
Geophysical Research Letters, 15, 589–591.
Deino, A., and Potts, R., 1992. Age-probability spectra from exam-
ination of single-crystal 40Ar/39Ar dating results: examples from
Olorgesailie, Southern Kenya rift. Quaternary International,
13(14), 47–53.
Deino, A., Renne, P. R., and Swisher, C. C., 1998. 40Ar/39Ar dating
in paleoanthropology and archeology. Evolutionary Anthropol-
ogy, 6, 63–75.
Dickin, A. P., 2005. Radiogenic isotope geology, 2nd edn. Cam-
bridge: Cambridge University Press, p. 492.
Dodson, M. H., 1973. Closure temperature in cooling geochrono-
logical and petrological systems. Contributions to Mineralogy
and Petrology, 40, 259–274.
Dong, H., Hall, C. M., Peacor, D. R., and Halliday, A. N., 1995.
Mechanisms of argon retention in clays revealed by laser
40
Ar-39Ar dating. Science, 267, 355–359.
Dong, H., Hall, C. M., Halliday, A. N., and Peacor, D. R., 1997.
Laser 40Ar-39Ar dating of microgram-size illite samples and
implications for thin-section dating. Geochimica et
Cosmochimica Acta, 61, 3803–3808.
Fitz Gerald, J. D., and Harrison, T. M., 1993. Argon diffusion
domains in K-feldspar I: microstructures in MH-10. Contribu-
tions to Mineralogy and Petrology, 113, 367–380.
Fleck, R. J., Sutter, J. F., and Elliot, D. H., 1977. Interpretation of
discordant 40Ar/39Ar age spectra of Mesozoic tholeiites from
Antarctica. Geochimica et Cosmochimica Acta, 41, 15–32.
Gaber, L. J., Foland, K. A., and Corbato, C. E., 1988. On the signif-
icance of argon release from biotite and amphibole during
40
Ar/39Ar vacuum heating. Geochimica et Cosmochimica Acta,
52, 2457–2465.
Gabunia, L., Vekua, A., Ferring, R., Justus, A., Lordkipanidze, D.,
Swisher, C. C., III, Tvalchrelidze, M., Antón, S. C., Bosinski,
G., de Lumley, M.-A., Majsuradze, G., Mouskhelishvili, A.,
and Nioradze, M., 2000. Earliest Pleistocene hominid cranial
remains from Dmanisi, Republic of Georgia, taxonomy, geo-
logic setting and age. Science, 288, 1019–1025.
Goodman, C., and Evans, R. D., 1941. Age measurements by radio-
activity. Geological Society of America Bulletin, 52, 491–544.
Halliday, A. N., 1978. 40Ar-39Ar stepheating studies of clay concen-
trates from Irish orebodies. Geochimica et Cosmochimica Acta,
42, 1851–1858.
Hanes, J. A., York, D., and Hall, C. M., 1985. An 40Ar/39Ar geo-
chronological and electron microprobe investigation of an
Archean pyroxenite and its bearing on ancient atmospheric com-
positions. Canadian Journal of Earth Sciences, 22, 947–958.
Hodges, K. V., Ruhl, K. W., Wobus, C. W., and Pringle, M. S., 2005.
40
Ar/39Ar thermochronology of detrital minerals. Reviews in
Mineralogy and Geochemistry, 58, 238–257.
Huneke, J. C., and Smith, S. P., 1974. The realities of recoil: 39Ar
recoil out of small grains and anomalous patterns in 39Ar–40Ar
dating. Geochimica et Cosmochimica Acta Supplement, 7(Pro-
ceedings of the 7th Lunar Science Conference), 1987–2008.
Jourdan, F., and Renne, P. R., 2007. Age calibration of the Fish Can-
yon sanidine 40Ar/39Ar dating standard using primary K–Ar
standards. Geochimica et Cosmochimica Acta, 71, 387–402.
Kelley, S., 2002. Excess argon in K–Ar and Ar–Ar geochronology.
Chemical Geology, 188, 1–22.
Kontak, D. J., Horne, R. J., Sandeman, H., Archibald, D., and Lee,
J. K. W., 1998. 40Ar/39Ar dating of ribbon-textured veins and
adjacent wallrock from the Meguma lode gold deposits: implica-
tions for timing and duration of vein formation of slate-belt-
hosted vein gold deposits. Canadian Journal of Earth Sciences,
35, 746–761.
Kramar, N., Cosca, M. A., and Hunziker, J. C., 2001. Heteroge-
neous 40Ar* distributions in naturally deformed muscovite: in
situ UV-laser ablation evidence for microstructurally controlled
intragrain diffusion. Earth and Planetary Science Letters, 192,
377–388.
Kuiper, K. F., Deino, A., Hilgen, F. J., Krijgsman, W., Renne, P. R.,
and Wijbrans, J. R., 2008. Synchronizing the rock clocks of earth
history. Science, 320, 500–504.
Lanphere, M. A., Champion, D., Melluso, L., Morra, V., Perrotta,
A., Scarpati, C., Tedesco, D., and Calvert, A., 2007. 40Ar/39Ar
ages of the AD 79 eruption of Vesuvius, Italy. Bulletin of Volca-
nology, 69, 259–263.
Leakey, M. G., Feibel, C. S., McDougall, I., Ward, C., and Walker,
A., 1995. New four-million-old hominid species from Kanapoi
and Allia Bay, Kenya. Nature, 376, 565–571.
Leakey, M. G., Feibel, C. S., McDougall, I., Ward, C., and Walker,
A., 1998. New specimens and confirmation of an early age for
Australopithecus anamensis. Nature, 393, 62–66.
Lee, J. K. W., Onstott, T. C., Cashman, K. V., Cumbest, R. J., and
Johnson, D., 1991. Incremental heating of hornblende in vacuo:
implications for 40Ar/39Ar geochronology and the interpretation
of thermal histories. Geology, 19, 872–876.
Lo, C. H., and Onstott, T. C., 1989. 39Ar recoil artifacts in chloritized
biotite. Geochimica et Cosmochimica Acta, 53, 2697–2711.
Lo, C. H., Lee, J. K. W., and Onstott, T. C., 2000. Argon release
mechanisms of biotite in vacuo and the role of short-circuit dif-
fusion and recoil. Chemical Geology, 165, 135–166.
Lovera, O. M., Richter, F. M., and Harrison, T. M., 1989. The
40
Ar/39Ar geothermometry for slowly cooled samples having
a distribution of diffusion domain sizes. Journal of Geophyscial
Research, 94, 17917–17935.
Mark, D. F., Rice, C. M., Lee, M. R., Fallick, A. E., Boyce, A.,
Trewin, N. H., and Lee, J. K. W., 2010. 40Ar/39Ar dating of
hydrothermal activity, biota, and gold mineralization in the
Rhynie hot-spring system, Aberdeenshire, Scotland.
Geochimica et Cosmochimica Acta, 75, 555–569.
McDougall, I., and Harrison, T. M., 1999. Geochronology and
thermochronology by the 40Ar/39Ar method, 2nd edn. New York:
Oxford University Press, p. 269.
McIntyre, G. A., Brooks, C., Compston, W., and Turek, A., 1966.
The statistical assessment of Rb–Sr isochrons. Journal of Geo-
physical Research, 71, 5459–5468.
Merrihue, C., and Turner, G., 1966. Potassium-argon dating by acti-
vation with fast neutrons. Journal of Geophysical Research, 71,
2852–2857.
Min, K., Mundil, R., Renne, P. R., and Ludwig, K. R., 2000. A test
for systematic errors in 40Ar/39Ar geochronology through com-
parison with U/Pb analysis of a 1.1 Ga rhyolite. Geochimica et
Cosmochimica Acta, 64, 73–98.
Nier, A. O., 1950. A redetermination of the relative abundances of
the isotopes of carbon, nitrogen, oxygen, argon, and potassium.
Physical Review, 77, 789–793.
Parsons, I., Brown, W. L., and Smith, J. V., 1999. 40Ar/39Ar
thermochronology using alkali feldspars: real thermal history
or mathematical mirage of microtexture? Contributions to
Mineralogy and Petrology, 136, 92–110.
Parsons, I., Fitz Gerald, J. D., Lee, J. K. W., Ivanic, T., and Golla-
Schindler, U., 2010. Time-temperature evolution of
microtextures and contained fluids in a plutonic alkali feldspar
during heating. Contributions to Mineralogy and Petrology,
160, 155–180.
Phillips, D., and Onstott, T. C., 1988. Argon isotope zoning in man-
tle phlogopite. Geology, 16, 542–546.
Phillips, D., Kiviets, G. B., Barton, E. S., Smith, C. B., Viljoen,
K. S., and Fourie L. F., 1999. 40Ar/39Ar dating of kimberlites
 ARCHAEOMAGNETIC DATING
and related rocks: problems and solutions. In 7th International
Kimberlite Conference v2, Red Roof Design, Cape Town,
pp 677–688.
Quang, C. X., Clark, A. H., Lee, J. K. W., and Guillén, B. J., 2003.
40
Ar–39Ar ages of hypogene and supergene mineralization in the
Cerro Verde – Santa Rosa porphyry Cu–Mo cluster, Arequipa,
Peru. Economic Geology, 98, 1683–1696.
Reichow, M. K., Pringle, M. S., Al'Mukhamedov, A. I., Allen,
M. B., Andreichev, V. L., Buslov, M. M., Davies, C. E.,
Fedoseev, G. S., Fitton, J. G., Inger, S., Medvedev, A. Y. A.,
Mitchell, C., Puchkov, V. N., Safonova, I. Y. U., Scott, R. A.,
and Saunders, A. D., 2009. The timing and extent of the eruption
of the Siberian Traps large igneous province: implications for the
end-Permian environmental crisis. Earth and Planetary Science
Letters, 277, 920.
Renne, P. R., Sharp, W. D., Deino, A. L., Orsi, G., and Civetta, L.,
1997. 40Ar/39Ar Dating into the historical realm: calibration
against Pliny the younger. Science, 277, 1279–1280.
Renne, P. R., Karner, D. B., and Ludwig, K. R., 1998. Absolute ages
aren’t exactly. Science, 282, 1840–1841.
Renne, P. R., Mundil, R., Balco, G., Min, K., and Ludwig, K. R.,
2010. Joint determination of 40K decay constants and
40
Ar*/40K for the Fish Canyon sanidine standard, and improved
accuracy for 40Ar/39Ar geochronology. Geochimica et
Cosmochimica Acta, 74, 5349–5367.
Renne, P. R., Deino, A. L., Hilgen, F. J., Kuiper, K. F., Mark, D. F.,
Mitchell, W. S., Morgan, L. E., Mundil, R., and Smit, J., 2013.
Time scales of critical events around the Cretaceous-Paleogene
boundary. Science, 339(6120), 684–687.
Sherlock, R., Lee, J. K. W., and Cousens, B. L., 2004. Geological
and geochronological constraints on the timing of mineralization
at the Nanisivik zinc-lead Mississippi Valley-type deposit, north-
ern Baffin Island, Nunavut, Canada. Economic Geology, 99,
279–293.
Sletten, V. W., and Onstott, T. C., 1998. The effect of the instability
of muscovite during in vacuo heating on 40Ar/39Ar step-heating
spectra. Geochimica Cosmochimica Acta, 62, 123–141.
Smits, F., and Gentner, W., 1950. Argonbestimmungen an Kalium-
Mineralien. I. Bestimmungen an tertiären Kalisalzen.
Geochimica et Cosmochimica Acta, 1, 22–27.
Spell, T. L., and McDougall, I., 1992. Revisions to the age of the
Brunhes-matayuma boundary and the Pleistocene geomagnetic
polarity timescale. Geophysical Research Letters, 19,
1181–1184.
Steiger, R. H., and Jäger, E., 1977. Subcommission on geochronol-
ogy: convention on the use of decay constants in geo- and
cosmochronology. Earth and Planetary Science Letters, 36,
359–362.
Stoenner, R. W., Schaeffer, O. A., and Katcoff, S., 1965. Half-lives
of argon-37, argon-39, and argon-42. Science, 148, 1325–1328.
Swisher, C. C., Curtis, G. H., Jacob, T., Getty, A. G., Suprijo, A.,
and Widiasmoro, 1994. Age of the earliest known hominids in
Java, Indonesia. Science, 263, 1118–1121.
Turner, G., and Cadogan, P. H., 1974. Possible effects of 39Ar recoil
in 40Ar–39Ar dating. Geochimica et Cosmochimica Acta Supple-
ment 5(Proceedings of the 5th Lunar Science Conference),
1601–1615.
Turner, G., Miller, L. A., and Grasty, R. L., 1966. The thermal his-
tory of the Bruderheim meteorite. Earth and Planetary Science
Letters, 1, 155–157.
Villa, I. M., 1997. Direct determination of the 39Ar recoil distance.
Geochimica et Cosmochimica Acta, 61, 689–691.
York, D., 1969. Least-squares fitting of a straight line with
correlated errors. Earth and Planetary Science Letters, 5,
320–324.
73
York, D., 1984. Cooling histories from Ar/Ar age spectra: implica-
tions for Precambrian plate tectonics. Annual Review of Earth
and Planetary Sciences, 12, 383–409.
Cross-references
Age of the Earth
Alpine Terranes (K–Ar/Ar–Ar)
Clays and Glauconites (K–Ar/Ar–Ar)
Extraterrestrial Materials (K–Ar/Ar–Ar)
Hominid Evolution Timescale
Kimberlites (K-Ar/Ar-Ar)
Mass Spectrometry
Metamorphic Terranes (K–Ar/40Ar/39Ar)
Minerals (40Ar–39Ar)
Noble Gas Mass Spectrometer
Single-Crystal Laser Fusion
Potassium–Argon (Argon–Argon), Structural Fabrics
Volcanic Rocks (Ar/Ar)
ARCHAEOMAGNETIC DATING
Cathy Batt
Archaeological Sciences, University of Bradford,
Bradford, UK
Definition
Archaeomagnetic dating is the study of the past geomag-
netic field as recorded by archaeological materials and
the interpretation of this information to date past events.
Introduction
Archaeomagnetic dating is a method for dating fired mate-
rials and sediments from archaeological sites, based on
their preserved magnetic remanence. The principles of
the method can be summarized as follows:
• The geomagnetic field changes significantly on archae-
ologically relevant timescales of decades and centuries
(Tarling, 1983, p. 145).
• Some archaeological materials contain magnetized par-
ticles, and certain events cause the geomagnetic field at
a particular moment in time to be recorded by these
particles.
• This recorded magnetization can be measured in the
laboratory.
• By comparing the recorded magnetization with a dated
record of changes in the geomagnetic field with time,
the event which caused the recording can be dated.
The application of archaeomagnetic dating is restricted
in time and location to regions where there is detailed
knowledge of the geomagnetic field for the period in ques-
tion. The strengths of archaeomagnetic dating are that it
dates fired clay and stone, for example, hearths, kilns,
ovens, and furnaces, which are frequently well preserved
on archaeological sites; it dates the last use of features,
74 ARCHAEOMAGNETIC DATING

providing a clear link to human activity; it can be cost-
effective and is potentially most precise in periods where
other dating methods, e.g., radiocarbon dating, are
problematic.
The physical basis
Archaeomagnetic dating is based on a comparison of the
past geomagnetic field, as recorded by archaeological
materials, with a dated record of changes in the Earth’s
field over time in a particular geographical area. The geo-
magnetic field changes both in direction (declination and
inclination) and in strength (intensity) (Lanza and Meloni,
2006, p. 34), and archaeomagnetic dating can be based on
changes in either direction or intensity or a combination.
For archaeological material to be suitable for dating, it
must contain sufficient magnetized particles, and an event
must have caused these particles to record the Earth’s
magnetic field. Many geologically derived materials,
e.g., soils, sediments, and clays, contain suitable iron
oxides, such as magnetite, hematite, and maghemite
(Merrill et al., 1998, p. 84). There are primarily two types
of archaeological events which may result in the Earth’s
magnetic field at a particular moment being recorded by
archaeological material: heating and deposition in water.
If fired materials have been heated above the Curie tem-
perature of the iron oxides (typically 650–700  C), they
easily become magnetized in the direction of the ambient
field, which is usually the geomagnetic field (Figure 1).
On cooling, they retain this thermoremanent magnetiza-
tion, which reflects the Earth’s magnetic field at the time
of last cooling. Suitable archaeological features include
hearths, kilns, ovens, and other fired structures. Water-lain
sediments may acquire a datable detrital remanent magne-
tization from the alignment of their magnetic grains by the
ambient field during deposition (Figure 2). The magne-
tized grains then settle and become immobilized, retaining
a record of the direction of the geomagnetic field at their
deposition. This effect allows deposits in wells, ditches,
and streams to be dated. However, uncertainty over the
exact moment at which the magnetization is acquired
(Evans and Heller, 2003, p. 86) and postdepositional dis-
turbance of the magnetization by bioturbation and diagen-
esis of sediments (Batt, 1999) means that almost all
archaeomagnetic studies focus on fired materials.
The method in practice
Archaeomagnetic dating by direction requires the exact
position of the archaeological material to be recorded in
relation to the present geomagnetic field, and so the mate-
rial must be undisturbed and sampled in situ. Careful eval-
uation of structural remains is required to identify any
post-firing movement and to select undisturbed areas for
sampling. In addition, the magnetic field in which the fea-
ture cooled may be disturbed by the magnetization of
faster cooling parts of the feature (Schurr et al., 1984).
The presence of magnetic materials, such as iron slag
(Abrahamsen et al., 2003), can also cause disturbance.
Sometimes such disturbances are not visible on site but
will be revealed by statistical analysis of the measure-
ments of remanent magnetization, and so, where possible,
different parts of the feature must be sampled and com-
pared (Figure 3). Samples of robust fired materials can
be taken by coring, attaching plastic reference buttons
using a fast-setting epoxy resin, or encasing samples in
plaster of Paris before lifting (Linford, 2006; Trapanese
et al., 2008). Sediments and friable fired materials are
sampled by insertion of plastic cylinders. Approximately
15 small (c. 1 cm3) samples are needed from each feature,
and it may be possible to select sampling locations to min-
imize the visual impact. In all cases, the orientation of the
in situ samples must be measured, usually by leveling the
sample horizontally and marking the north direction using
a magnetic or sun compass or a gyrotheodolite. Measure-
ments of archaeointensity have the significant advantage
that in situ samples are not required, allowing the dating
of bricks from buildings or non-orientated pottery frag-
ments (Shaw et al., 1999). These determinations can be
carried out on much smaller samples, depending on the
measuring instrument.
In the laboratory, the remanent magnetization of each
sample is measured in a magnetometer, and the stability
of this magnetization is evaluated by alternating magnetic
field or thermal demagnetization (Merrill et al., 1998,
p. 90). Demagnetization is necessary as the thermo- or
detrital remanent magnetization acquired in the past may
be overprinted by a viscous remanent magnetization,
caused by magnetically soft materials changing with the
geomagnetic field. The distribution of the stable magnetic
components is assessed using Fisher statistics (Fisher,
1953) to calculate a mean magnetic direction, together
with its angular spread (a95).
Dating by direction is relatively experimentally
straightforward as the magnetic direction recorded by the
sample is directly comparable to the past magnetic field
direction. However, dating by intensity is more complex
as the strength of magnetization of the sample is
influenced not only by the strength of the past field but
also by the mineralogical composition and firing regime
(Merrill et al., 1998, p. 102). Laboratory techniques have
been developed to correct for these factors (Tauxe and
Yamazaki, 2007; Gallet et al., 2009), but the determination
of past intensity involves significantly more measure-
ments than direction, and there are many more potential
sources of uncertainty (Gallet et al., 2009). These chal-
lenges mean that many past studies have focused on mag-
netic direction; however, the determination of intensity is
becoming routine, allowing definition of the entire mag-
netic vector (Sternberg, 2001).
Having obtained the past magnetic field, the date is
obtained by comparison with a secular variation record
of changes in the Earth’s field (Figure 4). This record is
compiled from direct measurements of the field, where
available, and, prior to that, from archaeomagnetic mea-
surements of features or sediments dated by other
methods. As the field varies spatially, there needs to be
 ARCHAEOMAGNETIC DATING 75

Archaeomagnetic Dating, Figure 1 The acquisition of thermoremanent magnetization. Before heating, the magnetic domains
within the material are randomly orientated within the ambient field and cancel out. During heating, some domains gain sufficient
energy to reorientate in the direction of the ambient field and retain this orientation on cooling, producing an induced
magnetization. As time passes, the ambient field changes, but the magnetic domains retain the magnetization at the time of cooling
(Adapted from Linford (2006), Figure 2).

sufficient data with 1,000 km (Noel and Batt, 1990), and
the development of secular variation records has been
the subject of significant recent research. Early studies
relied on a visual comparison of magnetic measurements
with hand-interpolated calibration curves (e.g., Clark
et al., 1988). Subsequent developments in the data avail-
able and statistical approaches have led to better quantifi-
cation of errors in both the magnetic direction to be dated
and the secular variation record (Batt, 1997). This has
been followed by the use of Bayesian statistical methods
to calculate confidence limits and combine the results for
magnetic declination, inclination, and intensity (Lanos
et al., 1999; Lanos, 2004). New modeling techniques
allow the consideration of data over larger regions,
reducing the need for regionally specific calibration
curves (Korte and Holme, 2003). For example, Pavón-
Carrasco et al. (2009) used a spherical cap harmonic
model to produce a secular variation record for Western
Europe. Global models can also be produced (e.g., Korte
et al., 2009) using worldwide data, but, at present, these
do not have the precision needed for archaeological
dating.
Applications
Archaeomagnetic studies have been used to investigate
a wide variety of archaeological features. Particularly
notable case studies include the dating of medieval and
76 ARCHAEOMAGNETIC DATING
Archaeomagnetic Dating, Figure 2 The acquisition of detrital
(or depositional) remanent magnetization. Some sediment
particles may be weakly magnetized, usually along their long
axes. As they settle through the water column, they are free to
rotate and align with the ambient magnetic field. They retain
this orientation on settling and become fixed in position by the
weight of sediment accumulating above them. They are no
longer able to rotate and retain the magnetic direction of the
ambient field at the time they became locked in place (Adapted
from Linford (2006), Figure 3).
Tudor glassmaking furnaces at Bagot’s Park, UK (Linford
and Welch, 2004), brick kilns (Casas et al., 2007), pottery
kilns (Reinders et al., 1999), ovens (Schnepp et al., 2003),
and canal sediments (Eighmy and Howard, 1991). The
precision of archaeomagnetic dates is crucial in their use
in addressing archaeological questions. A number of fac-
tors will influence this precision: recording of the past
field, post-firing or post-depositional disturbance, preci-
sion of sampling and laboratory measurements, precision
of the secular variation record and the method of compar-
ison, and spatial variation of the geomagnetic field. Of
these, the most significant is usually the precision of the
secular variation curve, and, as this varies with time, the
precision of the date obtained will depend on the archaeo-
logical period. As the geomagnetic field has occasionally
had the same direction at different times, it is also possible
to obtain two or more alternative dates for a single archae-
ological event. In most cases, the archaeological evidence
will indicate the most likely. Given the number of
Archaeomagnetic Dating, Figure 3 Archaeomagnetic
sampling of a pottery kiln, showing samples taken by the button
method from walls, floor, and pedestal.
contributing factors, it is not possible to define the general
precision of archaeomagnetic dates, but there will usually
be an error margin of at least 50 years (Linford, 2006). It
is important to note that the secular variation record
improves as more measurements become available; hence,
features that cannot be dated or give broad age ranges now
may be datable in the future.
Considerable research effort has been focused on build-
ing up secular variation records, making archaeomagnetic
dating a routine dating tool for some archaeological
periods and regions. This includes large parts of Europe,
most notably Bulgaria (Kovacheva et al., 2009), the UK
(Zananiri et al., 2007), France (Gallet et al., 2002), Ger-
many (Schnepp and Lanos, 2005), Iberia (Gómez-Paccard
et al., 2006), Greece (Tema and Kondopoulou, 2011), and
Hungary (Marton and Ferencz, 2006). There have also
been major studies in the American Southwest, where
independent dates are provided by dendrochronology
(Sternberg and McGuire, 1990; Doyel and Eighmy,
1994). Work is progressing in other regions, for example,
North Africa (Gómez-Paccard et al., 2012), New Zealand
(Stark et al., 2010), and Syria (Speranza et al., 2006).
Longer-term changes in the geomagnetic field include
magnetic reversals (Merrill et al., 1998, p. 163), and these
major changes can be used to date early hominid sites, if
they are recorded in associated sediments (e.g., Parés
and Pérez-Gonzalez, 1995). In addition, archaeomagnetic
studies have been used in estimating firing temperatures
(Linford and Platzman, 2004) and firing duration (Meng
and Noel, 1989) and in provenancing studies (Williams-
Thorpe et al., 1996; Rasmumussen, 2001). Wider applica-
tions are discussed more fully in Sternberg (2001) and
Tarling (1983, p. 157).
 ARCHAEOMAGNETIC DATING 77

Archaeomagnetic Dating, Figure 4 Calibration of a mean archaeomagnetic direction in the UK using RenDate (Lanos et al., 2005;
Zananiri et al., 2007). The upper graph shows dates when the geomagnetic field had the same inclination as the measured direction;
the middle graph shows dates when the declinations were the same. The lower graph shows the combined probability density where
the geomagnetic field displayed the same declination and inclination as that recorded by the samples. The black area represents the
date range with a 95 % confidence interval, in this case either before 950 BC or AD 1000–1160. Additional evidence is required to
select from the two possibilities.

Conclusions and future directions
Archaeomagnetic dating has been shown to provide valu-
able archaeological information and supplement the suite
of chronological tools available for interpreting archaeo-
logical sites. In addition, the development of
archaeomagnetic dating as a method contributes to the
wider understanding of the Earth’s magnetic field (Gallet
et al., 2009). As noted by Sternberg (2001),
archaeomagnetic dating is most effective as
a collaborative venture between archaeologists and geo-
physicists. Independently dated features from an archaeo-
logical excavation can provide geomagnetic field
information; once the secular variation pattern is
established, it can be used to provide archaeomagnetic
dates.
Key to improving the accuracy and reliability of
archaeomagnetic dates is the availability of magnetic
measurements on dated archaeological material. Collabo-
rative sampling ventures such as AARCH
(Archaeomagnetic Applications for the Rescue of Cultural
Heritage, http://dourbes.meteo.be/aarch.net/frameset_en.
html) have increased the data collected, and they are made
widely available through international databases, for
example, MagIC (Magnetics Information Consortium,
78 ARCHAEOMAGNETIC DATING
http://earthref.org/MAGIC/). In many regions, there has
been a focus on directional rather than intensity methods.
However, experimentally innovative methods for recover-
ing and interpreting archaeointensity (e.g., Casas et al.,
2005; Gallet et al., 2009) have led to an increased focus
on intensity studies, which permit the full magnetic vector
to be used in dating studies, thus improving the precision
of dates, and allowing the dating of unorientated material.
Bibliography
Abrahamsen, N., Jacobsen, B. H., Koppelt, U., De Lasson, P.,
Smekalova, T., and Voss, O., 2003. Archaeomagnetic investiga-
tions of iron age slags in Denmark. Archaeological Prospection,
10, 91–100.
Batt, C. M., 1997. The British archaeomagnetic calibration curve:
an objective treatment. Archaeometry, 39, 153–168.
Batt, C. M., 1999. Preliminary investigations into the acquisition of
remanence in archaeological sediments. In Tarling, D. H., and
Turner, P. (eds.), Palaeomagnetism and Diagenesis of Sedi-
ments. London: Geological Society. Special Publication of the
Geological Society of London, 151, pp. 9–19.
Casas, L., Shaw, J., Gich, M., and Share, J., 2005. High quality
microwave intensity determinations from an early C18th AD
English brick kiln. Geophysical Journal International, 161,
653–661.
Casas, L., Linford, P., and Shaw, J., 2007. Archaeomagnetic dating
of Dogmersfield Park brick kiln (Southern England). Journal of
Archaeological Science, 34, 205–213.
Clark, A. J., Tarling, D. H., and Noel, M., 1988. Developments in
archaeomagnetic dating in Britain. Journal of Archaeological
Science, 15(6), 645–667.
Doyel, D. E., and Eighmy, J. L., 1994. Archaeomagnetic dating and
the Bonito Phase chronology. Journal of Archaeological Sci-
ence, 21, 651–658.
Eighmy, J. L., and Howard, J. B., 1991. Direct dating of prehistoric
canal sediments using archaeomagnetism. American Antiquity,
56, 88–102.
Evans, M. E., and Heller, F., 2003. Environmental Magnetism.
Oxford: Academic.
Fisher, R. A., 1953. Dispersion on a sphere. Proceedings of the
Royal Society of London A, 217, 295–305.
Gallet, Y., Genevey, A., and Le Goff, M., 2002. Three millennia of
directional variations of the Earth’s magnetic field in western
Europe as revealed by archaeological artefacts. Physics of the
Earth and Planetary Interiors, 131, 81–89.
Gallet, Y., Genevey, A., Le Goff, M., Warmé, N., Gran-Aymerich,
J., and Lefèvre, A., 2009. On the use of archaeology in geomag-
netism, and vice versa: recent developments in
archaeomagnetism. C. R. Physique, 10, 630–648.
Gómez-Paccard, M., Lanos, P., Chauvin, A., McIntosh, G., Osete,
M. L., Catanzariti, G., Ruiz-Martínez, V. C., and Núñez, J. I.,
2006. The first Archaeomagnetic secular variation curve for the
Iberian Peninsula. Comparison with other data from western
Europe and with global geomagnetic field models. Geochemis-
try, Geophysics, Geosystems, 7, Q12001, doi:10.1029/
2006GC001476.
Gómez-Paccard, M., McIntosh, G., Chauvin, A., Beamud, E.,
Pavón-Carrasco, F. J., and Thiriot, J., 2012. Archaeomagnetic
and rock magnetic study of six kilns from North Africa (Tunisia
and Morocco). Geophysical Journal International, 189,
169–186, doi:10.1111/j.1365-246X.2011.05335.x.
Korte, M., and Holme, R., 2003. Regularisation of spherical cap
harmonics. Geophysical Journal International, 153, 253–262.
Korte, M., Donadini, F., and Constable, C., 2009. Geomagnetic
field for 0-3ka: 2. A new series of time-varying global models.
Geochemistry, Geophysics, Geosystems, 10, Q06008,
doi:10.1029/2008GC002297.
Kovacheva, M., Boyadziev, Y., Kostadinova-Avramova, M.,
Jordanova, N., and Donadini, F., 2009. Updated archeomagnetic
data set of the past 8 millennia from the Sofia laboratory, Bul-
garia. Geochemistry, Geophysics, Geosystems, 10, Q05002.
Lanos, P., 2004. Bayesian inference of calibration curves: applica-
tion to archaeomagnetism. In Buck, C. E., and Millard, A. R.
(eds.), Tools for Constructing Chronologies: Crossing Disciplin-
ary Boundaries. Berlin: Springer. Lecture Notes in Statistics,
177, pp. 43–82.
Lanos, P., Kovacheva, M., and Chauvin, A., 1999.
Archaeomagnetism, methodology and applications: implemen-
tation and practice of the archaeomagnetic method in France
and Bulgaria. European Journal of Archaeology, 2(3),
365–392.
Lanos, P., Le Goff, M., Kovacheva, M., Schnepp, E., 2005. Hierar-
chical modelling of archaeomagnetic data and curve estimation
by moving average technique. Geophysical Journal Interna-
tional, 160, 440–476, doi:10.1111/j.1365-246X.2005.02490.x.
Lanza, R., and Meloni, A., 2006. The Earth’s Magnetism. Berlin:
Springer.
Linford, P., 2006. Archaeomagnetic Dating. Guidelines on produc-
ing and interpreting archaeomagnetic dates. Swindon: English
Heritage, p. 31.
Linford, N. T., and Platzman, E., 2004. Estimating the approximate
firing temperature of burnt archaeological sediments through an
unmixing algorithm applied to hysteresis data. Physics of the
Earth and Planetary Interiors, 147(2–3), 197–207.
Linford, P., and Welch, C., 2004. Archaeomagnetic analysis of
glassmaking sites at Bagot’s Park in Staffordshire, England.
Physics of the Earth and Planetary Interiors, 147, 209–221.
Marton, P., and Ferencz, E., 2006. Hierarchical versus stratification
statistical analysis of archaeomagnetic directions: the secular
variation curve for Hungary. Geophysical Journal International,
164, 484–489.
Meng, Z., and Noel, M., 1989. Archaeomagnetic evidence for age
and duration of firing of mediaeval hearths from Coffee Yard,
York. Geophysical Journal International, 97, 357–359.
Merrill, R. T., McElhinny, M. W., and McFadden, P. L., 1998. The
Magnetic Field of the Earth. San Diego, CA: Academic.
Noel, M., and Batt, C. M., 1990. A method for correcting geograph-
ically separated remanence directions for the purpose of
archaeomagnetic dating. Geophysical Journal International,
102, 753–756.
Parés, J. M., and Pérez-Gonzalez, A., 1995. Paleomagnetic age for
hominid fossils at Atapuerca archaeological site, Spain. Science,
269, 830–832.
Pavón-Carrasco, F. J., Ostete, M. L., Torte, J. M., and Gaya-Piqué,
L. R., 2009. A regional archaeomagnetic model for Europe for
the last 3000 years, SCHA.DIF.3K: applications to
archaeomagnetic dating. Geochemistry, Geophysics and
Geosystems, 10, Q03013.
Rasmumussen, K. L., 2001. FOCUS: provenance of ceramics
revealed by magnetic susceptibility and thermoluminescence.
Journal of Archaeological Science, 28, 451–456.
Reinders, J., Hambach, U., Krumsiek, K., Sanke, M., and Strack,
N., 1999. An archaeomagnetic study on pottery kilns from
Bruhl-Pingsdorf (Germany). Archaeometry, 41, 413–420.
Schnepp, E., and Lanos, P., 2005. Archaeomagnetic secular varia-
tion in Germany during the past 2500 years. Geophysical Jour-
nal International, 163, 479–490.
Schnepp, E., Pucher, R., Goedicke, C., Manzano, A., Müller, U.,
and Lanos, P., 2003. Paleomagnetic directions and thermolumi-
nescence dating from a bread oven-floor sequence in Lübeck
(Germany): a record of 450 years of geomagnetic secular varia-
tion. J Geophys Res, 108, 2078, doi:10.1029/2002JB001975, B2.
 ARCHAEOMAGNETIC DATING
Schurr, K., Becker, H., and Soffel, H. C., 1984. Archaeomagnetic
study of medieval fireplaces and ovens and the problem of mag-
netic refraction. Journal of Geophysics, 56, 1–8.
Shaw, J., Yang, S., Rolph, T. C., and Sun, F. Y., 1999. A comparison
of archaeointensity results from Chinese ceramics using micro-
wave and conventional Thellier’s and Shaw’s methods. Geo-
physical Journal International, 136, 714–718.
Speranza, F., Maritan, L., Mazzoli, C., Morandi Bonacossi, D., and
D’Ajello Caracciolo, F., 2006. First directional archaeomagnetic
results from Syria: evidence from Tell Mishrifeh/Qatna. Geo-
physical Journal International, 165, 47–52, doi:10.1111/
j.1365-246X.2006.02914.x.
Stark, F., Cassidy, J., Hill, M. J., Shaw, J., and Sheppard, P., 2010.
Establishing a first archaeointensity record for the SW
Pacific. Earth and Planetary Science Letters, 298, 113–124.
Sternberg, R. S., 2001. Magnetic properties and archaeomagnetism.
In Brothwell, D. R., and Pollard, A. M. (eds.), Handbook of
Archaeological Sciences. Chichester: Wiley, pp. 73–79.
Sternberg, R. S., and McGuire, R. H., 1990. Archeomagnetic secu-
lar variation in the American Southwest, AD 700–1450. In
Eighmy, J. L., and Sternberg, R. S. (eds.), Archaeomagnetic Dat-
ing. Tucson, AZ: The University of Arizona Press, pp. 199–225.
79
Tarling, D. H., 1983. Palaeomagnetism: Principles and Applica-
tions in Geology, Geophysics and Archaeology. London: Chap-
man and Hall.
Tauxe, L., and Yamazaki, T., 2007. Palaeointensities. In Kono, M.
(ed.), Geomagnetism, Vol. 5, pp. 509–563.
Tema, E., and Kondopoulou, D., 2011. Secular variation of the
Earth’s magnetic field in the Balkan region during the last
8 millennia based on archaeomagnetic data. Geophysical Jour-
nal International, 186(2), 603–614.
Trapanese, A., Batt, C., and Schnepp, E., 2008. Sampling methods
in archaeomagnetic dating: a comparison using case studies from
Wörterberg, Eisenerz and Gams Valley (Austria). Physics and
Chemistry of the Earth, 33, 414–426.
Williams-Thorpe, O., Jones, M. C., and Thorpe, R. S., 1996. Mag-
netic susceptibility variations at Mons Claudianus and in Roman
columns: a method of provenancing to within a single quarry.
Archaeometry, 35, 15–42.
Zananiri, I., Batt, C. M., Lanos, P., Tarling, D. H., and Linford, P.,
2007. Archaeomagnetic secular variation in the UK during
the past 4000 years and its application to archaeomagnetic dat-
ing. Physics of the Earth and Planetary Interiors, 160(2),
97–107.
 B
BAND STRUCTURE
Jeroen W. Thompson
Department of Medical Physics and Applied Radiation
Sciences, McMaster University, Hamilton, ON, Canada
Synonyms
Electronic band structure
Definition
Configuration of allowed and disallowed electron states.
Crystals (structures consisting of regularly repeating
units) typically have “energy bands” in which an electron
may have any energy within a continuous, or nearly con-
tinuous, energy range. Energy bands are separated by
“band gaps” which specify a range of energies which are
impossible for an electron in that system to have.
A variety of models and mathematical tools can be used
to determine the nature of a particular crystalline mate-
rial’s electronic band structure from the basic principles
of quantum mechanics. For discussion here, we will con-
cern ourselves only with the highest-energy fully or par-
tially occupied band (“valence band”) and lowest-energy
unoccupied band (“conduction band”).
In an insulator, the ground state consists of a fully occu-
pied valence band and an empty conduction band. The
random excitations due to ambient temperatures are not
sufficient to add enough energy to an electron so that it
J.W. Rink, J.W. Thompson (eds.), Encyclopedia of Scientific Dating Methods, DOI 10.1007/978-94-007-6304-3,
© Springer Science+Business Media Dordrecht 2015
may occupy a state in the conduction band, where it is free
to move. In contrast, the ground state of a metal consists of
a partially occupied valence band. Here there is no energy
gap, and even thermal excitations are sufficient to popu-
late some higher-energy states in the valence, leaving
behind unoccupied states of lower energy. Electrons may
therefore absorb energy, be promoted to higher-energy
states, move (conduct) through the crystal, and finally
release energy when recombining with a so-called hole.
The physical basis of electron spin resonance, optically
stimulated luminescence, and thermoluminescence dating
(so-called “trapped charge” techniques) rely on defects in
the otherwise perfect pattern of a crystal. These defects
can introduce localized energy states in the band gap,
and transitions between localized defect states and
nonlocalized band states are responsible for the underly-
ing physics of these techniques.
Bibliography
Ashcroft, N.W. and Mermin, N.D., 1972. Solid State Physics.
New York: Holt, Rinehart and Winston.
Cross-references
Electron Spin Resonance (ESR) Dating, General Principles
Luminescence Dating
Quartz Defects, Optically Stimulated Luminescence and
Thermoluminescence
Radiation Defect
82 BETA COUNTER
BETA COUNTER
Jeroen W. Thompson
Department of Medical Physics and Applied Radiation
Sciences, McMaster University, Hamilton, ON, Canada
Definitions
Beta counter: Instrument for the detection of beta particles
and hence quantification of beta-emitting radioactive
material.
Beta particle (b
): Electron emitted from an atomic
nucleus during a nuclear decay in which the atomic number
(Z) increases by 1 (essentially, a proton is converted to
a neutron with emission of an electron from the nucleus).
Proportional counter: A gas-filled radiation detector in
which electrons are accelerated towards an anode, creating
secondary ionization events in the gas, which in turn cause
additional ionization events, etc. This cascade (Townsend
avalanche; charge amplification) results in electronic
pulses which are proportional to the energy originally
deposited within the detector by ionizing radiation.
Liquid scintillation counter: A liquid-filled radiation detec-
tor in which a radioactive sample is dissolved into a liquid
scintillator (typically based on an organic solvent such as
toluene). Deposition of energy through ionizing radiation
results in the emission of light by the liquid scintillator, with
detection by one or more photomultiplier tubes.
Beta counters are used in “conventional” 14C dating to
measure the activity of 14C. Regardless of detection tech-
nique, a measured count rate (decays/second [Bq] or equiva-
lent) is related to quantity through the known half-life of 14C.
Beta counting can also be used to directly measure envi-
ronmental beta activity in luminescence dating dose rate
determinations (Sanderson, 1988; Roberts et al., 2003).
The relatively low energy of the beta particles emitted by
14
C requires attention to the detector counting efficiency.
One approach is to convert the sample carbon to the gas
CO2, which is mixed with a fill gas and introduced into
a gas proportional counter. Alternately, the carbon can be
converted to benzene and then mixed with a solvent and liq-
uid scintillation cocktail for counting of light pulses emitted
as a result of energy deposited by individual beta particles.
Bibliography
Knoll, G. F., 2000. Radiation Detection and Measurement, 3rd edn.
New York: Wiley.
Roberts, H. M., Muhs, D. R., Wintle, A. G., Duller, G. A. T., and
Bettis, E. A., III, 2003. Unprecedented last-glacial mass accumu-
lation rates determined by luminescence dating of loess from
western Nebraska. Quaternary Research, 59, 411–419.
Sanderson, D. C. W., 1988. Thick Source Beta Counting (TBSC):
a rapid method for measuring beta dose rates. International Jour-
nal of Radiation Applications and Instrumentation Part D:
Nuclear Tracks and Radiation Measurements, 14, 203–207.
Cross-references
Radiation and Radioactivity
Radiocarbon Dating
BIG BANG
Gary R. Huss
Hawai’i Institute of Geophysics and Planetology,
University of Hawai’i at Mānoa, Honolulu, HI, USA
Definition
The most widely accepted model for how the universe
came to be is the Big Bang model. According to this
model, the universe expanded from an extremely hot,
dense state of essentially zero size (a singularity). From
this state, the universe expanded rapidly and cooled. As
the temperature dropped, energy was converted to sub-
atomic particles, which then formed protons, neutrons,
and electrons. Simple atomic nuclei formed at ~3 min after
the Big Bang, but it took ~379,000 years for neutral atoms
to form. Over the next billion years, slightly denser
regions of the nearly uniformly distributed matter gravita-
tionally attracted nearby matter and began to form clouds,
stars, galaxies, and other structures we observe today.
The Big Bang theory is based on three lines of evi-
dence. The first is Edwin Hubble’s observation that the
farther away a galaxy is, the more redshifted its light is.
The spectrum of a light source moving away from us at
high speed is shifted to longer wavelengths (toward the
red), while that of an object moving toward us will be
shifted to shorter wavelengths (toward the blue). Distance
can be reliably determined from Type 1a supernovae,
which occur in binary star systems where a white dwarf
accretes matter from its companion until it reaches the
Chandrasekhar limit of 1.4 solar masses. The star is
completely disrupted in a thermonuclear runaway that
makes a supernova as bright as an entire galaxy. Because
the precursor stars of Type 1a supernovae have similar
mass, the brightness is always about the same, providing
a “standard candle” against which the observed brightness
can be calibrated to give distance. The general observation
is that galaxies are receding from us in all directions, with
those farther away receding more rapidly, much as the
points on the surface of a balloon all become farther apart
as the balloon is inflated.
The second line of evidence for the Big Bang is the cos-
mic microwave background that permeates the universe.
The microwave background was discovered by accident
in 1964 by Arno Penzias and Robert Wilson, who were
conducting diagnostic observations at Bell Laboratories
using a new microwave receiver. During the initial stages
 BIOSTRATIGRAPHY
of the Big Bang, all of the particles and photons in the uni-
verse were in thermal equilibrium. Atoms and electrons
were unbound and photons were continuously scattered
by the electrons. Under these conditions, electromagnetic
radiation had a black-body spectrum. When the tempera-
ture of the universe fell sufficiently, electrons and protons
began to combine to form atoms. The scattering of pho-
tons ceased and photons could travel unimpeded. As the
universe expanded, the spectrum of photons was red-
shifted. Today, this primordial radiation falls into the
microwave region of the electromagnetic spectrum and
corresponds to a temperature of ~3 K.
The third line of evidence for the Big Bang is the rela-
tive abundances of H and He. The Big Bang model makes
specific predictions about how much H and He (and Li,
Be, and B) should have been produced as the temperature
dropped during expansion. The observed abundances of
H and He in the universe match the predictions very well.
The age of the universe can be estimated from the
expansion rate of the universe estimated from the rate at
which galaxies are receding from one another and from
the temperature of the microwave background. Models
of the evolution of the universe from the Big Bang until
today have become very sophisticated, giving an age of
the universe of about 13.8 billion years.
Bibliography
Hawking, S. W., 1988. A Brief History of Time. New York: Bantam
Books, 198 pp.
Hawley, J. F., and Holcomb, K. A., 2005. Foundations of Modern
Cosmology. New York: Oxford University Press, 554 pp.
Singh, S., 2004. Big Bang: The Origin of the Universe. New York:
Harper Collins, 560 pp.
Wineberg, S., 1993. The First Three Minutes. New York: Basic
Books, 191 pp.
BIOSTRATIGRAPHY
Marie-Pierre Aubry
Department of Earth and Planetary Sciences, Rutgers
University, Piscataway, NJ, USA
Definition and introduction
Biostratigraphy (from Gr. bios, life; L. stratum, bed;
Gr. grapho, write (Brown, 1991)) is arguably one of the
most broadly used stratigraphic methods in Earth Sciences
and concurrently one of the most misrepresented/misun-
derstood disciplines. In a strict sense, biostratigraphy is
the objective subdivision of the stratigraphic record based
on its paleontological content for the purpose of relative
dating and correlation, and, as such, it is described in inter-
national stratigraphic guides and national codes.
83
In the broadest sense, biostratigraphy encompasses the
whole spectrum of disciplines that reconstruct various
facets of Earth history from its sedimentary record, for
academic and industrial purposes. In this view, biostratig-
raphy is central to the reconstructions of paleocea-
nographic and paleoclimatic history, the formulation of
evolutionary theories, the description of sedimentary
architecture and of sea level history, and, foremost, the dis-
covery of geological time (or geochronology). This has
been emphatically demonstrated by McGowran (2005)
who illustrated the unprecedented role of microfossil
biostratigraphy in the discovery of Earth history.
In this entry, biostratigraphy is discussed as a method,
and I define its components and discuss the biostrati-
graphic correlations. This leads to the problem of diach-
rony and the importance of biostratigraphy in
determining the occurrence of stratigraphic gaps in sedi-
mentary sections. Finally I examine the significance of
integrated biostratigraphy to soundly interpret strati-
graphic successions and discuss a few examples of its role
as a means to interpret the paleontologic record. The ver-
tical succession of strata constitutes a clear evidence of the
passage of geological time, as Steno (1669) discovered,
leading to the formulation of the three principles that
would become the cornerstone of Earth Sciences. But what
are the relations of strata to time? Is there a one-to-one
relationship between rock and time, so that any stratigraphic
interval (thickness, measured in meters [m]) can be readily
converted into an equivalent amount of time (duration (for
distinction between Ma and Myr, the reader is referred to
Aubry (2009)), measured in million years [Myr]), and any
stratigraphic level (positioned with reference to Meter 0)
converted into a geological date (positioned as a point
expressed as mega-annum [Ma])? Or is the stratigraphic
record discontinuous, with concealed unconformities that
complicate the extraction of information about time from
stratal successions?
Biostratigraphy and paleontology
Biostratigraphy is intimately related to paleontology. This
is because biostratigraphy involves mostly the study of
extinct species, and for this reason, it was first known as
paléontologie stratigraphique (stratigraphic paleontol-
ogy) (D’Orbigny, 1852).
Species have a life span, which is the collective life
span of all the specimens in it, from the moment when
the species evolved to the moment when it became extinct.
Most of the species that ever lived are now extinct, but
many are known from their fossils, which are named based
on the Linnaean nomenclature of genus and species names
just as any living species. Not all fossils are of skeleton-
ized organisms, but the majority are, which results in bio-
stratigraphy being only compulsory for the relative dating
84 BIOSTRATIGRAPHY

Lithology Long-ranging Species Occurrences of Marker Species HO/LO Biozones

500m
B8

Sp. H
Lowest
Occurrence

Sp. G
Sp. 7
Interval
Zone
(LOIZ)
B7

Partial Range

Sp. 6
Zone
(PRZ)

B6
Highest
Occurrence
Sp. 5

Interval

Sp. B
Zone
(HOIZ)
B5

Sp. D Taxon-Range
Sp. 4

Sp. A

Zone
(TRZ)

B4
Highest to Lowest
Interval
Sp. E
Sp. 3

Sp. F

Occurrence
Zone
(HLOIZ)
B3

Concurrent
Sp. 2

Range Zone
(CRZ)

B2
Lowest
Occurrence
Sp. 1

Sp. C

Interval
Zone
(LOIZ)
B1

0m
B = Biohorizon Z = (Bio)Zone LO = Lowest Occurrence
Sp. = Species HO = Highest Occurrence

Biostratigraphy, Figure 1 An example of biostratigraphic subdivision of a section based on the range of selected species.
Biostratigraphic subdivision is arbitrary in the sense that a few species are selected among many to construct a biozonal framework.
In this example, the HO of Sp. F may have been chosen rather than the LO of Sp. H to delineate biohorizon B8.
 BIOSTRATIGRAPHY 85

Biostratigraphy, Table 1 Main taxonomic groups involved in biostratigraphy. Biostratigraphy relies on eukaryotic organisms,
those that have evolved a diversity of skeletal morphologies, and these comprise protists, invertebrates and vertebrates animals,
and (mostly) vascular plants. It can be readily inferred from this that not all fossils have equal potential with regard to
biostratigraphic usefulness. Species that, alive, had a narrow geographic distribution, as most benthic species did, do not
constitute good stratigraphic markers, at least not over long distances. (The term biomarker does not designate stratigraphic
markers but chemical molecules characteristic of specific organisms.) Benthic invertebrates, such as bivalves and echinoderms, are
biostratigraphically useful, but only regionally. This is true also of benthic protists such as large benthic foraminifera (e.g., the
Paleozoic fusulinids, Mesozoic Orbitoides, Cenozoic Nummulites) of warm water epicontinental deposits and of green algae such as
charophytes in brackish and lacustrine deposits. The distribution of vertebrates and vascular plants is also limited by climatic
barriers. Fossilized parts (shells, tests, gyrogonites, teeth, leaves) of these organisms usually serve for regional biostratigraphic
subdivision, but their trace fossils (burrows, foot tracks) may occasionally be helpful as well (Olsen et al., 2002)

Biological group Lifestyle Phyla/groups Biostratigraphic applications

Invertebrate animals Benthic Brachiopoda Regional

Echinodermata
Mollusca
Planktonic Hemichordata (graptolites) Broad
Vertebrate animals Terrestrial Regional
Marine Chordata (conodonts) Global
Nonvascular plants Brackish lacustrine Charophyta Regional
Vascular plants Terrestrial Regional
Unicellular organisms (“protists”) Marine, benthic Large benthic foraminifera Regional
Marine, planktonic Coccolithophores Global
Diatoms High latitudes
Dinoflagellates Epicontinental
Foraminifera Global
Radiolarians Low latitudes

Biostratigraphy, Figure 2 Morphologic trends in coccoliths (here discoasters) and relative age assignment. Each morphologic group is
symbolized by a species (left) (Modified after Martini, 1971b, Figure 39.1). Occasionally, when preservation in a stratigraphic section is
poor, there are no other means than broad morphology to determine its relative age. Also anomalous mixture of morphologies (e.g.,
morphologies represented by E. surculus and H. multiradiatus) is indicative of reworking (or contamination, depending on circumstances).
86 BIOSTRATIGRAPHY

Pleistocene

Birefringent ceratoliths

NCRZ
AZ3
Pleistocene
Upper

Non-Birefringent ceratoliths
Pleistocene
Lower

A sp.2

CRZ
C sp.1

AZ2
A sp.1

LOIZ
AZ1
Miocene
Upper

Biostratigraphy, Figure 3 The use of the optical properties of some coccoliths (here ceratoliths) helps delineate three biostratigraphic
intervals that are correlatable to chronostratigraphic units. This may be useful when coccolith assemblages have been strongly affected
by dissolution. The three biostratigraphic intervals can be regarded as assemblage zones, the older one comprising only species of
non-birefringent ceratoliths (genus Amaurolithus) and only them; the younger zone comprising only species of birefringent ceratoliths
(genus Ceratolithus) and only them; and the intermediate zone comprising species of both birefringent and non-birefringent ceratoliths.
Note that the lower two intervals could also be described in succession as follows: (1) lowest occurrence interval zone (between the LO of
the oldest species of Amaurolithus [A. sp. 1] and the LO of the oldest species of Ceratolithus [C. sp. 1]) and (2) concurrent-range zone
(from the LO of C. sp. 1 to the HO of the last representative of the genus Amaurolithus [A. sp. 2]).

and correlation of the Phanerozoic sedimentary strata. Its
usefulness extends, however, to the strata just below the
Cambrian, assigned to the Vendian series, in which soft-
body animals and traces of their activities (burrows in par-
ticular) are common and diverse.
The vertical extent over which a species occurs in sed-
imentary rocks is called its range, which is representative
of the life span of that species in the area where its fossils
are found. The range is delineated between two strati-
graphic horizons (Figure 1). A species (e.g., species D in
Figure 1) is absent below the lower (and older)
stratigraphic horizon but is present above it. It is
also absent above the upper horizon but is present
below it. Stratigraphers refer to the presence of a taxon
at a stratigraphic level as occurrence (the meaning of
“occurrence” seems to have changed through time. The
Roget’s II, 1980 (p. 645), gives three definitions for
occurrence:
1. “something that happens” [¼ circumstance];
2. “something significant that happens” [¼event]; and
3. “the condition or fact of being present” [¼ presence].
In contrast, the Roget’s II 2014 (http://thesaurus.com/
Roget-alpha-index.html) gives two definitions of
occurrence:
1. “the action, fact, or instance of occurring” and
2. “something that happens: event, incident”. I stress here
that “stratigraphic occurrence” should be understood as
“stratigraphic presence”).
Thus, the lower (older) horizon contains the lowest
occurrence (LO) of the taxon, which is the oldest strati-
graphic level in a section where the species is present,
while the upper (younger) horizon contains the highest
occurrence (HO), which is the youngest stratigraphic
level in the same section where the species is present.
LOs and HOs are biohorizons. The LO of a macrofossil
such as an ammonite species may appear significant
enough to qualify as a biohorizon, whereas the LO of a
5 mm coccolith may seem rather insignificant. Yet, the
same principle applies to microfossil (~63–250 mm) and
nannofossil (<63 mm to 2 mm) taxa as to macrofossils.
LOs and HOs are stratigraphic terms, positioned in a sec-
tion with respect to a specific point of reference.
Fossils were first treated as unrelated objects, as buttons
are, for instance, some sharing similar ranges, whereas
others would not. The concept of faunal succession
(Smith, 1815) was derived from this basic observation
and prior to the acknowledgement of the evolution of
organisms. But fossils are related to one another through
 BIOSTRATIGRAPHY 87

Assemblage Zone Abundance Zone

Taxon-Range Zone Lineage Zone

sp. B

sp. B

sp. A

Concurrent Range Zone Oppel Zone

sp. B

sp. A
Highest Occurrence Interval Zone Lowest Occurrence Interval Zone

sp. B

sp. B sp. A

sp. A

Highest to Lowest Occurrence Interval Zone Partial Range Zone

sp. B

sp. C

sp. A

Biostratigraphy, Figure 4 (Continued)

88 BIOSTRATIGRAPHY

Biostratigraphy, Table 2 Classification of biostratigraphic zones in the International Stratigraphic Guide (1976, 1984) and the North
American Stratigraphic Code (2005). Note the change in position of the lineage zone, from a type of range zone in Hedberg
(ed., 1976) to an independent type in Salvador (ed., 1994), the abandonment of the Oppel zone, and the agreement between the
current guide and code (North American Commission on Stratigraphic Nomenclature, 2005). Although the term is not used in
Hedberg (ed., 1976, p. 61), the concept of partial range zone is explicit in it, and Salvador (ed., 1994, p. 60) refers to its introduction
in the report of the Stratigraphical Code Subcommittee, Geological Society of London, 1967, p. 85

Hedberg 1976 Salvador, ed., 1994 NACSN, 2005

Assemblage zone Assemblage zone Assemblage biozone

Range zone Taxon-range zone Range zone Taxon-range zone Range biozone Taxon-range biozone
Concurrent-range zone Concurrent-range zone Concurrent-range zone
Oppel zone
Lineage zone Lineage zone Lineage biozone
Acme zone Abundance zone Abundance biozone
Interval zone Interval zone “Highest occurrence zone” Interval zone
“Lowest occurrence zone”
“Partial range zone”

organic evolution, which explains their enormous mor-
phologic diversity and the organization of this diversity
in specific stratigraphic intervals marked by the success
of particular groups or clades (thus the “Age of Fishes,”
those “of Reptiles,” “of Dinosaurs,” and “of Mammals”)
and the arrow of time that is perceived through the
unfolding of this diversity. The arrow of time confers upon
biostratigraphy its uniqueness: it is the only non-iterative
method of relative dating and the counterpart of the radio-
isotopic methods in numerical dating. Biostratigraphy is
the only means of relative dating that is self-sufficient,
although its integration with other stratigraphic means
(magnetostratigraphy, isotope stratigraphy, sequence stra-
tigraphy) enhances its potentialities.
The ideal species in biostratigraphy are those with
abundant and widespread geographic occurrences dur-
ing their life, regardless of latitudes and depth of
deposition (Table 1). They are marine planktonic spe-
cies, among which are the Paleozoic graptolites
(up to several mm long, Hemichordata), the phosphatic
conodont elements (~200 mm to 5 mm; Chordata), the
Paleozoic and Mesozoic cephalopods (Mollusca), and
the minute Mesozoic and Cenozoic planktonic protists
that belong to unrelated clades but are preserved
together in deep-sea sediments and among which the
planktonic foraminifera and coccolithophores occupy
a prominent role. Both are abundant in deep-sea oozes
and chalks, and they also occur in epicontinental
deposits; additionally, coccoliths constitute the fine
fraction of many such deposits. Examples in the fol-
lowing discussion are borrowed from these two
groups, because they have contributed more to bio-
stratigraphic thinking than any other group, but the
same principles and concepts in biostratigraphy apply
to all paleontological groups.
Basic concepts in biostratigraphy
The use in any form of fossils for relative dating and cor-
relation is part of biostratigraphy. For instance, morpho-
logic trends in a genus or differences in optical

Biostratigraphy, Figure 4 Types of biostratigraphic zones. An assemblage zone (AZ) is a stratigraphic interval characterized by the
co-occurrence of several taxa in combinations that may vary laterally and vertically. The taxa may be long ranging, with only one or
two having its LO or HO in the zone. AZs are mostly defined in shallow water or land deposits on macrofossils. An abundance zone
(AZ) is a stratigraphic interval characterized by the notable abundance of a specific taxon. The placement of the zonal boundaries may
be a matter of discussion, and the percentage of a taxon in the zone may vary broadly, depending on its usual frequency. A taxon-
range zone (TRZ) is defined by the total range of a taxon. A lineage zone (LZ) is defined between two horizons that record the
evolutionary transition of the species from its ancestor (base of the zone) to its descendent (top of the zone). TRZ and LZ are similar in
being characterized by a single taxon, but differ fundamentally in the nature of their boundaries. A concurrent-range zone (CRZ) is the
interval defined between the LO of one taxon and the HO of another taxon. A CRZ is marked by the co-occurrence of two species. An
Oppel zone is a hybrid between a concurrent-range and an assemblage zone in that it is comprised of several taxa (as an AZ) whose
ranges are mostly restricted to the zone (unlike an AZ and unlike a CRZ which may be comprised of two long-ranging taxa but whose
respective lower and upper ranges overlap). Oppel zones were mostly used in ammonite-based stratigraphy. Interval range biozones
(IRZ) are those defined by the range of one taxon to the exclusion of another one and may correspond to the interval between (1) the
HO of one species and the HO of another species, (2) the LO of one species and the LO of another species, and (3) the HO of a species
and the LO of another species. A partial range zone (PRZ) is defined by the occurrence of a selected species between the HO of a
species and the LO of another species. A PRZ is a constrained interval zone of the latter type.
 BIOSTRATIGRAPHY 89

Biostratigraphy, Table 3 Recent classification of biostratigraphic zones. The terms “base” and “top” for LO and HO are also used in
the method of graphic correlation

Berggren and Pearson (2005)

and Wade et al. (2011) Backman et al. (2012) This work

Taxon-range zone Taxon-range zone Range zone Taxon-range zone

Concurrent-range zone Concurrent-range zone Concurrent-range zone
Lowest occurrence zone Base zone Nonconcurrent-range zone Lowest occurrence zone
Highest occurrence zone Top zone Highest occurrence zone
Partial range zone Partial range zone Exclusion-range zone Partial range zone
Highest to lowest occurrence zone
Lineage zone
Abundance zone

Super- Biostratigraphy, Table 4 Examples of codifications for plank-

Biozone Subbiozone tonic calcareous microfossils
biozone
Biozone 7 Code Full name of zone Author

Biozone 6 N Neogene planktonic Banner and Blow,

foraminifera 1965
P Paleogene planktonic Blow, 1969;
Biozone 5 foraminifera Berggren, 1969
P1 to P5 Paleocene planktonic Berggren and
foraminifera Pearson, 2005
E1-E16 Eocene planktonic Berggren and
foraminifera Pearson, 2005
O1-O5 Oligocene planktonic Berggren and
Biozone 4 foraminifera Pearson, 2005
NP1-NN25 Nannoplankton Paleogene Martini, 1971a
NN1-NN21 Nannoplankton Neogene Martini, 1971b
Superzone

CP1-CP19 Coccolith Paleogene Okada and Bukry,

1980
Subbiozone 3 CN1-CN19 Coccolith Neogene Okada and Bukry,
1980
NTp1-NTp14 Nannoplankton, tertiary, Varol, 1989
Biozone 3 Subbiozone 2 Paleocene
NNTp1–NNTp North Sea, nannoplankton, Varol, 1989
tertiary, Paleocene
Subbiozone 1

Biozone 2

Biozone 1
Biostratigraphy, Figure 5 Superbiozones, biozones, and
subbiozones. These constitute a biostratigraphic hierarchy so
that the boundaries of a superbiozone are the lower and the
upper boundary of, respectively, its lowermost and uppermost
biozones. The division of a subbiozone necessitates the
introduction of an additional stratigraphic horizon to define the
subzonal boundaries, but the base and top of the lower and
upper subbiozones are those of the undivided zone.
properties among closely related species may be used to
characterize specific chronostratigraphic intervals
(Figures 2 and 3). These methods result, however, in broad
age assignments. Greater resolution is achieved with the
subdivision of stratigraphic succession into biostrati-
graphic zones (Figures 1 and 4, Tables 2 and 3) based on
the stratigraphic ranges of selected species or other
selected criteria. Occasionally, poor preservation of fossil
assemblages hampers fine resolution, leading to the
recourse of less satisfactory methods (Figures 2 and 3).
Biozone
Concept
The basic unit of biostratigraphy is the biostratigraphic
zone, or biozone for short, which is a stratigraphic interval
measured in terms of thickness (generally in meters and its
derivatives; occasionally feet in the industry) and charac-
terized by selected aspects of its fossil content.
A biozone is delineated between two biostratigraphic
90 BIOSTRATIGRAPHY

Zone
Biohorizon Epoch
Code Full Name

Early
NN2 Discoaster druggi Zone Miocene
NN1 Triquetrorhabdulus carinatus Zone

NP 25 Sphenolithus ciperoensis Zone

Late

Oligocene
NP 24 Sphenolithus distentus Zone

Early
NP 23 Sphenolithus predistentus Zone

NP 22 Helicopontosphaera reticulata Zone

NP 21 Ericsonia ? subdisticha Zone

NP19 Sphenolithus pseudoradians Zone

NP20 Isthmolithus recurvus Zone

Late
NP18 Chiasmolithus oamaruensis Zone

NP17
Discoaster saipanensis Zone

NP16 Discoaster tani nodifer Zone

Eocene
Middle

c
NP15

b Chiphragmalithus alatus Zone

b
NP14

Discoaster sublodoensis Zone

a

NP13 Discoaster lodoensis Zone

Early

NP12 Marthasterites contortus Zone

Biostratigraphy, Figure 6 Examples of calcareous nannoplankton biozonation. Martini’s zonation (1971a), here represented for the
lower Eocene (partim) through Oligocene, is still in broad use. NP ¼ nannoplankton Paleogene. The names of the zone and their code
are given together with the types of biohorizons that are defined by species ranges.
 BIOSTRATIGRAPHY 91

Planktonic Foraminifera

Series

Stage
(Sub) Tropical Transitional (Sub) Antarctic
Bow
Berggren, 1995 (1969) Berggren et al., 1983a, 1995 Berggren, 1992

Glb. nepenthes IZ
Gt. cibaoensis - N

Pliocene

Zanclian
b Gt. puncticulata IZ

Gt. tumida -
Glb. nepenthes ISZ

Lower
19
PL1 Gt. tumida -
Gt. cibaoensis N Gt. sphericomiozea
a
IRZ
18 IZ

Messinian
Neoglobquadrina
M14 Gt. lenguaensis - Gt. tumida IZ Gt. conomiozea/

pachyderma
Gt. mediterranea -
Mt10
Gd. extremus/ Gt. sphericomiozea AN

TRZ
N
Gt. plesiotumida- 17 IZ 7
b Gt. lenguaensis
ISZ

Upper
M13
P. mayeri -
N. acostaensis- Mt9

Tortonian
Gt. conomiozea
Gd. extremus/ N IZ

Miocene
a Gt. plesiotumida AN Gt. scitula
16 6 PRZ
ISZ

P. mayeri - N. acostaensis
M12 IZ
N15
M11 Glb. nepenthes/P. mayeri Conc. RZ N14 Mt8 Glb. nepenthes/P. mayeri Conc. Rz AN N. nympha
M10 Gt. F. robusta - Glb. nepenthes IZ
N13 5 TRZ
M9 b Gt. f. robusta ToT. RZ Gt. f. lobata-
Gt. f. robusta
a Gt. f. lobata Lin. Z IZ N12
Gt. peripheroronda -

Serravallian
M8 Gt. fohsi s.s. Lin. Z N11
Mt7 Glb. nepenthes
IZ
Gl. peripheroacuta

Middle
M7 N
Lin. Z
10 Orb. suturalis/
Mt6 Gt. peripheroronda AN Gt. miozea
M6 O. suturalis - Gl. peripher. IZ N9 Conc. RZ 4 PRZ Langhian
Pr. glomerosa -
Pr. sicana -
O. suturalis
Pr. sicana -
O. suturalis

Pr. glomerosa -
IZ

M5 b O. suturalis N8 Mt5 b O. suturalis

IZ

ISZ ISZ

Biostratigraphy, Figure 7 Examples of planktonic foraminiferal biozonal schemes. Note the loss of biostratigraphic resolution with
increasing latitude. In general, maximum biozonal resolution is obtained at low latitudes where species diversity is highest. Species
diversity decreases with increasing latitudes, resulting in the loss of marker species in high-latitude stratigraphies. In this example of
middle Miocene to lower Pliocene planktonic foraminiferal zonations for different latitudes, more than half of the biohorizons are lost
between low and high latitudes. The upper Serravallian–lower Tortonian stratigraphic interval can be subdivided into seven zones
and eight biozones at low latitude (M7–M13) but in only three biozones at mid latitudes and one biozone at high southern latitudes
(Modified from Berggren et al., 1995, Figure 5).

surfaces (biohorizons) that determine its base and top. considerably increases or across which a taxon is seen to
Most horizons are the LOs and HOs of chosen taxa (see have arisen from an ancestral species. Superposed
above). However, other biohorizons may be used, such biozones are separated by zonal boundaries so that the
as a surface across which the abundance of a species top of the underlying zone is also the bottom of the
92 BIOSTRATIGRAPHY

Biostratigraphy, Table 5 Strengths and weaknesses of specific types of biozones (From Hedberg 1976; Salvador, ed., 1994; North
American Commission on Stratigraphic Nomenclature, 2005)

Biozone type Definition Strength Weakness

Taxon-range zone TRZ Range of a species Presence of species required

Concurrent-range zone CRZ
Lowest occurrence interval
zone LOIZ
Highest occurrence interval
zone HOIZ
Highest to lowest occurrence
interval zone HLOIZ
Partial range zone PRZ
Lineage zone LZ
Abundance zone AZ
Overlap between the older range
of one species and the younger
range of another species
Interval between the LO of a
species and the LO of another
species
Interval between the HO of a
species and the HO of another
species
Interval between the HO of a
species and the LO of another
species
Partial range of a taxon between The introduction of a third taxon
the HO of a species and the LO
of another species
Interval between the two
phylogenetic transitions
Interval marked by the abundance
of a taxon
Presence of both species required
One of the types needed between
a TRZ and a CRZ
Successions of LOIZ associated
with times of increasing
diversification
Useful in drilled material
One of the types needed between
a TRZ and a CRZ
Successions of HOIZ associated
with times of decreasing
diversification
One of the types needed between Identification is difficult, based
a TRZ and a CRZ on the whole assemblage, if
isolated sample is to be dated
or if section encompasses only
one or two zones
constrains the HLOIZ
Potentially best first-order means of Taxonomy may be subjective.
chronostratigraphic correlations Requires quantification of
lineage between species
Abundance may vary between
regions; strict quantification is
not possible

overlying zone. A sequence of biozones based on the same
paleontological group is referred to as a biozonation or
biozonal scheme.
Types of biozones
Biozones differ from one another in two aspects. One is
the means used for defining boundaries, in which the base
and the top of a zone may be characterized by LO(s) and
HO(s) of a taxon (or taxa). The other is the content of
the zone itself, including occurrence and abundance of
one or several taxa. The combination of both characters
results in ten types of biozones distributed in two catego-
ries (interval and range zones, Figure 4). In assemblage
and abundance zones, the emphasis is placed on the con-
tent of the zone. In lineage and interval zones, the empha-
sis is placed on the boundaries, whereas in concurrent-
range, Oppel, and taxon-range zones, it is placed on both
content and boundaries. The number of taxa involved in
the definition of biozones varies from a single one
(abundance and taxon-range zones) to several
(assemblage and Oppel zones). Three species are involved
in the definition of a lineage zone (or phylozone) and a
partial range zone. A minimum of two species is needed
for defining concurrent-range and interval zones. Each
type of zone has strengths and weaknesses (see below),
and it is important to know the type of zone used for
biozonal assignment in order to evaluate its reliability.
Biozones are (or should be) established based on mor-
phologically characteristic species whose LOs and HOs
can be consistently delineated in sections over broad geo-
graphic areas. They may be subdivided into subzones and
regrouped into superzones (Figure 5). A biozone is
subdivided into subzones when species involved in the
subzonal definition have inconsistent occurrences, for
instance, exhibiting consistent LO, HO, or both in some
sections although not present in others. Superzones are
introduced when the older and the younger boundary of
the oldest and youngest biozones, respectively, are partic-
ularly prominent (easily identified and geographically
consistent).
Naming a biozone
Biozones (and super- and subzones) are formally defined
and named stratigraphic units. The procedures for their
establishment are explained in the International Strati-
graphic Guide and do not need repetition here. However,
it is important to emphasize that the naming implies that
each biozone is unique and should be used as defined.
Its designation includes the type of zone that is being
defined and the name(s) of the species that serves for its
 BIOSTRATIGRAPHY 93

definition. The (Miocene) Eudiscoaster quinqueramus

Zone a b c
Taxon-Range Zone (Martini, 1971a), the (lower
Eocene) Pseudohasterigerina wilcoxensis/Morozovella
velascoensis Concurrent-Range Zone (Berggren and Pear-
son, 2005), and the (Cambrian) Trichophycus pedum
Assemblage Zone (Landing et al., 2013) may serve as
examples.
Long taxonomic names in Latin do not, however, ease

CRZ
TRZ
AZ
LZ
communication among scientists other than specialists of

a b c

HOIZ
HLIZ
HOIZ6
LOIZ6

LZ6
HOIZ5
LOIZ5

CRZ
LZ5
HOIZ4
LOIZ4

LZ4

HLIZ
HOIZ3
LOIZ3

LZ3

TRZ
HOIZ2
LOIZ2

LZ2
HOIZ1
LOIZ1

Biostratigraphy, Figure 9 Taxon-range and concurrent-range

LZ1

zones cannot be in consecutive order, but are necessarily

Biostratigraphy, Figure 8 Biostratigraphic zonations based on:
(a) lowest occurrence interval zones, (b) highest occurrence
interval zones, and (c) lineage zones. Any one type allows
construction of a full biozonal scheme. This is different from
concurrent-range zones and taxon-range zones that can only be
used in combination with other biozones (Figure 9).
separated by either a highest–lowest occurrence interval zone
(HLOI zone, shown in a) or a partial range zone when only one
zone separates them or by a combination of a HLOI and one
lowest or highest occurrence interval zone when two zones
separate them. An HLOI zone or a highest occurrence interval
zone (b) necessarily intervenes between them and an overlying
zone of any type (c). Note that LO in (b) might be the biohorizon
defining the base of the taxon-range zone or the concurrent-
range zone in (c).

Biostratigraphy, Table 6 General difficulties encountered during biozonal correlations (Compiled from Pearson (1998))

Criteria Problem Consequences

Samples Density is too low Insufficient stratigraphic resolution

Size is too small Insufficient fossil recovery
Preservational biases Fossils are scarce Marker species are not found
Extraction of fossils is very difficult
Preservation variable between beds in same section Dissolution, abrasion, internal and external molds
Variable between sections
Displacement Reworking HOs are difficult to delineate
Displacement by fluids Dubious occurrences
Contamination In the field Anomalous occurrences
In the laboratory
Downhole contamination during drilling
94 BIOSTRATIGRAPHY

a
Extinction

Contraction
Contraction

Contraction Dispersal

Biochron
Time

Dispersal Dispersal

Speciation

Geographic Range

Pseudoextinction
biohorizon
Time

Extinction
biohorizon

Pseudospeciation
biohorizon

a b c
Biostratigraphy, Figure 10 Dispersal, extinction, pseudospeciation, and pseudoextinction biohorizons in a taxon-range zone
(Modified after Pearson, 1996, Figures 5.4 and 5.7). (a) Relationship of a taxon-range zone (species S) to space and time. Allopatric
speciation of species S is followed by broad dispersal, itself followed by regional geographic stability or contraction, and finally
geographic contraction until extinction of the species. In this model, the moment of speciation can only be known from one location;
whereas the dispersal of the species constitutes a diachronous biohorizon (or a series of discrete biohorizons), as does the
contraction of the biogeographic range. The extinction of species S is also known from a single location. (b) Possible relationships
between typological species and evolutionary lineages. (a) As a first approach, ranges of morphospecies are linked in a dendrogram
to show possible phylogenetic relationships. (b) Subsequent analyses show that transitional forms momentarily occur between
morphospecies (blue shading). (c) Phylogenetic reconstructions, showing when cladogenetic and extinction events occurred.
Combined, the cladogram and dendrogram allow distinction of pseudospeciation and pseudoextinction events that, although
meaningless with regard to evolutionary patterns, may be useful as biohorizons.

the groups whose species are involved in zonal defini- specific zonal scheme, associated with a number
tions. Thus, the most commonly used biozonations are corresponding to the position of the zone in the numerical
codified with acronyms representing the interval, taxo- succession of zones as counted in stratigraphic succession
nomic group, and/or geographic area concerned by a (Table 4). The Zone NP1 to Zone NP21 and Zone NN1 to
 BIOSTRATIGRAPHY 95

Magn.
Abundance Patterns of
Lithology Biozones Ranges of Selected Coccolithophores #/100 mm
2
Selected Coccolithophores

Samples
Itzurun Fm (Selandian)

0 200 400 600 800

59 V

NP5 NTp9 V
V
V
V
V
V
V
V
50 V
V
NTp8c A
V
V
A
NTp8b
0 V
A
R

40
0 30 0 16 0.0 2.0 0 6 0 8
C26n

R
(Aitzgorri Limestone Formation (Danian)

NTp8a

R
30
–5 R x
F R x
R

R
F
F
NP4
20
B2 NTp7b
F
A A x x
C C x x
F F cf cf x
C cf cf cf
F A x cf x
C C cf cf cf
–10
F cf x cf
R
D. alata
C27n

10 F F x
F cf cf

Diantholitha spp.
Lithoptychius spp.

Lithoptychius spp.

F cf x
B1
Diantholitha spp.

D. mariposa

D. magnolia

D. magnolia
NTp7a

D. mariposa

D. alata
1
C27r

Biostratigraphy, Figure 11 Abundance patterns of coccoliths of the genera Diantholitha and Lithoptychius in the upper Danian
section at Zumaia, Spain. Abundance patterns may be expressed as number of specimens/area (of a smear slide, prepared in a way
similar to a blood sample) or as percentage of a species in a whole assemblage. Note that the same biohorizons are delineated based
on qualitative and semiquantitative methods when the qualitative analysis has been rigorous. Red crosses: reworked specimens.
Note that species may be rare in the lower part of their range (Data from J. Criscione, personal communication, January 2014).

Zone NN21 of Martini (1971a) and the Zone P1 to Zone Biohorizons

P4, Zone E1 to Zone E16, and Zone O1 to Zone O6 of Just as a stratigraphic horizon is “an interface indicative of
Berggren and Pearson (2005) are among the most com- particular position in a stratigraphic sequence” (Hedberg
monly used in deep-sea stratigraphy. Examples of 1976, p. 14), a biohorizon is a “surface, or interface across
biozonations are given in Figures 6 and 7. which there is a significant and distinctive change in bio-
Biostratigraphic zones are equivalent, in their own cat- stratigraphic characters” (Salvador, Ed., 1994, p. 56).
egory, to lithostratigraphic and chronostratigraphic units, The LO and HO of each species are potential biohorizons.
but they are not as strictly established. Whereas the latter However, only a few such stratigraphic occurrences
require stratotypes, citations of reference section are only achieve the status of useful biohorizon. There are many
encouraged for biostratigraphic zones. Composite refer- reasons for this. To qualify, species must be common
ence sections are generally necessary for zonations that where they occur, their morphology must remain suffi-
cover a large chronostratigraphic interval. In contrast, it ciently stable under different environmental conditions
is highly recommended that the taxa used in zonal defini- for the species to be identifiable in different settings, and
tions be described as clearly as possible and illustrated their stratigraphic range with respect to broadly
(see below).
96 BIOSTRATIGRAPHY

Biostratigraphy, Table 7 Biohorizons defined on the basis of abundance patterns

Biohorizon Acronym Definition

Highest occurrence HO Youngest horizon where the species has been recorded
Highest common occurrencea HCO Horizon above which the species does not occur consistently (continuous and/or
common)
Reentrancea RE Horizon where a species re-occurs
Temporary high occurrenceb THO Level above which a species is anomalously absent
Crossover in abundance X Shift in dominance of one taxon relative to another
Paracme enda PE Horizon below which a species is temporarily absent or present with scattered single
specimens
Paracme enda PB Horizon above which a species is temporarily absent or present with scattered single
specimens
End of acmea AE Horizon below which a species exhibits peaks in abundance
Beginning of acmea AB Horizon above which a species exhibits peaks in abundance
Lowest common occurrencea LCO Horizon above which the species occurs consistently (continuous and/or common)
Lowest continuous common LCtO Horizon above which the species occurs consistently but in low to very low numbers
occurrencec
Lowest rare occurrencec LRO Horizon above which the occurrence of the species is rare
Lowest occurrence LO Oldest horizon where the species has been recorded
a
Raffi et al. (2006)
b
Fornaciari et al. (2010)
c
Monechi et al. (2013)

Pseudocryptic sp. A Pseudocryptic sp. B Pseudocryptic sp. C Pseudocryptic sp. D

50

40
Number of Specimens

30

20

10

1 2 3 4 5 6 7 8 9 10
Size μm
Biostratigraphy, Figure 12 Species and pseudocryptic species. Integrated morphologic and genetic studies of coccolithophores
have shown that living species are in fact an aggregate of several pseudocryptic species (e.g., Saez et al., 2003; Hagino et al., 2009).
This implies that the broad intraspecific variability attributed to single extinct taxon may in fact represent the cumulative variations of
several extinct pseudocryptic species (the extinctions of which may constitute the pseudospeciation biohorizons of Pearson, 1998).
Size may be a good indicator of pseudospeciation. Counts according to size class minimize the subjectivity involved in
semiquantitative analysis of abundance patterns.
 BIOSTRATIGRAPHY 97

Biozones Section
Mag. Means of Relative Dating
1 2 A

B3
a b c

Z3

Relative age of Section A

cx

interval a
Zone B2

Reversed Polarity
B2

interval c
interval b
Zone B2

Zone B2
Zone Z2

Zone Z2
Z2
cy

B1

Z1

1, 2 - Paleontologic Groups

Biostratigraphy, Figure 13 Integration of magneto- and biostratigraphy. Although magnetic reversals occurred iteratively through
time, each can be characterized by a particular species or group of species that existed or coexisted only during individual
magnetochrons. Thus, biozone B2 may be a taxon-range zone defined by species A that lived only from late Chron Cyn to mid Chron
Cxn. Assignment of a magnetozone associated with this biozone to any other chronal succession would be erroneous. In turn overlap
between stratigraphic means allows to determine more precisely the relative age of the section. Based on paleontological group
1 alone (a), the section covers interval Ia; based on integration of paleontological groups 1 and 2 (b), it covers only interval Ib; adding
magnetostratigraphy (c), it is reduced to interval Ic.

co-occurring taxa must be consistent. Thus, the most
prominent biohorizons serve as zonal boundaries. How-
ever, as stratigraphic investigation continues and as atten-
tion is given to specific taxonomic entities, new
biohorizons are delineated, which increases the biostrati-
graphic resolution within zones and subzones.
Biohorizons may be the LOs and HOs of additional spe-
cies, but also any prominent morphologic change
(in size, shape, or ornamentation) or abundance change
(of a species or in diversity; or between two taxa, the abun-
dance of one species increasing while that of the other
decreasing [the crossover biohorizon in nannofossil stra-
tigraphy]) are recorded over long distances.
Biostratigraphic correlations
The method of biostratigraphic correlation consists in
recognizing in lateral sections biohorizons that have been
formally defined in a reference section and to draw lines of
correlation between them on this basis. In this manner,
biozones are extended laterally to locations as far as the
biohorizons can be identified. Beyond the furthest
98 BIOSTRATIGRAPHY

Stratigraphic Section Time locations, the biozonation is no longer applicable. In prac-

m Ma tice, however, it is one thing to define a biozonal scheme
100 0
based on well-preserved fossils of species in an extended
section and another to recognize the same zonal scheme
(Bio)zone (Bio)chron
Z3 B3
in sections elsewhere. When the species have been
soundly selected and preservation is as good in the lateral
HO LAD sections as in the reference section, the characterization of
the biozones and delineation of their boundaries are made
with high confidence. However, this is not always the
case, and difficulty may increase with increasing geo-
Observation graphic distance and in particular across latitudes even
50 (Bio)zone (Bio)chron
Z2 B2
with planktonic organisms (Figure 7). Also each type of
Discovery
biozone has its own set of strengths and weaknesses that
cause uncertainty in its lateral extension, recognition,
and correlation (Table 5). An insufficiently recognized
problem is that the possible combinations between the
seven commonly used types of biozone do not number
LO FAD
(Bio)zone (Bio)chron
49 as statistics would imply, but much fewer (12). The
Z1 B1 most reliable types are the taxon-range and the
0 5
concurrent-range zones because they are defined by con-
tent and boundaries (see above). However, the two can-
Biostratigraphy, Figure 14 Relationship between rock not be in immediate succession, but are necessarily
and time.
separated either by a highest occurrence interval zone, a
highest to lowest occurrence interval zone, or a partial
range zone, which are less reliable (Figures 8 and 9). In
other words, the least dependable types of biozone sepa-
rate the two types upon which stratigraphic sections can
be most firmly anchored to a biozonation. This places
major demand on the delineation of the biohorizons that
Biostratigraphy, Table 8 Biostratigraphic and biochronologic define taxon-range and concurrent-range zones (see
terminology. The dual stratigraphic versus temporal terminol- below).
ogy shown here is recommended although it is not always
followed. The term “biohorizon” is also referred to as surface, Aside from the latitudinal effect on correlation
horizon, level, limit, boundary, band, marker, index, datum, potentials and the uneven reliability within most
datum plane, and key horizon. First and last appearance data biozonal schemes that is inherent to the very types of
(FAD, LAD) are commonly used in synonymy with biohorizon. component zones, two categories of difficulties may
However, first and last are temporal terms (in opposition to arise in extending zonal boundaries away from the ref-
lower and upper that are stratigraphic terms) and are more erence sections. In the first category are general diffi-
appropriately used to designate temporal events, such as
evolutionary appearance and extinction (see discussion in
culties that have been discussed in detail by Pearson
Aubry (1995, 2007)) (1998; Table 6). It can be added that the use of more
than one taxon to define a zonal boundary introduces
Stratigraphic an uncertainty in its location because it is rarely the
terminology Temporal terminology case that two species have precisely the same LOs or
HOs in normal stratigraphic successions. In short, the
Terms Terms Synonyms greater the number of species involved in defining a
Biozone Biochron
Biohorizon Bioevent zonal boundary, the greater the chance of characterizing
Highest Last appearance Extinction of taxon it generally, but the less the precision in its delineation.
occurrence datum (LAD) This is a common problem with coccolithophore stra-
(HO) Last occurrence Cases of pseudoextinction tigraphy, for example. In the second category are diffi-
datum (LOD) culties relative to taxonomy.
Lowest First appearance Evolutionary appearance of a It cannot be overemphasized that the concept of
occurrence datum (FAD) taxon
(LO) First occurrence Dispersal, cases of biozone is strictly dependent of the taxonomic concept(s)
datum pseudospeciation of the species used to define it (e.g., Hedberg, 1976,
Abundance Acme p. 55). Yet, taxonomy is a main source of uncertainty in
Stratigraphic Evolutionary Speciation, extinction, stratigraphic correlation, and the more so that there is a
event event cladogenetic divergence general tendency to minimize this fact, in concert with
Biogeographic Contraction/expansion of the general disregard for taxonomic work in many circles
event geographic distribution
Range Span
today. The taxonomic problem is about consistency:
(1) how to consistently recognize a species throughout
 BIOSTRATIGRAPHY 99

Biostratigraphy, Table 9 Examples of diachrony exhibited by species of planktonic foraminifera (Wade et al., 2011) and
coccolithophores (Aubry, 1992; Raffi et al., 2006; Backman et al., 2012). For reference (Raffi et al., 2006), only reliable (“A”) data are
cited. Note that although ages in different works cited are based on different time scales and should not be compared, the
amplitude of diachrony varies little between time scales. LOD last occurrence datum, FOD first occurrence datum, EM Eastern
Mediterranean; (Aubry, 1992): latitudinal diachrony. Other works: different water masses at low latitude sites (Aubry, 1992;
Raffi et al., 2006; Wade et al., 2011; Backman et al., 2012)

Datum Age area 1 (Ma) Age area 2 (Ma) 6¼ (kyr) Age area 3 (Ma) 6¼ (kyr) References

LOD Dentoglobigerina altispira Atlantic: 3.13 Pacific: 3.46 330 Wade et al., 2011
LOD Sphaeroidinellopsis seminulum Atlantic: 3.16 Pacific: 3.57 410 Wade et al., 2011
FOD Globorotalia tumida Atlantic: 5.63 Pacific: 5.51 120 Wade et al., 2011
FOD Nickilithus amplificus Site 844B: 6.83 Site 845A: 6.75 80 Backman et al., 2012
FOD Amaurolithus primus Atlantic: 7.39 844B: 7.38 40 845A: 7.42 40 Backman et al., 2012
FOD Discoaster berggrenii Atlantic: 8.20 844C: 8.43 23 845: 8.56 36 Backman et al., 2012
LOD Pseudoemiliania lacunosa Atlantic: 0.440 Pacific: 0.436 40 EM: 0.467 31 Raffi et al., 2006
LOD Eudiscoaster brouweri Atlantic: 1.926 Pacific: 2.06 134 EM: 1.950 90 Raffi et al., 2006
LOD Reticulofenestra pseudoumbilicus Atlantic: 3.81 Pacific: 3.79 80 EM: 3.839 49 Raffi et al., 2006
LOD Eudiscoaster quinqueramus Atlantic: 5.54 Pacific: 5.59 50 EM: 5.54 50 Raffi et al., 2006
LOD Amaurolithus amplificus Atlantic: 5.978 Pacific:
EM: 5.939 39 Raffi et al., 2006
FOD Amaurolithus primus Atlantic: 7.362 Pacific:
EM: 7.424 62 Raffi et al., 2006
LCO Eudiscoaster kugleri Atlantic: 11.578 Pacific:
EM: 11.596 18 Raffi et al., 2006
LCO Reticulofenestra floridana Atlantic: 13.334 Pacific: 13.294 40 EM: 13.283 51 Raffi et al., 2006
LOD Sphenolithus delphix Atlantic: 23.065 Pacific:
EM: 23.089 24 Raffi et al., 2006
FOD Sphenolithus delphix Atlantic: 23.328 Pacific:
EM: 23.356 28 Raffi et al., 2006
LOD Reticulofenestra umbilicus LL: ~32.3 HL: ~31.3 1 Aubry, 1992
LOD Ericsonia formosa LL: ~32.8 HL: ~39.7 6.9 Aubry, 1992
LOD Heliodiscoaster saipanensis LL: ~ 34.2 HL: ~35.4 1.2 Aubry, 1992
LOD Heliodiscoaster barbadiensis LL: ~34.3 HL: ~39 4.7 Aubry, 1992
LOD Reticulofenestra reticulata LL: ~35 HL: ~36.1 0.9 Aubry, 1992

Locality Locality Locality

A B C its spatial and temporal range and (2) how to consistently
a
b obvious in fossil assemblages as it is in living populations.
= volcanic ash layer
= time line
Biostratigraphy, Figure 15 Testing for diachrony. In this classic
example, a paleontological event (extinction event) is
considered diachronous because of its inconsistent relationship
to a formational boundary (a). However, the finding of an ash
layer of characteristic mineral composition (one of the
shortest possible time markers) indicates that the extinction
level is synchronous, whereas the lithological contact is
diachronous (b).
place biohorizons so as to enhance precision in
correlation.
Pearson (1998) and McGowran (2005) addressed the
first question by discussing the conflict between the typo-
logical concept in paleontology and the population con-
cept in biology. Species are not morphologically stable
in the sense that no two individuals are alike, which is as
However, studies of living populations are mainly
concerned with spatial variations, whereas studies of
extinct species are concerned with both spatial and tempo-
ral variations. Morphologic evolution has been
documented for very few protistan species (e.g.,
Knappertsbusch, 2000, 2007; Hull and Norris, 2009).
These studies have shown complex evolutionary patterns,
but with long intervals of morphologic stasis interrupted
by brief intervals of high morphologic variability
(in close agreement with the theory of punctuated equilib-
rium, see Gould and Eldredge, 1993) which lends cre-
dence to the application of the typological approach to
biostratigraphy.
The second question is answered differently in the dif-
ferent disciplines of biostratigraphy, depending on the
morphological characteristics of the fossils, as shown by
the following two examples:
100 BIOSTRATIGRAPHY

Time
F E D A B C
(Ma)

BZ3
BZ3
BZ3 LAD BZ3

Strat. Gap
BZ3
Stratigraphic Gap

(Hiatus)
T
BZ3
(Hiatus)

HO ≠ LAD

HO ≠ LAD HO ≠ LAD
= LOD

Biochron (myr)
BZ2 BZ2 BZ2 BZ2 BZ2 BZ2

LO ≠ FAD LO ≠ FAD
= FOD
Stratigraphic Gap

LO ≠ FAD
(Hiatus)

Strat. Gap
(Hiatus)

T BZ1
BZ1
BZ1 FAD BZ1
BZ1
BZ1

Biostratigraphy, Figure 16 Comparison between diachronous stratigraphic patterns and patterns resulting from unconformities
(Modified after Aubry, 1995). Consider six sections (A–F) assigned to biozones BZ1 to BZ3, zone BZ2 being a taxon-range zone defined
by the range of the planktonic species Aglae xanthos, which is abundant and equally well preserved in all sections. Zones BZ1 and BZ2
are interval range zones. Aglae xanthos had a life span of 2 Myr, between its FAD at 54 Ma (=T1 in figure) and its LAD at 52 Ma (= T2;
FAD and LAD as defined in Table 8). The range of A. xanthos in section A represents the whole life span of the species, so that the LO
and HO in the section correspond to the FAD and LAD of the species, at 54 and 52 Ma, respectively. The LO of A. xanthos is delayed in
section B and even more in section C, whereas its HO occurs prematurely in the two sections, so that neither stratigraphic event
corresponds to the FAD and LAD of the species. The LO is <54 Ma in section B and >> 54 Ma in section C, and the HO is >52 and
>> 52 Ma in sections B and C, respectively. This stratigraphic pattern is typical of diachrony, the delayed occurrence in sections B and
C corresponding to dispersal following evolutionary divergence or to ecologically forced expansion of the species, whereas the
premature regional extinction results from contraction of the geographic distribution of the species. Algae xanthos exhibits a
stratigraphic pattern in sections D to F that mirrors the diachronous pattern in sections A to C, but this latter results from truncation of
its stratigraphic range in sections E and F. The important point here is that the biohorizons in sections B and C have significance with
regard to the evolutionary and/or biogeographic history of the species and represent regional events (first and last occurrence),
whereas the same biohorizons in sections E and F do not carry any evolutionary or paleobiogeographic significance. Blue arrow
indicates migration; red arrow indicates contraction of biogeographic distribution.

1. The archetypal shape in the planktonic foraminifera
is a sphere, and the chambers that constitute the
test are modified spheres arranged around an axis.
This is the cause for an “infinite” morphologic
diversity in the group and high intraspecific mor-
phologic variability within lineages and species.
Except at high rank, taxonomy in this group is
dominated by morphology over structure. Special-
ists of foraminifera are thus confronted with a
broad morphologic diversity whose significance in
terms of evolutionary history must be determined.
Pearson (1998) has thus advocated the use of
biohorizons reflecting the evolutionary history of
lineages: dispersal, extinction, pseudospeciation,
and pseudoextinction biohorizons (Figure 10).
2. Coccoliths are the articulated or loosely arranged skel-
etal pieces of the coccospheres of coccolithophores.
Their shape varies little, but their structure differs
markedly between lineages, so that taxonomy down
to the genus level can be rigorously based on structure,
species determinations being based on small morpho-
logical differences. The specialist here is confronted
with having to determine the range of tiny fossils that
may be few in the lower part of their range (possibly
during species dispersal), reworked in its upper part,
and whose identification is made ambiguous by poor
 BIOSTRATIGRAPHY 101

500

1
1
400 2
2
3
3
4
4
1
1
300 5
5
Section Y

6 2
2 6
5
5 - 8 -7 - 9 - 10
3 10 9 7 8
200
3 6
6
4
4

100 7
7

8
8
9 10 9 7 8 6 5 4 3 2 1
9
9 8 7 4 6 3 5 2 1

0 100 200 300 400 500

(0) (1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
Standard Reference Section X
Biostratigraphy, Figure 17 Method of graphic correlation. Distance between stratigraphic events in a section is compared to their
distance in a standard reference section. Note the terrace indicative of a stratigraphic gap. Numbered circles and crosses represent
LOs and HOs, respectively [base and top in the method] (Modified from Miller, 1977, Text-Figure 5).

preservation. Specialists have thus introduced semi-
quantitative methods (Backman and Shackleton,
1983, Figure 11), where abundance patterns help
determine reliable biohorizons. The methodology is
broadly applied but encourages a cumbersome termi-
nology that may obscure its benefit (Table 7). It also
raises a conundrum. If the typological definition of
species is used in biostratigraphy and if the intraspe-
cific variability of species is not known, how can abun-
dance patterns be reliable? There can only be a large
amount of subjectivity involved in taxonomic determi-
nation. This will need to be addressed, particularly in
view of the existence of pseudocryptic species
(Figure 12).
Biostratigraphy and stratigraphic interpretation
Biostratigraphy is not only the most readily available
means of relative dating and correlation of sedimentary
rocks, it is also the cheapest and the fastest. For this rea-
son, micropaleontologists are the first specialists onboard
ship or on-site to receive sediments retrieved from core
catchers during scientific and industrial drilling.
Biostratigraphy also provides control on the identification
of iterative stratigraphic information such as magnetostra-
tigraphy and sequence stratigraphy (Figure 13). At the
same time, their integration results in significant refine-
ment of biostratigraphic assignment and correlations
(Figure 13). This procedure is at the root of the ever-
growing role of biostratigraphy in geochronology, which
is the development of a chronology (a calendar indeed)
in which biohorizons are converted into biotic events. In
this manner, the LO and HO of species, both measured in
terms of height/depth in a stratigraphic section, are
converted into evolutionary events—e.g., evolutionary
appearance and extinction— expressed in terms of dates
(from which durations are derived). It is highly suitable that
a dual terminology be used for biostratigraphy
and biochronology, the former being the product of obser-
vation and the latter an inference (Figure 14 and Table 8).
The fundamental problem in biostratigraphy is the tem-
poral significance of biohorizons. Does every biohorizon
record a specific event that can be singled out in the life
span of a species? For instance, does every LO and HO
correspond to, respectively, an FAD or an LAD? With
Hedberg (1976. Ed., p. 63), we must recognize that
102 BIOSTRATIGRAPHY

LITHOL.UNITS
DEPTH
BIOZONES
CORE
(m)
TIME (Ma)

(1) (2) (3) 56 55 54 53 52 51 50 49 48 47

C.gigas
NP16
26x P12-

20
(nannofossil ooze and chalk)
P14 Core 27x-5
240 27x
28x Core 27x-6
P12
29x
260
UNIT 2

30x
P11

31x (≥5cm/103yr)
NP15b

280
21

32x
33x
300
34x
35x
320 P10
36x
Core 37-3, 81cm
37x
NP14b

340
38x N. fulgens C. gigas
39x R. inflata
(porcellanitic foraminiter nannofossil chalk to porcellanite)

360 40x
NP14a

41x ? (≡6.5cm/10 3yr)

380 42x
P9

43x D. sublodoensis
400 44x
45x
420 46x (≡8cm/103yr)
47x
UNIT 3

NP13

440 48x base Pg •

49x

460 50x
51x
24

52x
480
P7 – P8

53x T. orthostylus
54x
500
55x
NP12

56x
520
57x Lowest occurrence
or 25

D. lodoensis
26

58x Highest occurrence

540
59x No recovery
60x NP11 P6b Core 60x-cc
560 61x MAESTRICHTIAN Core 61x

Biostratigraphy, Figure 18 (continued)

 BIOSTRATIGRAPHY 103

PLANKTONIC FORAMINIFERA

CALCAREOUS NANNOPLANKTON
CUTTING SAMPLES
DEPTH (FEET)

ZONE
FORMATION
LITHOLOGY

ZONE
GAMMA LOG

0 (API) 150 TIME (MA)

23 22 21 20 19 18 17 16 15 14
3000-3050

INDET.
3000'

3350-3400
3500'
3620-3650 3620´ Upper sample: 2100´-2150´
3800-3820 Lower sample: 11750´-11760´
S. heteromorphus (3820´)
Scale:1:15000
INDETERMINATE
4000' 4000-4020

4200-4220 Highest Occurrence (HO)

4420-4440
4500' Lowest Occurrence (LO)
4600-4620

4880-4890
5000'
5110-5140

~ cm/10 3yr)
(=79
5500' NN5

S. schlumbergeri
5880-5910

C. cancellata
6000'
6090-6130
6200-6210

6420-6450
6500'
6600-6630

6810-6840 6810´
CARAPITA

7000' 7020-7050
7020´ Unconformity 2 at~ 7410´
N9/ H. ampliaperta (7410´)
7210-7240
M6 7410´ 7410´
7410-7440
7500' 7410´
O. suturalis (7410´)
C. subglobosus

7620-7650 7620´
N8/
R. multicostata

7810-7840 7810´
8000'
M5
A. pozonensis

8210´
8210-8230 P. sicana (8210´)
8380-8400
8500'
8610-8640

8820-8850 N7/
9000' M4 ~ cm/103 yr)
(=47
9000-9030
9200-9220 NN4
9400-9420
9500' 9600´
9600-9620 G. stainforthi (9600´)
9800-9820
R. transversa

10000'
10020-10050
B. pisciformis
B. floridana

10210-10240
N6/
N. incisum

10500'
10410-10440 M3
10620-10650
E. scabra

10800-10830 11800´ Unconformity1 at~ 11260´

11000' 11020-11050 11020´
A. beccarii

11220-11245 11220´
11300-11310 N5?/ 11300´
11400-11405 M2 NN3?
11500' 11500-11510
11500´ 11400´
NARICUAL

11600-11610 11500´ 11500´

11750-11760 S. belemnos(11300´)
ANALIZED
ANALIZED

NOT
NOT

11400´

Biostratigraphy, Figure 18 Method of sedimentation rate curves. Sedimentation rate curves are constructed by comparing thickness
between biostratigraphic events to duration between biochronologic events. When sedimentation is continuous, the thickness of
biozones is essentially proportional to duration, and changes in rates result in changes in the slope of the curves. Stratigraphic gaps are
marked by terraces (as in graphic correlation) corresponding to the overlap at the same stratigraphic level of multiple biohorizons.
(a) Lower–middle Eocene section recovered from Deep Sea Drilling Site 612 (From Aubry, 1995) and (b) lower–middle Miocene section
from an exploration well (From Sanchez et al., 2014).
104 BIOSTRATIGRAPHY

CHRONOLOGY

BIOCHRONS
MAGNETO-
TEMPORAL
INTERPRETATION

TIME
SECTION SECTION SECTION
A B C

Car
a SECTION t6
T

Cbn
A

BZ3
0m
t6
SECTION t6
C
Mxn

0m t5
BZ3

BZ3
Mxn
t6 SECTION
B t6 t5 t4
10m 0m
PB T+1
Mxn

t5
t6 BZ3
t5 t4
t4
10m
t5

Cbr
t4

BZ2
PB t4 t4

BZ2
20m 10m
t3
BZ2
BZ2

t3
t3 T+2
PB

t2 t3 t5
20m t4
BZ1

t1 t2
Mvn

t2 t2 t3
30m 20m t2
t2
BZ1

t1 t1 t1 t1 t1
T+3
Mvn

BZ1
Ccn
BZ1
Mvn

30m

40m

Biostratigraphy, Figure 19 Stratigraphic resolution versus temporal resolution and pseudo-diachrony. Consider three stratigraphic
sections of different thicknesses, exhibiting a major paleobiological event [PB]. PB occurs in zone BZ2 in sections A and B, but in the
middle of the zone in section A and in its lower part in section B. PB occurs at the BZ2/BZ3 zonal boundary in section C. Whereas at face
value event PB is diachronous, temporal interpretation (cf. Figure 18) of the sections using integrated magnetobiostratigraphy shows
that correlative unconformities in sections B and C and associated with different hiatuses are responsible for apparent diachrony. Note
also that the temporal resolution considerably differs in the three sections (Modified from Aubry, 1998, Figure).

“Biostratigraphic correlation is not necessarily time
correlation.” This touches on the common problem of
diachrony (¼diachroneity) that has plagued biostratigra-
phy by casting doubt on its reliability. Diachrony concerns
delayed LOs and premature HOs (or other biostratigraphic
events) in stratigraphic sections (see McGowran, 2005, for
a thorough discussion). The dispersal of species may not
be instantaneous, or the geographic range of a species
may contract or expand momentarily (Figure 10). Latitu-
dinal diachrony generally leaves a clear signature, and it
involves long durations measurable in million years
(Table 9), whereas ecological diachrony related to differ-
ent water masses is of less amplitude. Tests allow determi-
nation of diachrony (Figure 15), and orbital chronology is
used to measure it (Table 9).
Diachrony is a rationalized process that immediately
comes to mind to explain unexpected biostratigraphic pat-
terns. However, unconformities produce patterns that
mimic diachrony (Aubry, 1995, Figure 16), and failure
to decipher unconformities leads to misinterpretation of
biohorizons, which leads, in turn, to miscorrelation and
incorrect age assignments and dating. Methodologies in
two categories have been devised to delineate unconfor-
mities in stratigraphic sections. The first category,
reviewed by Gradstein (2004), comprises deterministic
and probabilistic quantitative methods that are applicable
to all Phanerozoic stratigraphies and are generally used
where classical biostratigraphic means are not applicable,
in particular in industry. Graphic correlation (Shaw, 1964;
Edwards, 1984) is classical among them (Figure 17). The
other methodologies are based on the analysis of sedimen-
tation rates curves, in which terraces are indicative of
stratigraphic gaps (Figure 18). Terraces are formed at
stratigraphic levels where LOs and HOs are juxtaposed
although they markedly differ in age. This is at the origin
of the method of temporal interpretation (Aubry, 1995) in
which stratigraphic sections are projected into a temporal
framework so that the duration (hiatus) represented by
unconformities is fully accounted for.
The best documentation of evolutionary histories con-
cerns the planktonic protists, and this can be attributed to
their abundant, generally well-preserved, broadly distrib-
uted (the ocean occupies ~70 % of the surface of the
planet) fossil record. In contrast, reconstitution of the evo-
lutionary history of marine invertebrates and, to a lesser
extent, benthic protists, has been problematic, largely
because of their narrow dependence on specific environ-
mental parameters. Their paleontologic records are biased
 BIOSTRATIGRAPHY 105

TIME TIME

CCn
CCn
Shelf
Faunal

Pd

Pd
T A T
History
B
C
NPe

NPe

CCr
CCr

T D T
Pc

Pc
Unconformable part
NPd

NPd

CBn
CBn

of the
T Sequence Boundary Conformable part T
of the Sequence Boundary
(STRATIGRAPHIC GAP/HIATUS)
Pb

Pb
CBr
CBr

NPc

NPc
T T
NPa NPb

NPa NPb

CAn
CAn
Pa

Pa
T
a T

CCn
CCn
Pd

Pd
T A T
B
C
NPe

NPe

CCr
CCr

T D T
Pc

Pc
Unconformable part
NPd

NPd

CBn
CBn

of the
T Sequence Boundary Conformable part T
of the Sequence Boundary
(STRATIGRAPHIC GAP/HIATUS)
Pb

Pb
CBr
CBr

NPc

NPc
T T
NPa NPb

NPa NPb

CAn
CAn
Pa

Pa
T
b T

CCn
CCn

C
Pd

Pd
T B T
A
NPe

NPe

CCr
CCr

T T
Pc

Pc

Unconformable part
NPd

NPd

CBn
CBn

of the
T Sequence Boundary

(STRATIGRAPHIC GAP/HIATUS)
?
Conformable part
of the Sequence Boundary
? ? T
Pb

Pb
CBr
CBr

NPc

NPc

T T
NPa NPb

NPa NPb

CAn
CAn
Pa

Pa

T T
c
Biostratigraphy, Figure 20 Integrated magnetobiostratigraphic studies of sections carefully located along the sequence boundaries
should, in principle, help resolve evolutionary patterns among organisms living on epicontinental shelves. In this example, if the
hiatus associated with a sequence boundary decreases basinward, study of a continuous (temporally complete) section across the
correlative conformity will allow determination of stepwise (a) or abrupt (b) extinctions. However, if the hiatus increases or remains
constant basinward, the event cannot be constrained (c).
106 BIOSTRATIGRAPHY
by several factors, including artificial range truncation and
stratigraphic gaps, which has led to formulation of models
to predict their evolutionary patterns (e.g., Signor and
Lipps, 1982; Springer, 1990; Holland, 1995). The method
of temporal interpretation has permitted illustration of the
role of unconformities in altering the record of paleobio-
logical events as suspected by paleontologists and has
shown that temporal resolution does not necessarily match
stratigraphic resolution (Figure 19). It has also strongly
indicated that unconformities at sequence boundaries on
continental margins extend to the deep sea (Aubry, 1991,
1995), which has notable implications for the documenta-
tion of paleobiological events among marine invertebrates
(Figure 20). Sequence stratigraphy postulates that the hia-
tus associated with sequence boundaries decreases along a
coast to basinal transect so as to be nonexistent at the cor-
relative conformity (e.g., Van Wagoner et al., 1998;
Catuneanu et al., 2009). If the model of sequence stratigra-
phy is correct in this, paleobiological studies along depth
transects on continental margins will help verify or sup-
plant models of evolutionary patterns (Figure 19a, b).
However, if the model of sequence stratigraphy is incor-
rect with regard to the decreasing hiatus down dip, tests
of evolutionary models through studies of selected tran-
sects will not yield decisive information (Figure 19c).
Conclusions
Biostratigraphy is the only non-iterative method of rela-
tive dating. Based on firm foundations, capable of evolv-
ing through the developments of new techniques, the rise
of new concepts and their integration with other strati-
graphic means it is the most dependable means of
stratigraphic correlation and dating available to the
stratigrapher. It is at the root of biochronology, and it
makes possible insights into the rates at which earth and
life processes occur on geological time scales. As such it
is a major part of the framework upon which the history
of the Earth system is being deciphered.
Some immediate and fundamental problems that bio-
stratigraphy can help address concern the completeness
of stratigraphic sections. Although Ager (1973) cautioned
that the sedimentary record is highly discontinuous every-
where, overconfidence is increasing, unchallenged, that
biostratigraphic records are reliable markers of time, seis-
mic patterns are chronostratigraphic, and evolutionary
modes of life can be restored from direct recovery of strat-
igraphic patterns and without tests of temporal continuity.
These are fundamental questions that biostratigraphic
studies must continue to address in the future.
Acknowledgments
I am most grateful to Jack Rink for inviting me to contrib-
ute to this entry on biostratigraphy; to W. A. Berggren,
D. Bord, P. N. Pearson, W. Si, and J. A. Van Couvering
for the discussion and their review of this entry; and to
Sarah Klingler for the artwork.
Bibliography
Ager, D. V., 1973. The Nature of the Stratigraphical Record.
New York: Macmillan.
Aubry, M.-P., 1991. Sequence stratigraphy: eustasy or tectonic
imprint? Journal of Geophysical Research, 96(B4), 6641–6679.
Aubry, M.-P., 1992. Late Paleogene calcareous nannoplankton evo-
lution: a tale of climatic deterioration. In Prothero, D., and
Berggren, W. A. (eds.), The Eocene-Oligocene climatic and
biotic changes. Princeton: Princeton University Press,
pp. 272–309.
Aubry, M.-P., 1995. From chronology to stratigraphy: interpreting
the stratigraphic record. In Berggren, W. A., Kent, D. V., Aubry,
M.-P., and Hardenbol, J. (eds,), Geochronology, Time scales and
Global Stratigraphic Correlations: A Unified Temporal Frame-
work for an Historical Geology. Society for Sedimentary Geol-
ogy Special, Tulsa, Oklahoma, Vol. 54, pp. 213–274.
Aubry, M.-P., 1998. Stratigraphic (dis)continuity and temporal res-
olution in the upper Paleocene-lower Eocene deep sea record.
In Aubry, M.-P., Lucas, S., and Berggren, W. A. (eds.), Late
Paleocene-Early Eocene Climatic and Biotic Events in the
Marine and Terrestrial Records. New York: Columbia Univer-
sity Press, pp. 37–66.
Aubry, M.-P., 2007. Chronostratigraphic terminology: building on
principles. Stratigraphy, 7, 117–125.
Aubry, M.-P., 2009. Thinking of deep time. Stratigraphy, 6(2),
93–99.
Aubry, M.-P., and Van Couvering, J. A., 2004. Buried time:
chronostratigraphy as a research tool. In Koutsoukos, E. (ed.),
Applied Stratigraphy. Cambridge: Cambridge University Press.
Aubry, M.-P., Van Couvering, J. A., Christie-lick, N., Landing, E.,
Pratt, B. R., Owen, D. E., and Ferrusquía, I.-V., 2009. Terminol-
ogy of geological time: establishment of a community standard.
Stratigraphy, 6(2), 100–105.
Backman, J., and Shackleton, N. J., 1983. Quantitative
biochronology of Pliocene and early Pleistocene calcareous
nannofossils from the Atlantic, Indian and Pacific Oceans.
Marine Micropaleontology, 8, 141–170.
Backman, J., Raffi, I., Rio, D., Fornaciari, E., and Pälike, H., 2012.
Biozonation and biochronology of Miocene through Pleistocene
calcareous nannofossils from low and middle latitudes. Newslet-
ters in Stratigraphy, 45, 221–244.
Banner, F. T., and Blow, W. H., 1965. Progress in the planktonic
foraminiferal biostratigraphy of the Neogene. Nature, 208,
1164–1166.
Berggren, W. A., 1969. Paleogene biostratigraphy and planktonic
foraminifera of northern Europe. In Brönnimann, P., and Renz,
H. H. (eds.), Proceedings of the First International Conference
on Planktonic Microfossils, Geneva, 1967. Leiden: E.J. Brill,
Vol. 1, pp. 121–160.
Berggren, W. A., and Pearson, P. N., 2005. Revised tropical to sub-
tropical Paleogene planktonic foraminiferal zonation. Journal of
Foraminiferal Research, 35, 279–298.
Berggren, W. A., Kent, D. V., Swisher, C. C., III, and Aubry, M. -P.,
1995. A revised cenozoic geochronology and chronostra-
tigraphy. In Berggren, W. A., Kent, D. V., Aubry, M. -P., and
Hardenbol, J. (eds.), Geochronology, Time scales and Global
Stratigraphic Correlations: A Unified Temporal Framework
for an Historical Geology. Society of Economic Geologists and
Mineralogists Special, Vol. 54, pp. 129–212.
Blow, W. H., 1969. Late Middle Eocene to Recent planktonic fora-
minifera biostratigraphy. In Bronnimann, P., and Renz, H. H.
(eds.), Proceedings of the First International Conference on
Planktonic Microfossils (Geneva, 1967). Leiden: E.J. Brill,
Vol. 1, pp. 199–421.
Brown, R. W., 1991. Composition of Scientific Words. Washington,
DC: Smithsonian Institution Press.
 BIOSTRATIGRAPHY
Catuneanu, O., Abreu, V., Bhattacharya, J. P., Blum, M. D., Dalrym-
ple, R. W., Eriksson, P. G., Fielding, C. R., Fisher, W. L., Gallo-
way, W. E., Gibling, M. R., Giles, K. A., Holbrook, J. M., Jordan,
R., Kendall, C. G., St, C., Macurda, B., Martinsen, O. J., Miall,
A. D., Neal, J. E., Nummedal, D., Pomar, L., Posamentier, H.
W., Pratt, B. R., Sarg, J. F., Shanley, K. W., Steel, R. J., Strasser,
A., Tucker, M. E., and Winker, C., 2009. Towards the standard-
ization of sequence stratigraphy. Earth-Science Reviews, 92,
1–33.
Catuneanu, O., Bhattacharya, J. P., Blum, M. D., Dalrymple, R. W.,
Eriksson, P. G., Fielding, C. R., Fisher, W. L., Galloway, W. E.,
Gianolla, P., Gibling, M. R., Giles, K. A., Holbrook, J. M., Jor-
dan, R., Kendall, C. G., St, C., Macurda, B., Martinsen, O. J.,
Miall, A. D., Nummedal, D., Pomar, L., Posamentier, H. W.,
Pratt, B. R., Shanley, K. W., Steel, R. J., Strasser, A., and Tucker,
M. E., 2010. Sequence stratigraphy: common ground after three
decades of development. First Break, 28, 21–34.
D’Orbigny., 1852. Paris: Victor Masson.
Edwards, L., 1984. Insights on why graphic correlation (Shaw’s
method) works. Journal of Geology, 92, 583–597.
Fornaciari, R., Agnini, C., Catanzariti, R., Rio, D., Bolla, E. M., and
Valvasoni, E., 2010. Mid-latitude calcareous nannofossil biostra-
tigraphy and biochronology across the middle and late Eocene
transition. Stratigraphy, 7, 229–264.
Gould, S. J., and Eldredge, N., 1993. Punctuated equilibrium comes
to age. Nature, 366, 223–227.
Gradstein, F. M., 2004. Quantitative methods for applied microfossil
biostratigraphy. In Koutsoukos, E. A. M. (ed.), Applied Stratigra-
phy. Cambridge: Cambridge University Press, pp. 359–386.
Hagino, K., Takano, Y., and Horiguchi, T., 2009. Pseudo-cryptic
speciation in Braarudosphaera bigelowii (Gran and Braarud)
Deflandre. Marine Micropaleontology, 72, 210–221.
Hedberg, H. D. (ed.), 1976. International stratigraphic guide.
A guide to stratigraphic classification, terminology, and proce-
dure. New York: Wiley, 200 pp.
Holland, S. M., 1995. The stratigraphic distribution of fossils.
Paleobiology, 21, 92–109.
Hull, P. M., and Norris, R. D., 2009. Evidence for abrupt speciation
in a classic case of gradual evolution. Proceedings of the
National Academy of Sciences, 106, 21224–21229.
Knappertsbusch, M., 2000. Morphological evolution of the
coccolithophorid Calcidiscus leptoporus from the early miocene
to recent. Journal of Paleontology, 74, 712–730.
Knappertsbusch, M., 2007. Morphological variability of
Globorotalia menardii (planktonic foraminifera) in two DSDP
cores from the Caribbean Sea and the Eastern Equatorial Pacific.
Carnets de Géologie / Notebooks on Geology, Brest, Article
2007/04 (CG2007_A04), 1-34
Landing, E., Geyer, G., Brasier, M. D., and Bowring, S. A., 2013.
Cambrian evolutionary radiation: context, correlations, and
chronostratigraphy – overcoming deficiencies of the first
appearance datum (FAD) concept. Earth-Science Reviews, 123,
133–177.
Martini, E., 1971a. Standard tertiary and quaternary calcareous nan-
noplankton zonation. In Farinacci, A. (ed.), Proceedings of the II
Planktonic Conference, Roma, 1970, Rome: Ediz. Tecnoscienza,
Vol. 2, pp. 739–785.
Martini, E., 1971b. The occurrence of pre-quaternary calcareous
nannoplankton in the oceans. In Funnell, B. M., and Riedel,
W. R. (eds.), The Micropalaeontology of Oceans. Cambridge:
Cambridge University Press, pp. 535–544.
McGowran, B., 2005. Biostratigraphy. Microfossils and Geological
Time. Cambridge: Cambridge University Press, 459 pp.
Miller, F. X., 1977. The graphic correlation method in biostratigra-
phy. In Kauffman, E. G., and Hazel, J. E. (eds.), Concepts and
Methods in Biostratigraphy. Stroudsburg, PA: Dowden,
Hutchington and Ross, Inc, pp. 165–186.
107
Monechi, S., Reale, V., Bernaola, G., and Balestra, B., 2013. The
Danian/Selandian boundary at site 1262 (South Atlantic) and
in the Tethyan region: biomagnetostratigraphy, evolutionary
trends in fasciculiths and environmental effects of the latest
Danian event. Marine Micropaleontology, 98, 28–40.
North American Commission on Stratigraphic Nomenclature, 2005.
North American stratigraphic code. Bulletin of the American
Association of Petroleum Geologists, 89, 1547–1591.
Okada, H., and Bukry, D., 1980. Supplementary modification and
introduction of code numbers to the low–latitude coccolith bio-
stratigraphic zonation (Bukry, 1973; 1975). Marine Micropale-
ontology, 5, 321–325.
Olsen, P.E., Kent, D.V., Sues, H.-D., Koeberl, C., Huber, H.,
Montanari, A., Rainforth, E. C., Fowell, S. J., Szajna, M. J.,
and Hartline, B. W., 2002. Ascent of dinosaurs linked to an irid-
ium anomaly at the triassic-jurassic boundary. Science, 296,
1305–1307.
Pearson, P. N., 1998. Evolutionary concepts in biostratigraphy. In
Doyle, P., and Bennett, M. R. (eds.), Unlocking the stratigraphic
record: Advances in Modern Stratigraphy. New York: John
Wiley and Sons Ltd., pp. 123–144.
Raffi, I., Backman, J., Fornaciari, E., Pälike, H., Rio, D., Lourens,
L., and Hilgen, F., 2006. A review of calcareous nannofossil
astrochronology encompassing the past 25 million years.
Quaternary Science Reviews, 25, 3113–3137.
Saez, A.G., Probert, I., Geisen, M., Quinn, P., Young, J.R., and
Medlin, L.K., 2003. Pseudo-cryptic speciation in
coccolithophores. Proceedings of the National Academy of Sci-
ences of the United States of America, 100, 7163–7168.
Salvador, A. (ed.), 1994. International Stratigraphic Guide:
A Guide to Stratigraphic Classification, Terminology, and Pro-
cedure, 2nd edn. Trondheim, Norway /Boulder, CO: Interna-
tional Union Geological Sciences. and Geological Society,
pp. iii–xix, 214 pp.
Sanchez, D., Aubry, M.-P., and Liska, R. D., 2014. Neogene history
of the Carapita Formation, Eastern Venezuela Basin. Stratigra-
phy, 11(1), 1–37.
Shaw, A. B., 1964. Time in Stratigraphy. New York: McGraw-Hill.
Signor, P. W., and Lipps, J. H., 1982. Sampling bias, gradual extinc-
tion patters, and catastrophes in the fossil record. Geological
Society of America, Special Paper, 190, 291–296.
Smith, W., 1815. A Delineation of the Strata of England and Wales
with Part of Scotland. London: Privately published.
Springer, M. S., 1990. The effect of random range truncations on pat-
terns of evolution in the fossil record. Paleobiology, 16, 512–520.
Steno, N., 1669/1916. De Solido Intra Solidum Naturaliter Content
Dissertationis Prodromus, Florence. The Prodromus of Nicolaus
Steno’s Dissertation Concerning a Solid Body Enclosed by Pro-
cess of Nature within a Solid. An English Version with an Intro-
duction and Explanatory Notes by John Garrett Winter,
University of Michigan. New York: The Macmillan Company.
Van Wagoner, J. C., Posamentier, H. W., Mitchum, R. M., Vail,
P. R., Sarg, J. F., Loutit, T. S., and Hardenbol, J., 1998. An over-
view of the fundamentals of sequence stratigraphy and key def-
initions. In Wilgus, C. K., Hastings, B. S., Kendall, C. G., St,
C., Posamentier, H. W., Ross, C. A., and Van Wagoner, J. C.
(eds.), Sea-Level Changes: An Integrated Approach. Tulsa,
Okhlahoma: Society for Sedimentary Geology. Special Publica-
tions, Vol. 42, pp. 39–45.
Varol, O., 1989. Palaeocene calcareous nannofossil
biostratigraphy. In Crux, J. A., and van Heck, S. E. (eds.),
Nannofossils and their applications. Chichester: Ellis Horwood,
pp. 267–310.
Wade, B. S., Pearson, P. N., Berggren, W. A., and Pälike, H., 2011.
Review and revision of Cenozoic tropical planktonic foraminiferal
biostratigraphy and calibration to the geomagnetic polarity and
astronomical time scale. Earth-Science Reviews, 104, 111–142.
108 BIVALVE SCLEROCHRONOLOGY
BIVALVE SCLEROCHRONOLOGY
Donna M. Surge1 and Bernd R. Schöne2
1
Department of Geological Sciences, University of North
Carolina, Chapel Hill, NC, USA
2
Institute of Geosciences, University of Mainz, Mainz,
Germany
Synonyms
Bivalve shell chronometer; Related to dendrochronology
Definition
Sclerochronology: The study of incremental growth pat-
terns in hard part remains of organisms that grow by accre-
tion (i.e., adding on discrete growth layers throughout the
life of the organism). The term “sclerochronology” was
introduced for the study of growth patterns in calcareous
exoskeletons or shells and applied to study the rate and
patterns of coral growth (Buddemeier et al., 1974). It is
analogous to the earlier established approach of dendro-
chronology. The application of the term
“sclerochronology” has since broadened to the hard parts
of many other aquatic taxa (e.g., bivalves, limpets, fish,
and coralline sponges).
Bivalve: Bivalves are in the class Bivalvia of the phylum
Mollusca and include freshwater, marine, and estuarine
clams, mussels, quahogs, scallops, and oysters. They have
a wide biogeographic distribution, extending from the
equator to the poles, and occur from shallow to deep water.
Furthermore, bivalves have a long geologic history and
are archaeologically important. Fossil bivalves document
more than 500 million years of evolutionary history.
Bivalves from archaeological deposits (e.g., shell mid-
dens) provide information on food-gathering activities of
historic and prehistoric people and insights into human-
climate interactions. Many taxa are relatively short-lived
(e.g., Mercenaria spp., Spisula spp., Chione, spp.,
Crassostrea virginica); however, many other bivalve spe-
cies are extremely long-lived (e.g., Arctica islandica,
Neopycnodonte zibrowii, and Margaritifera
margaritifera). Their shells are composed of calcium car-
bonate (CaCO3), most commonly in the form of aragonite
and/or calcite. The ultrahigh-resolution (daily, seasonal,
annual) records of growth increments enable bivalve
shells to serve as unique skeletal diaries.
Introduction
Many hard part remains of aquatic organisms form by
accretionary growth, implying that the accretion of new
material is periodically retarded or interrupted which
results in the formation of growth lines. Growth lines
delimit growth increments, i.e., periods of fast growth.
Growth increments represent time slices on annual, fort-
nightly (tidal), circadian (24-h cycle), circalunidian (lunar
day; on average, 24 h and 50 min), and ultradian (period
shorter than a day, minutes to hours) scales. Together,
growth lines and increments can be used to estimate
ontogenetic age and season of growth and add a calendar
date to discrete shell portions. Biological clocks entrained
by environmental pacemakers (e.g., light/dark cycles,
tidal cycles, food availability) apparently control the regu-
lar formation of growth lines and increments. This entry
focuses on the use of sclerochronology in bivalve shells.
Bivalve sclerochronology can provide environmental
and ecological data from many different spatial, temporal,
and cultural settings. Combining sclerochronology with
geochemical analyses (e.g., oxygen and carbon isotope
ratios; d18O and d13C values, respectively) can provide
additional environmental and ecological information,
such as water temperature, precipitation, glacial meltwater
pulses, primary productivity, etc. The key criterion for
chronological research, however, is precise time
control. The following section provides an overview on
the formation of periodic growth patterns in bivalves and
focuses on daily, fortnightly, and annual increments. We
refer the reader to Schöne and Surge (2012) for a more
detailed discussion of bivalve sclerochronology and
geochemistry.
Shell growth patterns
Shell growth patterns result from changes in the rate of
deposition associated with variations in chemical compo-
sition and crystallographic properties. Growth lines sepa-
rate the growth pattern into time slices referred to as
growth increments. Together, growth lines and increments
form a shell calendar, providing an ideal method to mea-
sure time. They are typically studied in the outer or middle
shell layers of cross-sectioned valves cut along the maxi-
mum axis of growth. Several methods, such as staining
and acetate peels, have been developed to visualize
growth patterns (Clark, 1980; Kennish et al., 1980; Tevez
and Carter, 1980; Richardson, 1987; Schöne et al., 2005a).
The outer shell layer and portions of the middle shell
layer provide a complete and undisturbed shell record,
assuming no diagenesis or taphonomic alteration. New
shell material is deposited along the shell margins includ-
ing inner surfaces when the shell is open (gaping). Shell
dissolution may occur along the inner shell layer and por-
tions of the middle shell layer when the valves are closed
for extended periods of time (e.g., during seasonal anoxia,
beyond temperature or salinity tolerances) (Crenshaw,
1980). Therefore, these portions of the shell should be
avoided for calendar-dated studies. The outer shell layer
and outer portion of the middle layer are not affected by
dissolution during shell closure.
Many bivalve shells contain daily microgrowth incre-
ments, which are ideally studied in fast-growing, young
portions of the shells. Under optimal growth conditions,
some pectinids and giant clams (Tridacna spp.) form
increments corresponding to the number of solar days that
elapsed during that growth interval (Clark, 1974, 1975;
Parsons et al., 1993; Watanabe and Oba, 1999; Chauvaud
et al., 2005). This finding suggests that such microgrowth
increments form with circadian periodicity (Clark 1975).
 BIVALVE SCLEROCHRONOLOGY
Some bivalves from intertidal habitats produce distinct,
tidally controlled growth patterns in their shells. Growth
cessation occurs when they are aerially exposed during
low tide and resumes during high tide when they are sub-
merged (House and Farrow, 1968; Evans, 1972; Ohno,
1989; Goodwin et al., 2001; Miyaji et al., 2007). It should
be noted, however, that most bivalves form circalunidian
increments regardless of whether or not they are aerially
exposed at low tide. Under semidiurnal tidal conditions,
two growth lines and increments (two circatidal growth
patterns) can form per lunar day. Growth lines produced
during spring tides are more prominent than those formed
during neap tides. The distance from the shoreline influ-
ences the duration of aerial exposure and, hence, the num-
ber of growth lines and increments. For example, bivalves
living in the high intertidal zone can be exposed for sev-
eral days during extreme spring tides without forming
growth lines and increments. In comparison, individuals
farther away from the shoreline are submerged for longer
intervals and can have significantly more growth lines
and increments relative to specimens from the high inter-
tidal zone (Ohno, 1989). Moreover, shells from the low
intertidal zones typically have broader growth increments
than shells from mid-intertidal settings. This pattern is
influenced by the duration of submergence. Similarly,
shorter submergence times during spring tides produce
narrow growth increments relative to wider increments
formed during neap tides. Tidally influenced lines and
increments form distinct bundles of ~13–15. Schöne and
Surge (2012) provide a more detailed explanation about
the mechanisms controlling daily and semidiurnal shell
growth.
Most bivalve species produce annual growth lines or
increments in their shells (Hall et al., 1974; Jones, 1980;
Brey and Mackensen, 1997). Prominent dark growth lines
or increments are often visible with the naked eye on the
outer shell surface or in cross section along the maximum
axis of growth. Under high magnification, annual growth
increments appear as bundles of closely spaced daily
growth lines (Barker, 1964; Hall et al., 1974). Daily
growth lines gradually decrease in width (i.e., growth
slows) toward annual growth lines and increase gradually
afterward as growth rates increase. This important feature
distinguishes annual growth lines from disturbance lines.
Disturbance lines are characterized by abruptly changing
microgrowth increment widths and can occur when the
animal is exposed to a sudden environmental stress, such
as a major storm. Some species (e.g., Mercenaria
mercenaria, M. campechiensis, Spisula solidissima) form
couplets of dark and light increments visible under
reflected light (translucent and opaque under transmitted
light), which together represents 1 year of growth (Lutz
and Rhoads, 1980; Jones, 1983; Peterson et al., 1983;
Jones et al., 1990; Arnold et al., 1998). Dark/translucent
increments occur during periods of slow growth, whereas
light/opaque increments form during fast growth intervals
(Jones, 1983; Jones et al., 1990). It should be noted that
not all bivalve species produce obvious periodic growth
109
lines or increments. Surge et al. (2001) reported that
annual cycles in shells of the American oyster,
Crassostrea virginica, are only apparent through stable
isotope analysis (see, however, Kirby et al., 1998).
Possible mechanisms that produce annual growth line
formation include reproductive processes (i.e., gonad
development and spawning), food supply, and minimum
or maximum temperature thresholds (Pannella and
MacClintock, 1968; Lutz and Rhoads, 1977; Jones,
1980, 1983; Jones et al., 1990; Cerrato et al., 1991; Jones
and Quitmyer, 1996; Brockington and Clarke, 2001;
Surge et al., 2001; Schöne et al., 2005b; García-March
et al., 2011). Thermal tolerance is an important control
for annual growth line/increment formation in many shal-
low-water mollusks (e.g., Kraeuter and Castagna, 2001
and references therein; Beukema et al., 1985; Brown,
1988; Tanabe and Oba, 1988; Schöne et al., 2003b; Surge
et al., 2013). In their seminal paper on isotope
sclerochronology, Jones and Quitmyer (1996) reported
that the timing of slow growth (dark/translucent incre-
ments) in Mercenaria mercenaria and M. campechiensis
shells from the east coast of North America and Gulf of
Mexico occurs during cold winter temperature in mid lat-
itudes (cold-temperate zone), whereas specimens from
low latitudes (warm-temperate zone) slow their shell
growth during summer (see, however, Elliot et al., 2002;
Henry and Cerrato, 2007). They noted that at the transition
between the warm- and cold-temperate zones, both winter
and summer growth increments are present. Despite the
importance of thermal tolerance on shell growth rates,
the formation of annual growth lines/increments is proba-
bly a function of a combination of environmental factors,
physiology, and endogenous rhythms (Brockington and
Clarke, 2001). In many cases the underlying reason for
annual growth formation remains obscure. For example,
at different latitudes of the North Atlantic, the ocean qua-
hog, A. islandica, spawns at different times of the year or
even year-round (Thórarinsdóttir, 2000), and the repro-
ductive cycles of different populations are not synchro-
nized (Fritz, 1991). Yet, according to d18O data, annual
growth lines always form about a month after the seasonal
temperature maximum (Weidman et al., 1994; Witbaard
et al., 1994; Schöne et al., 2005b; Schöne et al., 2005c).
It appears likely that circadian/circalunidian biological
clocks are used to measure time that elapsed after the sum-
mer temperature peak in order to determine the timing of
seasonal growth retardation.
Applications of shell growth pattern analysis
Sclerochronology offers a wide range of potential applica-
tions in a variety of disciplines, such as biology, ecology,
archaeology, and climate research. Here we provide exam-
ples of selected applications.
Two approaches have been used to estimate the timing,
duration, and rate of seasonal shell growth. The translu-
cent-opaque staging technique identifies the stage of
110 BIVALVE SCLEROCHRONOLOGY
terminal growth as being within either the translucent
(dark) increment or the opaque (light) increment on
a cross-sectioned valve cut along the maximum axis of
growth (Jones et al., 1990; Arnold et al., 1998). Next,
the growth phase is determined within each stage of termi-
nal growth based on the relative thickness of the terminal
growth band (Quitmyer and Jones, 1992; Quitmyer
et al., 1997; Arnold et al., 1998; Quitmyer and Jones,
2012; Quitmyer, 2013). Another approach counts the
number of daily growth lines between annual incre-
ments to estimate the duration of the growing season
and assign precise dates to each shell portion (assuming
the date of the last-formed growth line is known). This
method is potentially accurate to the nearest 2–4 weeks
(Hallmann et al., 2009). If maximum and minimum
temperature thresholds of a species are known, such
data can be used to determine changes in seasonal tem-
perature variability and climate through time and space
(Ansell, 1968; Tanabe and Oba, 1988; Jones and
Quitmyer, 1996; Schöne et al., 2004; Lohmann and
Schöne, 2013).
When combined with oxygen and carbon isotope ratios
(discussed more fully in the next section), isotope
sclerochronology can be used to address ecological ques-
tions in modern and fossil shells to interpret life history
and to assess the timing of the initial introduction of inva-
sive species (Jones and Quitmyer, 1996; Schöne et al.,
2002; Henry and Cerrato, 2007; Goodwin et al., 2010;
Haveles and Ivany, 2010). For example, it was recently
discovered that bivalves are among the longest-lived soli-
tary animals with some species living for many centuries
(Thompson et al., 1980; Shaul and Goodwin, 1982;
Mutvei and Westermark, 2001; Schöne et al., 2005b;
Wanamaker et al., 2008; Wisshak et al., 2009; Butler
et al., 2013). Such long-lived species have been used to
generate master chronologies for long-term climate recon-
struction (Noakes and Campbell, 1992; Marchitto et al.,
2000; Schöne et al., 2003a; Strom et al., 2004, 2005;
Wanamaker et al., 2008; Butler et al., 2010, 2013).
Paleoclimate, paleoceanographic, and paleoenvir-
onmental reconstruction based on isotope
sclerochronology provides critical information on season-
ality in seawater temperature, North Atlantic Oscillation
(NAO) variability, paleoproductivity/upwelling, drought
conditions, changes in river discharge, and aquatic pollu-
tion (Schöne et al., 2003b; Dettman et al., 2004; Schöne
et al., 2005b; Dunca et al., 2005; Carroll et al., 2009;
Schöne and Fiebig, 2009; Harding et al., 2010;
Wanamaker et al., 2011; Wang et al., 2011; Sadler et al.,
2012). Sclerochronologic analysis of shells from archaeo-
logical midden deposits (trash heaps) allows interpretation
of human activity, such as site occupation, season of har-
vest, and intensification of resource exploitation (Bailey
et al., 1983; Milner, 2001; Milner et al., 2007; Carré
et al., 2009; Andrus, 2011; Gutiérrez-Zugasti, 2011;
Andrus and Thompson, 2012; Reitz et al., 2012 and refer-
ences therein; Burchell et al., 2013a; Burchell et al.,
2013b).
Isotope sclerochronology
Stable carbon isotope ratios of bivalve shells are reported
in delta notation:
013 13
1
C C
B12 shell
12 VPDBC
B C C C
d13 Cshell ¼ B
B 13
C1, 000
C
@ C
VPDB A
12
C
where VPDB is the Vienna Pee Dee Belemnite standard
and values are in per mil units (%). In principle, d13Cshell
values can provide information on the carbon isotope ratio
of dissolved inorganic carbon (DIC) and, hence, serve as
a proxy for primary productivity and remineralization
(e.g., respiration, oxidation of organic matter) (Mook
and Vogel, 1968). Plants preferentially incorporate 12C
in their organic tissue, thereby enriching the remaining
carbon pool in 13C. Thus, increases in photosynthesis
results in higher d13CDIC values, whereas increased respi-
ration lowers d13CDIC values. Bivalves sample the carbon
isotope ratio of the ambient water and in theory record
changes in the d13C value of DIC. For example, higher
d13Cshell values would reflect increased primary produc-
tivity, whereas lower d13Cshell values suggest increased
respiration. However, bivalves also incorporate up to
ca. 10 % metabolically derived CO2 into their shells
(McConnaughey and Gillikin, 2008). Many species seem
to incorporate increased amounts of respiratory CO2
through their lifetime, resulting in more negative d13Cshell
values as the animal ages (Gillikin et al., 2007;
McConnaughey and Gillikin, 2008). Therefore, d13Cshell
values should be treated with caution. More recent studies
have reported that long-lived bivalves that grow slowly
during their youth are not hampered by such ontogenetic
effects on the carbon isotope ratio of shell carbonate. In
these cases, studying past carbon cycle dynamics in the
ocean or tracking the anthropogenic increase in atmo-
spheric and oceanic CO2 (i.e., the Suess effect) is possible
(Butler et al., 2011; Schöne et al., 2011).
Unlike stable carbon isotope ratios, metabolic effects
(also known as vital effects) do not seem to impact oxygen
isotope ratios recorded in bivalve shells. Oxygen isotope
ratios are reported in similar delta notation as stable car-
bon isotope ratios:
018 18
1
O O
B18 shell
18 VPDBC
B O O C
d Oshell ¼ B
18
B 18
C1, 000
C
@ O
VPDB A
18
O
They are the most widely used geochemical proxy of
bivalve shells and are a function of variations in the ambi-
ent water temperature and d18Owater value (Urey, 1947;
Epstein et al., 1953). Many studies have demonstrated that
almost all bivalves precipitate their shells in oxygen isoto-
pic equilibrium with the ambient water (Mook and Vogel,
 BIVALVE SCLEROCHRONOLOGY
1968; Surge et al., 2001; Elliot et al., 2003; Hallmann
et al., 2009; see, however, Owen et al., 2002, 2008;
Hallmann et al., 2008; Yan et al., 2012). Values of d18Oshell
have been used to estimate salinity (assuming temperature
can be held constant and the mixing relation between
marine and freshwater is known; Ingram et al., 1996)
or water temperature (assuming the d18Owater value
can be held constant; Jones et al., 1989; Wefer and Berger,
1991; Weidman et al., 1994; Dettman et al., 1999; Surge
et al., 2001; Carré et al., 2005; Schöne et al., 2005b;
Schöne et al., 2005c; Wanamaker et al., 2006; Ivany
et al., 2008). Several experimental studies have demon-
strated a 1 % change in d18Oshell value that represents
a temperature change of ~4.3 ºC (Epstein et al., 1953;
Grossman and Ku, 1986; Dettman et al., 1999). Higher
d18Oshell values correspond to cold temperatures, whereas
relatively lower values indicate warmer temperatures.
Advances in microsampling techniques enable high-
resolution isotopic records that can produce temperature
reconstructions on seasonal to subdaily time scales,
depending on shell growth rates, to the nearest 0.35  C
(Dettman and Lohmann, 1995). Sclerochronologic analy-
sis ensures that the reconstructed temperature record from
d18Oshell values is placed within a precise temporal
context.
Other geochemical proxies in bivalve shells used to
reconstruct environmental conditions with varying
degrees of success include clumped isotopes, nitrogen iso-
topes, and minor and trace elemental ratios. We refer the
reader to Schöne and Surge (2012) and Schöne and
Gillikin (2013) (and papers therein) for more details and
examples regarding these topics.
Summary
Bivalve sclerochronology developed from the observa-
tions of visible growth lines and increments on the outer
shell layer and along cross-sectioned shells cut along the
maximum axis of growth. Such growth patterns are often
visible with the naked eye, but in some species these fea-
tures are more clearly revealed with staining or acetate
peels. Sclerochronologic records from long-lived bivalves
provide master chronologies used to reconstruct climate
change on decadal to centennial time scales. Records from
short-lived bivalves can be resolved on seasonal time
scales. In combination with geochemical proxies,
sclerochronologic records can produce time series of envi-
ronmental variability that can be quantified (e.g., estimat-
ing water temperature, salinity). Their abundance in fossil
and archaeological deposits and their wide biogeographic
distribution make bivalve shells unique and valuable
archives for a wide range of interdisciplinary study.
Bibliography
Andrus, C. F. T., 2011. Shell midden sclerochronology. Quaternary
Science Reviews, 30(21–22), 2892–2905. http://dx.doi.org/
10.1016/j.quascirev.2011.07.016.
Andrus, C. F. T., and Thompson, V. D., 2012. Determining the hab-
itats of mollusk collection at the Sapelo Island shell ring
111
complex, Georgia, USA using oxygen isotope sclerochronology.
Journal of Archaeological Science, 39(2), 215–228. http://dx.
doi.org/10.1016/j.jas.2011.08.002.
Ansell, A. D., 1968. The rate of growth of the hard clam Mercenaria
mercenaria (L.) throughout the geographical range. Conseil Per-
manent International pour l'Exploration de la Mer, Journal du
Conseil, 31(3), 364–409.
Arnold, W. S., Bert, T. M., Quitmyer, I. R., and Jones, D. S., 1998.
Contemporaneous deposition of annual growth bands in
Mercenaria mercenaria (Linnaeus), Mercenaria campechiensis
(Gmelin), and their natural hybrid forms. Journal of Experimen-
tal Marine Biology and Ecology, 223, 93–109.
Bailey, G. N., Deith, M. R., and Shackleton, N. J., 1983. Oxygen iso-
tope analysis and seasonality determinations: limits and potential
of a new technique. American Antiquity, 48(2), 390–398.
Barker, R. M., 1964. Microtextural variation in pelecypod shells.
Malacologia, 2, 69–86.
Beukema, J. J., Knol, E., and Cadee, G. C., 1985. Effects of temper-
ature on the length of the annual growing season in the tellinid
bivalve Macoma balthica (L.) living on tidal flats in the Dutch
Wadden Sea. Journal of Experimental Marine Biology and Ecol-
ogy, 90, 129–144.
Brey, T., and Mackensen, A., 1997. Stable isotopes prove shell
growth bands in the Antarctic bivalve Laternula elliptica to be
formed annually. Polar Biology, 17, 465–468.
Brockington, S., and Clarke, A., 2001. The relative influence of
temperature and food on the metabolism of a marine inverte-
brate. Journal of Experimental Marine Biology and Ecology,
258(1), 87–99. http://dx.doi.org/10.1016/S0022-0981(00)
00347-6.
Brown, J. R., 1988. Multivariate analyses of the role of environmen-
tal factors in seasonal and site-related growth variation in the
Pacific oyster, Crassostrea gigas. Marine Ecology: Progress
Series, 45, 225–236.
Buddemeier, R. W., Maragos, J. E., and Knutson, D. W., 1974.
Radiographic studies of reef coral exoskeletons: rates and pat-
terns of coral growth. Journal of Experimental Marine Biology
and Ecology, 14(2), 179–199. http://dx.doi.org/10.1016/0022-
0981(74)90024-0.
Burchell, M., Cannon, A., Hallmann, N., Schwarcz, H. P., and
Schöne, B. R., 2013a. Inter-site variability in the season of shell-
fish collection on the Central Coast of British Columbia. Journal
of Archaeological Science, 40, 626–636.
Burchell, M., Hallmann, N., Martindale, A., Cannon, A., and
Schöne, B. R., 2013b. Seasonality and intensity of shellfish
harvesting in the north coast of British Columbia. Journal of
Coastal and Island Archaeology, 8, 152–169.
Burchell, M., Cannon, A., Hallmann, N., Schwarcz, H. P., and
Schöne, B. R., 2013a. Inter-site variability in the season of shell-
fish collection on the Central Coast of British Columbia. Journal
of Archaeological Science, 40, 626–636.
Burchell, M., Hallmann, N., Martindale, A., Cannon, A., and
Schöne, B. R., 2013b. Seasonality and intensity of shellfish
harvesting in the north coast of British Columbia. Journal of
Coastal and Island Archaeology, 8, 152–169.
Butler, P. G., Richardson, C. A., Scourse, J. D., Wanamaker, A. D.,
Jr., Shammon, T. M., and Bennell, J. D., 2010. Marine climate in
the Irish Sea: analysis of a 489-year marine master chronology
derived from growth increments in the shell of the clam Arctica
islandica. Quaternary Science Reviews, 29(13–14), 1614–1632.
http://dx.doi.org/10.1016/j.quascirev.2009.07.010.
Butler, P. G., Wanamaker, A. D., Jr., Scourse, J. D., Richardson,
C. A., and Reynolds, D. J., 2011. Long-term stability of d13C
with respect to biological age in the aragonite shell of mature
specimens of the bivalve mollusk Arctica islandica.
Palaeogeography, Palaeoclimatology, Palaeoecology, 302
(1–2), 21–30. http://dx.doi.org/10.1016/j.palaeo.2010.03.038.
112 BIVALVE SCLEROCHRONOLOGY
Butler, P. G., Wanamaker, A. D., Jr., Scourse, J. D., Richardson,
C. A., and Reynolds, D. J., 2013. Variability of marine climate
on the North Icelandic Shelf in a 1357-year proxy archive based
on growth increments in the bivalve Arctica islandica.
Palaeogeography, Palaeoclimatology, Palaeoecology, 373,
141–151.
Carré, M., Bentaleb, I., Blamart, D., Ogle, N., Cardenas, F.,
Zevallos, S., Kalin, R. M., Ortlieb, L., and Fontugne, M., 2005.
Stable isotopes and sclerochronology of the bivalve Mesodesma
donacium: potential application to Peruvian paleoceanographic
reconstructions. Palaeogeography, Palaeoclimatology,
Palaeoecology, 228(1–2), 4–25. http://dx.doi.org/10.1016/j.
palaeo.2005.03.045.
Carré, M., Klaric, L., Lavallée, D., Julien, M., Bentaleb, I.,
Fontugne, M., and Kawka, O., 2009. Insights into early Holo-
cene hunter-gatherer mobility on the Peruvian Southern Coast
from mollusk gathering seasonality. Journal of Archaeological
Science, 36(5), 1173–1178. http://dx.doi.org/10.1016/j.
jas.2009.01.005.
Carroll, M. L., Johnson, B. J., Henkes, G. A., McMahon, K. W.,
Voronkov, A., Ambrose, W. G., Jr., and Denisenko, S. G.,
2009. Bivalves as indicators of environmental variation and
potential anthropogenic impacts in the southern Barents Sea.
Marine Pollution Bulletin, 59, 193–206, doi:10.1016/j.
marpolbul.2009.02.022.
Cerrato, R. M., Wallace, H. V. E., and Lightfoot, K. G., 1991. Tidal
and seasonal patterns in the chondrophore of the soft-shell clam
Mya arenaria. Biological Bulletin, 181(2), 307–311,
doi:10.2307/1542102.
Chauvaud, L., Lorrain, A., Dunbar, R. B., Paulet, Y.-M., Thouzeau,
G., Jean, F., Guarini, J.-M., and Mucciarone, D., 2005. Shell of
the great scallop Pecten maximus as a high-frequency archive
of paleoenvironmental changes. Geochemistry Geophysics
Geosystems, 6(8), doi:10.1029/2004GC000890.
Clark, G. R., II, 1974. Growth lines in invertebrate skeletons.
Annual Review of Earth and Planetary Sciences, 2, 77–99.
Clark, G. R. I., 1975. Periodic growth and biological rhythms in
experimentally grown bivalves. In Rosenberg, G. D., and
Runcom, S. K. (eds.), Growth Rhythms and the History of the
Earth’s Rotation. London: Wiley, pp. 103–117.
Clark, G. R. I., 1980. Study of molluscan shell structure and growth
lines using thin sections. In Rhoads, D. C., and Lutz, R. A. (eds.),
Skeletal Growth of Aquatic Organisms: Biological Records of
Environmental Change. New York/London: Plenum Press,
pp. 603–606.
Crenshaw, M. A., 1980. Mechanisms of shell formation and disso-
lution. In Rhoads, D. C., and Lutz, R. A. (eds.), Skeletal Growth
of Aquatic Organisms: Biological Records of Environmental
Change. New York/London: Plenum Press, pp. 115–132.
Dettman, D. L., and Lohmann, K. C., 1995. Microsampling carbon-
ates for stable isotope and minor element analysis: physical sep-
aration of samples on a 20 micrometer scale. Journal of
Sedimentary Research, A65(3), 566–569.
Dettman, D. L., Reische, A. K., and Lohmann, K. C., 1999. Con-
trols on the stable isotope composition of seasonal growth bands
in aragonitic fresh-water bivalves (Unionidae). Geochimica et
Cosmochimica Acta, 63(7/8), 1049–1057.
Dettman, D. L., Flessa, K. W., Roopnarine, P. D., Schone, B. R., and
Goodwin, D. H., 2004. The use of oxygen isotope variation in
shells of estuarine mollusks as a quantitative record of seasonal
and annual Colorado river discharge. Geochimica et
Cosmochimica Acta, 68(6), 1253–1263.
Dunca, E., Schöne, B., and Mutvei, H., 2005. Freshwater bivalves
tell of past climates: but how clearly do shells from polluted
rivers speak. Palaeogeography, Palaeoclimatology,
Palaeoecology, 228, 43–45, doi:10.1016/j.palaeo.2005.03.050.
Elliot, M., deMenocal, P. B., Linsley, B. K., Howe, S. S.,
Guilderson, T., and Quitmyer, I. R. (2002). Late Holocene envi-
ronmental and hydrologic conditions in northwestern Florida
derived from seasonally resolved profiles of d18O and Sr/Ca of
fossil bivalves. EOS Transactions, American Geophysical
Union. San Francisco, CA: American Geophysical Union.
Elliot, M., deMenocal, P. B., Linsley, B. K., and Howe, S. S., 2003.
Environmental controls on the stable isotopic composition of
Mercenaria mercenaria: potential application to paleoenvir-
onmental studies. Geochemistry, Geophysics, Geosystems,
4(7), 1–16.
Epstein, S., Buchsbaum, R., Lowenstam, H. A., and Urey, H. C.,
1953. Revised carbonate water isotopic temperature scale. Geo-
logical Society of America Bulletin, 2, 417–425.
Evans, J. W., 1972. Tidal growth increments in the cockle
Clinocardium nuttalli. Science, 176(4033), 416–417,
doi:10.2307/1734401.
Fritz, K. W., 1991. Seasonal condition change, morphometrics,
growth and sex ratio of the ocean quahog, Arctica islandica
(Linneaus, 1767) off New Jersey, USA. Journal of Shellfish
Research, 10, 79–88.
García-March, J. R., Surge, D., Lees, J. M., and Kersting, D. K.,
2011. Ecological information and water mass properties in the
Mediterranean recorded by stable isotope ratios in Pinna nobilis
shells. Journal of Geophysical Research: Biogeosciences, 116
(G2), G02009, doi:10.1029/2010jg001461.
Gillikin, D. P., Lorrain, A., Meng, L., and Dehairs, F., 2007. A large
metabolic carbon contribution to the d13C record in marine ara-
gonitic bivalve shells. Geochimica et Cosmochimica Acta,
71(12), 2936–2946. http://dx.doi.org/10.1016/j.
gca.2007.04.003.
Goodwin, D. H., Flessa, K. W., Schöne, B. R., and Dettman, D. L.,
2001. Cross-calibration of daily growth increments, stable iso-
tope variation, and temperature in the Gulf of California bivalve
mollusk Chione cortezi: implications for paleoenvironmental
analysis. Palaios, 16, 387–398.
Goodwin, D. H., Cohen, A. N., and Roopnarine, P. D., 2010. Foren-
sics on the half shell: a sclerochronological investigation of
a modern biological invasion in San Francisco Bay, United
States. Palaios, 25(11), 742–753, doi:10.2110/palo.2010.p10-
015r.
Grossman, E. L., and Ku, T., 1986. Oxygen and carbon isotope frac-
tionation in biogenic aragonite: temperature effects. Chemical
Geology, 59, 59–74.
Gutiérrez-Zugasti, I., 2011. Coastal resource intensification across
the Pleistocene–Holocene transition in Northern Spain: evi-
dence from shell size and age distributions of marine gastropods.
Quaternary International, 244(1), 54–66. http://dx.doi.org/
10.1016/j.quaint.2011.04.040.
Hall, C. A., Jr., Dollase, W. A., and Corbató, C. E., 1974. Shell
growth in Tivela stultorum (Mawe, 1823) and Callista chione
(Linnaeus, 1758) (Bivalvia): annual periodicity, latitudinal dif-
ferences, and diminution with age. Palaeogeography, Palaeocli-
matology, Palaeoecology, 15(1), 33–61. http://dx.doi.org/
10.1016/0031-0182(74)90036-4.
Hallmann, N., Schöne, B. R., Strom, A., and Fiebig, J., 2008. An
intractable climate archive – sclerochronological and shell oxy-
gen isotope analyses of the Pacific geoduck, Panopea abrupta
(bivalve mollusk) from Protection Island (Washington State,
USA). Palaeogeography, Palaeoclimatology, Palaeoecology,
269, 115–126.
Hallmann, N., Burchell, M., Schöne, B. R., Irvine, G. V., and Max-
well, D., 2009. High-resolution sclerochronological analysis of
the bivalve mollusk Saxidomus gigantea from Alaska and Brit-
ish Columbia: techniques for revealing environmental archives
and archaeological seasonality. Journal of Archaeological
 BIVALVE SCLEROCHRONOLOGY
Science, 36(10), 2353–2364. http://dx.doi.org/10.1016/j.
jas.2009.06.018.
Harding, J. M., Spero, H. J., Mann, R., Herbert, G. S., and Sliko,
J. L., 2010. Reconstructing early 17th century estuarine drought
conditions from Jamestown oysters. Proceedings of the National
Academy of Sciences of the United States of America, 107,
10549–10554, doi:10.1073/pnas.1001052107.
Haveles, A. W., and Ivany, L. C., 2010. Rapid growth explains large
size of mollusks in the Eocene Gosport Sand, United States Gulf
Coast. Palaios, 25(9), 550–564, doi:10.2110/palo.2009.p09-
148r.
Henry, K. M., and Cerrato, R. M., 2007. The annual macroscopic
growth pattern of the northern quahog [¼hard clam, Mercenaria
mercenaria (L.)], in Narragansett Bay, Rhode Island. Journal of
Shellfish Research, 26(4), 985–993, doi:10.2983/0730-
8000(2007)26[985:tamgpo]2.0.co;2.
House, M. R., and Farrow, G. E., 1968. Daily growth banding in the
shell of the cockle, Cardium edule. Nature, 219(5161),
1384–1386.
Ingram, B. L., Conrad, M. E., and Ingle, J. C., 1996. Stable isotope
and salinity systematics in estuarine waters and carbonates: San
Francisco Bay. Geochimica et Cosmochimica Acta, 60(3),
455–467.
Ivany, L. C., Lohmann, K. C., Hasiuk, F., Blake, D. B., Glass, A.,
and Moody, R. M., 2008. Eocene climate record of a high
southern latitude continental shelf: Seymour Island,
Antarctica. Bulletin of the Geological Society of America, 120,
659–678.
Jones, D. S., 1980. Annual cycle of shell growth increment forma-
tion in two continental shelf bivalves and its paleoecologic sig-
nificance. Paleobiology, 6(3), 331–340, doi:10.2307/2400349.
Jones, D. S., 1983. Sclerochronology: reading the record of the mol-
luscan shell. American Scientist, 71, 384–391.
Jones, D. S., and Quitmyer, I. R., 1996. Marking time with bivalve
shells: oxygen isotopes and season of annual increment forma-
tion. Palaios, 11(4), 340–346.
Jones, D. S., Arthur, M. A., and Allard, D. J., 1989. Sclerochro-
nological records of temperature and growth from shells of
Mercenaria mercenaria from Narragansett Bay, Rhode Island.
Marine Biology, 102, 225–234.
Jones, D. S., Quitmyer, I. R., Arnold, W. S., and Marelli, D. C.,
1990. Annual shell banding, age, and growth rate of hard clams
(Mercenaria spp.) from Florida. Journal of Shellfish Research,
9, 215–225.
Kennish, M. J., Lutz, R. A., and Rhoads, D. C., 1980. Preparation of
acetate peels and fractured sections for observation of growth
patterns within the bivalve shell. In Rhoads, D. C., and Lutz,
R. A. (eds.), Skeletal Growth of Aquatic Organisms: Biological
Records of Environmental Change. New York/London: Plenum
Press, pp. 597–601.
Kirby, M. X., Soniat, T. M., and Spero, H. J., 1998. Stable isotope
sclerochronology of Pleistocene and recent oyster shells
(Crassostrea virginica). Palaios, 13(6), 560–569.
Kraeuter, J. N., and Castagna, M. (eds.), 2001. Biology of the Hard
Clam. Developments in Aquaculture and Fisheries Science. New
York: Elsevier.
Lohmann, G., and Schöne, B. R., 2013. Climate signatures on
decadal to interdecadal time scales as obtained from mollusk
shells (Arctica islandica) from Iceland. Palaeogeography,
Palaeoclimatology, Palaeoecology, 373, 152–162. http://dx.
doi.org/10.1016/j.palaeo.2012.08.006.
Lutz, R. A., and Rhoads, D. C., 1977. Anaerobiosis and a theory of
growth line formation. Science, 198, 1222–1227.
Lutz, R. A., and Rhoads, D. C., 1980. Growth patterns within the
molluscan shell: an overview. In Rhoads, D. C., and Lutz,
R. A. (eds.), Skeletal Growth of Aquatic Organisms: Biological
113
Records of Environmental Change. New York/London: Plenum
Press, pp. 203–254.
Marchitto, T. A., Jones, G. A., Goodfriend, G. A., and Weidman,
C. R., 2000. Precise temporal correlation of Holocene mollusk
shells using sclerochronology. Quaternary Research, 53,
236–246.
McConnaughey, T., and Gillikin, D., 2008. Carbon isotopes in mol-
lusk shell carbonates. Geo-Marine Letters, 28(5–6), 287–299,
doi:10.1007/s00367-008-0116-4.
Milner, N., 2001. At the cutting edge: using thin sectioning to deter-
mine season of death of the European oyster, Ostrea edulis. Jour-
nal of Archaeological Science, 2001(28), 861–873.
Milner, N., Barrett, J., and Welsh, J., 2007. Marine resource intensi-
fication in Viking Age Europe: the molluscan evidence from
Quoygrew, Orkney. Journal of Archaeological Science, 34(9),
1461–1472. http://dx.doi.org/10.1016/j.jas.2006.11.004.
Miyaji, T., Tanabe, K., and Schöne, B. R., 2007. Environmental
controls on daily shell growth of Phacosoma japonicum
(Bivalvia: Veneridae) from Japan. Marine Ecology: Progress
Series, 336, 141–150, doi:10.3354/meps336141.
Mook, W. G., and Vogel, J. C., 1968. Isotopic equilibrium between
shells and their environment. Science, 159, 874–875.
Mutvei, H., and Westermark, T., 2001. How environmental infor-
mation can be obtained from Naiad shells. In Bauer, G., and
Wächtler, K. (eds.), Ecology and Evolution of the Freshwater
Mussels Unionoidea. Berlin/Heidelberg: Springer. Ecological
Studies, Vol. 145, pp. 367–379.
Noakes, D. J., and Campbell, A., 1992. Use of geoduck clams to
indicate changes in the marine environment of Ladysmith Har-
bor, British Columbia. Environmetrics, 3, 81–97.
Ohno, T., 1989. Palaeotidal characteristics determined by
microgrowth patterns in bivalves. Palaeontology, 32, 237–263.
Owen, R., Kennedy, H., and Richardson, C., 2002. Isotopic
partitioning between scallop shell calcite and seawater: effect
of shell growth rate. Geochimica et Cosmochimica Acta,
66(10), 1727–1737.
Owen, E. F., Wanamaker, A. D., Jr., Feindel, S. C., Schöne, B. R.,
and Rawson, P. D., 2008. Stable carbon and oxygen isotope frac-
tionation in bivalve (Placopecten magellanicus) larval aragonite.
Geochimica et Cosmochimica Acta, 72(19), 4687–4698. http://
dx.doi.org/10.1016/j.gca.2008.06.029.
Pannella, G., and MacClintock, C., 1968. Biological and environ-
mental rhythms reflected in Molluscan shell growth. Memoir
(The Paleontological Society), 2, 64–80.
Parsons, G. J., Robinson, S. M. C., Roff, J. C., and Dadswell, M. J.,
1993. Daily growth rates as indicated by valve ridges in
postlarval giant scallop (Placopecten magellanicus) (Bivalvia:
Pectinidae). Canadian Journal of Fisheries and Aquatic
Sciences, 50(3), 456–464, doi:10.1139/f93-053.
Peterson, C. H., Duncan, P. B., Summerson, H. C., and Safrit,
J. G. W., 1983. A mark-recapture test of annual periodicity of
internal growth band deposition in shells of hard clams,
Mercenaria mercenaria, from a population along the southeast-
ern United States. Fishery Bulletin, 81, 765–779.
Quitmyer, I. R., 2013. Precolumbian site seasonality and harvest of
estuarine resources at the Pineland Site Complex, Charlotte Har-
bor, Florida. In Marquardt, W. H., Walker, K. J., Marquardt,
W. H., and Walker, K. J. (eds.), The Archaeology of Pineland:
A Coastal Southwest Florida Site Complex, A.D. 50–1700.
Gainesville: University of Florida. Institute of Archaeology and
Paleoenvironmental Studies, Monograph, Vol. 4, pp. 349–372.
Quitmyer, I. R., and Jones, D. S., 1992. Calendars of the coast: sea-
sonal growth increment patterns in shells of modern and archae-
ological southern quahogs, Mercenaria campechiensis, from
Charlotte Harbor, Florida. In Marquardt, W. H., and Marquardt,
W. H. (eds.), Culture and Environment in the Domain of the
114 BIVALVE SCLEROCHRONOLOGY
Calusa. Gainesville: Institute of Archaeological Studies, Univer-
sity of Florida. Institute of Archaeology and Paleoenvironmental
Studies, Monograph, Vol. 1, pp. 247–264.
Quitmyer, I. R., and Jones, D. S., 2012. Annual incremental shell
growth patterns in hard clams (Mercenaria spp) from St. Catherines
Island, Georgia: a record of seasonal and anthropogenic impact on
zooarchaeological resources. In Reitz, E. J., Quitmyer, I. R., and
Thomas, D. H. (eds.), Seasonality and Human Mobility along the
Georgia Bight. New York: Anthropological Papers of the American
Museum of Natural History, pp. 135–148.
Quitmyer, I. R., Jones, D. S., and Arnold, W. S., 1997. The
sclerochronology of hard clams, Mercenaria spp., from the
South-Eastern U.S.A.: a method of elucidating the
zooarchaeological records of seasonal resource procurement
and seasonality in prehistoric shell middens. Journal of Archae-
ological Science, 24, 825–840.
Richardson, C. A., 1987. Tidal bands in the shell of the clam Tapes
philippinarum (Adams and Reeve, 1850). Proceedings of the
Royal Society of London B, 230, 367–387.
Reitz, E. J., Quitmyer, I. R., and Thomas, D. H., (eds.), 2012. Season-
ality and Human Mobility along the Georgia Bight. Anthropolog-
ical Papers of the American Museum of Natural History, Number
97. New York: American Museum of Natural History, 236pp.
Sadler, J., Carré, M., Azzoug, M., Schauer, A. J., Ledesma, J.,
Cardenas, F., Chase, B. M., Bentaleb, I., Muller, S. D., Mandeng,
M., Rohling, E. J., and Sachs, J. P., 2012. Reconstructing past
upwelling intensity and the seasonal dynamics of primary pro-
ductivity along the Peruvian coastline from mollusk shell stable
isotopes. Geochemistry, Geophysics, Geosystems, 13(1),
Q01015, doi:10.1029/2011gc003595.
Schöne, B. R., and Fiebig, J., 2009. Seasonality in the North Sea
during the Allerød and late Medieval Climate Optimum using
bivalve sclerochronology. International Journal of Earth Sci-
ences, 98, 83–98, doi:10.1007/s00531-008-0363-7.
Schöne, B. R., and Gillikin, D. P., 2013. Unraveling environmental
histories from skeletal diaries – advances in sclerochronology.
Palaeogeography, Palaeoclimatology, Palaeoecology, 373,
1–5. http://dx.doi.org/10.1016/j.palaeo.2012.11.026.
Schöne, B. R., and Surge, D., 2012. Chapter 14, Bivalve
sclerochronology and geochemistry. Part N, Bivalvia, revised,
volume 1. Treatise Online, 46, 1–24. Seldon, P., Hardesty, J.,
and Carter, J. G., (coordinator). Lawrence, KS: University of
Kansas, Paleontological Institute.
Schöne, B. R., Goodwin, D. H., Flessa, K. W., Dettman, D. L., and
Roopnarine, P. D., 2002. Sclerochronology and growth of the
bivalve mollusks Chione (Chionista) fluctifraga and C.
(Chionista) cortezi in the northern Gulf of California, Mexico.
The Veliger, 45(1), 45–54.
Schöne, B. R., Oschmann, W., Rössler, J., Castro, A. D. F., Houk,
S. D., Kröncke, I., Dreyer, W., Janssen, R., Rumohr, H., and
Dunca, E., 2003a. North Atlantic Oscillation dynamics recorded
in shells of a long-lived bivalve mollusk. Geology, 31(12),
1037–1040, doi:10.1130/g20013.1.
Schöne, B. R., Tanabe, K., Dettman, D. L., and Sato, S., 2003b.
Environmental controls on shell growth rates and d18O of the
shallow-marine bivalve mollusk Phacosoma japonicum in
Japan. Marine Biology, 142, 473–485.
Schöne, B. R., Oschmann, W., Tanabe, K., Dettman, D., Fiebig, J.,
Houk, S. D. S. D., and Kanie, Y., 2004. Holocene seasonal envi-
ronmental trends at Tokyo Bay, Japan, reconstructed from
bivalve mollusk shells–implications for changes in the East
Asian monsoon and latitudinal shifts of the Polar Front. Quater-
nary Science Reviews, 23(9–10), 1137–1150.
Schöne, B. R., Dunca, E., Fiebig, J., and Pfeiffer, M., 2005a.
Mutvei’s solution: an ideal agent for resolving microgrowth
structures of biogenic carbonates. Palaeogeography, Palaeocli-
matology, Palaeoecology, 228(1–2), 149–166.
Schöne, B. R., Fiebig, J., Pfeiffer, M., Gleß, R., Hickson, J., John-
son, A. L. A., Dreyer, W., and Oschmann, W., 2005b. Climate
records from a bivalved Methuselah (Arctica islandica,
Mollusca; Iceland). Palaeogeography, Palaeoclimatology,
Palaeoecology, 228(1–2), 130–148.
Schöne, B. R., Houk, S. D., Freyre Castro, A. D., Fiebig, J.,
Oschmann, W., Kröncke, I., Dreyer, W., and Gosselck, F.,
2005c. Daily growth rates in shells of Arctica islandica: assessing
sub-seasonal environmental controls on a long-lived bivalve mol-
lusk. Palaios, 20(1), 78–92, doi:10.2110/palo.2003.p03-101.
Schöne, B. R., Wanamaker, A. D., Jr., Fiebig, J., Thébault, J., and
Kreutz, K., 2011. Annually resolved d13Cshell chronologies of
long-lived bivalve mollusks (Arctica islandica) reveal oceanic
carbon dynamics in the temperate North Atlantic during recent
centuries. Palaeogeography, Palaeoclimatology, Palaeoecology,
302(1–2), 31–42. http://dx.doi.org/10.1016/j.palaeo.2010.02.002.
Shaul, W., and Goodwin, L., 1982. Geoduck (Panope generosa:
Bivalvia) age as determined by internal growth lines in the shell.
Canadian Journal of Fisheries and Aquatic Sciences, 39, 632–636.
Strom, A., Francis, R. C., Mantua, N. J., Miles, E. L., and Peterson,
D. L., 2004. North Pacific climate recorded in growth rings of
geoduck clams: a new tool for paleoenvironmental reconstruc-
tion. Geophysical Research Letters, 31(6), L06206,
doi:10.1029/2004gl019440.
Strom, A., Francis, R. C., Mantua, N. J., Miles, E. L., and
Peterson, D. L., 2005. Preserving low-frequency climate
signals in growth records of geoduck clams (Panopea abrupta).
Palaeogeography, Palaeoclimatology, Palaeoecology, 228(1–2),
167–178. http://dx.doi.org/10.1016/j.palaeo.2005.03.048.
Surge, D. M., Lohmann, K. C., and Dettman, D. L., 2001. Controls
on isotopic chemistry of the American oyster, Crassostrea
virginica: implications for growth patterns. Palaeogeography,
Palaeoclimatology, Palaeoecology, 172, 283–296.
Surge, D., Wang, T., Gutiérrez-Zugasti, I., and Kelley, P. H., 2013.
Isotope sclerochronology and season of annual growth line for-
mation in limpet shells (Patella vulgata) from warm- and cold-
temperate zones in the eastern North Atlantic. Palaios, 28(6),
386–393, doi:10.2110/palo.2012.p12-038r.
Tanabe, K., and Oba, T., 1988. Latitudinal variation in shell growth
patterns of Phacosoma japonicum (Bivalvia: Veneridae) from
the Japanese coast. Marine Ecology: Progress Series, 47, 75–82.
Tevez, M. J. S., and Carter, J. G., 1980. Chapter 4, Study of annual
growth bands in Unionacean bivalves. In Rhoads, D. C., and
Lutz, R. A. (eds.), Skeletal Growth of Aquatic Organisms:
Biological Records of Environmental Change. New York/London:
Plenum Press, pp. 613–614.
Thompson, I., Jones, D. S., and Dreibelbis, D., 1980. Annual inter-
nal growth banding and life history of the ocean quahog Arctica
islandica (Mollusca: Bivalvia). Marine Biology, 57(1), 25–34,
doi:10.1007/bf00420964.
Thórarinsdóttir, G. G., 2000. Annual gametogenic cycle in ocean
quahog, Arctica islandica from north-western Iceland. Journal
of the Marine Biological Association of the United Kingdom,
80, 661–666.
Urey, H. C., 1947. The thermodynamic properties of isotopic sub-
stances. Journal of the Chemical Society, 562–581,
doi:10.1039/jr9470000562.
Wanamaker, A. D., Kreutz, K. J., Borns, H. W., Introne, D. S.,
Feindel, S., and Barber, B. J., 2006. An aquaculture-based
method for calibrated bivalve isotope paleothermometry. Geo-
chemistry, Geophysics, Geosystems, 7(9), Q09011,
doi:10.1029/2005gc001189.
Wanamaker, A. D., Jr., Heinemeier, J., Scourse, J. D., Richardson,
C. A., Butler, P. G., Eiríksson, J., and Knudsen, K. L., 2008.
Very long-lived mollusks confirm 17th Century AD tephra-
based radiocarbon reservoir ages for North Icelandic shelf
waters. Radiocarbon, 50(3), 399–412.
 BOMB CARBON
Wanamaker, A. D., Jr., Kreutz, K. J., Schöne, B. R., and Introne,
D. S., 2011. Gulf of Main shells reveal changes in seawater tem-
perature seasonality during the Medieval Climate Anomaly and
the Little Ice Age. Palaeogeography, Palaeoclimatology,
Palaeoecology, 302, 43–51, doi:10.1016/j.palaeo.2010.06.005.
Wang, T., Surge, D., and Walker, K. J., (2011). Seasonal climate
change across the Roman Warm Period/Vandal Minimum transi-
tion using isotope sclerochronology in archaeological shells and
otoliths. Quaternary International. 308–309, 230–241. http://
dx.doi.org/10.1016/j.quaint.2012.11.013.
Watanabe, T., and Oba, T., 1999. Daily reconstruction of water tem-
perature from oxygen isotopic ratios of a modern Tridacna shell
using a freezing microtome sampling technique. Journal of Geo-
physical Research: Oceans, 104(C9), 20667–20674,
doi:10.1029/1999jc900097.
Wefer, G., and Berger, W. H., 1991. Isotope paleontology: growth
and composition of extant calcareous species. Marine Geology,
100, 207–248.
Weidman, C. R., Jones, G. A., and Lohmann, K. C., 1994. The long-
lived mollusc Arctica islandica: a new paleoceanographic tool
for the reconstruction of bottom temperatures for the continental
shelves of the northern North Atlantic Ocean. Journal of Geo-
physical Research: Oceans, 99, 18305–18314.
Wisshak, M., López Correa, M., Gofas, S., Salas, C., Taviani, M.,
Jakobsen, J., and Freiwald, A., 2009. Shell architecture, element
composition, and stable isotope signature of the giant deep-sea oys-
ter Neopycnodonte zibrowii sp. n. from the NE Atlantic. Deep Sea
Research, Part I: Oceanographic Research Papers, 56(3),
374–407. http://dx.doi.org/10.1016/j.dsr.2008.10.002.
Witbaard, R., Jenness, M. I., van der Borg, K., and Ganssen, G.,
1994. Verification of annual growth increments in Arctica
islandica L. from the North Sea by means of oxygen and carbon
isotopes. Netherlands Journal of Sea Research, 33, 91–101.
Yan, L., Schöne, B. R., and Arkhipkin, A., 2012. Eurhomalea
exalbida (Bivalvia): a faithful recorder of climate in southern
South America? Palaeogeography, Palaeoclimatology,
Palaeoecology, 250–252, 91–100.
Cross-references
Corals (Sclerochronology)
Molluscs, Foraminifera, and Other Carbonate Fossils
BOMB CARBON
George S. Burr
NSF Arizona Accelerator Mass Spectrometry Laboratory,
Department of Geosciences and Physics, University
of Arizona, Tucson, AZ, USA
Department of Geosciences, National Taiwan University,
Taipei, Taiwan
Synonyms
Bomb 14C; Bomb radiocarbon; Excess anthropogenic
radiocarbon; Excess radiocarbon
Definition
Bomb carbon refers to anthropogenic radiocarbon produced
and released into the atmosphere during aboveground
nuclear weapons testing, primarily between 1945 and 1963.
115
Introduction
When Willard Libby introduced the radiocarbon method
(Libby et al., 1949), global atmospheric radiocarbon
(14C) from nuclear weapons fallout was undetectable. This
soon changed, however, as weapons testing programs
accelerated during the 1950s and into the 1960s. At the
same time, new radiocarbon laboratories were being
established in Europe, North America, Africa, and Austra-
lia. It did not take long for researchers to discover a dra-
matic increase in atmospheric 14C from nuclear weapons
testing fallout, in both the Northern Hemisphere
(de Vries, 1958; Münnich and Vogel, 1958; Broecker
and Walton, 1959) and the Southern Hemisphere (Rafter
and Fergusson, 1957; Münnich and Vogel, 1958; Broecker
and Walton, 1959). This nuclear weapons-related radio-
carbon became known as bomb carbon. The production
of bomb carbon accelerated in the late 1950s and early
1960s as tests became more numerous and the explosive
power of nuclear weapons increased dramatically. By
1963, the radiocarbon content of the atmosphere reached
its peak when a nuclear test ban treaty put an end to above-
ground nuclear weapons testing.
Plotted versus time, atmospheric bomb carbon follows
a curve with a sharp increase in the 1950s and with a peak
production in the early 1960s. Atmospheric bomb carbon
peaks around 1964 and then declines rapidly (Figure 1).
This feature is often referred to as the radiocarbon bomb
pulse, bomb spike, or bomb curve. The decline of atmo-
spheric bomb carbon marks the end of aboveground
nuclear weapons testing and illustrates the subsequent
rapid removal of atmospheric 14CO2 by the oceans
(Craig, 1957; Revelle and Suess, 1957) and biosphere.
The oceanic uptake of bomb carbon has been extensively
documented by direct seawater analyses (Stuiver
et al., 1981; Key et al., 2004), and biospheric uptake is
reflected in tree-ring 14C records. These examples repre-
sent two primary pathways of the carbon cycle – a global
dynamic system that moves carbon within and between
the atmosphere, oceans, biosphere, and lithosphere. They
also illustrate the chief importance of bomb carbon to sci-
ence, as an isotopic tracer and time marker.
14
C production
Radiocarbon is produced naturally in the Earth’s atmo-
sphere, as cosmic rays collide with atmospheric mole-
cules. These collisions result in a variety of particles,
including secondary thermal neutrons. A portion of these
thermal neutrons (n) interact with atmospheric 14N and
produce 14C and protons (p) by the reaction:
14
N ðn, pÞ14 C: ð1Þ
This is the dominant pathway for the formation of natural
radiocarbon on Earth (Libby, 1955), and it is a conse-
quence of the relatively large thermal neutron cross
section (1.8 barns) for the 14N(n,p)14C reaction (Lal
and Peters, 1967). Aboveground nuclear weapons tests
produce carbon in the same manner as cosmic rays
116 BOMB CARBON
Bomb Carbon, Figure 1 The radiocarbon bomb pulse. Atmospheric 14CO2 values from northern Europe, expressed as fraction
modern carbon (F) values (see text for details) (Data from Levin and Kromer (2004)).
because the atmosphere is rich in nitrogen and both
fission- and fusion-based nuclear weapons provide an
ample supply of thermal neutrons.
Estimates for the total amount of bomb carbon pro-
duced during the years of aboveground nuclear weapons
testing are on the order of 1028 atoms. These values are
usually expressed in multiples of 1026 14C atoms, or in
kmol 14C (1 kmol 14C ¼ 6.02  1026 atoms). Since atmo-
spheric carbon rapidly exchanges with carbon in the ocean
and biosphere, it is not possible to directly measure the
integral total production of bomb radiocarbon during the
years of testing. An alternative is to estimate a value that
sums the output from individual weapons tests. Naegler
and Levin (2006) used this approach to estimate total pro-
duction from 1945 to 1980 at 598–632  1026 14C atoms,
based in part on estimates from Enting (1982), Enting and
Pearman (1982), Rath (1988), Hesshaimer et al. (1994),
Lassey et al. (1996), and Yang et al. (2000). The amount
of bomb carbon produced by a particular nuclear explo-
sion is related to the amount of energy released. The pre-
cise yield, however, depends on factors such as the type
and design of a given nuclear weapon, the altitude of the
blast, and environmental conditions (Glasstone and
Dolan, 1977). These factors lead to intrinsic uncertainties
in the calculation of the exact amount of total bomb carbon
produced by weapons testing (Tans, 1981; Naegler and
Levin, 2006), hence the broad range quoted above.
The natural production rate of 14C in the atmosphere is
currently about 2 atoms cm
2 s
1 (Castagnoli and Lal,
1980). Multiplying this by the surface area of the Earth
(5.1  1018 cm2) and multiplying by 1 year (3.15 
107 s) gives a ballpark average for the annual production
of natural radiocarbon on Earth, which is 3.21  1026
atoms. A comparison with the estimate from Naegler and
Levin (2006) implies that the total production of bomb
carbon amounts to approximately 190 years of natural
production.
Records of atmospheric bomb carbon
Atmospheric records of bomb carbon are important to cli-
mate modelers because the bulk of 14C produced by
nuclear weapons was injected into the stratosphere. Strato-
spheric measurements of bomb carbon using aircraft and
balloons have been compiled by Telegadas (1971). These
data provide constraints on the total bomb carbon produc-
tion estimates described above (Tans, 1981; Naegler and
Levin, 2006). For the purpose of radiocarbon dating, it is
the tropospheric radiocarbon values that are pertinent
because the troposphere is in direct contact with the plan-
etary boundary layer where plants and animals live.
Tropospheric records of bomb carbon are obtained
from direct measurements of atmospheric CO2 and from
measurements of trees. Trees can be used to reconstruct
radiocarbon time series because they incorporate carbon
directly from the atmosphere (through photosynthesis)
and because they form annual growth rings suitable for
age control. Tans (1981) compiled a composite record of
global atmospheric and marine bomb carbon using a vari-
ety of sources as a database for modelers. Levin and
Kromer (2004) published a detailed atmospheric bomb
carbon record from Europe, for the period 1953–2003,
based on measurements of atmospheric CO2. Hua and
Barbetti (2004) made a detailed comparison of atmo-
spheric records from around the world. Their results
showed a large difference between the two hemispheres,
with the highest values in the Northern Hemisphere and
lowest values in the Southern Hemisphere (Figure 2).
These differences are much greater than the well-known
natural differences between the two hemispheres
 BOMB CARBON
1000
900
800
700
600
D14 C (‰) 500
400
300
200
100
0
−100
1950 1960 1970
Year (AD)
Bomb Carbon, Figure 2 Atmospheric radiocarbon from nuclear weapons testing in the 1950s and 1960s. Note the different
amplitude and rate of increase in the Southern Hemisphere, as compared with the Northern Hemisphere. Zones represented in the
figure are NH1 Northern Hemisphere high-latitude sites, NH2 and NH3 Northern Hemisphere mid- to low-latitude sites, and SH
Southern Hemisphere (Figure from Hua and Barbetti (2004)).
described by McCormac et al. (1998) and Hogg
et al. (2002). Hua and Barbetti (2004) also documented
bomb carbon zonation within the Northern Hemisphere
and proposed four primary curves to describe the bomb
pulse: NH1, NH2, NH3, and SH (Figure 2). As shown
below, these records are critical to a bomb carbon age
determination because they represent the starting point
of a sample for a particular region.
Records of marine bomb carbon
The ocean is the world’s largest carbon reservoir (Craig,
1957; Revelle and Suess, 1957), and characterizing the
transfer of bomb carbon into the oceans and its subsequent
transport within individual water masses became the sub-
ject of intense study following the era of aboveground
nuclear weapons testing. The multinational Geochemical
Ocean Sections Study (GEOSECS) formally began in
1967 and included a large number of scientific cruises
across the Pacific and Atlantic Oceans (Craig, 1972,
1974; Craig and Turekian, 1976, 1980). A key element
of the GEOSECS program was to study the distribution
of 14C in marine surface waters and at depth (Stuiver
et al., 1981). A second such program, the World Ocean
Circulation Experiment (WOCE), was conducted in the
1990s (Key et al., 2004). The WOCE program initiated
the construction of a dedicated accelerator mass spectrom-
etry (AMS) facility in Woods Hole, MA, which eventually
produced over 13,000 radiocarbon measurements from
seawater (McNichol et al., 2000). These and other pro-
grams were instituted to provide the type of three-
dimensional data coverage and measurement precision
117
NH zone 1
NH zone 2
NH zone 3
SH zone
1980 1990 2000
necessary for climatological modeling. The records pro-
duced are also ideal for radiocarbon dating in the ocean.
A second significant source of marine bomb 14C
records comes from radiocarbon analyses of corals
(Druffel and Linick, 1978; Druffel, 1981, 1987). Certain
reef-building corals produce annual growth rings that
allow for the determination of calendrical ages. When
paired with radiocarbon measurements, corals offer an
excellent archive for the reconstruction of radiocarbon
time series, analogous to the terrestrial example of trees.
Many thousands of such dates have now been published,
and these feature much finer geographic and time resolu-
tion than seawater measurements collected during individ-
ual cruises.
Dating with bomb carbon
Dating with bomb carbon follows the same logic as tradi-
tional radiocarbon dating. A date is determined by com-
parison of the measured 14C content of a sample with
calibrated 14C time series records. The sharp rise and fall
of the bomb pulse makes it possible to date very accurately
(within a few years) events which occurred since the late
1950s. The 14C content of a sample can be measured using
decay counting or by AMS. AMS results are usually
expressed as fraction of modern carbon (F) values, defined
as (Donahue et al., 1990).
 
14
C=13 C
F  sample½
25 ; ð2Þ
14 C=13 C
1950½
25
118 BOMB CARBON
Bomb Carbon, Figure 3 Sample calculation of a bomb carbon date. The fraction modern carbon value (F ¼ 1.6000) is projected onto
the atmospheric 14CO2 data of Levin and Kromer (2004) along the dashed line. This gives two possible ages for the sample, at 1963
and 1967–1968. The width of the atmospheric curve includes two sigma uncertainties.
where (14C/13C)sample [
25] is the measured ratio of the sam-
ple, corrected to d13C ¼
25%, and (14C/13C)1950 [
25]
is the d13C-corrected ratio for the standard, adjusted for
decay to the year 1950 (the “zero” year for radiocarbon
dating).
Figure 3 shows an example of a radiocarbon age deter-
mination using bomb carbon for a sample with a fraction
modern carbon value of F ¼ 1.6000 0.0050. This exam-
ple employs the atmospheric curve of Levin and Kromer
(2004), appropriate for northern Europe. The age of the
sample is determined by projecting the measured F value
onto the bomb carbon curve. This gives two possible ages:
(1) in 1963 and (2) 1967/1968. The uncertainties in this
procedure can be obtained graphically or calculated. The
CALIBOMB program, supported by Queen’s University
Belfast, is freely available to calculate and plot bomb
radiocarbon dating results using the Levin and Kromer
(2004) dataset (www.calib.org).
Applications of bomb pulse radiocarbon dating
The application of bomb pulse radiocarbon dating cuts
across many disciplines. The common feature of these
studies is that they are concerned with events that tran-
spired within the past 50 years. Recent examples from
the literature include the following: (1) Fisheries Sciences,
where bomb carbon dating has been used to determine the
age of fishes as a critical component of population assess-
ments (Passerotti et al., 2010; Andrews et al., 2011);
(2) Forensic Sciences, where bomb carbon dating has
assisted in the identification of crime victims (Speller
et al., 2012) and to analyze the production to marketing
cycle of the illicit drug trade (Ehleringer et al., 2012);
(3) Earth Sciences, where bomb carbon has been used to
determine the growth rate of cacti in reconstructing
records of precipitation (English et al., 2010) and to deter-
mine the growth rate of speleothems (Hodge et al., 2011);
and (4) Environmental Sciences, where bomb carbon has
been used to determine rates of recent soil development
and shoreline accretion (Lovelock et al., 2010). Applica-
tions of bomb pulse radiocarbon dating have expanded
in recent years, and with radiocarbon’s relatively long
half-life (5,700 years), it is likely that this expansion will
continue into the foreseeable future.
Summary
Aboveground nuclear weapons testing conducted between
1945 and 1963 nearly doubled the amount of 14C in the
atmosphere with the addition of bomb carbon. After the
nuclear test ban treaty was signed in 1963, this excess
atmospheric radiocarbon rapidly diminished through
exchange with the oceans and biosphere. This rise and fall
of atmospheric radiocarbon created a global radiocarbon
bomb pulse that serves as a time marker for a wide range
of dating applications.
Bibliography
Andrews, A. H., Natanson, L. J., Kerr, L. A., Gurgess, G. H., and
Cailliet, G. M., 2011. Bomb radiocarbon and tag-recapture dat-
ing of sandbar shark (Carcharhinus plumbeus). Fishery Bulletin,
109(4), 454–465.
Broecker, W. S., and Walton, A., 1959. Radiocarbon from nuclear
tests. Science, 130, 309–314.
Castagnoli, G., and Lal, D., 1980. Solar modulation effects in terres-
trial production of carbon-14. Radiocarbon, 22(2), 133–158.
Craig, H., 1957. The natural distribution of radiocarbon and the
exchange time of carbon dioxide between atmosphere and sea.
Tellus, 9(1), 1–17.
Craig, H., 1972. The GEOSECS program: 1970–1971. Earth and
Planetary Science Letters, 16, 47–49.
 BOMB CARBON
Craig, H., 1974. The GEOSECS program: 1972–1973. Earth and
Planetary Science Letters, 23, 63–64.
Craig, H., and Turekian, K. K., 1976. The GEOSECS program:
1973–1976. Earth and Planetary Science Letters, 32, 217–219.
Craig, H., and Turekian, K. K., 1980. The GEOSECS program:
1976–1979. Earth and Planetary Science Letters, 49, 263–265.
de Vries, H., 1958. Atomic bomb effect: variation of radiocarbon in
plants, shells, and snails in the past 4 years. Science, 128,
250–251.
Donahue, D. J., Linick, T. W., and Jull, A. J. T., 1990. Isotope-ratio
and background corrections for accelerator mass spectrometry
radiocarbon measurements. Radiocarbon, 32(2), 135–142.
Druffel, E. R. M., 1981. Radiocarbon in annual coral rings from the
eastern tropical Pacific Ocean. Geophysical Research Letters, 8,
59–62.
Druffel, E. R. M., 1987. Bomb radiocarbon in the Pacific: annual
and seasonal timescale variations. Journal of Marine Research,
45, 667–698.
Druffel, E. R. M., and Linick, T. W., 1978. Radiocarbon in annual
coral rings of Florida. Geophysical Research Letters, 5,
913–916.
Ehleringer, J. R., Casale, J. F., Barnette, J. E., Xu, X., Lott, M. J.,
and Hurley, J., 2012. 14C analyses quantify time lag between
coca leaf harvest and street-level seizure of cocaine. Forensic
Science International, 214(1–3), 7–12.
English, N., Dettman, D. L., and Williams, D. G., 2010. A 26-year
stable isotope record of humidity and El Niño-enhanced precip-
itation in the spines of saguaro cactus, Carnegiea gigantea.
Palaeogeography, Palaeoclimatology, Palaeoecology, 293,
108–119.
Enting, I., 1982. Nuclear weapons data for use in carbon cycle mon-
itoring. CSIRO Division of Atmospheric Physics Technical
Paper No. 44, Melbourne.
Enting, I., and Pearman, G. I., 1982. Description of a
one-dimensional global carbon cycle model. CSIRO Division
of Atmospheric Physics Technical Paper No. 42, Melbourne.
Glasstone, S., and Dolan, P. J., 1977. The Effects of Nuclear
Weapons. Washington, DC: U. S. Department of
Defense/U. S. Department of Energy.
Griffith, D. R., Martin, W. R., and Eglinton, T. I., 2010. The radio-
carbon age of organic carbon in marine surface sediments.
Geochimica et Cosmochimica Acta, 74, 6788–6800.
Hesshaimer, V., and Levin, I., 2000. Revision of the stratospheric
bomb 14CO2 inventory. Journal of Geophysical Research, 105,
11,641–11,658.
Hesshaimer, V., Heimann, M., and Levin, I., 1994. Radiocarbon evi-
dence for a smaller oceanic carbon dioxide sink than previously
believed. Nature, 370, 201–203.
Hodge, E., McDonald, J., Fischer, M., Redwood, D., Hua, Q.,
Levchenko, V., Drysdale, R., Waring, C., and Fink, D., 2011.
Using the C-14 bomb pulse to date young speleothems. Radio-
carbon, 53(2), 345–357.
Hogg, A. G., McCormac, F. G., Higham, T. F. G., Reimer, P. J., Bail-
lie, M. G. L., and Palmer, J. G., 2002. High-precision radiocar-
bon measurements of contemporaneous tree-ring dated wood
from the British Isles and New Zealand: AD 1850–950. Radio-
carbon, 44(3), 633–640.
Hua, Q., and Barbetti, M., 2004. Review of tropospheric bomb 14C
data for carbon cycle modeling and age calibration purposes.
Radiocarbon, 46(3), 1273–1298.
Key, R. M., Kozyr, A., Sabine, C. L., Lee, K., Wanninkhof, R.,
Bullister, J. L., Feely, R. A., Millero, F. J., Mordy, C., and Peng,
T.-H., 2004. A global ocean carbon climatology: results from
Global Data Analysis Project (GLODAP). Global Biogeochem-
ical Cycles, 18, 1–23, doi:10.1029/2004GB002247.
Lal, D., and Peters, B., 1967. Cosmic ray produced radioactivity on
the Earth. Handbuch der Physik, 46(2), 551–612.
119
Lassey, K. R., Enting, I., and Trudinger, C. M., 1996. The earth’s
radiocarbon budget: a consistent model of the global carbon
and radiocarbon cycles. Tellus, Series B, 48, 487–501.
Levin, I., and Kromer, B., 2004. The tropospheric 14CO2 level in
mid-latitudes of the Northern Hemisphere (1959–2003). Radio-
carbon, 46(3), 1261–1272.
Libby, W. F., 1955. Radiocarbon Dating, 2nd edn. Chicago:
Chicago University Press.
Libby, W. F., Anderson, E. C., and Arnold, J. R., 1949. Age determi-
nation by radiocarbon content: world-wide assay of natural
radiocarbon. Science, 109, 227–228.
Lovelock, C. E., Sorrell, B. K., Hancock, N., Hua, Q., and Swales,
A., 2010. Mangrove forest and soil development on a rapidly
accreting shore in New Zealand. Ecosystems, 13, 437–451,
doi:10.1007/s10021-010-9329-2.
McCormac, F. G., Hogg, A. G., Higham, T. F. G., Lynch-Stieglitz,
J., Broecker, W. S., Baillie, M. G. L., Palmer, J., Xiong, L.,
Pilcher, J. R., Brown, D., and Hoper, S. T., 1998. Temporal var-
iation in the interhemispheric C-14 offset. Geophysical Research
Letters, 25, 1321–1324.
McNichol, A., Schneider, R. J., von Reden, K. F., Gagnon, A. R.,
Elder, K. L., NOSAMS, Key, R. M., and Quay, P. D., 2000.
Ten years after – the WOCE AMS radiocarbon program.
Nuclear Instruments and Methods B, 172, 479–484.
Münnich, K. O., and Vogel, J. C., 1958. Durch Atomexplosionen
erzeugter radiokohlenstoff in der Atmosphäre. Naturwis-
senschaften, 45, 327–329.
Naegler, T., and Levin, I., 2006. Closing the global radiocarbon
budget 1945–2005. Journal of Geophysical Research, 111,
1–14, doi:10.1029/2005JD006758.
Nydal, R., 1968. Further investigation on the transfer of radiocarbon
in nature. Journal of Geophysical Research, 73(12), 389–406.
Nydal, R., Lövseth, K., and Skogseth, F. H., 1980. Transfer of bomb
14
C to the ocean surface. Radiocarbon, 22(3), 626–635.
Passerotti, M. S., Carlson, J. K., Piercy, A. N., and Campana, S. E.,
2010. Age validation of great hammerhead shark (Sphyrna
mokarran), determined by bomb radiocarbon analysis. Fisheries
Bulletin, 108, 346–351.
Rafter, T. A., and Fergusson, G. J., 1957. “Atom bomb
effect”-recent increase of Carbon-14 content of the atmosphere
and biosphere. Science, 126, 557–558.
Rath, H., 1988. Simulation der globalen 85Kr- und 14CO2 –
Verteilung mit Hilfe eines zeitabhängigen, zweidimensionalen
modells der atmosphäre. PhD thesis, Heidelberg University.
Revelle, R., and Suess, H. E., 1957. Carbon dioxide exchange
between atmosphere and ocean and the question of an increase
of atmospheric CO2 during the past decades. Tellus, 9(1), 18–27.
Speller, C. F., Spalding, K. L., Buchholz, B. A., Hildebrand, D.,
Moore, J., Mathewes, R., Skinner, M. F., and Yang, D. Y.,
2012. Personal identification of cold case remains through com-
bined contribution from anthropological, mtDNA, and bomb-
pulse dating analyses. Journal of Forensic Sciences, 57(5),
1354–1360, doi:10.1111/j.1556-4029.2012.02223.x.
Stuiver, M., Östlund, H. G., and McConnaughey, T. A., 1981.
GEOSECS Atlantic and Pacific 14C distribution. In Bolin,
B. (ed.), Carbon Cycle Modeling (Scope 16). New York: Wiley,
pp. 201–221.
Tans, P., 1981. A compilation of bomb 14C data for use in global car-
bon model calculations. In Bolin, B. (ed.), Carbon Cycle Model-
ing (Scope 16). New York: Wiley, pp. 131–157.
Telegadas, K., 1971. The seasonal atmospheric distribution and
inventories of excess carbon-14 from March 1955 to July
1969. U. S. Atomic Energy Commission Report HASL-243.
Yang, X., North, R., and Romney, C., 2000. CMR nuclear explosion
data base (revision 3). CMR Technical Report 00/16. Arlington,
VA: Center for Monitoring Research, U. S. Army Space and
Missile Defense Command.
120 BONES (U-SERIES)
Cross-references
Accelerator Mass Spectrometry
Terrestrial Cosmogenic Nuclide Dating
BONES (U-SERIES)
Alistair W. G. Pike
Department of Archaeology, University of Southampton,
Southampton, UK
Synonyms
230
Th/U dating of bone; U-series disequilibrium dating of
bone; U–Th dating of bone
Definition
Uranium-series dating: A geochronological method that
uses intermediate decay products in the 238U or 235U
radioactive decay chains to measure the length of time
required to reach the current state of radioactive disequi-
librium from an initial condition.
Bone: A dense, semirigid, porous, calcified connective
tissue forming the major portion of the skeleton of most
vertebrates consisting of an organic matrix, largely colla-
gen, and a mineral phase of nonstoichiometric carbonate
hydroxyapatite Ca10(PO4)6(OH)2.
Introduction
Uranium-series disequilibrium dating of carbonate precip-
itates provides one of the most precise (and accurate) dat-
ing methods for the last 500 ka and has found application
in archaeology for dating flowstone floors that cap or
bracket archaeological deposits (e.g., Schwarcz and
Blackwell, 1992) or that are associated with cave paint-
ings (Pike et al., 2012). The restricted range of contexts
where suitable carbonate samples are associated with cul-
tural layers, however, limits its applicability to archaeo-
logical chronology. Because of this, considerable effort
has been made to attempt to apply the U-series method
directly on archaeological bone.
Living bone is a composite of protein (largely collagen)
and nonstoichiometric carbonate hydroxyapatite (often
referred to as bioapatite). It contains only traces of
uranium (<a few ppb), whereas archaeological bone can
contain many 10 s of ppm. Unlike calcite flowstones, bone is
an open system; it gains most of its uranium from the burial
environment, and any attempt to calculate a U-series date
from measurements of 234U, 238U, and 230Th requires the
application of a model of this uptake.
The early uptake assumption
Early attempts at U-series dating of bone simply treated it
as an approximation to a closed system, using the “early
uptake” (EU) assumption – that the uranium was taken
up by the bone initially over a short period of time relative
to its age. A comparison of radiocarbon dates (between
10 and 30 ka) with EU U-series dates on bones showed
an overall average underestimation in the U-series dates
of 2,700 years and appeared initially to support the EU
assumption (Szabo, 1980). Given most of the bones were
only slightly older than 10 ka, however, this is still a signif-
icant underestimation. While use of the EU assumption
was widespread in early attempts to date bone, often with
very mixed results (e.g., Szabo et al., 1973), a chemical or
physical mechanism for it was not formally proposed until
the work of Rae and Ivanovich (1986). It had previously
been proposed (e.g., Szabo, 1980) that the primary mech-
anism for incorporation of uranium in bone was by substi-
tution for Ca2+ in the bone mineral hydroxyapatite.
Because of differences in their ionic radii, this necessitated
the reduction of U6+ to U4+. In most burial environments,
uranium is in the mobile U6+ (uranyl) rather than the
immobile U4+form (Gascoyne, 1982), but Rae proposed
that the conditions needed for reduction would be present
in archaeological bone as a result of the decomposition of
the bone’s organic matrix – collagen – into its constituent
amino acids. This mechanism would explain not only how
the uranium was fixed in the bone through reduction to U4+
and substitution for Ca but also how uranium can be
incorporated early in the burial history of a bone after which
further uptake ceases. Collagen is rarely found in bones
older than a few tens of thousands of years and almost never
in bones older than 100 ka. Therefore, according to Rae’s
proposal, the uptake of uranium should be limited to the
early stages of burial history of a bone – the period while
the collagen is degrading. While this mechanism seemed
plausible, Rae herself found dates calculated using the EU
assumption to give mixed results, especially from open-air
(non-cave) sites (Rae et al., 1989), and also noted that
some bones appeared to give older than expected results,
probably as a result of the loss of uranium unaccounted
for in the EU model. The few further systematic studies
comparing EU U-series dates with expected ages were
unable to substantiate EU behavior as a general feature of
bone (e.g., see Pike and Pettitt, 2003).
Open-system modeling using 230Th/U
and 235U/231Pa
In parallel, there were a number of attempts to model the
open-system behavior of bone using combined 230Th/U
and 235U/231Pa measurements (e.g., Szabo and Malde,
1969; Szabo et al., 1973; Szabo and Collins, 1975; Szabo,
1979). Lacking a geochemical model of uranium-bone
interactions, these models were generally straightforward
mathematical assumptions, albeit with plausible physical
implications. For example, a model proposed by Szabo
and Rosholt (1969) with later modifications by Chen and
Yuan (1988) incorporated uranium uptake in two stages:
an initial period of rapid uptake followed by a second later
episode of either uptake or loss of uranium. Using isochro-
nous samples (e.g., multiple bones from a single archaeo-
logical layer) that were assumed to have been affected
by the second migration at the same time but to
 BONES (U-SERIES)
different extents, they calculated a corrected age from
230
Th/U and 235U/231Pa (see Cheng et al., 1998 for more
on U/Pa systematics). While this method was used to
provide chronologies for a number of key sites in China,
which in turn raised the intriguing possibility of
chronological overlap between Homo erectus and archaic
Homo sapiens (Chen and Zhang, 1991), subsequent
redating of the sites with U-series and other techniques
has shown the ages of the H. erectus sites to be much older
than estimated by Cheng and Yuan’s model (e.g., Grün
et al., 1998; Shen et al., 2001; Zhao et al., 2001).
The diffusion-adsorption model
A key development in the understanding of uranium
uptake mechanisms in bone was the work of Millard and
Hedges (1995, 1996). In contrast to Rae’s explanation
for EU which required the reduction of U4+ to U6+,
in vitro experiments by Millard and Hedges found uptake
of uranyl (i.e., U6+) in protein-free hydroxyapatite and
measured a partition coefficient between hydroxyapatite
and uranyl solution of 104–106, in agreement with estima-
tions from observations on archaeological bone. Drawing
on mathematical models of diffusion developed by Crank
(1975), they proposed an explicit geochemical model of
uranium uptake, the diffusion-adsorption (D-A) model,
where diffusion was the dominant transport mechanism
of uranyl ions into bone pores where it was subsequently
adsorbed onto the large (>100 m3 g
1) surface area of
bioapatite.
Using formulations given by Crank (1975), Millard and
Hedges proposed the diffusion equation for simultaneous
diffusion and adsorption as
@C D @ 2C
¼ many researchers had observed. After initial uptake,
@t ðR þ 1Þ @x2
where C is the uranium concentration at a point x in the
bone, D is the diffusion coefficient reduced for diffusion
in a porous medium such as bone, and R is the volumetric
partition coefficient which is related to the partition coef-
ficient (Kd) by R ¼ Kd/p where p is the specific porosity
of the bone. Both D and R can be related to the physical
structure of bone (e.g., volume, size distribution and tortu-
osity of bone pores, and the internal surface area).
Using a modification of Crank’s equation for diffusion
in an infinite planar slab and assuming a linear adsorption
isotherm, they derive the concentration (ppm), Z, at
a point x:
h i
4 X?
ð
1Þ n
Dð2nþ1Þ2 p2 t
Z ðxÞ ¼ pRC 1 1

2
e ðRþ1Þ4l
p n¼0 2n þ 1
 
ð2n þ 1Þp x
cos
2l
where x is the distance from the center of the slab (cm),
t is the time of burial, C1 is the environmental concentra-
tion of uranyl (ppm), and l is half the thickness of
121
the bone (cm). To simplify this, Millard and Hedges use
three reduced dimensionless parameters:
Z 0 ¼ Z=pRC 1 x0 ¼ ðx
lÞ=l
t 0 ¼ tD=ðR þ 1Þl 2
Under this scheme, Z0 represents the fraction of the
equilibrium concentration of uranium in the bone, since at
equilibrium the concentration of uranium, Z, is pRC1.
The parameter x0 represents the fractional distance
from the center of the bone section and takes a value of

1 to 1. The parameter t0 is a function of both time
and D/R. Young bones or bones with lower D/R give
smaller t0 .
Not only can the D-A model predict the rate of accumu-
lation of uranium in bone, but it also predicts the distribu-
tion (“profiles”) of uranium and uranium-series isotopes
transversely across a bone (Figure 1). Under constant geo-
chemical conditions, the model predicts the diffusion of
uranium from the outer and inner (medullary) surfaces of
the bone, leading to [-shaped uranium profiles, and an
increasing underestimation of closed-system U-series
dates toward the center of the bone. Over time, these ura-
nium concentration profiles gradually flatten until the
adsorbed uranium in the bone is at equilibrium with ura-
nium in the groundwater. The parameter t0 can be esti-
mated from the measured uranium concentration profiles
and U-series isotope profiles generated for different values
of t and compared to observations to yield a date.
Furthermore, since adsorption is a reversible process,
the D-A model can also predict the leaching of uranium
from the bone (e.g., in response to a decrease in uranium
concentration in the groundwater) and the overestimation
of U-series dates calculated using the EU assumption that
a decrease in uranium concentration in the groundwater
will lead to desorption of uranium and diffusion in the
opposite direction (i.e., out of the bone) and lead to \- or
M-shaped uranium concentration profiles. As
a consequence, 230Th/238U activity ratios become elevated
toward the surface of the bone due to the preferential loss
of U relative to Th resulting in overestimates of closed-
system U-series dates and elevated initial 234U/238U activ-
ity ratios. Similarly, an increase in the U concentration of
the groundwater will lead to increased uptake and charac-
teristic steep gradients of U profiles (Figure 2).
Pike et al. (2002) compared U concentration and
U-series isotope profiles in archaeological bone to those
predicted by the model. They found that where both ura-
nium concentration and U-series isotope profiles were
[-shaped and agreed with the profiles predicted by the
D-A model, open-system dates (calculating using the
D-A model) agreed well with control dates. However,
these bones were in the minority of those examined; many
did not appear to have experienced constant geochemical
environments. Bones that exhibited leached profiles as
well as those that had taken up uranium at an increasing
rate (deemed “recent uptake”) were present, as well as
122 BONES (U-SERIES)

i 1
t′=10
0.9 t′=1.0

Relative uranium concentration

0.8
0.7
0.6 t′=0.5
0.5
0.4
t′=0.3
0.3
0.2
0.1
t'=0.01 t′=0.1
0
–1.0 –0.8 –0.6 –0.4 –0.2 0.0 0.2 0.4 0.6 0.8 1.0
Relative distance from centre of bone

ii 100
t′=10
Apparent closed system date (ky)

90
80 t′=1.0

70
60
50 t'=0.1
40
30
20 t′=0.01
10
0
–1 –0.8 –0.6 –0.4 –0.2 0 0.2 0.4 0.6 0.8 1
Relative distance from centre of bone
Bones (U-Series), Figure 1 Simulated U concentration profiles (i) and “date profiles” (ii) across a transverse section of bone predicted
by the D-A model under constant geochemical conditions for a 100-ka-old bone. Different curves show the effects for different values
of the model parameter t0 , which is a function of time and parameters intrinsic to a particular bone. For a given bone, diffusion of
U from the edges of the bone leads to a [-shaped U profile that gradually flattens over time until an equilibrium concentration is
reached between U in the bone and U in the groundwater. The pattern of apparent closed-system dates shows increased
underestimation toward the center of the bone, except where bones equilibrate rapidly (i.e., t0
1).

those with irregular profiles that may have experienced be a mid-shaft section of long bone for uptake to
complex changes in burial geochemistry or diagenetic approximate to the one-dimensional diffusion in an
processes over long timescales. However, Pike’s infinite planar slab which is the basis of the D-A model.
et al. approach was to reject U-series results from these Other geometries (e.g., of teeth) have been considered
profiles and only accept results with profiles that (Pike and Hedges, 2001), and Crank (1975) gives
conformed to straightforward predictions of the D-A equations which could be incorporated to adapt the D-A
model. model to radial diffusion and other phenomena. Initially,
Pike et al. (2002) used inductively coupled plasma mass
Sampling and development of laser ablation spectrometric (ICP-MS) U concentration measurements
methods on 6–10 subsamples drilled from each cross section to
U concentration and isotope profiles were made on create uranium concentration profiles and screen bones
a transverse section, usually a few mm thick, and cut for U-series measurement. A second set of subsamples
from the bone using a diamond saw. Ideally, this would were drilled from selected bones for thermal ionization
 BONES (U-SERIES) 123

i
1.4

Relative uranium concentration

1.2 (C)
1 (A–1)

0.8
(A–2)
0.6
(B)
0.4
0.2 Inner (endosteal)
Outer (periosteal) surface of bone surface of bone
0
–1 –0.8 –0.6 –0.4 –0.2 0 0.2 0.4 0.6 0.8 1
Relative distance to centre of bone

130 ky 130 ky
ii 40000
35000 (B)
Apparent closed system date

30000
(A–1)
25000
20000 (A–2)

15000 (C)
10000
Inner (endosteal)
5000 Outer (periosteal) surface of bone surface of bone
0
–1 –0.8 –0.6 –0.4 –0.2 0 0.2 0.4 0.6 0.8 1
Relative distance to centre of bone
Bones (U-Series), Figure 2 Simulated U concentration profiles (i) and “date profiles” (ii) across a transverse section of bone predicted
by the D-A model for different U uptake regimes for a 30-ka-old bone. Shown in (i) are U concentration profiles predicted for diffusive
uptake under constant conditions with the bone nearing equilibrium with the groundwater, i.e., t0 ! 1 (A-1); diffusive uptake under
constant conditions with the bone far from equilibrium, i.e., t0  1 (A-2); leaching of uranium after initial uptake (B); and a recent
increase in the uptake of uranium, “recent uptake” (C). Simulated date profiles across the same bone are shown in (ii). Diffusive
U uptake under constant conditions (A-1) shows apparent closed-system dates decreasing toward the center of the bone section,
although dates are not as underestimated as for bones further from equilibrium with the groundwater (A-2). The leaching of U after
initial uptake increases the 230Th/U resulting in apparent closed-system dates that overestimate the true age, especially toward the
outer and inner surfaces of the bone (B). Recent increased U uptake (C) gives underestimated apparent closed-system dates and can
lead to a characteristic \ distribution of dates.

mass spectrometer (TIMS) measurement of U-series to data-dense measured profiles. The precision on
isotopes. This proved analytically costly, and slow, and modeled dates largely depends on the goodness of fit of
was subsequently replaced by a method that used laser- the measured profiles to the model’s predictions and is
ablation multi-collector ICP-MS (Eggins et al., 2003, typically 2–10 % (e.g., see Pike et al., 2005). Millard
2005). For bones with a few ppm of uranium or more, and Hedges’ model was later refined to allow for the evo-
U concentration and U-series isotope profiles could be lution of 234U/238U in the bone within a groundwater
measured simultaneously using laser ablation, at far with constant 234U/238U (Sambridge et al., 2012), though
higher spatial resolution than by drilling samples as yet it remains unclear to what extent there is exchange
(in some cases, several hundred 238U/230Th per bone sec- between adsorbed uranium and the groundwater, and
tion), and typically in less than an hour per bone. Preci- therefore the evolution of 234U/238U during uranium
sion on an individual measurement was far poorer than uptake remains a significant uncertainty which presum-
for TIMS, but this is outweighed by the increased preci- ably affects the accuracy of D-A dates where
sion and accuracy afforded by fitting predicted profiles 234
U/238U
1.0.
124 BONES (U-SERIES)
Applications
Because of their size and availability, the majority of
applications have been on faunal bones, for example, to
constrain the ages of biostratigraphic faunal assemblage
zones in the United Kingdom (e.g., Pike et al., 2005) or
as a proxy for the age of hominin bones or activity from
the same archaeological layer (e.g., Storm et al., 2013),
but the ability to date very small samples using laser abla-
tion has opened up the possibility of dating valuable
human fossils. These have included direct U-series dates
on the Omo-Kibish 1 modern human skull from Ethiopia,
confirming the Ar–Ar dating of sediments that suggest an
age of close to 195 ky (Aubert et al., 2012), supporting
genetic models of the appearance of modern humans in
Africa c. 120–220 ka ago (e.g., Ingman et al., 2000).
The method has also been used to produce dates of rele-
vance to understanding the migrations of modern humans
out of Africa, in particular in regions where the first
appearance of modern humans predate the c. 40–50 ka
limit of radiocarbon dating, such as a human metatarsal
dated to 67 1 ka in Callao Cave, Luzon, Philippines
(Mijares et al., 2010), or where collagen preservation is
poor, for example, for the Niah Cave deep skull from Bor-
neo dated to 35.2 0.3 ka (Barker et al., 2007; Reynolds
et al., 2013). In Europe, several studies have used the
method to constrain the timing of the disappearance of
the Neanderthals (Walker et al., 2008; Daura et al., 2010;
Zilhao et al., 2011), although some of the U-series results
were used only to corroborate other dating methods.
Future directions
One of the key obstacles to more widespread adoption of
U-series methods of dating bone is the number of bones
that are rejected as unsuitable for dating on the basis of
their measured profiles. A large proportion of these are
at equilibrium and show uniform U concentration and
U-series isotope profiles. A value for t0 cannot be esti-
mated from a uniform profile, and thus the rate of equili-
bration cannot be determined. This is important since
diagenetic changes in the bone may increase the rate at
which equilibrium is established, raising the possibility
that a bone may have re-equilibrated with geochemical
changes that would have resulted in the loss or further
uptake of uranium (phenomena normally spotted by char-
acteristic profiles). Thus dates from equilibrium profiles
are only of use to corroborate other dating methods (e.g.,
Walker et al., 2008). However, the recent development
of ICP methodologies to measure 231Pa at high precision
will enable combined U–Th-Pa concordance dating of
these samples (e.g., see Cheng et al., 1998) and even per-
haps the measurement of 235U/231Pa profiles to further
refine the D-A method.
Summary
Bone is an open system and gains (or loses) uranium from
the burial environment, and therefore a model of uranium
migration is required to calculate a U-series date from the
measured 230Th/U. Early models, based on simple mathe-
matical assumptions, did not yield consistent results, but
some success has been had using an explicit geochemical
model based on the processes of diffusion and adsorption
(the D-A model). The D-A model predicts the spatial dis-
tribution of U and U-series isotopes across a bone section,
and where these “profiles” indicate the bone has experi-
enced relatively constant geochemical conditions, an
open-system date can be calculated.
Bibliography
Aubert, M., Pike, A. W. G., Stringer, C., Bartsiokas, A., Kinsley, L.,
Eggins, S., Day, M., and Grün, R., 2012. Confirmation of a late
middle Pleistocene age for the Omo Kibish 1 cranium by direct
uranium-series dating. Journal of Human Evolution, 63(5),
704–710.
Barker, G., Barton, H., Bird, M., Daly, P., Datan, I., Dykes, A., Farr,
L., Gilbertson, D., Harrisson, B., Hunt, C., Higham, T.,
Kealhofer, L., Krigbaum, J., Lewis, H., McLaren, S., Paz, V.,
Pike, A., Piper, P., Pyatt, B., Rabett, R., Reynolds, T., Rose, J.,
Rushworth, G., Stephens, M., Stringer, C., Thompson, J., and
Turney, C., 2007. The ‘human revolution’ in lowland tropical
Southeast Asia: the antiquity and behaviour of anatomically
modern humans at Niah Cave (Sarawak, Borneo). Journal of
Human Evolution, 52, 243–261.
Chen, T., and Yuan, S., 1988. Uranium-series dating of bones and
teeth from Chinese Palaeolithic sites. Archaeometry, 30(1),
59–76.
Chen, T., and Zhang, Y., 1991. Paleolithic chronology and possible
coexistence of Homo erectus and Homo sapiens in China. World
Archaeology, 23(2), 147–154.
Cheng, H., Lawrence Edwards, R., Murrell, M. T., and Benjamin,
T. N., 1998. Uranium-thorium-protactinium dating systematics.
Geochimica et Cosmochimica Acta, 62(21/22), 3437–3452.
Crank, J., 1975. The Mathematics of Diffusion. Oxford: Oxford
University Press.
Daura, J., Sanz, M., Zilhao, J., Pike, A. W. G., Subirà, E., Fornós, F.,
Fullola, J. M., and Julià, R., 2010. Precise chronological dating
of the Neanderthal Mandible of Cova del Gegant (Sitges, Barce-
lona, Spain). Journal of Human Evolution, 59(1), 109–122.
Eggins, S., Grün, R., Pike, A., Shelley, M., and Taylor, L., 2003.
U and Th profiling and U-series isotope analysis of teeth by laser
ablation-ICPMS. Quaternary Science Reviews, 22, 1373–1382.
Eggins, S. M., Grün, R., McCulloch, M. T., Pike, A. W. G.,
Chappell, J., Kinsley, L., Mortimer, G., Shelley, M., Murray-
Wallace, C. V., Spötl, C., and Taylor, L., 2005. In situ U-series
dating by laser ablation multi-collector ICPMS: new prospects
for Quaternary geochronology. Quaternary Science Reviews,
24(23–24), 2523–2538.
Gascoyne, M., 1982. Geochemistry of the actinides and their daugh-
ters. In Ivanovich, M., and Harmon, R. S. (eds.), Uranium-Series
Disequilibrium: Application to Environmental Problems.
Oxford: Clarendon, pp. 33–55.
Grün, R., Huang, P. H., Huang, W., McDermott, F., Thorne, A.,
Stringer, C. B., and Yan, G., 1998. ESR and U-series analyses
of teeth from the palaeoanthropological site of Hexian, Anhui
Province, China. Journal of Human Evolution, 34(6), 555–564.
Ingman, M., Kaessmann, H., Paabo, S., and Gyllensten, U., 2000.
Mitochondrial genome variation and the origin of modern
humans. Nature, 408(6813), 708–713.
Mijares, A. S., Détroit, F., Piper, P., Grün, R., Bellwood, P., Aubert,
M., and Dizon, E., 2010. New evidence for a 67,000-year-old
human presence at Callao Cave, Luzon, Philippines. Journal of
Human Evolution, 59(1), 123–132.
 BONES (U-SERIES)
Millard, A. R., and Hedges, R. E. M., 1995. The role of the environ-
ment in uranium uptake by buried bone. Journal of Archaeolog-
ical Science, 22, 239–250.
Millard, A. R., and Hedges, R. E. M., 1996. A diffusion-adsorption
model of uranium uptake by archaeological bone. Geochimica et
Cosmochimica Acta, 60(12), 2139–2152.
Pike, A. W. G., and Hedges, R. E. M., 2001. Sample geometry and
U-uptake in archaeological teeth: implications for U-series and
ESR dating. Quaternary Science Reviews, 20(5–9), 1021–1025.
Pike, A. W. G., and Pettitt, P. B., 2003. U-series dating and human evo-
lution. Reviews in Mineralogy and Geochemistry, 52, 607–629.
Pike, A. W. G., Hedges, R. E. M., and Van Calsteren, P., 2002.
U-series dating of bone using the diffusion-adsorption model.
Geochimica et Cosmochimica Acta, 66(24), 4273–4286.
Pike, A. W. G., Eggins, S., Grün, R., Hedges, R. E. M., and Jacobi,
R. M., 2005. U-series dating of the Late Pleistocene fauna from
Wood Quarry (Steetley), Notinghamshire, UK. Journal of
Quaternary Science, 20(1), 59–65.
Pike, A. W. G., Hoffmann, D. L., García-Diez, M., Pettitt, P. B.,
Alcolea, J., De Balbín, R., González-Sainz, C., de las Heras,
C., Lasheras, J. A., Montes, R., and Zilhão, J., 2012. U-series
dating of Palaeolithic art in 11 caves in Spain. Science,
336(1409), 1409–1413.
Rae, A. M., and Ivanovich, M., 1986. Successful application of uranium
series dating of fossil bone. Applied Geochemistry, 1, 419–426.
Rae, A., Hedges, R. E. M., and Ivanovich, M., 1989. Further studies
for uranium-series dating of fossil bones. Applied Geochemistry,
4, 331–337.
Reynolds, T., Barker, G., Barton, H., Cranbrook, G., Hunt, C.,
Kealhofer, L., Paz, V., Pike, A., Piper, P., Rabett, R., Rushworth,
G., Stimpson, C., and Szabó, K., 2013. The first modern humans
at Niah, c. 50,000–35,000 years ago. In Barker, G. (ed.),
Rainforest Foraging and Farming in Island Southeast Asia.
The Archaeology of the Niah Caves, Sarawak. Cambridge:
McDonald Institute, pp. 135–172.
Sambridge, M., Grün, R., and Eggins, S., 2012. U-series dating of
bone in an open system: the diffusion-adsorption-decay model.
Quaternary Geochronology, 9, 42–53.
Schwarcz, H. P., and Blackwell, B. A., 1992. Archaeological appli-
cations. In Ivanovich, M., and Harmon, R. S. (eds.), Uranium
Series Disequilibrium Applications to Earth, Marine and
Environmental Sciences. Oxford: Clarendon, pp. 513–552.
Shen, G., Ku, T. L., Cheng, H., Edwards, R. L., Yuan, Z., and Wang,
Q., 2001. High-precision U-series dating of Locality 1 at
Zhoukoudian, China. Journal of Human Evolution, 41(6),
679–688.
Storm, P., Wood, R., Stringer, C., Bartsiokas, A., de Vos, J., Aubert,
M., and Grün, R., 2013. U-series and radiocarbon analyses of
125
human and faunal remains from Wajak, Indonesia. Journal of
Human Evolution, 64(5), 356–365.
Szabo, B. J., 1979. Dating fossil bone from Orange Free State,
South Africa. Journal of Archaeological Science, 6, 201–203.
Szabo, B. J., 1980. Results and assessment of uranium-series dating
of vertebrate fossils from quaternary alluviums in Colorado.
Arctic and Alpine Research, 12(1), 95–100.
Szabo, B. J., and Collins, D., 1975. Ages of fossil bones from
British interglacial sites. Nature, 254, 680–682.
Szabo, B. J., and Malde, H. E., 1969. Dilemma posed by uranium-
series dates on archaeologically significant bones from
Valsequillo, Puebla, Mexico. Earth and Planetary Science
Letters, 6, 237–244.
Szabo, B. J., and Rosholt, J. N., 1969. Uranium-series dating of
pleistocene molluscan shells from southern California – an open
system model. Journal of Geophysical Research, 74(12),
3253–3260.
Szabo, B. J., Stalker, A. M., and Churcher, C. S., 1973. Uranium
series ages from some quaternary deposits near Medicine Hat,
Alberta, Canada. Canadian Journal of Earth Sciences, 109,
1464–1469.
Walker, M. J., Gilbert, J., Lopez, M. V., Lombardi, A. V., Pérex-P-
érez, A., Zapata, J., Ortega, J., Higham, T., Pike, A.,
Schwenninger, J.-L., Zihão, J., and Trinkaus, E., 2008. Late
Neanderthals in Southeastern Iberia: Sima de las Palomas del
Cabexo Gordo, Murcia, Spain. Proceedings of the National
Academy of Sciences of the United States of America, 105(52),
20631–20636.
Zhao, J., Hu, K., Collerson, K. D., and Xu, H., 2001. Thermal ion-
ization mass spectrometry U-series dating of a hominid site near
Nanjing, China. Geology, 29(1), 27–30.
Zilhao, J., Cardoso, J. L., Pike, A. W. G., and Wenninger, B., 2011.
Gruta Nova da Columbeira (Bombarral, Portugal): site stratigra-
phy, age of the Mousterian sequence, and implications for the
timing of Neanderthal extinction in Iberia. Quartär, 58, 93–112.
Cross-references
Apatite
Ar–Ar and K–Ar Dating
Archaeomagnetic Dating
Carbonates, Speleothem Archaeological (U-Series)
Laser Ablation Inductively Coupled Mass Spectrometer
(LA ICP-MS)
Luminescence Dating of Archaeological Sediments
Mass Spectrometry
Thermal Ionization Mass Spectrometer (TIMS)
U-Series Dating
 C
14
C IN PLANT MACROFOSSILS
Christine Hatté1 and A. J. Timothy Jull2
1
Laboratoire des Sciences du Climat et de
l’Environnement, UMR8212 CEA-CNRS-UVSQ,
Domaine du CNRS, Gif-sur-Yvette, France
2
NSF-Arizona AMS Laboratory, Physics Building,
University of Arizona, Tucson, AZ, USA
Definition
We discuss 14C dating of plant macrofossils from different
contexts and consider the various complicating effects of
the history of the samples on the 14C measurement.
Introduction
Late Quaternary chronologies are commonly constructed
using AMS 14C dating of plant macrofossils, because they
are generally argued to provide the most reliable chronol-
ogy. Although plant macrofossils are often relatively
abundant and well preserved in a variety of site type, they
are still potentially subject to a range of complications. Be
they terrestrial or aquatic, plant macrofossils are prone to
absorb CO2 of mixed origin during photosynthesis, to be
reworked, contaminated by dissolved organic carbon and
by modern carbon due to inappropriate storage and analy-
sis. They can also be potentially impacted by measure-
ment effects relating to small sample sizes.
We will overview different macrofossils here, in dry
and then humid environments, will give few notes on
insects and invertebrates commonly retrieved besides
plant macrofossils and finally, we will highlight good
practices in the laboratory.
Dry environments
Numerous archeological and paleoclimatological studies
are performed in dry environments and chronological
J.W. Rink, J.W. Thompson (eds.), Encyclopedia of Scientific Dating Methods, DOI 10.1007/978-94-007-6304-3,
© Springer Science+Business Media Dordrecht 2015
frameworks are then established from (carbonized or
non-carbonized) vegetal remains. This section deals with
tree products, phytoliths, and pollen.
Tree products
Every year, tree growth adds wood to the outside part of
the trunk. The date measured on heartwood may be
already many centuries old by the time a tree was cut
down. This is particularly true for long-lived tree speci-
mens, such as oak and juniper. An ideal material to date,
if available, would be a twig or branchlet of small trees
that can integrate 5 years in the worst case. Seeds and other
food remains are also reliable material to be dated. Spe-
cific cultural or environmental conditions must equally
be seen as potentially impacting radiocarbon dates. This
is the case for arid, coastal, volcanic, and “industrial”
environments.
“Old wood effect”
The problem of the “old wood” effect in radiocarbon dat-
ing has long been recognized for charcoal and wood sam-
ples (e.g., timber), the age of which may be hundreds of
years older than expected. Besides this, the archeological
field faces another difficulty linked to possible time lags
between felling and final deposition. The timber may have
an extensive history of use and reuse, especially in arid
environments where hard trees are rare. We should be
aware that the date we are measuring is the death of the
tree and not its last use (e.g., Saliège et al., 2012).
Atmospheric CO2 of different origin and 14C signature
Costal environment. Significant effect from ocean
upwelling can be recorded in some coastal regions.
There, 14C-depleted CO2 degassing from deep ocean water
in an upwelling system can mix with free atmospheric
CO2 and be incorporated by coastal vegetation, which
results in older apparent ages, or “ageing.” There are
 14
128 C IN PLANT MACROFOSSILS
indications that there may be small but significant offsets
in Japan (Nakamura et al., 2007). This offset is difficult
to assess as it is directly linked to the highly fluctuating
ocean circulation and equilibrium.
Volcanic environment. It is also documented that equiva-
lent effects can be local variations close to volcanic vents,
where terrestrial macrofossil may be affected by magmatic
CO2 devoid of 14C (e.g., Pasquier-Cardin et al., 1999;
Cook et al., 2001a). Usually, such effects seem to be local-
ized and not be recognized around lakes under windy
influence (e.g., Hajdas et al., 1998).
Industrial modern environment. Industrial-scale
release of fossil fuel-derived CO2 can have a similar large
local effect, and this could be relevant for the use of radio-
carbon dating for historical (postindustrial revolution)
times.
Phytoliths
Phytoliths are increasingly used for isotopic studies
performed on organic matter embedded into the skele-
ton-building silicate (e.g., Hodson et al. (2008) and refer-
ences therein). Phytoliths trap trace amounts of organic
carbon (typically 0.1%wt) that is encased in silica
(Mulholland and Prior, 1993). This very small amount of
carbon available for a 14C measurement requires the
highest control of the blank quality (Santos et al., 2010).
In a recent study performed in highly controlled
conditions, Santos et al. (2010, 2012) showed that
phytolith carbon may being released from old soil organic
matter and might not be directly linked to the age of the
plant, but this is still under debate (Sullivan and Parr,
2013).
Pollen
Because many continental paleoclimatic studies are based
on pollen assemblages, it appears judicious to directly
base the chronology on pollen itself. However, very few
14
C dating attempts were performed on pollen grains
themselves (Long et al., 1992; Mensing and Southon,
1999), most are performed on pollen concentrate from
sediment (e.g., Brown et al., 1989; Kilian et al., 2002;
Piotrowska et al., 2004; Neulieb et al., 2013) and the
results are highly variable, from a clear rejuvenation, that
is, the age is too young, to some hundreds of years older,
or an age in agreement with larger identified macrofossil.
Hence, pollen dating is still in development.
Humid environments: lake, river, and peat
The main other types of carbon reservoirs, relevant to
radiocarbon dating, are lakes, rivers, and peats. Humid
environment sediments reflect a variety of different
deposits ranging on a scale from purely allochthonous to
purely autochthonous, minerogenic, and/or organic mate-
rial. They may contain, for example, precipitated and/or
in-washed mineral matter, terrestrial and aquatic plant
and animal remains, including algae, bacteria, and fungi,
as well as reworked older organic material. Such freshwa-
ter systems not only act as 14C reservoirs in their own
right and exchange CO2 with atmosphere but also
incorporate carbonate and organic carbon from surround-
ing catchments. This means that the radiocarbon concen-
tration can lie anywhere between the levels of the
atmosphere and those of the bedrock (zero for geological
carbonate).
Generalities on “freshwater reservoir effect” and
“hard water effect”
Aquatic cells photosynthesize subaquatically and hence
build carbon from dissolved inorganic carbon (DIC) into
their cellular material, so they reflect the 14C:12C ratios
of the water from which they grow. The DIC is influenced
by (1) exchange with the atmospheric CO2 reservoir,
(2) decomposition of organic matter, (3) residence time
of lake or peat bog water and in areas with calcareous bed-
rock and/or soils, and (4) dissolved carbonate from sur-
rounding limestone catchments. This means that the 14C
activity of DIC does not reflect the 14C activity of
atmosphere, but it is 14C depleted. This results in an
artificial ageing, the so-called freshwater reservoir effect
in soft water environments and the “hard water effect” in
calcareous areas, inherited by algae cells and which can
show wide variation (Macdonald et al., 1991; Fontana,
2005).
As example, in arctic oligotrophic lakes, it has been
suggested that 14C-depleted particulate and dissolved
organic carbon (POC and DOC) transported from soils
and peat in the watershed of the lakes may have a large
influence on the age of the surface sediments (e.g., Abbott
and Stafford, 1996) with reservoir ages of up to 1,000
years. In such extreme environments, with lakes of low
aquatic production, the allochthonous organic fraction
may make up a large part of the total organic matter and
may thus influence the composition of the sediments. If
the allochthonous fraction consists of old organic matter,
washed out from peats and soils, it will obviously increase
the age of the sediment. Such considerations may be very
important for paleolimnologic studies of lakes with low
productivity surrounded by a landscape rich in peat lands
and thick soils. Perennial lake ice cover, which seals off
the lake water from the atmosphere (e.g., Doran et al.,
1999) and/or inflow of glacial melt water (enrichment by
old CO2), may also act in favor of artificial ageing. How-
ever, there are some examples of peat bogs that do not
seem to be subjected to freshwater effect, such as the case
of lacustrine or palustrine environments developed on gla-
cial till (e.g., Hajdas et al., 1998; Walker et al., 2001;
Blaauw et al., 2004).
Furthermore, in fluvial systems, the actual radiocarbon
concentrations can vary seasonally and depend on details
of the recent weather conditions (e.g., Neff et al., 2006).
In larger lakes, the variation will generally be slower and
reflect long-term climatic trends.
 14
C IN PLANT MACROFOSSILS
Algae and aquatic moss macrofossils
As “freshwater reservoir effect” and “hard water effect”
vary according to environment specificities, their magni-
tudes affect the relevance of plant macrofossils for dating
that are recovered from sediments. Algal macrofossils
should thus be avoided if possible. If there is no other
choice, an evaluation of the reservoir age should be asso-
ciated with the measurements. This could be done by the
collection, in a “reference” level, of algae macrofossils
and subaerial plants. But once again, attention must be
brought to the choice of the reference level. Indeed, the
level chosen to determine the reservoir age should corre-
spond to an equivalent climatic context to the one to which
the correction will be done.
Relevance/match of some identified terrestrial plant
macrofossils
Based on the assumption that superior (terrestrial) plants
that photosynthesize from free gaseous CO2 instead of
DIC are less susceptible to contamination by inert
carbon than their aquatic counter parts, since they obtain
their carbon by subaerial photosynthesis, many scientists
have advocated the exclusive use of superior plant macro-
fossils for radiocarbon-dating lacustrine and palustrine
sediment sequences. However, they also are not free of
problems.
Potamogeton seeds and other floating superior plant
macrofossils
Floating superior plants photosynthesize from the CO2
available at the water surface that mostly results from
a mixture of free atmospheric CO2 and degassing DIC.
Indeed, a high content of mineral carbon in water can
result from strong limestone catchment erosion or from
high organic matter mineralization, such as is commonly
observed following a production bloom that may result
in DIC degassing. Consequently, floating superior plants
inherit part of the DIC 12C:14C composition, and even if
diluted, the “freshwater effect” affects the 14C activity of
Potamogeton seeds. The conversion of this activity into
a date would induce significant ageing (i.e., the sample
appears too old) (e.g., Hatté et al., 2013) or even rejuvena-
tion (too young) in the specific case of modern environ-
ments that absorbed high “bomb peak” 14C. Indeed, the
DIC may thus be imprinted by peak bomb organic carbon
and then the floating plants appear 14C enriched (e.g.,
Olsson and Kaup, 2001).
There are some examples of peat bogs where DIC
degassing is not significant, and dating obtained on
Potamogeton seeds gives a reliable chronological frame-
work (e.g., Walker et al., 2001). Nevertheless, since this
is linked to water organic production and thus to seasonal
cycle, the degassing can fluctuate greatly, and therefore,
a quantitative evaluation is difficult. It remains hazardous
to work on this kind of macrofossil. Therefore, it is better
to avoid the use of floating plant macrofossils for dating as
much as possible.
129
Sphagnum, terrestrial mosses, and smaller-sized superior
plant macrofossils
As ombotrophic peat deposits are very common and
sphagnum-formed, sphagnum macrofossils are often
used for 14C dating (Kilian et al., 1995). These superior
plants do not grow in the water context and are not
subjected to a “hard water effect” or DIC degassing influ-
ence; however, they are not free of old carbon effects.
Indeed, palustrine areas may be highly organic, especially
in ombotrophic bogs, and then very efficient environ-
ments for organic matter mineralization. As this minerali-
zation can be from recent as well as old organic matter, the
resulting mineralized CO2 emitted by the “soil” shows an
old apparent 14C age. Thus, at the ground surface, CO2
results in a mixture between free atmospheric CO2 and
mineralized organic matter CO2. At a lower scale, this
effect can be linked to the well-known canopy effect in
a tropical forest. Close to the surface, the smaller-sized
superior plants like sphagnum and mosses may thus show
14
C ageing (Jungner et al., 1995). Some mosses are
calciphilous and thus may incorporate limestone as
a carbon source; this effect may result in an ageing (Zazula
et al., 2006). These mosses may not constitute the most
reliable support to establish a chronology. Considering
the uncertainty ranges associated with 14C dating, this age-
ing can sometimes not be significant, and chronologies
based on 14C on sphagnum can be reliable. But with the
increasing precision of 14C dating, the ageing risk linked
to organic matter mineralization has to be taken into
account.
Tall superior plant macrofossils
Finally, larger superior plant macrofossils seem to be the
most suitable macrofossils for 14C dating. A plethora of
such macrofossils can be found in peat and lake sediment.
We can list needles of Pinus, Larix, and Picea; seeds of
Betula, Carex, and Eleocharis; and leaves of Salix, Dryas,
Erica, and Calluna as the most common.
On account of their high resistance to diagenesis, Carex
seeds were intensively used in support of chronological
frameworks. However, this effective resistance does not
permit verification of the state of preservation, and so
some possibly reworked or redeposited seeds might not
be recognized from indices on macrofossil shape and
entirety. This was the origin of some inconsistent chronol-
ogies based on macrofossils of several centimeter tall
superior plants (Turney et al., 2000). We would recom-
mend the collection of “fragile” macrofossils from
which it is easy to check that they are unlikely to have
been physically transferred within the sediment matrix
and/or otherwise reworked from the surrounding catch-
ment and/or reworked by fauna and/or the rhizosphere.
For example, intact Betula seeds surrounded by a fragile
voile or survival of the fragile Salix leaves can be ideal
candidates for 14C dating to establish a chronological
framework.
 14
130 C IN PLANT MACROFOSSILS
Insects, amphibians, and invertebrates
Beside plant macrofossils, insects are common and often
abundant in a range of sediment types. Examination of
their utility as dating materials has focused on bark beetles
and chironomids. Beetles sclerites (Elias et al., 1991;
Porch and Kershaw, 2010) and chironomid head capsules
(Jones et al., 1993; Fallu et al., 2004) extracted from lake
sediments have shown to provide reliable radiocarbon
dates. However, it should be remembered that some ani-
mals such as detritus feeding beetles will ingest carbon
that might have been photosynthesized some years to hun-
dreds of years before and are part of soil organic matter.
They are thus 14C depleted relatively to atmosphere and
will result in older than expected 14C ages. In lakes
affected by erosion of surrounding catchment, these bury-
ing beetle remains can become part of lake sediment and
thus also yield for 14C inversion. Finally, aquatic beetle
species that feed with algae mimic the addition of old car-
bon to their food.
Land snails might also exhibit a similar (even more
extreme) radiocarbon signature. They can use calcium
from limestone as major component of their shells and
incorporate at the same time the limestone carbon that,
mixed with ingested plant carbon, might result to very
high ageing (up to 2,400 years) especially in arid and
semiarid regions (Goodfriend et al., 1999). Furthermore,
snail shells are quite soluble and chemically interact with
the environment. Dissolution and recrystallization are
highly probable and would induce apparently rejuvenated
radiocarbon age. Humid phases that followed a dry fossil-
ization inevitably imply carbonate dissolution and recrys-
tallization. Dating land snail carbonate should be avoided.
Some animals spend time in different parts of the bio-
sphere. That is the case of amphibians that live in atmo-
sphere and water and of some fishes as salmons and eels
that spend part of their life in ocean and then freshwater.
They thus get their food from a whole variety of sources.
This means that the carbon they incorporate will ulti-
mately come from different primary reservoirs and that
animals integrate a large variety of 14C signatures (e.g.,
Cook et al., 2001b). Disentangling the resulting 14C signal
is impossible and this kind of support should be avoided.
Sample fossilization
Once an organism dies, its constituent parts are constantly
being broken and reformed, and usually little remains after
a few years. However, in some conditions, most notably
waterlogged deposits and very dry conditions, there is bet-
ter preservation. It is important to remember that this pres-
ervation is exceptional and therefore that in most cases,
the excavated sample composition is not representative
of its original composition.
An important factor to consider is mechanical distur-
bance and downward movement through bioturbation by
roots, burrowing animals and insects, and geological
instabilities. The reworking risk can be evaluated by ani-
mal and insect identification and by a detailed
sedimentological study. A reliable chronological analysis
cannot be carried out alone and has to be jointly conducted
in a collaborative mind.
The organic materials that survive intact for longest
with their original complement of carbon atoms are nor-
mally the polymers. The carbon from organisms that
degrade is not always lost, but can often present in form
of soluble molecules such as dissolved organic carbon or
the well-known but poorly characterized humic and fulvic
acids. Such components can be recovered from sediment
but are highly mobile and cannot however be related to
a single organism or to a well-defined photosynthesis
event. Even when carbon remains in some stable form
(non mobile form), there will usually be many other more
mobile carbon-containing compounds and other contami-
nants that become incorporated into the material remains
of an organism. These contaminants will have different
origins and so will not share the same radiocarbon signal.
Removing these contaminants will be a key step in the lab
processing.
In the lab
Sample preservation
The core storage and the preparation and identification of
small macrofossil samples are usually carried out in
a nonsterile environment; however Wohlfarth et al.
(1998) have shown that the long-term storage of wet mac-
rofossil samples appears to have a significant effect on
radiocarbon age, even when samples are kept cool. Fungal
spores and microorganisms can easily be incorporated into
the sample. As they might use ambient water carbon as
source of carbon besides from fossil leaf materials, they
therefore contaminate the sample. Similar findings were
previously evidenced for marine cores preserved at 4
C
that were contaminated by recent carbon from modern,
terrestrial bacteria (Geyh et al., 1974; Colman et al.,
1996). Wet storage of the selected macrofossils should
be avoided in order to prevent bacterial/fungal activity
and the sample should be either dried at 100–110
C as
quick as possible (Björk and Wohlfarth, 2001) or dried
under vacuum at lower temperature (60
C) (Gauthier
and Hatté, 2008). The choice between these procedures
should be done according to the sample sturdiness.
Dry samples can then be stored in a clean (850
C pre-
combusted) glass bottle or on clean (400
C pre-
combusted) aluminum foil.
Sample size
Very small samples are likely to be prone to contamina-
tions, especially lab contamination and allochthonous
input. To avoid this high risk, large amount of macrofos-
sils should be sampled if possible. For example, Bronk
Ramsey (2008) weighed up a rejuvenation of 840 years
consecutively to adjunction of 1%wt of modern contami-
nation in a 20 000-year sample. Such a high rejuvenation
that results in only 1 mg contaminating C in a 100 mg C
 14
C IN PLANT MACROFOSSILS
sample has to be put in regard to the weight of a 1 cm long
hair (# 50 mg) and of a pullover fiber (# 15 mg).
Furthermore, a larger than expected scatter in measured
values for subsamples lower than 0.3 mg C was reported
(Stuiver and Pearson, 1993; Oswald et al., 2005). Two fac-
tors contribute likely to these findings: the uncertain com-
pleteness of the graphitization reaction and the increasing
impact on contaminant carbon background (Brown and
Southon, 1997). An incomplete graphitization reaction
yields for 14C fractionation up to 60% that could be
interpreted in case of decoupled d13C/14C measurements
(d13C measured on gas sample, prior graphitization) as
an apparent older age up to 500 years.
The best compromise between the lab risk and the error
inherent to macrofossil nature (e.g., linked to freshwater or
burrowing effect) should be found. What is the best
between an isolated Betula seed of few mg and beetles
sclerites not perfectly identified in high quantity? That will
always remain the question.
Conclusion
It is impossible to suggest a universal strategy for
obtaining reliable radiocarbon dates, because each
approach has to be adapted to the type of soil, lake, peat
bog, and sediment material under investigation, as well
as to the objective of the study. However, before deciding
on the sampling and dating strategy, the following consid-
erations should be made:
1. What are the scientific objectives of the project in terms
of dating?
2. Which type of organic material is available and in
which quantity?
3. How will the present and the past environment and cli-
mate influence the radiocarbon measurement?
These questions are an important precondition to high-
light radiocarbon issues such as freshwater effect and its
variability within the study period, recrystallization risk,
and multiple 14C reservoirs, etc., and finally to adapt the
sampling to the chronological framework one needs to
obtain.
Bibliography
Abbott, M. B., and Stafford, T. W. J., 1996. Radiocarbon geochem-
istry of modern and ancient Arctic lake systems, Baffin Island,
Canada. Quaternary Research, 45, 300–311.
Björk, S., and Wohlfarth, B., 2001. 14C chronostratigraphic tech-
niques in paleolimnology. In Last, W. M., and Smol, J. P.
(eds.), Tracking Environmental Change Using Lake Sediments.
Dordrecht: Kluwer. Basin Analysis, Coring and Chronological
Techniques, Vol. 1.
Blaauw, M., van der Plicht, J., and van Geel, B., 2004. Radiocarbon
dating of bulk peat samples from raised bogs: non-existence of
a previously reported “reservoir effect”? Quaternary Science
Reviews, 23, 1537–1542.
Bronk, R. C., 2008. Radiocarbon dating: revolutions in understand-
ing. Archaeometry, 50, 249–275.
Brown, T. A., and Southon, J. R., 1997. Corrections for contamina-
tion background in AMS 14C measurements. Nuclear
131
Instruments & Methods In Physics Research Section B-Beam
Interactions With Materials and Atoms, 123, 208–213.
Brown, T. A., Nelson, D. E., Mathewes, R. W., Vogel, J. S., and
Southon, J. R., 1989. Radiocarbon dating of pollen by accelera-
tor mass spectrometry. Quaternary Research, 32, 205–212.
Colman, S. M., Jones, G. A., Rubin, M., King, J. W., Peck, J. A., and
Orem, W. H., 1996. AMS radiocarbon analyses from Lake
Baikal, Siberia: challenges of dating sediments from a large, oli-
gotrophic lake. Quaternary Science Reviews, 15, 669–684.
Cook, A. C., Hainsworth, L. J., Sorey, M. L., Evans, W. C., and
Southon, J. R., 2001a. Radiocarbon studies of plant leaves and
tree rings from Mammoth Mountain, CA: a long-term record of
magmatic CO2 release. Chemical Geology (Isotope Geoscience
Section), 177, 117–131.
Cook, G. T., Bonsall, C., Hedges, R. E. M., McSweeney, K.,
Boronean, V., and Pettin, P. B., 2001b. A freshwater diet-derived
14C reservoir effect at the Stone Age sites in the Iron Gates
gorge. Radiocarbon, 43, 453–460.
Doran, P. T., Berger, G. W., Lyons, W. B., Wharton, R. A., Jr.,
Davisson, M. L., Southon, J., and Dibb, J. E., 1999. Dating Qua-
ternary lacustrine sediments in the McMurdo Dry Valleys, Ant-
arctica. Palaeogeography Palaeoclimatology Palaeoecology,
147, 223–239.
Elias, S. A., Carrara, P. E., Toolin, L. J., and Jull, A. J. T., 1991.
Revised age of deglaciation of Lake Emma based on new radio-
carbon and macrofossil analyses. Quaternary Research, 36,
307–321.
Fallu, M.-A., Pietniz, R., Walker, I. R., and Overpeck, J. T., 2004.
AMS 14Cdating of tundra lake sediments using chironomid
head capsules. Journal of Paleolimnology, 31, 11–22.
Fontana, S. L., 2005. Holocene vegetation history and palaeoenvir-
onmental conditions on the temperate Atlantic coast of Argen-
tina, as inferred from multi-proxy lacustrine records. Journal of
Paleolimnology, 34, 445–469.
Gauthier, C., and Hatté, C., 2008. Effects of handling, storage, and
chemical treatments on delta C-13 values of terrestrial fossil
organic matter. Geophysics, Geochemistry and Geosystem, 9.
doi:10.1029/2008gc001967.
Geyh, M. A., Krumhein, W. E., and Kudrass, H. R., 1974.
Unreliable 14C dating of long-stored deep-sea sediment due to
bacterial activity. Marine Geology, 17, M45–M50.
Goodfriend, G. A., Lain Ellis, G., and Toolin, L., 1999. Radiocar-
bon age anomalies in land snail shells from Texas: ontogenetic,
individual, and geographic patterns of variation. Radiocarbon,
41, 149–156.
Hajdas, I., Bonani, G., Zolitschka, B., Brauer, A., and Negendank,
J. F. W., 1998. C-14 ages of terrestrial macrofossils from Lago
Grande di Monticchio (Italy). Radiocarbon, 40, 803–807.
Hatté, C., Bréheret, J.-G., Jacob, J., Argant, J., and Macaire, J.-J.,
2013. Refining the Sarliève paleolake (France) Neolithic chro-
nology by combining several radiocarbon approaches. Radio-
carbon, 55(2–3), 979–992.
Hodson, M. J., Parker, A. G., Leng, M. J., and Sloane, H. J., 2008.
Silicon, oxygen and carbon isotope composition of wheat
(Triticum aestivum L.) phytoliths: implications for
palaeoecology and archaeology. Journal of Quaternary Science,
23, 331–339.
Jones, V. J., Battarbee, R. W., and Hedges, R. E. M., 1993. The use
of chironomid remains for AMS 14C dating of lake sediments.
The Holocene, 3, 161–163.
Jungner, H., Sonninen, E., Possnert, G., and Tolonen, K., 1995. Use
of bomb-produced 14C to evaluate the amount of CO2 emanat-
ing from two peat-bogs in Finland. Radiocarbon, 37, 567–573.
Kilian, M. R., Van der Pflicht, J., and van Geel, B., 1995. Dating
raised bogs: new aspects of AMS 14C wiggle matching,
a reservoir effect and climatic change. Quaternary Science
Reviews, 14, 959–966.
132 CARBONATES, LACUSTRINE (U-SERIES)
Kilian, M. R., Van der Pflicht, J., van Geel, B., and Goslar, T., 2002.
Problematic 14C-AMS dates of pollen concentrates from Lake
Gosciaz (Poland). Quaternary International, 88, 21–26.
Kuzucuoglu, C., Pastre, J. F., Black, S., Ercan, T., Fontugne, M.,
Guillou, H., Hatté, C., Karabiyikoglu, M., Orth, P., and
Turkecan, A., 1998. Identification and dating of tephra layers
from Quaternary sedimentary sequences of Inner Anatolia, Tur-
key. Journal of Volcanology and Geothermal Research, 85,
153–172.
Long, A., Davis, O. K., and Delanois, J., 1992. Separation and C-14
dating of pure pollen from lake-sediments – nanofossil AMS
dating. Radiocarbon, 34, 557–560.
MacDonald, G. M., Beukens, R. P., and Kieser, W. E., 1991. Radio-
carbon dating of limnic sediments: a comparative analysis and
discussion. Ecology, 72, 1150–1155.
Mensing, S., and Southon, J. R., 1999. A simple method to separate
pollen for AMS radiocarbon dating and its application to lacus-
trine and marine sediments. Radiocarbon, 41, 1.
Mulholland, S. C., and Prior, C. A., 1993. AMS radiocarbon dating
of phytolith. In Pearsall, D. M. (ed.), Current Research in Phyto-
lith Analysis: Applications in Archeology and Paleoecology.
MASCA: Research Papers in Science and Archeology Univer-
sity of Pennsylvania Museum.
Nakamura, T., Miyahara, H., Masuda, K., Menjo, H., Kuwana, K.,
Kimura, K., Okuno, M., Minami, M., Oda, H., and Rakowski,
A., 2007. High precision 14C measurements and wiggle-match
dating of tree-rings at Nagoya University. Nuclear Instruments
& Methods In Physics Research Section B-Beam Interactions
With Materials and Atoms, 259, 408–413.
Neff, J. C., Finlay, J. C., Zimov, S. A., Davydov, S. P., Carrasco,
J. J., Schuur, E. A. G., and Davydova, A. I., 2006. Seasonal
changes in the age and structure of dissolved organic carbon in
Siberian rivers and streams. Geophysical Research Letters, 33,
L23401.
Neulieb, T., Levac, E., Southon, J., Lewis, M., Pendea, I. F.,
Chmura, G., 2013. Potential pitfalls of pollen dating. Radiocar-
bon, 55(4), 1142–1155.
Olsson, I. U., and Kaup, E., 2001. The varying radiocarbon activity
of some recent submerged estonian plants grown in the early
1990s. Radiocarbon, 43, 809–820.
Oswald, W. W., Anderson, P. M., Brown, T. A., Brubaker, L. B., Hu,
F. S., Lozhkin, A. V., Tinner, W., and Kaltenrieder, P., 2005.
Effects of sample mass and macrofossil type on radiocarbon dat-
ing of arctic and boreal lake sediments. The Holocene, 15,
758–767.
Pasquier-Cardin, A., Allard, P., Ferreira, T., Hatté, C., Coutinho, R.,
Fontugne, M. R., and Jaudon, M., 1999. Magma-derived CO2
emissions recorded in 14C and 13C content of plants growing
in Furnas Caldera, Azores. Journal of Volcanology and Geother-
mal Research, 92, 195–207.
Piotrowska, N., Bluszcz, A., Demske, D., Granoszewski, W., and
Heumann, G., 2004. Extraction and AMS radiocarbon dating
of pollen from Lake Baikal sediments. Radiocarbon, 46,
181–187.
Porch, Nick and Kershaw, A. Peter 2010, Comparative AMS 14C
dating of plant macrofossils, beetles and pollen preparations
from two late pleistocene sites in southeastern Australia, in
Altered ecologies (Terra Australis 32): fire, climate and human
influence on terrestrial landscapes, ANU E Press, Canberra, A.
C.T., pp. 395–404.
Saliège, J.-F., Zazzo, A., Hatté, C., and Gauthier, C., 2012. Radio-
carbon dating in Petra: limitations and potential in semi-arid
environments. In Mouton, M., and Schmid, S. (eds.), Men on
the Rocks – The Formation of Nabataean Petra. Berlin: Logos
Verlag.
Santos, G. M., Alexandre, A., Coe, H. H. G., Reyerson, P. E.,
Southon, J. R., and De Carvalho, C. N., 2010. The phytolith
14C puzzle: a tale of background determinations and accuracy
tests. Radiocarbon, 52, 113–128.
Santos, G. M., Alexandre, A., Southon, J. R., Treseder, K. K.,
Corbineau, R., and Reyerson, P. E., 2012. Possible source of
ancient carbon in phytolith concentrates from harvested grasses.
Biogeosciences, 9, 1873–1884.
Stuiver, M., and Pearson, G. W., 1993. High-precision decadal cal-
ibration of the radiocarbon time scale, AD 1950–500 BC and
2500–6000 BC. Radiocarbon, 35(1), 1–23.
Sullivan, L. A., and Parr, J. F., 2013. Comment on “Possible source
of ancient carbon in phytolith concentrates from harvested
grasses” by G.M. Santos et al. (2012). Biogeosciences, 10,
977–980.
Turney, C. S. M., Coope, G. R., Harkness, D. D., Lowe, J. J., and
Walker, D. A., 2000. Implications for the dating of Wisconsinan
(Weichselian) Late-Glacial events of systematic radiocarbon age
differences between plant macrofossils from a site in SW Ireland.
Quaternary Research, 53, 114–121.
Walker, M. J. C., Bryant, C. L., Coope, G. R., Harkness, D. D.,
Lowe, J. J., and Scott, E. M., 2001. Towards a radiocarbon chro-
nology of the late-glacial: sample selection strategies. Radiocar-
bon, 43, 1007–1019.
Wohlfarth, B., Skog, G,. Possnert, G., and Holmqvist, B.H., 1998.
Pitfalls in the AMS radiocarbon-dating of terrestrial macrofos-
sils. Journal of Quaternary Science, 13, 137–145.
Zazula, G. D., Schweger, C. E., Beaudouin, A. B., and McCourt,
G. H., 2006. Macrofossil and pollen evidence for full-glacial
steppe within an ecological mosaic along the Bluefish River,
eastern Beringia. Quaternary International, 142–143, 2–19.
Cross-references
Lacustrine Environments (14C)
Peat (14C)
Radiocarbon Dating
CARBONATES, LACUSTRINE (U-SERIES)
Mordechai Stein
Geological Survey of Israel, Jerusalem, Israel
Synonyms
Aragonitic lake sediments; Calcitic lake sediments; Lake
carbonates
Definition
Method of dating late Quaternary geological events
(typically up to ~600 ka) by U and Th isotopes in primary
calcium carbonates (e.g., aragonitic corals, calcitic
speleothems, aragonitic or calcitic lake sediments). The
method is based on the radioactive decay of 238U to its
daughters 234U and 230Th. The radiometric “clock” mea-
sures the time that is required for a particular carbonate
system to reach the condition of secular equilibrium after
chemical fractionation between soluble U and insoluble
Th in waters (e.g., seawater). The isotopes of U and Th
can be measured by a-counting or mass spectrometric
methods.
 CARBONATES, LACUSTRINE (U-SERIES)
Introduction
The U-series dating method is the principal means to
establish precise chronologies of late Quaternary geologi-
cal events by analyses of primary carbonates, those that
precipitate chemically from marine and continental water
bodies: corals, speleothems, lacustrine sediments, and
travertines. Examples are chronologies of sea level
changes, coastal tectonic movements, and coastal hydrol-
ogy (e.g., Edwards et al., 1986; Stein et al., 1993; Lazar
and Stein, 2011), as well as patterns of continental climate
changes by dating primary lacustrine carbonates or cave
speleothems (cf. Kaufman, 1971; 1988; Winograd
et al., 1992; Kaufman et al., 1998; Schramm et al., 2000;
Haase-Schramm et al., 2004; McGee et al., 2012;
Torfstein et al., 2013a; Torfstein et al., 2013b). The possi-
bility of obtaining precise U–Th ages on corals stems from
the nearly complete chemical separation between
U (soluble) and Th (insoluble) in seawater, which results
in “clean” precipitation of U within the coral aragonitic
skeleton (no initial 230Th and negligible amount of detri-
tus). Lake carbonates, however, typically contain signifi-
cant amounts of non-authigenic U and Th, and possibly
authigenic (hydrogenous) Th, which must be considered
in age evaluations (cf. Ku and Liang, 1984; Luo and Ku,
1991; Kaufman, 1993; Lin et al., 1996).
U–Th dating of “pure” and “dirty” carbonates
The 234U–230Th dating method is based on the deviation
of the intermediate daughter isotopes in the 238U decay
series from the state of secular equilibrium. The initial state
of the disequilibrium condition is produced when U and Th
fractionate in a hydrologic environment such as seawater or
lake water. This fractionation reflects the distinctly different
chemical affinity of U and Th. While U forms soluble com-
plexes in water with various salinities and pH values, Th
typically is removed by absorption onto surfaces of sedi-
mentary particles. Thus, the [230Th/238U] activity ratio
approaches zero in calcium carbonate (CaCO3) precipitated
from the water. Furthermore, the carbonate incorporates
U according to the specific partition coefficient that
depends on the type of the CaCO3 mineral phase (e.g., ara-
gonite or calcite) and environmental factors such as water
temperature and pH (cf. Lazar et al., 2004).
The condition of secular equilibrium in the decay series
238
U!234 U!230 Th ! . . . !206 Pb
is achieved because the half-life of the parent isotope 238U
is much longer than the intermediate daughter isotopes in
the series. For the pair 234U–230Th, the condition of secu-
lar equilibrium is typically achieved after ~500 ka. The
234
U–230Th “clock” measures the time elapsed from the
incorporation of U within the CaCO3 and the present day
(provided that it is less than the time required to achieve
secular equilibrium). The incorporation of U in the carbon-
ate mineral phases follows the chemical separation between
U and Th in seawater or lake water that leaves the U in solu-
tion (as uranyl ion complexes) and removes Th attached to
133
settling particles. Assuming that the precipitated carbonate
remained closed with regard to U and Th mobility and that
the initial 230Th/238U is negligible (this may not be the case
in system of “dirty carbonates” an issue elaborated below),
the equations governing the decay of 238U and 230Th ages
that allow the calculation of single ages are (modified after
Kaufman and Broecker, 1965):
h i
230
Th=238 U act ¼ 1  expðl230 TÞ
þ d234 Uð0Þ103  ½l230 =ðl230  l234 Þ
 ½1  expððl234  l230 ÞTÞ
ð1Þ
d234 UðTÞ ¼ d234 Uð0Þ expðl234 TÞ ð2Þ
where T represents the age, in years, since mineral forma-
tion, [230Th/238U]act represents the ratio of 230Th and 238U
activities, and l230 and l234 represent radioactive decay
constants in years1 for 230Th and 234U, respectively.
d234U(0) is defined as the measured activity value
[234U/238U]act recalculated relative to the secular equilib-
rium value of 1.00 and given in per mil units (%; i.e.,)
d234U ¼ ([234U/238U]actsample/1.00) – 1)  1,000). Simi-
larly, d234U(T) is the initial value of [234U/238U]act calcu-
lated at the time of isolation of the specific mineral system
(e.g., precipitation of lacustrine aragonite or formation of
aragonitic coral, or calcitic speleothem, or travertine).
In general, we expect that, for samples representing
a simple closed-system evolution, the initial d234U(T)
value should reflect the value present in the ambient
hydrologic system. Thus corals deposited in seawater
will have d234U(T) of 150 (Edwards et al., 1986; Chen
et al., 1991; Stein et al., 1993), whereas primary
aragonite deposited from the hypersaline lakes in the Dead
Sea basin will have d234U(T) of 500 (Haase-Schramm
et al., 2004; Torfstein et al., 2013b). Decay constants used
in the calculation of the ages are l230 ¼ 9.158  106;
l232 ¼ 4.9475  1011, l234 ¼ 2.8263  106, and
l238 ¼ 1.55125  1010 per year (Le Roux and
Glendenin, 1963; Jaffey et al., 1971; Cheng et al., 2000).
The basic assumptions behind Eq. (1) are No initial
230
Th is present in the sample and all of the U is derived
from the ambient water in the depositional habitat. These
assumptions hold well for corals, which typically contain
negligible amounts of detrital Th (typically in the range
of 0.3–1 pmol/g e.g., Chen et al., 1991; Stein et al., 1993),
but rarely holds for other carbonates. Possible interfer-
ences in obtaining a true age from measured isotope ratios
are U and Th contributed from admixed detritus present in
the sample and the possible presence of initial Th in the
authigenic (or primary) component.
Correction for the “detrital” U and Th by the
“isochron” method
Correction of the U and Th concentrations in “dirty car-
bonates” is often done by plotting the U and Th activity
134 CARBONATES, LACUSTRINE (U-SERIES)
ratios of a set of coeval samples containing various
amounts of detritus on the Rosholt- or Osmond-type “iso-
chron” diagrams. These diagrams are used to identify the
[230Th/238U] and [234U/238U] activities of the “pure pri-
mary carbonate.” The [230Th/238U] and [234U/238U] activ-
ity ratios of the “detritus-free carbonate” is retrieved
from the slopes of “Rosholt”-type diagrams (plots
of 230Th/232Th vs. 238U/232Th and 234U/232Th vs.
238 232
U/ Th activity ratios, respectively) or the
y-intercepts of “Osmond”-type diagrams (plots of
232
Th/238U vs. 230Th/238U and 232Th/238U vs. 234U/238U
activity ratios, respectively). Alternatively, a “single sam-
ple” correction method subtracts the detrital U and Th con-
tributions using either an assumed or measured isotopic
composition from individual sample data. The “isochron”
approaches have advantages over the single sample cor-
rection method when there is a large variation in the
mixing proportions of coeval detritus and carbonate sam-
ples (cf. Bischoff and Fitzpatrick, 1991; Kaufman, 1993;
Ludwig and Titterington, 1994; Haase-Schramm
et al., 2004; Torfstein et al., 2013a). Such a set of samples
may display a large range of values along a line in “iso-
chron” diagrams and the composition of the detrital
end-member may not have to be explicitly assumed.
Correction for initial hydrogenous Th
Correction for initial hydrogenous Th in the primary car-
bonate phase cannot be done by the isochron method. This
component might be determined empirically by using
a sample of known age determined by independent means
(e.g., by 14C ages of organic material). Any apparent dif-
ference in the U–Th age could be attributed to the pres-
ence of initial 230Th (after correcting for the detritus
component). In this case, the extra initial 230Th is
corrected by subtracting this component from the
detritus-corrected values. Alternatively, the fraction of
the hydrogenous Th can be retrieved from analyses of
trace element chemistry of specific samples (e.g., Th/Zr
ratios, as detailed in Haase-Schramm et al. (2004)).
The following sections describe two case studies where
U–Th chronologies were established for lacustrine
sequences: the last glacial Lake Lisan in the Levant-Dead
Sea basin and Lake Bonneville in the western US Great
Basin. Other examples of lacustrine deposits that were
dated by the U–Th are described by Israelson et al. (1997),
Ku et al. (1998), and Lin et al. (1996), among others.
U-series dating of Lake Lisan (last glacial Dead Sea)
Lake Lisan occupied the tectonic depressions along the
Dead Sea Transform fault in the Levant region of the east-
ern Mediterranean during the last glacial period. The lake
comprises Ca-chloride brine that requires the supply of
bicarbonate ions to precipitate primary carbonate. This is
furnished by the annual freshwater input. Turbulent
mixing across the epilimnion (upper water layer) and
hypolimnion resulted in precipitation of primary aragonite
(Stein et al., 1997). The excellent preservation and high
U concentrations (~3 ppm) of the aragonite make it useful
for 234U–230Th dating. The pioneering attempts in apply-
ing this method to Lake Lisan sediments were made by
Kaufman and colleagues using a-counting techniques
(Kaufman, 1971; Kaufman et al., 1992). They showed
that, unlike coral samples, the Lisan aragonites are associ-
ated with a detrital U and Th component that should be
corrected out prior to calculation of 234U–230Th and
applied the “isochron” methods to perform these correc-
tions. With the development of the TIMS method for ana-
lyzing U and Th isotopes in small carbonate samples,
Schramm et al. (2000) and Haase-Schramm et al. (2004)
measured U and Th abundances in a stratigraphic section
of the Lisan Formation (PZ1 section at Perazim Valley),
corrected for detritus U and Th and hydrogenic Th and
established an age-height chronological model for the
time interval of 70–14 ka BP. Recently, Torfstein
et al. (2013a) used the plasma machine (ICP-MS-MC) to
analyze the U and Th isotope in several stratigraphic sec-
tions of the Lisan Formation in the Dead Sea-Jordan Val-
ley and established an integrated multi-site U–Th
chronology for the Lisan Formation. The U–Th Lisan
chronology has been used for paleoclimate and
paleoseismic studies and for calibrating the radiocarbon
time scale in the last glacial time interval (Bartov
et al., 2003; Haase-Schramm et al., 2004; Kolodny
et al., 2005; Torfstein et al., 2013a; Torfstein et al., 2013b).
U-series dating of Lake Bonneville
Lake Bonneville, which occupied the Bonneville Basin of
the northeastern Great Basin (USA), experienced a lake
level history very similar to the Levant-Lake Lisan,
approaching its highest stands during the last glacial
~26 ka BP and commenced its deglacial retreat after
~16 ka BP (Oviatt et al., 1990, 1992). Nevertheless, the
reconstruction of lake paleohydrology and
paleolimnology suffered from lack of accurate chronolo-
gies (e.g., uncertainties in the radiocarbon reservoir ages).
Recently, new chronological constraints on the glacial and
postglacial history of the lake were achieved by U–Th dat-
ing of lacustrine cave carbonates (McGee et al., 2012).
The cave carbonates provide a promising new archive of
past hydrologic changes in the Bonneville Basin. The car-
bonates precipitated within caves (e.g., Cathedral and
Craners caves), crevices, and other protected spaces
flooded by Lake Bonneville during its high stand in the
last glacial period. The caves are located at similar eleva-
tions approximately 100 m above the modern Great Salt
Lake and almost 200 m below Lake Bonneville’s
highstand shoreline. Precise U–Th ages determined on
the cave carbonates indicate a minimal (~200 y) radiocar-
bon reservoir age in the lake and allows 14C dating to be
used for age control in portions of the deposits less suit-
able for U–Th dating. The combined U–Th and radiocar-
bon data document the timing of the lake’s transgression
between 26 and 18 ka and a large influx of fresh water dur-
ing Heinrich Stadial 2. This was reflected by a hiatus in
 CARBONATES, LACUSTRINE (U-SERIES)
calcite deposition. Calcite deposition resumed at ~16.4 ka,
suggesting that basin overflow had ceased by this time.
The lake’s deglacial regression began ~16 ka. Cessation
of this second phase of deposition at 14.7  0.2 ka, coinci-
dent with the Bølling-Allerod warming, may reflect the
lake’s drop below Cathedral Cave’s elevation. Overall,
the Great Basin (USA) Bonneville lacustrine system and
the Dead Sea-Lake Lisan (Levant) show very similar
responds to the global climate, which demonstrates the
great potential in performing precise U–Th dating on the
lacustrine carbonate deposits.
Conclusions
U–Th dating of lacustrine (authigenic or primary) calcium
carbonates is a prime method of achieving accurate and
high-resolution chronologies for the geochemical-
limnological evolution of late Pleistocene lakes. However,
the lake carbonates are typically “contaminated” by detri-
tal U and Th and initial hydrogenous Th. Application of
the “isochron methods” (e.g., Rosholt- or Osmond-type
diagrams), combined with stratigraphical considerations,
independent dating methods, such as radiocarbon over
the past 50 ka, and oxygen isotope chronologies, helps
to derive reliable and high-resolution chronologies for
the lacustrine sequences.
Bibliography
Bartov, Y., Goldstein, S. L., Stein, M., and Enzel, Y., 2003. Cata-
strophic arid episodes in the Eastern Mediterranean linked with
the North Atlantic Heinrich events. Geology, 31, 439–442.
Bischoff, J. L., and Fitzpatrick, J. A., 1991. U-series dating of
impure carbonates: an isochron technique using total-sample dis-
solution. Geochimica et Cosmochimica Acta, 55, 543–554.
Chen, J. H., Curran, H. A., White, B., and Wasserburg, G. J., 1991.
Precise chronology of the last interglacial period: 234U-230Th
data from fossil coral reefs in the Bahamas. Geological Society
of America Bulletin, 103, 82–97.
Cheng, H., Edwards, R. L., Hoff, J., Gallup, C. D., Richards, D. A.,
and Asmerom, Y., 2000. The half-lives of uranium-234 and
thorium-230. Chemical Geology, 169, 17–33.
Edwards, R. L., Chen, J. H., and Wasserburg, G. J., 1986.
238
U-234U-230Th-232Th systematics and the precise measurement
of time over the past 500,000 years. Earth and Planetary Science
Letters, 81, 175–192, doi:10.1016/0012 -821X(87)90154-3.
Haase-Schramm, A., Goldstein, S. L., and Stein, M., 2004. U-Th
dating of Lake Lisan (late Pleistocene Dead Sea) aragonite and
implications for glacial east Mediterranean climate change.
Geochimica et Cosmochimica Acta, 68, 985–1005.
Israelson, C., Björck, S., Hawkesworth, C. J., and Possnert, G.,
1997. Direct U-Th dating of organic- and carbonate-rich lake
sediments from southern Scandinavia. Earth and Planetary Sci-
ence Letters, 153, 251–263.
Jaffey, A. H., Flynn, K. F., Glendenin, L. E., Bentley, W. C., and
Essling, A. M., 1971. Precision measurement of half-lives and
specific activities of 235U and 238U. Physical Review C, 4,
1889–1906.
Kaufman, A., 1971. U-Series dating of Dead Sea Basin carbonates.
Geochimica et Cosmochimica Acta, 35, 1269–1281.
Kaufman, A., 1993. An evaluation of several methods for determin-
ing ages in impure carbonates. Geochimica et Cosmochimica
Acta, 57, 2303–2317.
135
Kaufman, A., and Broecker, W. S., 1965. Comparison of 230Th and
14
C ages for carbonate material from lakes Lahanton and
Bonneville. Journal of Geophysical Research, 70, 4039–4054.
Kaufman, A., Yechieli, Y., and Gardosh, M., 1992. Reevaluation of
the lake-sediment chronology in the Dead Sea basin, Israel,
based on new 230Th/U dates. Quaternary Research, 38,
292–304.
Kaufman, A., Wasserburg, G. J., Porcelli, D., Bar-Mathews, M.,
Ayalon, A., and Halicz, L., 1998. U-Th isotope systematics from
the Soreq cave, Israel and climatic correlations. Earth and Plan-
etary Science Letters, 156, 141–155.
Kolodny, Y., Stein, M., and Machlus, M., 2005. Sea-rain–lake
relation in the last glacial east Mediterranean revealed by
d18O – d13C in Lake Lisan aragonites. Geochimica et
Cosmochimica Acta, 69, 4045–4060.
Ku, T.-L., and Liang, Z. C., 1984. The dating of impure carbonates
with decay-series isotopes. Nuclear Instruments and Methods in
Physics Research, 223, 563–571.
Ku, T.-L., Luo, S., Lowenstein, T. K., Li, J., and Spencer, R. J.,
1998. U-Series chronology of lacustrine deposits in Death Val-
ley, California. Quaternary Research, 50, 261–275.
Lazar, B., and Stein, M., 2011. Freshwater on the route of hominids
out of Africa revealed by U-Th in Red Sea corals. Geology, 39,
1067–1070, doi:10.1130/G32257.1.
Lazar, B., Enmar, R., Schossberger, M., Bar-Matthews, M., Halicz,
L., and Stein, M., 2004. Diagenetic effects on the distribution of
uranium in live and Holocene corals from the Gulf of Aqaba.
Geochimica et Cosmochimica Acta, 68, 4583–4593.
Le Roux, L. J., and Glendenin, L. E., 1963. Half-life of 232Th. In
Proceedings of the National Meeting on Nuclear Energy, Preto-
ria, South Africa.
Lin, J. C., Broecker, W. S., Anderson, R. F., Hemming, S.,
Rubenstone, J. L., and Bonani, G., 1996. New 230Th/U and 14C
ages from Lake Lahontan carbonates, Nevada, USA, and
a discussion of the origin of initial thorium. Geochimica et
Cosmochimica Acta, 60, 2817–2832.
Ludwig, K. R., and Titterington, D. M., 1994. Calculation of iso-
chrons, ages, and errors. Geochimica et Cosmochimica Acta,
58, 5031–5042.
Luo, S., and Ku, T.-L., 1991. U-series isochron dating: a generalized
method employing total-sample dissolution. Geochimica et
Cosmochimica Acta, 55, 555–564.
McGee, D., Quade, J., Edwards, R. L., Broecker, W. S., Cheng, H.,
Reiners, P. W., and Evenson, N., 2012. Lacustrine cave carbon-
ates: novel archives of paleohydrologic change in the Bonneville
Basin (Utah, USA). Earth and Planetary Science Letters,
351–352, 182–194.
Oviatt, C., Currey, D., and Miller, D., 1990. Age and paleoclimatic
significance of the Stansbury shoreline of Lake Bonneville,
northeastern Great Basin. Quaternary Research, 33, 291–305.
Oviatt, C. G., Currey, D. R., and Sack, D., 1992. Radiocarbon chro-
nology of Lake Bonneville, Eastern Great Basin, USA.
Palaeogeography Palaeoclimatology Palaeoecology, 99,
225–241.
Schramm, A., Stein, M., and Goldstein, S. L., 2000. Calibration of
the 14C time scale to >40 ka by 234U-230Th dating of Lake Lisan
sediments (last glacial Dead Sea). Earth and Planetary Science
Letters, 175, 27–40.
Stein, M., Wasserburg, G. J., Aharon, P., Chen, J. H., Zhu, Z. R.,
Bloom, A., and Chappell, J., 1993. TIMS U-series dating
and stable isotopes of the last interglacial event in Papua
New Guinea. Geochimica et Cosmochimica Acta, 57,
2541–2554.
Stein, M., Starinsky, A., Katz, A., Goldstein, S. L., Machlus, M.,
and Schramm, A., 1997. Strontium isotopic, chemical, and sed-
imentological evidence for the evolution of Lake Lisan and the
Dead Sea. Geochimica et Cosmochimica Acta, 61, 3975–3992.
136 CARBONATES, MARINE CARBONATES (U-SERIES)
Torfstein, A., Goldstein, S. L., Kagan, E., and Stein, M., 2013a.
Multi-site integrated U-Th chronology of the last glacial Lake
Lisan. Geochimica et Cosmochimica Acta, 104, 210–234.
Torfstein, A., Goldstein, S. L., Enzel, Y., and Stein, M., 2013b.
Impacts of abrupt climate changes in the Levant from last glacial
Dead Sea levels. Quaternary Science Reviews, 69, 1–7.
Winograd, I. J., Szabo, B. J., Coplen, T. B., and Riggs, A. C., 1988.
A 250,000 year climatic record from Great Basin vein calcite:
implications for Milankovitch theory. Science, 242, 1275–1280.
Winograd, I. J., Coplen, T. B., Landwehr, J. M., Riggs, A. C.,
Ludwig, K. R., Szabo, B. J., Kolesar, P. T., and Reversz,
K. M., 1992. Continuous 500,000 year climate record from vein
calcite in Devils Hole, Nevada. Science, 258, 189–368.
Cross-references
Alpha Spectroscopy
Carbonates, Marine Carbonates (U-Series)
Carbonates, Pedogenic (U-Series)
Carbonates, Speleothem Climatic (U-Series)
Lacustrine Environments (14C)
Luminescence Dating, Deep-Sea Marine and Lacustrine
Marine Varves
Mass Spectrometry
Thermal Ionization Mass Spectrometer (TIMS)
U-Series Dating
CARBONATES, MARINE CARBONATES (U-SERIES)
Claudine H. Stirling
Department of Chemistry, University of Otago, Dunedin,
New Zealand
Synonyms
Coral dating; 238U-234U-230Th dating; U-series chronol-
ogy; U-Th dating
Definition
U-series nuclides. Nuclides of the 238U, 235U, and 232Th
radioactive decay chains, which begin with the above
radioactive nuclides and end with a stable isotope of lead.
Fractionation. The fractional gain or loss of one element
relative to another (elemental fractionation) or one isotope
relative to another of the same element (isotopic fraction-
ation) during geological events.
Decay constant. The proportionality constant of
a radionuclide defining the number of atoms decaying
over a given period of time compared with the number
of atoms present initially.
Mass spectrometry. An analytical technique that produces
spectra of the masses of molecules or atoms used to deter-
mine the isotopic composition of a sample. Particles are
ionized in the instrument source followed by physical
separation based on their mass-to-charge ratio.
Introduction
For the last 800 millennia, climate has oscillated between
cool ice ages and warmer interglacial conditions with
a periodicity of about 100 thousand years. For example,
during the “Last Deglaciation” between about 21 and 6
thousand years ago (ka), more than 50 million cubic kilo-
meters of ice melted from the major continental ice sheets,
raising the sea level by ~130 m (Lambeck et al., 2002).
The waxing and waning of the ice sheets, fall and rise in
sea level, and the cooling and warming of Earth’s climate
are faithfully preserved as compositional changes in
marine sediment records (Lisiecki and Raymo, 2005)
(Figure 1a). These archives provide continuous, high-
resolution records of past climate change but lack critical
information on absolute timing because they cannot be
directly dated, relying instead on model chronologies
(Milankovitch, 1941) that require further validation. As
a result, the exact mechanisms driving the Earth’s natural
climate cycles remain uncertain. Obtaining reliable, abso-
lutely dated records of past climate change is critical for
understanding how Earth’s natural climate cycles work.
These data can provide a firmer basis for assessing the role
of anthropogenic effects, such as greenhouse gas emis-
sions, in modifying climate, thus improving twenty-first-
century climate projections.
In-built clocks based on the decay of radioactive ele-
ments that are present in minute quantities in rocks and
sediments allow an absolute chronology to be assigned
to geological records. One of the key dating methods used
in paleoclimate research takes advantage of the natural
radioactive decay of uranium (U) to thorium (Th) in
U-bearing minerals, such as calcium carbonate. This tech-
nique, commonly referred to as the U-series dating
method (Broecker, 1963), has been widely applied to the
following shallow-water marine carbonate archives in
the reconstruction of past climate. Results have provided
constraints on the precise timing, duration, and magnitude
of key climate episodes:
1. Shallow-water reef-building corals. These archives
grow close to the sea surface, providing valuable
information on both sea-level and ocean-water tem-
perature fluctuations, particularly during interglacial
sea-level high stands when reefs grow prolifically
(Broecker, 1963; Edwards et al., 1987). The strati-
graphic and morphological structures of the reef,
combined with the height-age relationships of fossil
corals within the reef framework, can be used to
reconstruct the reef’s growth history and, in turn,
derive local sea-level curves for the study locality
(Figure 1b).
2. Calcium carbonate slope sediments. Calcium carbon-
ate sediments, such as those swept from the banks of
the Bahamas, yield continuous sea-level and climate
records, especially during interglacial periods when
the banks are flooded and sediment accumulation rates
are high. Records are derived by combining the
U-series ages of the carbonate fraction of the sediment
with the oxygen stable isotope signatures of foraminif-
era (marine plankton) that have been deposited simul-
taneously (Slowey et al., 1996).
 CARBONATES, MARINE CARBONATES (U-SERIES) 137

a MIS 5 MIS 7 MIS 9 MIS 11

Deep sea sediments
(Lisiecki & Raymo, 2005)
3 Last
Interglacial
Last MIS 13 MIS 15
Deglaciation

δ18O (‰)
4

5 MIS 2 MIS 6

0 100 200 300 400 500 600 700

Age (thousand years ago)

NW
Australia

Carbonates, Marine Carbonates (U-Series), Figure 1 (a) The LR04 deep-sea sediment oxygen stable isotope (d18O) record, which
comprises a stack of 57 globally distributed deep-sea records and provides a combined signal of deep-ocean temperature and global
sea level for the past 5.3 million years (Lisiecki and Raymo 2005). The past 700 thousand years of this record is shown and has been
assigned a timescale by direct correlation with model chronologies. Glacial-interglacial climate episodes are defined by “marine
isotope stage” (or MIS) terminology, whereby even- and odd-numbered MIS intervals refer to glacial and interglacial periods,
respectively. Distinct oscillations between warm interglacial periods (high d18O values) and cool glaciations (low d18O values) are
apparent. Centennial- and millennial-scale climate variability is also evident. (b) Photograph of an elevated “Last Interglacial” fossil
coral reef, within a few meters of the present shoreline, along the coast of Western Australia which formed around 125,000 years ago.
Fossil corals within the reef framework can be directly dated using the U-series dating method.

3. Submerged coastal speleothems (stalagmites and sta-
lactites). These archives are currently found below
sea level in coastal caves. Calcium carbonate
growth occurs during sea-level low stands when
the caves are emergent, whereas growth ceases dur-
ing flooding by rising sea levels. Important infor-
mation on the timing, duration, and elevation of
sea-level low stands and marine transgressions can
be obtained by dating the carbonate preserved on
either side of a growth hiatus (Richards and Dorale,
2003).
The U-series dating method has also been applied to
deepwater marine carbonates, including deep-sea
solitary corals and chimney deposits (Teichert
et al., 2003; Robinson et al., 2005).
138 CARBONATES, MARINE CARBONATES (U-SERIES)
Basic principles of the U-series chronometer
The U-series dating method utilizes the natural radioactive
decay of 238U to its longest-lived intermediate daughters,
234
U and 230Th, and subsequent decay, via a series of
short-lived nuclides, to stable 206Pb as follows:
238
U! 234
Th ! 234
Pa ! 234
U! 230
Th ! . . . : !206 Pb
Half-life 4.5 billion 24 days 1 min 245 thousand 76 thousand Stable
years years years
The respective 245,000 and 76,000 year half-lives of
234
U and 230Th make the U-series system particularly use-
ful for dating climate events spanning the last
600,000 years.
U-series nuclides in the modern global ocean
For the U-series system, a state of radioactive equilibrium
occurs when the activities, or rates of decay or in-growth,
of 234U and 230Th, and all other intermediate daughters in
the 238U decay chain, are equal and identical to the activity
of 238U. This condition is achieved after five or six half-
lives of the daughter isotope, provided that the system is
left undisturbed without being subjected to parent-
daughter fractionation events. In this state, the activity
ratios between 238U and its 234U and 230Th daughters,
referred to as [234U/238U]act and [230Th/238U]act, respec-
tively, are invariant and equal to unity, irrespective of the
time elapsed. After this point, the system cannot be used
as a dating tool. Importantly, the U-series dating of marine
carbonates becomes possible because chemical and
nuclear processes continuously disrupt the state of
U-series radioactive equilibrium in modern hydrological
systems. This gives rise to the following two conditions
of radioactive disequilibrium in the marine environment,
allowing the [234U/238U]act and [230Th/238U]act ratios of
marine minerals to evolve to a new equilibrium state as
a predictable function of time:
1. There is an extreme chemical fractionation between
U and Th in the oxygenated modern ocean.
In particular, U is present in its highest U(VI) oxidation
state and forms uranyl complexes that are highly solu-
ble in seawater. In contrast, Th exists as relatively
insoluble species in the Th(IV) oxidation state, yield-
ing negligible levels of dissolved Th in the global
ocean (Ivanovitch and Harmon, 1992).
2. The state of radioactive equilibrium between 238U and
234
U is also disrupted due to internal damage to mineral
grains during the energetic decay of 238U. As a result,
the 234U daughter nuclide is often only loosely bound
to the mineral structure, making it more susceptible
to preferential leaching during weathering than
lattice-bound 238U (Ivanovitch and Harmon, 1992).
Consequently, seawater contains an ~15 % excess
of 234U over 238U (i.e., seawater [234U/238U]act ¼ 1.15)
compared to the radioactive equilibrium value of 1.00
due to steady-state input by river water loaded with
dissolved U from continental weathering processes
(Edwards et al., 2003).
U-series nuclides in modern marine carbonates
During the formation of marine carbonates, the 238U,
234
U, and 230Th isotopes are removed from seawater,
together with other dissolved metals, and incorporated into
the mineral structure. Thus, newly formed corals,
speleothems, and aragonitic slope sediments contain
U with a “disequilibrium” initial [234U/238U]act ratio of
1.15, but no significant 230Th (Edwards et al., 2003). Fol-
lowing calcium carbonate formation, the U-series nuclides
become isolated from the steady-state ocean (where
decaying nuclides are continuously resupplied from river-
ine and other sources), allowing the 234U and 230Th inter-
mediate daughters to evolve to a new equilibrium state
with their 238U parent as a predictable function of the time
elapsed since the mineral developed. Thus, for every
marine carbonate sample, both the radioactive decay of
234
U and the simultaneous in-growth of 230Th toward radio-
active equilibrium with 238U can be expressed as a function
of its U-series or 230Th age, denoted by T, provided that
(1) [234U/238U]act and [230Th/238U]act can be reliably mea-
sured and (2) the decay constants for 234U and 230Th (l234
and l230, respectively) are accurately known. This relation-
ship is shown in Figure 2 and Eq. (1), and T can be calcu-
lated iteratively (Edwards et al., 1987):
 230   234  
Th U
1  238 ¼ el230 T  238 U
 1
U act
  act 
l230
1  eðl234 l230 ÞT
l230  l234
ð1Þ
The 230Th age will accurately reflect the “true” age of the
sample, provided that (1) no initial 230Th existed in the sam-
ple at the time of its formation and (2) no further gain nor
loss of the U-series isotopes takes place other than by
“closed-system” radioactive in-growth and decay
(Edwards et al., 1987). The latter assumption is sometimes
violated, especially in older samples affected by secondary
alteration, which tends to disturb the U-series isotopes, off-
setting the 230Th age from the true sample age.
Analysis of the U-series isotopes
Chemical procedures using ion-exchange resins to selec-
tively adsorb and desorb the elements of interest are typi-
cally adopted to isolate “pure” U and Th from the calcium
carbonate matrix of the sample (Goldstein and Stirling,
2003). This purification procedure enables the
[234U/238U]act and [230Th/238U]act ratios to be reliably
determined, free of analytical artifacts such as isobaric or
molecular interferences.
 CARBONATES, MARINE CARBONATES (U-SERIES) 139

a
1.15
Seawater E
volution C
urve

0
1.10

,00
00
>5
[234 U/ 238U]act 1.05

years
1.00

50,000

000

00

00

ine
0,0
0,0
0.95

100,

el
30
20

ag
/U
0.90

Th 0
23
0.85

ite
fin
In
0.80
0 0.2 0.4 0.6 0.8 1 1.2
[230 Th/ 238U]act

b 1.16 c 1.0
equilibrium
[230 Th/ 238U]act 0.8
[234 U/ 238U]act

1.12

1.08 0.6

0.4
1.04
0.2
1.00 equilibrium
0.0
0 400 800 1200 0 100 200 300
Age Age
(thousand years ago) (thousand years ago)
Carbonates, Marine Carbonates (U-Series), Figure 2 (a) Plot showing the isotopic evolution of [234U/238U]act and [230Th/238U]act
values as a function of time (T). Isochrons (straight, steeply sloping lines) are shown at 50, 100, 200, 300, and 500 ka. The red curve
starting at [234U/238U]act ¼ 1.15 and [230Th/238U]act ¼ 0 gives the “seawater evolution path” for a sample formed in the marine
environment with U equivalent to the [234U/238U]act composition of seawater but no Th at the time of its formation. (b) Isotope
evolution curve for [234U/238U]act starting at an initial value of 1.15 evolves by radioactive decay towards the radioactive equilibrium
value of 1.0 with time. (c) Isotope evolution curve for [230Th/238U]act starting at an initial value of 0 evolves by radioactive in-growth
towards the radioactive equilibrium value of 1.0. Age resolution of the U-series dating method decreases with increasing sample
age due to the progressive decrease in separation between isochrons, as demonstrated by the two data points shown in (a) for
125,000- and 200,000-year-old samples measured with identical levels of precision.

Analytical measurements of [234U/238U]act and
[ Th/238U]act have an associated uncertainty, which, by
230
convention, is reported at the 2-sigma level. Propagation
of these uncertainties through the above U-series age
equation results in 230Th-age errors that become propor-
tionally larger with increasing sample age, as radioactive
equilibrium is approached and the rate of radioactive
decay and in-growth of 234U and 230Th slows down
(Figure 2). This progressive loss in the resolution of the
U-series chronometer with increasing age demonstrates
the importance of achieving the highest analytical preci-
sions possible.
To this end, the U-series dating method has benefitted
enormously from ongoing innovations in technology that
have occurred over the past 30 years. In particular, the
advent of thermal ionization mass spectrometry (TIMS)
in the 1980s and, more recently, multiple-collector induc-
tively coupled plasma mass spectrometry (MC-ICPMS) in
the late 1990s to U-series isotopic analysis dramatically
reduced sample-size requirements, shortened analysis
times, and improved measurement precision compared
with earlier decay-counting techniques using alpha spec-
trometry (Goldstein and Stirling, 2003). Ongoing design
improvements in instrumentation over the last 10 years
have continued to enhance U-series analytical
performance (Stirling and Andersen, 2009). As a result,
230
Th-age uncertainties of 800 years in 120,000-year-
old samples and 3,000 years in 300,000-year-old
samples can now be routinely achieved. Some recent
studies have further improved 230Th-age uncertainties to
better than 1,000 years in 300,000-year-old samples
(Stirling and Andersen, 2009).
140 CARBONATES, MARINE CARBONATES (U-SERIES)
U-series dating to marine carbonates:
windows into past climate
Over the past two decades, the U-series dating of marine
carbonates has provided crucial new constraints on the
absolute timing and duration of key climate intervals
throughout the past 600 thousand years of the Earth’s cli-
mate history. At the young end of the dating range,
U-series measurements have been used to determine the
timing and frequency of El Niño events, occurring over
the past few centuries (Edwards et al., 2003). In the
<50,000-year range, coupled U-series and 14C/12C mea-
surements for deep-sea corals have been used to constrain
ventilation and circulation patterns of oceanic water
masses and how they regulate global climate (Robinson
et al., 2005). Furthermore, U-series measurements for
drilled fossil coral reefs have been used to calibrate the
14
C chronometer for the past 40,000 years (Edwards
et al., 2003). These drill-core coral data also provide
a continuous record of sea-level rise and climate change
for the “Last Deglaciation,” extending from the peak of
the last glacial maximum to the onset of the present inter-
glacial warm period, from around 21 to 6 thousand years
ago (Edwards et al., 2003; Deschamps et al., 2012). These
results are discussed in more detail as follows:
The last deglaciation
Collectively, records from core drilled at coral reefs show
evidence for several distinct “meltwater pulses” rather
than continuous melting during the Last Deglaciation
due to episodic periods of rapid ice sheet retreat. The larg-
est of these events, referred to as “meltwater pulse”
(MWP)-1A, occurring around 14 thousand years ago, pro-
duced an extremely rapid sea-level rise of ~20 m over
300 years, which is in excess of 40 m/1,000 years and
approximately ten times faster than present melting rates.
This was succeeded by the abrupt “Younger Dryas”
cooling event, involving a transient return to full ice age
conditions at ~12 ka before renewed sea-level rise. These
abrupt centennial- to millennial-scale oscillations in cli-
mate and sea level appear synchronous with large-scale
reorganization of North Atlantic thermohaline circulation,
an oceanic conveyor belt exporting heat from the tropics
to the poles. However, there is no consensus on the geo-
graphic source of meltwater during deglaciation nor exact
underlying mechanisms controlling rapid sea-level and
climate change. Recent U-series dating of drill core from
fossil corals at different locations worldwide provides
high-resolution “fingerprinting” of MWP-1A and impli-
cates a substantial contribution of meltwater from the ice
sheets of both northern and southern hemispheres
(Deschamps et al., 2012). This suggests that melting of
both the Greenland and Antarctic ice sheets will be highly
sensitive to sustained global warming in the coming
decades and may already be contributing to present day
sea-level rise.
Prior to the last 50,000 years, a wealth of information is
available only for the Last Interglacial period, occurring
around 125 thousand years ago (discussed below). This
is primarily because marine carbonates that formed close
to modern shorelines during this time are relatively acces-
sible for study today (Stirling and Andersen, 2009).
The last interglacial period
The Last Interglacial interval, occurring from approxi-
mately 128 to 116 thousand years ago (Stirling and Ander-
sen, 2009), represents that the last time global sea levels
were at or near modern levels and, by inference, the last
time global climatic conditions, as well as the size and
configuration of the major ice sheets, were similar to those
in the present day. The Last Interglacial period has there-
fore been identified as a suitable analogue of the present
climate system in order to speculate on future climate
trends. Importantly, the Last Interglacial temperatures are
likely to have been similar to, or slightly warmer, than
the preindustrial state. Therefore, a reliable assessment
of the timing, duration, and amplitude of global sea levels
during the Last Interglacial is crucial for assessing the sta-
bility of the today’s ice sheets in response to climate
warming of a few degrees. Dutton and Lambeck (2012)
have recently compiled all available U-series data for the
Last Interglacial fossil reefs, demonstrating that the Last
Interglacial global sea levels peaked between 5.5 and
9 m above the present sea level. These amplitudes require
smaller ice sheets in both Greenland and Antarctica rela-
tive to today, implying a strong sensitivity of the cryogenic
system to only minor climate warming.
Marine carbonates formed earlier than 125,000 years
ago are prone to secondary alteration. Going forward,
one of the key challenges faced by the U-series commu-
nity is to acquire reliably dated climatic records for earlier
interglacial periods that could represent more suitable ana-
logues of the present climate system than the Last Intergla-
cial period, such as “marine isotope stage” 9, 11, or
15 (Figure 1) (Stirling and Andersen, 2009).
Summary
The U-series dating of marine carbonates is based on the
natural radioactive decay of 238U to its longest-lived inter-
mediate daughters, 234U and 230Th. This technique has
been used to provide reliable, accurately dated records of
past sea-level and climate change for the past
600,000 years of the Earth’s history, especially for the past
130,000 years of the last glacial-interglacial cycle. This, in
turn, has facilitated a more complete understanding of the
mechanisms driving the Earth’s natural climate cycles,
which have helped to inform projections of twenty-first-
century climate change.
Bibliography
Broecker, W. S., 1963. A preliminary evaluation of uranium series
inequilibrium as a tool for absolute age measurement on marine
carbonates. Journal of Geophysical Research, 68, 2817–2834.
Deschamps, P., Durand, N., Bard, E., Hamelin, B., Camoin, G.,
Thomas, A. L., Henderson, G. M., Okuno, J., and Yokoyama, Y.,
 CARBONATES, PEDOGENIC (U-SERIES)
2012. Ice-sheet collapse and sea-level rise at the Bølling
warming 14,600 years ago. Nature, 483, 559–564.
Dutton, A., and Lambeck, K., 2012. Ice volume and sea level during
the Last Interglacial. Science, 337, 216–219.
Edwards, L. R., Chen, J. H., and Wasserburg, G. J., 1987.
238
U-234U-230Th-232Th systematics and the precise measurement
of time over the past 500,000 years. Earth and Planetary Science
Letters, 81, 175–192.
Edwards, R. L., Gallup, C. D., and Cheng, H., 2003. Uranium-series
dating of marine and lacustrine carbonates. In Bourdon, B.,
Henderson, G. M., Lundstrom, C. C., and Turner, S. P. (eds.),
Uranium-Series Geochemistry. Washington, DC: Mineralogical
Society of America. Reviews in Mineralogy & Geochemistry,
Vol. 52, pp. 363–405.
Goldstein, S. J., and Stirling, C. H., 2003. Techniques for measuring
uranium-series nuclides: 1992–2002. In Bourdon, B., Hender-
son, G. M., Lundstrom, C. C., and Turner, S. P. (eds.), Uranium-
Series Geochemistry. Washington, DC: Mineralogical Society of
America. Reviews in Mineralogy & Geochemistry,
Vol. 52, pp. 23–57.
Ivanovitch, M., and Harmon, R. S. (eds.), 1992. Uranium-series
Disequilibrium: Applications to Earth, Marine, and Environ-
mental Sciences, 2nd edn. Oxford, UK: Oxford University Press,
p. 910.
Lambeck, K., Esat, T. M., and Potter, E. K., 2002. Links between
climate and sea levels for the past three million years. Nature,
419, 199–206.
Lisiecki, L. E., and Raymo, M. E., 2005. A Pliocene-Pleistocene
stack of 57 globally distributed benthic d18O records. Paleocea-
nography, 20: PA1003, 17 p.
Milankovitch, M. M., 1941. Canon of Insolation and The Ice
Age Problem. Königlich Serische Akademie, Beograd
(English translation by the Israel Prog. for Sci. Translations,
Washington, DC).
Richards, D. A., and Dorale, J. A., 2003. Uranium-series chronol-
ogy and environmental applications of speleothems. In Bourdon,
B., Henderson, G. M., Lundstrom, C. C., and Turner, S. P. (eds.),
Uranium-Series Geochemistry. Washington, DC: Mineralogical
Society of America. Reviews in Mineralogy & Geochemistry,
Vol. 52, pp. 407–460.
Robinson, L. F., Adkins, J. F., Keigwin, L. D., Southon, J.,
Fernandez, D. P., Wang, S. L., and Scheirer, D. S., 2005. Radio-
carbon variability in the western North Atlantic during the last
deglaciation. Science, 310, 1469–1473.
Slowey, N. C., Henderson, G. M., and Curry, W. B., 1996. Direct
U-Th dating of marine sediments from the two most recent inter-
glacial periods. Nature, 383, 242–244.
Stirling, C. H., and Andersen, M. B., 2009. Uranium-series dating
of fossil coral reefs: extending the sea-level record beyond the
last glacial cycle. Earth and Planetary Science Letters
(Frontiers), 284, 269–283.
Teichert, B. M. A., Eisenhauer, A., Bohrmann, G., Haase-Schramm,
A., Bock, B., and Linke, P., 2003. U/Th systematics and ages of
authigenic carbonates from Hydrate Ridge, Cascadia Margin:
recorders of fluid flow variations. Geochimica et Cosmochimica
Acta, 67, 3845–3857.
Cross-references
Alpha Spectroscopy
Carbonates, Speleothem Climatic (U-Series)
Corals (Sclerochronology)
Lacustrine Environments (14C)
Luminescence Dating, Deep-Sea Marine and Lacustrine
Marine Isotope Stratigraphy
Marine Varves
Mass Spectrometry
141
Radiocarbon Dating
Radiocarbon Dating of Marine Carbonates
Sea Level Change (U-Series)
Thermal Ionization Mass Spectrometer (TIMS)
U-Series Dating
CARBONATES, PEDOGENIC (U-SERIES)
Warren D. Sharp
Berkeley Geochronology Center, Berkeley, CA, USA
Definition
Secondary material consisting of authigenic carbonate
and/or silica formed in the shallow vadose zone as part
of soil development, typically in arid to semiarid climates.
Geochronological significance
Pedogenic carbonate, if less than about 0.5 Myr old,
dense, and relatively pure, is amenable to uranium-series
dating. Such dates can provide useful estimates of the
depositional age of the host rock and a time axis for
paleoclimate proxy records contained in the carbonate
itself.
Origin of pedogenic carbonate
Pedogenic carbonate (the term “carbonate” is used here for
brevity, but silica is commonly also present either as
a dispersed phase or in discrete layers) generally forms
in regions with less than 750–1,000 mm annual precipita-
tion, is common in a wide range of climates over more
than half of the Earth’s land surface, and generally accu-
mulates most markedly within soil profiles at depths of
0.3–2.0 m (Birkeland, 1984). The primary sources of cal-
cium ions (Ca2+) for pedogenic carbonate are detrital car-
bonate or Ca-bearing silicates present in the parent
material, wind-deposited carbonate-bearing dust, and Ca
ions in rainwater (Machette, 1985; Capo and Chadwick,
1999). The latter two sources are sufficient for pedogenic
carbonate formation in carbonate-poor parent materials.
The requisite carbonate ions (CO22) form in the soil from
a complex mixture of atmospheric CO2, plant-respired
CO2, and carbon released by decomposing organic matter
(Cerling, 1999). The relative contributions of these
sources vary with local conditions as revealed by carbon
isotope studies (Amundson et al., 1994). During pedogen-
esis, Ca-bearing minerals are leached from the upper part
of the soil profile along with trace amounts of uranium
and other elements. At greater depths, typically concen-
trated between 0.5 and 2 m below the surface, pedogenic
carbonate precipitation is caused by increasing ion
strength resulting from evaporation, transpiration or freez-
ing, degassing of CO2 from the soil water, or changing sol-
ubility of CaCO3 as solutions move along temperature
gradients in soils (Birkeland, 1984). In gravelly soils
where substantial pore space may initially surround the
142 CARBONATES, PEDOGENIC (U-SERIES)
sedimentary clasts, carbonate dissolved in the soil solu-
tions is generally precipitated as the consequence of
evapotranspiration. In soils as old as a hundred thousand
years or more, up to six stages of pedogenic carbonate
development have been described (Gile et al., 1966;
Machette, 1985).
Carbonate-rich clast coatings that begin to form on the
bottoms of clasts in the earliest stages of carbonate accu-
mulation in gravelly soils can consist of dense, pure, lam-
inated carbonate that may be highly suitable for uranium-
series (hereafter, U-series) dating (e.g., Ludwig and Paces,
2002; Sharp et al., 2003). Datable thicknesses of such clast
coatings (0.5 mm or less) may accumulate in as little as
a few hundred years after host deposition. Accordingly,
U-series dating of early formed pedogenic carbonate clast
coatings can provide useful estimates for the age of grav-
elly alluvium deposits, which may be otherwise difficult
to date.
U-series dating applied to pedogenic carbonate
The basis for U-series dating using the 230Th/U system is
derived from the relation among 230Th, 234U and 238U,
all nuclides of the 238U decay series. 230Th (half-life
75.69 kyr) is the alpha-decay product of 234U (half-life
245.25 kyr), which in turn forms by alpha decay of 238U
(half-life 4.4683 Gyr) through two short-lived intermedi-
ate daughter isotopes (Jaffey et al., 1971; Cheng
et al., 2000). A closed uranium-bearing system will
evolve toward a state of secular equilibrium in which the
number of decays per unit time (the activity) of each
nuclide in the decay series becomes equal. Newly formed
pedogenic carbonate usually contains ppm levels of ura-
nium and is depleted in 230Th relative to secular equilib-
rium because uranium is immensely more soluble than
thorium in oxygenated, infiltrating surface waters and
soil waters. The initial concentration of 234U in pedo-
genic carbonate generally exceeds the value obtained
under secular equilibrium conditions because 234U is
enriched in soil waters by two processes: (1) ejection
from particles by recoil upon its formation by alpha
decay from 238U and (2) enhanced solubility relative
to 238U because 234U is located in lattice sites damaged
by alpha decay. Accordingly, U-series dating using the
230
Th/U system relies on the ingrowth of 230Th and
the decay of 234U toward secular equilibrium in
a closed system (Ivanovich et al., 1992). Carefully
selected, dense pedogenic carbonate has been shown
to remain closed with respect to the 230Th/U system
for up to >500 kyr (Ludwig and Paces, 2002).
The U-series method is sensitive to the presence of ini-
tial 230Th, which will result in anomalously old apparent
ages if left uncorrected. Mineralogically pure carbonate
precipitated from oxygenated soil waters is likely to have
negligible initial Th contents due to its extremely low sol-
ubility relative to U. This inference is supported by high
230
Th/232Th activity ratios – that is, the ratio of uranogenic
230
Th to common 232Th (the abundant Th isotope, with
half-life ~14 Gyr) – observed in some visibly pure pedo-
genic carbonate samples (e.g., Sharp et al., 2003). These
high 230Th/232Th activity ratios must reflect in situ decay
of uranium in initially thorium-poor material
(rather than incorporation of extraneous 230Th) because
coherent microstratigraphic relations are preserved among
subsamples with widely varying 230Th/232Th ratios.
Nonetheless, most pedogenic carbonate contains some
232
Th, probably reflecting contamination of the authigenic
carbonate cements by fine-grained detritus, implying
the presence of detrital 230Th, 234U, and 238U. This
detrital component must be subtracted from the measured
concentrations to obtain the composition of the pure
authigenic component, which has age significance.
U-series dating of pedogenic carbonate was pioneered
by Ku et al. (1979), who used 5–8 g samples from the
inner 2–3 mm of pedogenic carbonate rinds developed
in soils on Pleistocene gravels of the arid Vidal Valley,
California. Samples were leached with 1N HCL and both
leachates and insoluble residues were analyzed using
decay counting by alpha spectrometry. Correction for
detrital U and Th isotopes was made using the residue ana-
lyses, which had U-Th isotopic ratios close to secular
equilibrium and an average atomic 232Th/238U of 4 – sim-
ilar to average upper continental crust (Taylor and
McLennan, 1985). More recently, carefully selected,
milligram-size portions of pedogenic carbonate rinds that
were totally dissolved and analyzed by mass spectrometry
have been shown to be highly suitable for U-series dating
(e.g., Ludwig and Paces, 2002; Sharp et al., 2003). Most
such samples have sufficiently high 230Th/232Th activity
ratios that correction for initial isotopes can be made using
a model detritus composition (Ludwig and Paces, 2002).
This “single sample, single date” approach precludes
meaningless 230Th/U ages that can sometimes result from
chemical fractionation of Th from U during laboratory
processing by leach-residue techniques (e.g., Bischoff
and Fitzpatrick, 1991). Furthermore, the greatly reduced
sample size of the latter approach helps to minimize aver-
aging of long (>105 years), and sometimes visibly discon-
tinuous, depositional histories of individual rinds and
allows ages to be tested against microstratigraphic
relations.
Highly sensitive analyses of U and Th in carbonates by
ICP-MS (inductively coupled mass spectrometry; e.g.,
Shen et al., 2012) now make it possible to precisely ana-
lyze samples of pedogenic carbonate of sub-milligram
size. Such analytical improvements, along with laser abla-
tion ICP-MS analyses of carbonates (Hellstrom
et al., 2010), facilitate studies of texturally complex,
slowly formed pedogenic carbonate microstratigraphic
sequences (e.g., Brock and Buck, 2005). Where pure ped-
ogenic silica is present, its higher U concentrations
(>50 ppm) make analyses by SIMS (secondary ion mass
spectrometry; i.e., ion probe) feasible, allowing dating at
a spatial resolution of ~30 mm, albeit at relatively low pre-
cision (age uncertainties of 10–20 %; Maher
et al., 2007).
 CARBONATES, PEDOGENIC (U-SERIES)
Examples of application of U-series dating to
pedogenic carbonate
U-series dating of pedogenic carbonate has been used to
date landforms offset by faults, thereby establishing the
timing and rate of fault motions in Nevada (Ludwig and
Paces, 2002), Tibet (Blisniuk and Sharp, 2003; Gold
et al., 2011), and along the San Andreas Fault system in
southern California (Fletcher et al., 2010, 2011). Nuriel
et al. (2012) studied and dated calcite veins and striations
associated with the Dead Sea fault in Israel, thereby
constraining the timing of mineralization and related
deformation on the fault. Dating of Late Quaternary allu-
vial fans using U-series on pedogenic carbonate coordi-
nated with 10Be exposure dating has shown that the two
techniques provide consistent dates. Combining results
from both methods provides enhanced reliability relative
to dates determined using a single technique (e.g., Behr
et al., 2010; Blisniuk et al., 2012). U-series dating of ped-
ogenic carbonate has been applied to landforms of cli-
matic and geomorphic significance; for example, dates
on glaciofluvial terraces in Wyoming help to constrain
the timing of the penultimate glaciation in the Rocky
Mountains (Sharp et al., 2003). U-series on pedogenic car-
bonate has been applied to calcic soils in Spain to quantify
ages of landforms hosting such soils characterized by
advanced stages of pedogenic carbonate development
including the formation of calcretes (Candy et al., 2004,
2005).
Summary
Pedogenic carbonate (silica) is widely developed in soils
formed in arid to semiarid climates. U-series dating of
dense, relatively pure pedogenic carbonate that commonly
forms in coarse clastic deposits such as alluvial gravels
can provide reliable ages for Middle and Late Quaternary
calcic soils and useful estimates of the ages of host
deposits and landforms.
Bibliography
Amundson, R., Wang, Y. R., Chadwick, O. A., Trumbore, S. E.,
McFadden, L., McDonald, E., Wells, S., and DeNiro, M.,
1994. Factors and processes governing the 14C content of car-
bonate in desert soil. Earth and Planetary Science Letters, 125,
385–405.
Behr, W. M., Rood, D. H., Fletcher, K. E., Guzman, N., Finkel, R.,
Hanks, T. C., Hudnut, K. W., Kendrick, K. J., Platt, J. P., Sharp,
W. D., Weldon, R. J., and Yule, J. D., 2010. Uncertainties in slip-
rate estimates for the Mission Creek strand of the southern San
Andreas fault at Biskra Palms Oasis, southern California. Geo-
logical Society of America Bulletin, 122, 1360–1377.
Birkeland, P. W., 1984. Soils and Geomorphology. New York:
Oxford University Press.
Bischoff, J. L., and Fitzpatrick, J. A., 1991. U-series dating of
impure carbonates: an isochron technique using total-sample dis-
solution. Geochimica et Cosmochimica Acta, 55, 543–554.
Blisniuk, P. M., and Sharp, W. D., 2003. Rates of late Quaternary
normal faulting in central Tibet from U-series dating of pedo-
genic carbonate in displaced fluvial gravel deposits. Earth and
Planet Science Letters, 215, 169–186.
143
Blisniuk, K., Oskin, M., Fletcher, K., Rockwell, T., and Sharp, W.,
2012. Assessing the Reliability of U-series and 10Be dating tech-
niques on Alluvial Fans in the Anza Borrego Desert, California.
Quaternary Geochronology, 13, 26–41.
Brock, A. L., and Buck, B. J., 2005. A new formation process for
calcic pendants from Pahranagat Valley, Nevada, USA, and
implication for dating Quaternary landforms. Quaternary
Research, 63, 359–367.
Candy, I., Black, S., and Sellwood, B. W., 2004. Quantifying time
scales of pedogenic calcrete formation. Sedimentary Geology,
170, 177–187.
Candy, I., Black, S., and Sellwood, B. W., 2005. U-series isochron
dating of immature and mature calcretes as a basis for
constructing Quaternary landform chronologies for the Sorbas
basin, southeast Spain. Quaternary Research, 64, 100–111.
Capo, R. C., and Chadwick, O. A., 1999. Sources of strontium and
calcium in desert soils and calcrete. Earth and Planetary Science
Letters, 170, 61–72.
Cerling, T. E., 1999. Stable carbon isotopes in paleosol carbonates.
Special Publication of the International Association of Sedimen-
tologists, 27, 43–60.
Cheng, H., Edwards, R. L., Hoff, J., Gallup, C. D., Richards, D. A.,
and Asmerom, Y., 2000. The half-lives of uranium-234 and
thorium-230. Chemical Geology, 169, 17–33.
Fletcher, K. E., Sharp, W. D., Kendrick, K. J., Behr, W. M., Hudnut,
K. W., and Hanks, T. C., 2010. 230Th/U dating of a late Pleisto-
cene alluvial fan along the southern San Andreas fault. Geolog-
ical Society of America Bulletin, 122, 1347–1359.
Fletcher, K. E. K., Rockwell, T. K., and Sharp, W. D., 2011. Late
Quaternary slip rate of the southern Elsinore fault, Southern Cal-
ifornia: dating offset alluvial fans via 230Th/U on pedogenic car-
bonate. Journal of Geophysical Research, 116, FO2006, 11 pp.
Gile, L. H., Peterson, F. F., and Grossman, R. B., 1966. Morpholog-
ical and genetic sequences of carbonate accumulation in desert
soils. Soil Science, 101, 347–360.
Gold, R. D., Cowgill, E., Arrowsmith, J. R., Chen, X. H., Sharp,
W. D., Cooper, K. M., and Wang, X. F., 2011. Faulted terrace
risers place new constraints on the late Quaternary slip rate for
the Altyn Tagh fault, northwest Tibet. Geological Society of
America Bulletin, 123, 958–978.
Hellstrom, J., Sharp, W. D., Ludwig, K. R., Hergt, J. M., Woodhead,
J. D., Amundson, R., and Chadwick, O. A., 2010. Th-230/U dat-
ing of pedogenic carbonate by laser ablation ICP-MS.
Geochimica et Cosmochimica Acta, 74, A396.
Ivanovich, M., Latham, A. G., and Ku, T. L., 1992. Uranium-series
disequilibrium applications in geochronology. In Ivanovich, M.,
and Harmon, R. S. (eds.), Uranium-series Disequilibrium:
Applications to Earth, Marine and Environmental Sciences.
Oxford: Oxford Science, pp. 62–89.
Jaffey, A. H., Flynn, K. F., Glendenin, L. E., Bentley, W. C., and
Essling, A. M., 1971. Precise measurement of half-lives and spe-
cific activities of 235U and 238U. Physical Review C, 4,
1889–1906.
Ku, T. L., Bull, W. B., Freeman, S. T., and Knauss, K. G., 1979.
230
Th–234U dating of pedogenic carbonates in gravelly desert
soils of Vidal Valley, southern California. Geological Society of
America Bulletin, 90, 1063–1073.
Ludwig, K. R., and Paces, J. B., 2002. Uranium-series dating of
pedogenic silica and carbonate, Crater Flat, Nevada. Geochimica
Cosmochimica Acta, 66, 487–506.
Machette, M., 1985. Calcic soils of the southwestern United States.
In Weide, D. L., and Faber, M. L. (eds.), Soils and Quaternary
Geology of the Southwestern United States. Boulder: Geological
Society of America. Geological Society of America Special
Paper, 203, pp. 1–21.
Maher, K., Wooden, J. L., Paces, J. B., and Miller, D. M., 2007.
230
Th–U dating of surficial deposits using the ion microprobe
144 CARBONATES, SPELEOTHEM ARCHAEOLOGICAL (U-SERIES)
(SHRIMP-RG): a microstratigraphic perspective. Quaternary
International, 166, 15–28.
Nuriel, P., Rosenbaum, G., Zhao, J.-X., Feng, Y., Golding, S. D.,
Villemant, B., and Weinberger, R., 2012. U-Th dating of striated
fault planes. Geology, 40, 647–650.
Sharp, W. D., Ludwig, K. R., Chadwick, O. A., Amundson, R., and
Glaser, L. L., 2003. Dating fluvial terraces by 230Th/U on pedo-
genic carbonate, Wind River Basin, Wyoming. Quaternary
Research, 59, 139–150.
Shen, C.-C., Wu, C., Cheng, H., Edwards, L., Hsieh, Y., Gallet, S.,
Chang, C., Li, T., Lam, D., Kano, A., Hori, M., and
Spotl, C., 2012. High-precision and high-resolution
carbonate 230Th dating by MC-ICP-MS with SEM protocols.
Geochimica et Cosmochimica Acta, 99, 71–86.
Taylor, S. R., and McLennan, S. M., 1985. The Continental Crust:
Its Composition and Evolution. Oxford: Blackwell Scientific
Publications.
Cross-references
Carbonates, Lacustrine (U-Series)
Faults (U-Series)
Laser Ablation Inductively Coupled Mass Spectrometer
(LA ICP-MS)
Mass Spectrometry
Secondary Ion Mass Spectrometry (SIMS)
Thermal Ionization Mass Spectrometer (TIMS)
Uranium Series, Opal
U-Series Dating
CARBONATES, SPELEOTHEM ARCHAEOLOGICAL
(U-SERIES)
Dirk Hoffmann
Department of Human Evolution, Max Planck Institute for
Evolutionary Anthropology, Leipzig, Germany
Definition
Deposits of secondary carbonate minerals (usually calcite
or aragonite consisting of CaCO3) that formed in caves
containing artefacts or other evidence of past human or
hominid occupation
Introduction
U-series dating such as U-Th (or 230Th/U) is a well-
established, accurate, and precise method with a wide
range of applications in earth sciences, palaeoclimate
research, and archaeology (Bourdon et al., 2003).
U-series dating can be used to determine the age of mate-
rials that formed with a well-constrained radioactive dis-
equilibrium between U and its daughter isotopes,
including precipitates of calcium carbonate (CaCO3) in
cave environments, i.e., speleothems like stalagmites, sta-
lactites, or flowstones. U-series dating is essential for
palaeoclimate research using speleothems but is less com-
monly used for archaeological applications. However,
cave environments are often associated with archaeology
and archaeological finds. Excavations in caves often
exhibit stratigraphic relationships with speleothem
formation. Where a stratigraphy can be unambiguously
established between archaeology and CaCO3 formation,
U-series dating provides a powerful chronological tool
and can be applied to constrain minimum and/or maxi-
mum ages for associated archaeological finds.
Basic principles of U-series dating of speleothems
U-Th dating can be applied to materials like secondary
CaCO3 that incorporate disequilibrium between U (238U
and 234U) and the daughter isotope 230Th at the time of for-
mation. For example, speleothems such as stalagmites or
flowstones precipitate from percolating waters entering
a cave. The water also contains trace elements dissolved
from host rock above the cave, including U, which are
subsequently incorporated in the CaCO3. The key for
U-Th dating is the difference in solubility between U and
Th which leads to elemental fractionation in the percolat-
ing water. In contrast to U, Th is largely insoluble and thus
not incorporated in secondary carbonates when they form
from drip waters in cave environments. Thereafter, 230Th
starts to build up until radioactive equilibrium is reached.
The return of isotope activities to equilibrium allows
quantification of time, i.e., the present 230Th/238U and
234 238
U/ U activity ratios enable calculation of the time
since formation.
The reliability of U-Th ages strongly depends on
whether the dated material has remained as a so-called
closed system, i.e., U and Th isotopes are neither lost
nor gained after precipitation, and whether any initial
230
Th was present. Presence of initial (or detrital) Th is
tested by determining the abundance of common Th
(232Th) in the sample. In case of significant amounts of
232
Th, a correction for initial 230Th needs to be applied
by mathematically subtracting an assumed or measured
detrital component or by using “isochron” methods
(Ludwig and Titterington, 1994).
A single U-Th age does not provide a means of
confirming that the dated material represents a closed sys-
tem. However, there are ways to check reliability of U-Th
ages of speleothems with respect to closed-system
behaviour. The most rigorous test is to employ the
alternative 235U-series chronometer and measure
231
Pa/235U in addition to 230Th/U on the same material
(Cheng et al., 1998). However, U-Pa dating requires much
larger sample sizes due to the significantly smaller concen-
trations of 235U relative to 238U (235U/238U ¼ 0.0073) and
thus cannot be employed in all cases, especially where sam-
ple size is limited. Alternatively, where open-system behav-
iour cannot be excluded, it is recommended to perform
repeat U-Th analyses of coeval samples. If dating results
on different subsamples are concordant, the system can be
regarded as closed. Stalagmites, flowstones, or other carbon-
ates that form with internal stratigraphic constraints provide
an additional possibility to check reliability by comparing
ages of subsamples precipitated successively along
a growth axis, which must become progressively younger
from bottom to top.
 CARBONATES, SPELEOTHEM ARCHAEOLOGICAL (U-SERIES)
U-series isotope measurements and sample sizes
The applicability of U-series methods depends on availabil-
ity of sufficient suitable material for dating. In many cases
only small samples of less than 100 mg CaCO3 are avail-
able, making the sample size critical. Recent advances in
thermal ionization mass spectrometry (TIMS) and multi-
collector inductively coupled plasma mass spectrometry
(MC-ICPMS) instrumentation and protocols not only led
to very high precision measurements of U-series ratios
(Andersen et al., 2004; Potter et al., 2005) but also greatly
reduced the sample size needed for a precise U-Th age. Cur-
rently, sample sizes between 10 and 100 mg are sufficient in
most cases (Hoffmann, 2008; Hoffmann et al., 2009). Over-
all, the ability to work on small samples in the range of
10 mg is now the key to many applications in archaeology
where only very small sections or thin layers of CaCO3 are
found to have direct stratigraphic relations to archaeological
finds.
U-series age constraints for excavations in cave
environments
U-series dating has been applied to speleothems from
archaeological sites for over 40 years employing alpha
spectrometry (e.g., Fornaca-Rinaldi, 1968; Schwarcz and
Blackwell, 1983; Blackwell et al., 1983), TIMS (e.g.,
Shen et al., 2001; Zhao et al., 2001) or more recently
MC-ICPMS techniques (e.g., Mercader et al., 2009;
Clark-Balzan et al., 2012; Hoffmann et al., 2013). Gener-
ally, sediment fillings in caves containing archaeology can
often be found on top of flowstone formations or are cov-
ered by a layer of flowstone or in some cases both. Flow-
stone thus formed before or after artefact-hosting
sediments accumulated, and U-series dating of such
speleothems provides maximum or/and minimum ages
for bracketed sediments. Similarly, stalagmites that form
locally on top of sediments at a drip site or that are covered
by sediment fill also may provide age constraints for the
sediments. The stratigraphy between carbonate deposits
and archaeology bearing sediments is essential; only car-
bonates with unambiguous relationships provide diagnos-
tic ages of the target sediments and should be analysed.
For capping flowstones, the lowest layer is closest in
age to the underlying sediment, and for underlying flow-
stones, the uppermost layer. However, the layers in contact
with sediments are also most prone to incorporation of sil-
icate detritus (dirt) within the carbonate matrix and there-
fore do not necessarily yield the best dating results. Thus,
in most cases, subsamples of detritus-poor carbonate
should be taken from a layer within several mm of the
flowstone–sediment boundary. It is generally
recommended to target subsamples from the most pristine
and cleanest parts of the speleothem. It is also essential to
perform repeat analyses of a single growth layer and/or
a suite of subsamples along the growth axis to verify the
presence of closed-system behaviour. In samples that con-
tain high detrital components, “isochron” methods may
need to be applied (Ludwig and Titterington, 1994).
145
Carbonates, Speleothem Archaeological (U-Series),
Figure 1 Example of subsamples and U-series results for
a section of flowstone (see text for details).
An example of a 3 cm thick flowstone section is shown
in Figure 1. A cross section of the flowstone was cut and
polished for better visibility of growth layers which can
clearly be distinguished by colour differences between
lighter and darker bands. The top of the flowstone is less
dense with a microcrystalline structure and dirt inclusions
indicating that this part of the specimen is not suitable for
U-Th dating. The bottom of the flowstone is denser but
also shows substantial amounts of dirt incorporated in
the calcite. Five subsamples were obtained from this sec-
tion by sawing cuts parallel to growth layers using
a diamond wire saw and breaking off small solid pieces
using a scalpel (subsamples 3–7). Two additional powder
samples (subsamples 1 and 2) were drilled from the top of
the section with a handheld micro drill.
The measured 230Th/232Th activity ratios shown in
Figure 1 indicate the degree of contamination with detrital
material. Typical bulk earth detritus has a 230Th/232Th
activity ratio of 0.8  0.4. The value of 1.5 for the topmost
subsample confirms that this part of the section has high
dirt content. Consequently, the resulting U-Th isotope data
are dominated by the detrital component, and a reliable
U-series age cannot be calculated for this subsample. Sub-
sample 2 also shows a high detrital component; however,
an age can be calculated for this analysis using a bulk earth
Th-correction, although the large correction leads to high
dating uncertainty. The results of subsamples 3–6 have
230
Th/232Th activity ratios >400 indicating that negligible
detritus is included. Calculated ages require little correc-
tion and are in stratigraphic order indicating reliable
results. In the case of this flowstone, U-Th dating con-
strains the underlying sediments to be older than 500 ka
and the overlying to be younger than 150 ka.
Buried stalagmites, stalactites, or soda straws have also
been suggested to provide age constraints for sediment
fills in caves (Schwarcz and Blackwell, 1992; St Pierre
et al., 2009). Speleothem fragments such as broken soda
straws can often be found mixed into cave sediments and
must have formed prior to the sediment accumulation.
U-series ages of these materials thus can provide
a maximum age of the sediment in which they are found.
146 CARBONATES, SPELEOTHEM ARCHAEOLOGICAL (U-SERIES)

Carbonates, Speleothem Archaeological (U-Series), Figure 2 Sampling of calcite crusts covering cave art: after identification of
suitable calcite formation with clear stratigraphic relationship to the cave painting, first the surface of the section is scraped off to
remove potential dust contamination and expose the calcite surface for inspection of mineral structure. Then calcite sample is
scraped off with a scalpel and collected either directly in a pre-cleaned sample container (left) or onto a clean tray. Scraping is stopped
as soon as paint becomes visible underlying the sample position (Photos: C. Hoffmann, J. Zilhão).

U-series age constraints for artefacts and cave art
In cave environments, artefacts (bones, tools) or archaeo-
logically relevant sections such as painted cave walls may
become exposed to percolating waters and subsequently
covered by secondary carbonates. In most cases, only very
thin layers of CaCO3 are found on artefacts, which poses
a major restriction for application of U-series dating. Reli-
able U-series ages can only be obtained where sufficient
amounts of CaCO3 can be removed from the artefacts
and where the CaCO3 has unambiguous stratigraphic rela-
tions to the artefacts. Since the calcite formation postdates
the archaeology, its age provides a minimum age con-
straint. For example, Frank et al. (2002) constrained the
age of Trojan artificial water-supply tunnels by dating cal-
cite that formed on the tunnel walls. Fu et al. (2008) used
calcite coatings to constrain the age of human skeleton
fragments from northeastern China.
U-series dating of calcite formations covering cave art
was first suggested by Schwarcz and Blackwell (1992)
as a future application to provide minimum ages for the
underlying art. In most cases, the amount of calcite coating
cave art with unambiguous stratigraphic relationship is
limited, and availability of sufficient sample size was the
major obstruction of U-series dating of cave art. A few
studies on cave art dating using TIMS have been
published, e.g., Bischoff et al. (2003), Plagnes et al. (2003),
or Pike et al. (2005). Dating cave art using U-series
methods is now possible for a much wider range of paint-
ings (motifs) again due to the significant reduction of sam-
ple material needed for precise U-series measurements
(Hellstrom, 2012). This enables to work on very small
carbonate samples allowing dating of tiny samples
scraped from calcite layers covering cave art (Figure 2).
For example, calcite crusts coating cave paintings from
over 50 motifs in northern Spain have been dated using
this method yielding minimum ages between 500 a and
40,800 a (Pike et al., 2012).
Summary and conclusions
In cave environments, archaeologically relevant sections
or artefacts can be found with stratigraphic relations to
secondary calcite deposits which can be dated by
U-series methods. State-of-the-art U-series techniques
minimize sample sizes and thus allow selection of the
most suitable calcite formations to provide many possibil-
ities for employing this technique in archaeological con-
texts. The stratigraphic relations between calcite
formation and archaeology need to be unambiguous in
order to use U-series ages of calcite to constrain the age
of archaeological finds and provide minimum or
maximum ages.
Bibliography
Andersen, M. B., Stirling, C. H., Potter, E. K., and Halliday, A. N.,
2004. Toward epsilon levels of measurement precision on
U-234/U-238 by using MC-ICPMS. International Journal of
Mass Spectrometry, 237, 107–118.
Bischoff, J., Garcia Diez, M., Gonzalez Morales, M. R., and
Sharp, W., 2003. Aplicacion del metodo de series de uranio al
grafismo rupestre de estilo paleolitico: el caso de la cavidad de
Covalanas. Veleia, 20, 143–150.
Blackwell, B., Schwarcz, H. P., and Debenath, A., 1983. Absolute
dating of hominids and paleolithic artifacts of the cave of La
 CARBONATES, SPELEOTHEM CLIMATIC (U-SERIES)
Chaise-de-Vouthon (Charente), France. Journal of Archaeologi-
cal Science, 10, 493–513.
Bourdon, B., Henderson, G. M., Lundstrom, C. C., and Turner, S. P.,
2003. Uranium-series Geochemistry. Washington, DC: Mineral-
ogical Society of America. Reviews in Mineralogy and Geo-
chemistry, Vol. 52.
Cheng, H., Edwards, R. L., Murrell, M. T., and Benjamin, T. M.,
1998. Uranium-thorium-protactinium dating systematics.
Geochimica et Cosmochimica Acta, 62, 3437–3452.
Clark-Balzan, L. A., Candy, I., Schwenninger, J. L., Bouzouggar,
A., Blockley, S., Nathan, R., and Barton, R. N. E., 2012.
Coupled U-series and OSL dating of a Late Pleistocene cave sed-
iment sequence, Morocco, North Africa: significance for
constructing Palaeolithic chronologies. Quaternary Geochro-
nology, 12, 53–64.
Fornaca-Rinaldi, G., 1968. 230Th/234U dating of cave concretions.
Earth and Planetary Science Letters, 5, 120–122.
Frank, N., Mangini, A., and Korfmann, M., 2002. Th-230/U dating
of the Trojan ‘water quarries’. Archaeometry, 44, 305–314.
Fu, R. Y., Shen, G. J., He, J. N., Ren, H. K., Feng, Y. X., and Zhao,
J. X., 2008. Modern Homo sapiens skeleton from Qianyang
Cave in Liaoning, northeastern China and its U-series dating.
Journal of Human Evolution, 55, 349–352.
Hellstrom, J., 2012. Absolute dating of cave art. Science, 336,
1387–1388.
Hoffmann, D. L., 2008. 230Th isotope measurements of femtogram
quantities for U-series dating using multi ion counting (MIC)
MC-ICPMS. International Journal of Mass Spectrometry, 275,
75–79.
Hoffmann, D. L., Spötl, C., and Mangini, A., 2009. Micromill and
in situ laser ablation sampling techniques for high spatial resolu-
tion MC-ICPMS U-Th dating of carbonates. Chemical Geology,
259, 253–261.
Hoffmann, D. L., Pike, A. W. G., Wainer, K., and Zilhão, J., 2013.
New U-series results for the speleogenesis and the Palaeolithic
archaeology of the Almonda karstic system (Torres Novas, Por-
tugal). Quaternary International, 294, 168–182.
Ludwig, K. R., and Titterington, D. M., 1994. Calculation of 230Th/
U isochrons, ages, and errors. Geochimica et Cosmochimica
Acta, 58, 5031–5042.
Mercader, J., Asmerom, Y., Bennett, T., Raja, M., and Skinner, A.,
2009. Initial excavation and dating of Ngalue Cave: A Middle
Stone Age site along the Niassa Rift, Mozambique. Journal of
Human Evolution, 57, 63–74.
Pike, A. W. G., Gilmour, M., Pettitt, P., Jacobi, R., Ripoll, S., Bahn,
P., and Munoz, F., 2005. Verification of the age of the
Palaeolithic cave art at Creswell Crags, UK. Journal of Archae-
ological Science, 32, 1649–1655.
Pike, A. W. G., Hoffmann, D. L., García-Diez, M., Pettitt, P. B.,
Alcolea, J., De Balbín, R., González-Sainz, C., De Las Heras,
C., Lasheras, J. A., Montes, R., and Zilhão, J., 2012. U-series
dating of paleolithic art in 11 caves in Spain. Science, 336,
1409–1413.
Plagnes, V., Causse, C., Fontugne, M., Valladas, H., Chazine, J. M.,
and Fage, L. H., 2003. Cross dating (Th/U-C-14) of calcite cov-
ering prehistoric paintings in Borneo. Quaternary Research, 60,
172–179.
Potter, E. K., Stirling, C. H., Andersen, M. B., and Halliday, A. N.,
2005. High precision Faraday collector MC-ICPMS thorium iso-
tope ratio determination. International Journal of Mass Spec-
trometry, 247, 10–17.
Schwarcz, H. P., and Blackwell, B., 1983. Th-230/U-234 age of
a Mousterian site in France. Nature, 301, 236–237.
Schwarcz, H. P., and Blackwell, B., 1992. Archaeological applica-
tions. In Ivanovich, M., and Harmon, R. S. (eds.), Uranium-series
Disequilibrium: Applications to Earth, Marine, and Environmen-
tal Sciences. Oxford: Oxford University Press, pp. 513–552.
147
Shen, G. J., Ku, T. L., Cheng, H., Edwards, R. L., Yuan, Z. X., and
Wang, Q., 2001. High-precision U-series dating of Locality 1 at
Zhoukoudian, China. Journal of Human Evolution, 41,
679–688.
St Pierre, E., Zhao, J. X., and Reed, E., 2009. Expanding the utility
of Uranium-series dating of speleothems for archaeological
and palaeontological applications. Journal of Archaeological
Science, 36, 1416–1423.
Zhao, J. X., Hu, K., Collerson, K. D., and Xu, H. K., 2001. Thermal
ionization mass spectrometry U-series dating of a hominid site
near Nanjing, China. Geology, 29, 27–30.
Cross-references
Carbonates, Speleothem Climatic (U-Series)
Mass Spectrometry
Sediments, Terrestrial (Paleomagnetism)
Thermal Ionization Mass Spectrometer (TIMS)
U-Series Dating
CARBONATES, SPELEOTHEM CLIMATIC (U-SERIES)
David A. Richards
School of Geographical Sciences, University of Bristol,
Bristol, UK
Synonyms
Carbonates: calcite, aragonite; Climatic: proxy climate
data; Speleothem: cave deposit, stalactite, stalagmite,
flowstone; U-series: 230Th/U dating, U-series disequilibrium
Definition
Carbonates. Strictly, a salt of the anion group CO32, but
here refers specifically to calcium carbonates, generally
associated with limestones (i.e., calcite and aragonite).
Speleothem. A secondary carbonate precipitate formed in
a cave void by degassing of CO2 from supersaturated
waters. By far the most common speleothems used for
the investigation of past climate are gravity-fed drip-
stones, such as stalagmites or flowstones formed in the
vadose zone, as opposed to phreatic examples formed at
or below the water table.
Uranium-series dating. Chronological tool based on the
state of disequilibrium in the U-series decay chains. Gen-
erally refers to 238U–234U–230Th and 235U–231 Pa
methods but can also be applied to U–Pb methods.
Introduction
Secondary carbonate precipitates in caves (speleothems)
provide an important archive of past environmental condi-
tions (see extensive review in Fairchild and Baker, 2012)
that can be related directly to past climate and accurately
dated by U-series (or U–Th–Pa and U–Pb disequilibrium)
methods (Richards and Dorale, 2003).
The presence of speleothem deposits is unambiguous
evidence that conditions in the karst catchment at the time
of formation were suitable for outgassing of carbonate-
148 CARBONATES, SPELEOTHEM CLIMATIC (U-SERIES)
saturated water; that certain threshold conditions, such as
permafrost (e.g., Vaks et al., 2013) or arid conditions
(e.g., Wang et al., 2004), did not prevail for significant
periods; or that cave passages were not flooded
(by rising sea levels or water table, e.g., Richards
et al., 1994). Thus, the presence or absence of calcite can
be indicative of past climate. Also, successive growth
layers in speleothems can preserve geochemical
(or proxy) evidence of the state of past climate conditions,
and a wide range of variables have been investigated at
resolutions from submicron to cm scale, including Mg,
Sr, Ba, and U elemental concentrations (e.g., Johnson
et al., 2006), molecular organic matter (e.g., Blyth
et al., 2008), stable isotopes of O, C (McDermott, 2004),
Mg (Galy et al., 2002), and Ca (Reynard et al., 2011) in
the solid phase and also stable isotopes within fluid inclu-
sions (e.g., Arienzo et al., 2013). Particularly important for
both geochronology and climatic interpretation is the fact
that most vadose speleothems comprise calcite, although
not exclusively so, which is a dense, crystalline, stable
mineral that is much less susceptible to postdepositional
alteration than other climate archives, such as corals, sed-
iments, and shells.
To successfully test models of past climate patterns in
both the spatial and temporal domain, we need robust
and accurate high-resolution evidence of leads, lags, or
synchronous change in proxy climate signals in widely
distributed archives. Increased attention on speleothems
in the past decade (Henderson, 2006) with a motivation
to obtain ever higher precision ages on smaller subsamples
has demanded constant reassessment of the basic assump-
tions of this dating technique, that is, accurate assessment
of the state of initial conditions, closed-system behavior,
and decay constants. Also, for most cases, the determina-
tion of axial growth history relies on a subsampling distri-
bution that is less dense for dating than for the climate
proxy of interest, which can be as high as sub-annual.
For this reason, interpolation and extrapolation methods
become just as important, in terms of age precision, as
the age determinations themselves.
Initial conditions
Uranium-series geochronology relies on the extreme
fractionation in the hydrosphere of the parent U isotopes
(238U, 235U, and 234U) from their long-lived daughters
231
Pa and 230Th. Uranium is highly mobile in the meteoric
environment as either the uranyl ion or stable carbonate
complexes, while the daughter products are readily
hydrolyzed and either precipitated or adsorbed on
detrital particulates (inorganic or organic), clay minerals, or
Fe-oxyhydroxides. Because of these very distinct chemical
behaviors in aqueous solutions, precipitation of calcite and
other carbonate minerals such as aragonite will incorporate
a state of disequilibrium in the 238U and 235U decay chains.
With time, the system will evolve toward secular equilib-
rium (when 238Ul238U ¼ 234Ul234U ¼ 230Thl230Th;
235
Ul235U ¼ 231Pal231Pa). With achievable precision for
U and Th isotope ratios of sub-permil (<1 %) using state-
of-the-art multi-collector mass spectrometers, finite ages
>500 ka are now routinely possible; however, the maximum
possible finite age is dependent on the amount of 234U
excess present at time of formation. While ages approaching
~0.8 Ma are possible with current technology, researchers
are more likely to turn to alternative techniques including
electron spin resonance, paleomagnetism, or U–Pb methods
for material of this age and older.
Critical for the U–Th–Pa disequilibrium suite of
methods is the assumption of negligible or known concen-
tration of daughter products at the time of formation. It is
recognized that there is always a component of Th or Pa
adsorbed on detrital clays or colloidal particles, and these
can be trapped in the calcite. The extent of detrital, colloi-
dal, or hydrogenous 230Th contamination is estimated by
determining the 232Th concentration and correcting for a
typical initial 230Th/232Th activity ratio. For many samples,
where silicates and clays are the principal contaminant, a
bulk Earth activity ratio of 0.8  0.4 is commonly used.
However, this assumption should be tested by using three-
dimensional isochron techniques to derive the local initial
230
Th/232Th ratio (Ludwig and Titterington, 1994). In
numerous settings, radiogenic activity ratio values have
been derived, as much as 10–100 times greater than bulk
Earth (see Beck et al., 2001; Meckler et al., 2012).
A wide range of U concentrations, perhaps five orders
of magnitude, has been observed in groundwaters, and this
is reflected in the range of concentrations in secondary
carbonate minerals, including speleothems. In general,
aragonitic speleothems have higher U concentrations, by
virtue of the higher partition coefficient, but calcite
speleothems with concentrations >100 mg g1 have been
analyzed. Not surprisingly, higher U concentrations
enable smaller sample sizes for the same effective preci-
sion. Indeed, researchers are now turning to laser ablation
methods, especially in higher-concentration materials, to
take advantage of the improved spatial resolution of
analysis.
Closed-system behavior
Speleothems comprise a wide variety of forms and miner-
alogical compositions; indeed, some samples exhibit
repeated shifts in mineralogy associated with changing
hydrological and/or climatological conditions. Many sub-
samples are not suitable for age determination because
they exhibit, in section, zones of recrystallization or disso-
lution fronts. Not all samples provide primary, dense, crys-
talline calcite or aragonite throughout the growth axis, so
the analyst must be careful when selecting specific hori-
zons to date. Occasionally, primary aragonite may be
recrystallized to a more stable form of calcite. If this
occurs rapidly, the apparent age may be useful (e.g., Frisia
et al. 2006). It is expected that material closest to hiatuses
in growth might be susceptible to postdepositional alter-
ation, yet this is often the most important calcite because
it constrains the timing of initiation or cessation of
 CARBONATES, SPELEOTHEM CLIMATIC (U-SERIES)
continuous growth. Also, because sample sizes routinely
used for multi-collector ICP-MS measurements are small
(<30 mg for material with U concentrations of 100 ng g1),
zones of alteration can easily bias the apparent age. For
this reason, dense sampling and duplicate analyses from
cogenetic horizons are expected to ensure closed-system
behavior of U, Th, and Pa and provide the most accurate
determination of the growth histories.
Decay constants
Uranium-series dating relies on accurate determination of
the decay constants involved and is ultimately limited, in
terms of precision, by the uncertainties associated with
these values. Different strategies have been used to deter-
mine these values, the most fundamental of which are
direct decay counting or ingrowth. However, a recent
update to the decay constants for 234U and 230Th relied
on a different strategy (Cheng et al., 2013) that measured
the U–Th isotopic ratios in numerous materials that were
assumed to be at secular equilibrium and calibrated those
measurements against an accurately determined gravimet-
ric standard derived from U and Th metals of known
purity. The suite of samples used included speleothems
from the Alps, Australia, and South Dakota with ages
ranging from 2 to 26 Ma based on U–Pb dating methods
and also on vein and mega-crystalline calcite assumed to
be much older. Using these improved values for decay
constants, age uncertainties of 5–10 ka are achievable
for materials older than 500 ka if U concentrations are
>1 ppm. This degree of precision is sufficient to test the
phasing relationships between records of stable isotopic
composition (specifically, d18O) in speleothem climate-
proxy records and calculations of insolation forcing in
the middle Pleistocene (Cheng et al., 2013; see Figure 1).
Measurement techniques
Alpha decay counting is still utilized in a few laboratories,
but by far the most commonly used platform for analysis is
mass spectrometry because of the reduced sample sizes
and greater achievable precision. Ionization methods
include thermal or inductively coupled plasma (with solu-
tion or laser introduction), and a combination of Faraday
and ion counters is used for collection in static or peak
jumping mode, depending on ion beam intensity and
required precision. Ionization and transmission efficiency
are such that U and Th analysis can be conducted on
speleothem subsamples <50 mg (based on a typical U con-
centration of 300 ng g1). U and Th are separated from the
carbonate matrix using extraction chromatography or
ion-exchange methods prior to mass spectrometric analy-
sis. Age uncertainty is a function of a multitude of param-
eters including sample size, age, U concentration, beam
stability, decay constant uncertainty, and blank
contribution.
Laser ablation techniques offer some distinct advan-
tages over solution introduction protocols (see Eggins
et al., 2005), although age precisions cannot be matched.
149
If U concentrations are suitable, subsample resolution
can be significantly reduced, such that individual domains
of <100 mm can be analyzed. Also, sample preparation is
more straightforward (no sample digestion or chemical
purification). High-resolution maps of Sr, Ba, Mg,
and U based on laser ablation can also reveal the internal
architecture of samples, or growth layers, where none
were visible petrographically (Drysdale et al., 2012).
These can then guide micro-milling of samples at the sub-
millimeter scale for subsequent high-precision U-series
solution work.
Constructing growth histories
Chronologies of past climate based on speleothems rely,
for the most part, on multiple ages for subsamples that
are strategically selected from along the growth axis (see
Figure 1). Subsampling density for the climate proxy of
interest is generally at a much greater resolution than that
used for dating, and thus, interpolation is required to esti-
mate the age for each lamination formed along the axis.
A variety of statistical strategies are employed, including
linear interpolation, automatic smoothing splines, Monte
Carlo simulation, and Bayesian methodology (for discus-
sion, see Scholz et al., 2012). There is no standard method,
and each has advantages and disadvantages with respect to
outliers, treatment of hiatuses, and rapid changes in
growth rates. Subsamples close to hiatuses are the most
susceptible to alteration or contamination because they
may have been exposed for extended periods of time;
however, they are often the most important because they
define the timing of thresholds in local hydrological or cli-
mate regime (arid/humid; submergence/emergence).
Robust ages have been obtained for submillimeter wafers
from former surfaces of flowstones and stalagmites
deposited in underwater caves, where the calcite would
have been submerged for many thousands of years (see
Figure 2). In spite of the potential problems, many robust
chronologies have been developed for terrestrial stalag-
mite climate-proxy records that match or improve upon
the resolutions of records determined from ice and ocean
sediment cores, for example.
Comparison with other dating methods
Ideally, to determine the accuracy of or improve confi-
dence in the constructed growth histories of speleothem
climate-proxy records, results determined by U–Th
methods should be compared with other dating techniques
for the same material. In practice, this is difficult because
of the much greater precisions achievable using U-series
methods. Radiocarbon methods would be useful in this
regard for ages <50 ka (e.g., Hoffmann et al., 2010), but
the precisions of 14C dates in carbonates are not compara-
ble because of the variability in dead carbon fraction cor-
rection that must be accommodated. Having said this,
U–Th dated records (corals and speleothems) play a sig-
nificant role in calibration of the atmospheric radiocarbon
curves, particularly beyond the limit of overlapping tree-
150 CARBONATES, SPELEOTHEM CLIMATIC (U-SERIES)

Carbonates, Speleothem Climatic (U-Series), Figure 1 Sanbao d18O time series, based on the U–Th age-distance relationship using
newly determined half-lives (Cheng et al., 2013). The climate-proxy data broadly matches the 23 ka cycles in the summer
insolation curve (60
N June, Berger, and Loutre, 1991). Routine uncertainties of <5 ka are required to test the phasing of insolation
forcing and climate response. Also, the good agreement illustrated is a strong support for the accuracy of the recent
determinations of 234U and 230Th half-lives.

ring records (~15 ka). Additional dating methods that can
be applied to speleothems include paleomagnetism, elec-
tron spin resonance, and Pb-210. Note that U–Pb dating
methods can be applied to speleothems and are also
included in the suite of disequilibrium methods (see
Woodhead et al., 2006).
Conclusions
Uranium-series dating has provided excellent chronologi-
cal constraints for high-resolution climate, sea level, and
permafrost records preserved in secondary carbonates,
especially speleothems. These have been compared
directly with other climate-proxy records from ice cores,
ocean cores, and lake sediments, for example, to test our
understanding of the dynamics of past states of the Earth
system. Mass spectrometric methods provide sufficient
sensitivity to give uncertainties to the nearest year for his-
torical archives and, perhaps, a few thousand years for
ages up to ~500 ka. The technique can be applied for as
long as a state of radioactive disequilibrium can be distin-
guished using high-precision analytical techniques on
closed-system material and can be extended to include
U–Pb methods, which have no effective age limit, if mate-
rial has relatively high U and low Pb concentrations.
 CARBONATES, SPELEOTHEM CLIMATIC (U-SERIES)
Carbonates, Speleothem Climatic (U-Series), Figure 2 Hiatus
in speleothem flowstone sequence (GB-89-25-5A) collected
from 18 m below present sea level in a Bahamas cave
(Sagittarius; see Richards et al., 1994). The upper surface of
calcite that grew during Marine Isotope 6 (MIS 6; penultimate
glaciation >140 ka) was exposed to seawater, freshwater, and air
in this inland blue hole prior to reinitiation of growth at ~80 ka.
In spite of this, crystal terminations remain well preserved (see
arrow) and U–Th results show no sign of age inversion (Scale bar,
1 mm width).
Bibliography
Arienzo, M. M., Swart, P. K., and Vonhoff, H. B., 2013. Measure-
ment of d18O and d2H values of fluid inclusion water in
speleothems using cavity ring-down spectroscopy compared
with isotope ratio mass spectrometry. Rapid Communications
in Mass Spectrometry, 27, 2616–2624.
Beck, J. W., Richards, D. A., Edwards, R. L., Silverman, B. W.,
Smart, P. L., Donahue, D. J., Hererra-Osterheld, S., Burr, G. S.,
Calsoyas, L., Jull, A. J. T., and Biddulph, D., 2001. Extremely
large variations of atmospheric 14C concentration during the last
glacial period. Science, 292, 2453–2458.
Berger, A., and Loutre, M. F., 1991. Insolation values for the climate
of the last 10 million years. Quaternary Science Reviews, 10,
297–317.
Blyth, A. J., Baker, A., Collins, M. J., Penkman, K. E. H., Gilmour,
M. A., Mosse, J. S., Genty, D., and Drysdale, R. N., 2008. Molec-
ular organic matter in speleothems and its potential as an environ-
mental proxy. Quaternary Science Reviews, 27, 905–921.
Cheng, H., Edwards, R. L., Shen, C.-C., Polyak, V. J., Asmerom, Y.,
Woodhead, J., Hellstrom, J., Wang, Y., Kong, X., Spötl, C.,
Wang, X., and Alexander, E. C., 2013. Improvements in 230Th
dating, 230Th and 234U half-life values, and U–Th isotopic mea-
surements by multi-collector inductively coupled plasma mass
spectrometry. Earth and Planetary Science Letters, 371–372,
82–91.
Drysdale, R. N., Paul, B. T., Hellstrom, J. C., Couchoud, I., Greig,
A., Bajo, P., Zanchetta, G., Isola, I., Spötl, C., Baneschi, I.,
Regattieri, E., and Woodhead, J. D., 2012. Precise
microsampling of poorly laminated speleothems for U-series
dating. Quaternary Geochronology, 14, 38–47.
Eggins, S. M., Grün, R., McCulloch, M. T., Pike, A. W. G.,
Chappell, J., Kinsley, L., Mortimer, G., Shelley, M., Murray-
Wallace, C. V., Spötl, C., and Taylor, L., 2005. In situ U-series
dating by laser-ablation multi-collector ICPMS: new prospects
151
for Quaternary geochronology. Quaternary Science Reviews,
24, 2523–2538.
Fairchild, I. J., and Baker, A., 2012. Speleothem Science: From Pro-
cess to Past Environments. Chichester: Wiley-Blackwell.
416 pp. ISBN 978-1-4051-9620-8.
Frisia, S., Borsato, A., Fairchild, I. J., McDermott, F., and Selmo,
E. M., 2002. Aragonite-calcite relationships in speleothems
(Grotte de Clamouse, France): environment, fabrics and carbon-
ate geochemistry. Journal of Sedimentary Research, 72,
687–699.
Galy, A., Bar-Matthews, M., Halicz, L., and O’Nions, R. K., 2002.
Mg isotopic composition of carbonate: insight from speleothem
formation. Earth and Planetary Science Letters, 201, 105–115.
Henderson, G. M., 2006. Caving in to new chronologies. Science,
313, 620–622.
Hoffmann, D. L., Beck, J. W., Richards, D. A., Smart, P. L.,
Singarayer, J. S., Ketchmark, T., and Hawkesworth, C. J.,
2010. Towards radiocarbon calibration beyond 28 ka based on
speleothems from the Bahamas. Earth and Planetary Science
Letters, 289, 1–10.
Johnson, K. R., Hu, C., Belshaw, N. S., and Henderson, G. M.,
2006. Seasonal trace-element and stable-isotope variations in a
Chinese speleothem: the potential for high-resolution
paleomonsoon reconstruction. Earth and Planetary Science Let-
ters, 244, 394–407.
Ludwig, K. R., and Titterington, D. M., 1994. Calculation of
230
Th/U isochrons, ages and errors. Geochimica et
Cosmochimica Acta, 58, 5031–5042.
McDermott, F., 2004. Palaeo-climate reconstruction from stable
isotope variations in speleothems: a review. Quaternary Science
Reviews, 23, 901–918.
Meckler, A. N., Clarkson, M. O., Cobb, K. M., Sodemann, H., and
Adkins, J. F., 2012. Interglacial hydroclimate in the tropical West
Pacific through the late Pleistocene. Science, 336, 1301–1304.
Reynard, L. M., Day, C. C., and Henderson, G. M., 2011. Large
fractionation of calcium isotopes during cave-analogue calcium
carbonate growth. Geochimica et Cosmochimica Acta, 75,
3726–3740.
Richards, D. A., and Dorale, J. A., 2003. Uranium-series chronol-
ogy and environmental applications of speleothems. In Bourdon,
B., Henderson, G. M., Lundstrom, C. C., and Turner, S. P. (eds.),
Uranium-series Geochemistry. Washington, DC: Mineralogical
Society of America. Reviews in Mineralogy and Geochemistry,
Vol. 52, pp. 407–460.
Richards, D. A., Smart, P. L., and Edwards, R. L., 1994. Maximum
sea levels for the last glacial period from U-series ages of sub-
merged speleothems. Nature, 367, 357–360.
Scholz, D., Hoffmann, D. L., Hellstrom, J., and Bronk Ramsey,
C. B., 2012. A comparison of different methods for speleothem
age modelling. Quaternary Geochronology, 14, 94–104.
Vaks, A., Gutareva, O. S., Breitenbach, S. F. M., Avirmed, E.,
Mason, A. J., Thomas, A. L., Osinzev, A. V., Kononov, A. M.,
and Henderson, G. M., 2013. Speleothems reveal 500,000-year
history of Siberian permafrost. Science, 340, 183–186.
Wang, X. F., Auler, A. S., Edwards, R. L., Cheng, H., Christalli,
P. S., Smart, P. L., Richards, D. A., and Shen, C.-C., 2004. Wet
periods in northeastern Brazil over the past 210 kyr linked to dis-
tant climate anomalies. Nature, 432, 740–743.
Woodhead, J., Hellstrom, J., Maas, R., Drysdale, R., Zanchetta, G.,
Devine, P., and Taylor, T., 2006. U–Pb geochronology of
speleothems. Quaternary Geochronology, 1, 208–221.
Cross-references
210
Pb Dating
Carbonates, Marine Carbonates (U-Series)
Carbonates, Speleothem Archaeological (U-Series)
152 CHEMICAL WEATHERING (U-SERIES)
Corals (Sclerochronology)
Dendrochronology, Progress
Electron Spin Resonance (ESR) Dating, General Principles
Laser Ablation Inductively Coupled Mass Spectrometer
(LA ICP-MS)
Magnetic Chronology
Mass Spectrometry
Radiocarbon Dating of Marine Carbonates
Radiocarbon Dating of Terrestrial Carbonates
Sea Level Change (U-Series)
U-Series Dating
CHEMICAL WEATHERING (U-SERIES)
Anthony Dosseto
Wollongong Isotope Geochronology Laboratory, School
of Earth and Environmental Sciences, University of
Wollongong, Wollongong, NSW, Australia
Synonyms
Regolith formation rates; Soil formation rates
Definition
Uranium-series (U-series) isotopes are radionuclides that
compose the 238U and 235U decay chains. Measurements
of U-series isotopes in soils and river sediments can be
used to infer timescales of chemical weathering and derive
quantitative estimates of soil age and production rates as
well as sediment transport rates through catchments.
Historical material
U-series isotopes in geological materials have received
significant interest since the 1960s (Adams et al., 1959;
Pliler and Adams, 1962; Cherdyntsev et al., 1964;
Rosholt et al., 1966; Syromyatnikov et al., 1967; Hansen
and Stout, 1968; Scott, 1968; Syromyatnikov and
Ivanova, 1968; Cherdyntsev, 1969; Titayeva and Veksler,
1977) or even earlier, when investigations were mostly
supported by governmental programs motivated by the
development of raw materials for nuclear weapons and
power (Adams et al., 1959). The first studies were inter-
ested with the mobility of actinides in weathering profiles
(Pliler and Adams, 1962; Rosholt et al., 1966; Hansen and
Stout, 1968; Hansen and Huntington, 1969; Moreira-
Nordemann, 1977, 1980), while others were investigating
the radioactive disequilibrium associated with uranium
ore deposits (Cherdyntsev et al., 1964; Syromyatnikov
et al., 1967; Lively et al., 1979), in groundwater
(Osmond and Cowart, 1976, 1982; Cowart and Osmond,
1977; Krishnaswami et al., 1982; Osmond et al., 1983)
or in streams and estuaries (Scott, 1968; Martin
et al., 1978; Mangini et al., 1979). Early studies in nuclear
physics examined the mechanisms for U-series isotope
fractionation (e.g., Linhard and Scharff, 1961; Harvey,
1962). Later work would explore how these processes
could account for radioactive disequilibrium in geological
materials with a focus on recoil and preferential leaching
effects (Kigoshi, 1971; Fleischer and Raabe, 1978a;
Fleischer and Raabe, 1978b; Fleischer, 1980, 1982a;
Fleischer, 1982b, 1983; Sheng and Kuroda, 1984; Hussain
and Lal, 1986; Dran et al., 1988; Adloff and Roessler,
1991). Despite a large body of early work on U-series iso-
tope fractionation during chemical weathering, the 1980s
and 1990s saw a decrease in interest for this application.
However, the topic has regained interest since the late
1990s and early 2000s, in part thanks to improved chemi-
cal separation methods (Horwitz et al., 1993; Chabaux
et al., 1994; Horwitz, 1996; Bourdon et al., 1999) and
the development of plasma source mass spectrometry
allowing the precise measurement of low abundance
U-series isotopes (e.g., Turner et al., 2001).
Essential concepts and applications
Uranium-238 (238U) decays through a series of radioactive
products to 206Pb as the final, stable isotopic product
(Figure 1). The 238U decay chain is composed of a series
of daughter-parent systems where each daughter nuclide
is the result of alpha or beta disintegration of the parent
nuclide. Although “uranium-series” refer to both 238U and
235
U decay chains, most studies focus on the decay prod-
ucts in the 238U decay series, in particular three isotopes:
238
U, the parent nuclide of the decay series; 234U, which
is produced after alpha decay of 238U into 234Th followed
by rapid decays to 234 Pa (by beta emission of 234Th with
a half-life of 24.1 days) and 234U (by beta emission of 234
Pa with a half-life of 6.67 h); and 230Th, produced by alpha
decay of 234U with a half-life of 2.46  105 a.
For any geological system that remains closed to mass
exchange for more than a million years, all radioactive
systems in the decay chain are in secular equilibrium,
i.e., the activity of the daughter nuclide equals that of the
parent. The activity of a nuclide is its rate of decay, i.e.,
the product between its abundance and its decay constant.
Results are generally expressed in terms of daughter-to-
parent activity ratios, for instance, (234U/238U) and
(230Th/238U), with the use of parentheses indicating activ-
ities. Thus, if a system is in secular equilibrium, these
ratios are equal to unity.
A variety of geological processes induce fractionation
between U-series isotopes, termed radioactive disequilib-
rium, in which case the activity ratio deviates from unity.
Once a disequilibrium is produced, the parent-daughter sys-
tem will tend to return to secular equilibrium via radioactive
decay, over a timescale that is about five times the half-life
of the daughter nuclide (for instance, 1.2 Ma for 238U-234U,
375 ka for 234U-230Th). Thus, the value of the activity ratio
in a geological sample is a function of (i) the behavior of
isotopes during geological processes and (ii) the time
elapsed since fractionation for a discrete process or the rate
of fractionation for a continuous process.
Mechanisms of U-series disequilibrium
Uranium can be found in two stable oxidation states,
4 and 6, while the only stable oxidation state for thorium
 CHEMICAL WEATHERING (U-SERIES) 153

Chemical Weathering (U-Series), Figure 1 Uranium-238 decay chain. Each nuclide half-life is shown under its name and
expressed in years (yr), days (d), hours (hr), or minutes (min). Two types of radioactive decay occur in the chain: a (decay energy
shown under the symbol) and b.

is tetravalent. At low oxygen fugacity (i.e., reducing con- magmas (Harmon and Rosholt, 1982; Bourdon
ditions as it is common in the Earth’s interior), uranium is et al., 2003; Lundstrom, 2003; Turner et al., 2003b;
tetravalent and has properties similar to thorium, although Dosseto et al., 2010). Under oxidizing conditions (as it is
differences exist which yield 230Th-238U disequilibrium in commonly the case at the Earth’s surface), U is present
154 CHEMICAL WEATHERING (U-SERIES)

in its hexavalent state and forms the uranyl ion UO22+.

This ion can form stable carbonyl complexes, which
accounts for uranium’s solubility in natural waters. In con-
trast, thorium is relatively insoluble at pH values typical of
surface waters (5–8), and it is soluble only in the presence
of organic acids at low pH (<3), via binding to organic
colloids (Langmuir and Herman, 1980; Viers
et al., 1997). As a consequence of these chemical behav-
iors, solutes (e.g., groundwater, soil pore water, and
stream water) are generally enriched in U relative to Th
and are characterized by (230Th/238U) activity ratios less
than 1. Weathering products (hereby restricted to solid
weathering products, i.e., weathered rocks, soils, stream
sediments) are generally enriched in Th relative to U and
show (230Th/238U) activity ratios greater than
1 (Syromyatnikov and Ivanova, 1968; Titayeva and
Veksler, 1977; Latham and Schwarcz, 1987).
Although 234U and 238U theoretically have the same
chemical behavior, as illustrated by the absence of signif-
icant fractionation in igneous rocks, fractionation between
these two isotopes occurs at the Earth’s surface, resulting
in an enrichment of 234U in natural waters and a depletion
in weathered rocks, soils, and sediments (Syromyatnikov
and Ivanova, 1968; Titayeva and Veksler, 1977; Latham
and Schwarcz, 1987). This fractionation occurs as a conse-
quence of two processes: (i) Direct recoil of 234Th out of
the mineral grain during alpha decay of 238U and subse-
quent beta decays into its granddaughter 234U (Figure 1; Chemical Weathering (U-Series), Figure 2 Schematic
Kigoshi, 1971). This occurs because when 238U decays representation of various scenarios of alpha decay. In (a), case
into 234Th, the daughter is displaced by 15–35 nm (the (a1) occurs when 234Th (black dot) is directly recoiled out of the
source mineral (red circle). In case (a2), the direction of recoil
recoil length) in most minerals (Hashimoto et al., 1985). from 238U (gray dot) is such that 234Th is embedded further into
If this occurs within a recoil length from the mineral sur- the source mineral. In case (a3), 234Th is embedded in an
face, a fraction of 234Th can be lost to the surrounding adjacent mineral (blue circle). In all cases, 234Th then decays
medium (air or water; Figure 2 case (a1)). (ii) Preferential rapidly into 234U. In (b), a fluid fills in the pore space. In case (b1),
leaching of 234U embedded in recoil tracks (Fleischer, the 234Th is directly recoiled into the fluid which readily
1980, 1982b). The fraction of recoiled 234Th that is not transports the subsequently formed 234U. In case (b2), mineral
dissolution (light red area) exposes a recoil track such that the
directly ejected into the surrounding medium is embedded embedded 234U can be leached out of the track. In case (b3), the
in recoil tracks either deeper the source mineral (Figure 2 recoil track in the adjacent mineral is exposed to the fluid and
case (a2)) or into adjacent minerals (Figure 2 case (a3)) the embedded 234U leached away.
especially when the pore space is filled with air (Sun and
Furbish, 1995). When a solution fills the pore space, the
234
U produced by decay is leached out of the exposed
234
recoil tracks in adjacent minerals (Figure 2 case (b3)) U in residual solids (as it is the case below), such differ-
(Fleischer, 1980; Andersen et al., 2009) or from newly entiation is not necessary. Indeed, assessments of recoiled
234
exposed recoil tracks as a result of dissolution of the Th can overestimate the actual amount of 234U lost to
source mineral (Figure 2 case (b2)). Complete leaching the solution if a significant proportion is embedded into
of embedded nuclides occurs over a timescale as short as adjacent grains; however, it is exactly because these
200 years (Fleischer, 1980). embedded nuclides are subsequently leached from recoil
Initially, it was proposed that by determining the frac- tracks that eventually all recoiled 234Th end up being lost
tion of 234Th directly recoiled out of minerals, one could from the minerals (whether directly recoiled in the water
quantify the supply rate of 234U to the solution leaching or embedded to another grain and later leached).
these minerals (Kigoshi, 1971). However, it was later pos- As mentioned above, radioactive disequilibrium is gen-
tulated that preferential leaching of embedded 234U is erally reported as the ratio of daughter-to-parent activity,
another important mechanism for the delivery of 234U to e.g., (234U/238U). The activity ratios used to report U-Th
solutions (Fleischer, 1980). It is important to note that disequilibrium in the literature vary: some studies report
although differentiating both mechanisms is important to (230Th/234U), while others use (230Th/238U). Since 230Th
accurately study the enrichment of solutions in 234U over is the produce of 234U decay, it would seem more sensible
238
U, when studying the complementary depletion in to report (230Th/234U) ratios. However, if the aim is to
 CHEMICAL WEATHERING (U-SERIES)
study the U-Th disequilibrium imparted by the difference
in mobility of U and Th during mineral dissolution, this
ratio may not be appropriate: 234U is sensitive not only
to U mobilization during dissolution but also to direct
recoil and preferential leaching of 234U embedded in recoil
tracks. 230Th is also sensitive to these two processes, but
because of the high reactivity of Th with particle surfaces,
it is highly likely that the 230Th released by recoil or pref-
erential leaching will rapidly re-adsorb on mineral grains.
As a consequence, the (230Th/238U) activity ratio may be a
more appropriate measure of the U-Th disequilibrium that
occurs during mineral dissolution.
Occurrence of U-series nuclides in weathering products
Weathering products can be simplified as the mixture of
four components: (i) primary minerals (i.e., mineral
phases from the parent rock), (ii) secondary minerals pro-
duced during incongruent dissolution of primary minerals,
(iii) secondary minerals produced by precipitation from
soil pore water, and (iv) organic matter. Our understanding
of the occurrence of U-series disequilibrium is bound to
the distribution of U in these components.
As reviewed in Chabaux et al. (2003), uranium can be
sorbed onto mineral surfaces with a sorption efficiency
ranked as follows: Fe oxides and silica gels > clays and
micas > opals (Ames et al., 1983a). For Fe oxides,
the sorption efficiency is ordered as follows: amorphous
Fe oxyhydroxides > goethite > hematite (Ames
et al., 1983a, c; Hsi and Langmuir, 1985; Manceau
et al., 1992; Waite et al., 1994). For clays, U sorption effi-
ciency is in the following order:
montmorillonite > illite > kaolinite (Borovec, 1981;
Ames et al., 1983a; Shirvington, 1983). U can also be
sorbed on micas, mostly on muscovite, with a low sorption
efficiency for biotite and phlogopite (Ames et al., 1983b).
Near-neutral pH conditions tend to favor U sorption, while
in acidic conditions, U sorption is enhanced by the pres-
ence of humic acids (Lenhart and Honeyman, 1999).
It has been shown that in some cases, organic acids can
control the U budget of weathering products (Plater
et al., 1992; Porcelli et al., 1997; Andersson et al., 1998;
Dequincey et al., 2002; Chabaux et al., 2003;
Vigier et al., 2005; Dosseto et al., 2006a; Dosseto
et al., 2006b). Plater et al. (1992) showed that for bed load
sediments from the Witham River in the UK, the organic
fraction accounts for 2–12 wt% of the total U budget,
while amorphous and crystalline Fe-Mn oxyhydroxides
represent 10–20 wt%. They found that resistate minerals,
i.e., primary minerals, concentrate 62–85 wt% of the U in
the sediments.
In primary minerals, U is mostly found in accessory
phases with zircon, xenotime, or monazite exhibiting con-
centrations as high as several 1,000s of ppm, while ilmen-
ite and magnetite have U concentrations between a few
ppm and several 10s of ppm, and apatite between 5 and
150 ppm (Adams et al., 1959). Major minerals like quartz
or feldspars contain sub-ppm to ppm levels of U, most
likely in liquid inclusions or submicroscopic inclusions
155
of accessory minerals (Adams et al., 1959). Other major
minerals like micas or amphiboles can contain up to 10s
of ppm of U. Interstitial U, i.e., at grain boundaries, along
cleavages, and in microfractures, is another significant
pool of uranium in igneous parent rocks (Tieh and Ledger,
1981; Guthrie and Kleeman, 1986). Guthrie and Kleeman
(1986) showed that during the early stages of granite
weathering, “background” U (i.e., contained in quartz
and feldspars) is redistributed to the interstitial pool, and
a net loss of U occurs via (i) decrease of the interstitial
U pool and (ii) dissolution of accessory minerals. While
the model abundance of these accessory minerals
decreases, their U concentration remains constant.
As shown below, U-series disequilibrium has been used
in weathering products to determine the time elapsed since
conversion of the parent rock into soils and sediments. In
this respect, it is important to note that the behavior of pri-
mary minerals during weathering is as important as their
U content: to impart radioactive disequilibrium to the
weathering product, a mineral must not only (i) contain a
significant amount of U but must also (ii) actively dissolve
yet (iii) survive chemical weathering. On the one hand, for
a U-rich mineral like zircon, because it is highly resistant
to chemical weathering, it is likely that no significant loss
of nuclides occur via dissolution (Figure 3). However, it
has been shown that intense metamictization can favor zir-
con weatherability (Balan et al., 2001). Note also that
recoil can implant significant amounts of 234U and 230Th
to adjacent minerals that may actively dissolve and this
can result in a net loss of these nuclides. On the other hand,
a U-rich mineral that readily dissolves and is consumed
early during weathering will have no contribution to the
(234U/238U) and (230Th/238U) ratios of the weathering
product since all its U-series nuclides will have been lost
to the solution (Figure 3). Thus, the U-series isotope com-
position of primary minerals and their products is con-
trolled by the U-rich minerals that actively dissolve but
survive the weathering process (Figure 3). This was
recently confirmed by a study of soil profiles in southeast-
ern Australia which found a correlation between the soil
residence time inferred from U-series isotopes and the soil
muscovite content (Suresh et al., 2011, 2013).
Models of U-series disequilibrium in weathering products
The time-sensitive property of U-series isotopes offers the
perspective of being able to determine time constraints on
weathering processes. Early studies investigated the qual-
itative potential of mechanisms to account for observed
U-series isotope compositions in soils (e.g., Rosholt
et al., 1966; Rosholt, 1982). Latham and Schwarcz
(1987) and Scott et al. (1992) later developed quantitative
models that would describe the evolution of nuclide abun-
dances in weathering products.
In Latham and Schwarcz (1987), a uranium-leach
model was proposed to account for (234U/238U) 1 and
(230Th/238U) >1 in weathered granitic rocks. In this
model, the abundance of a daughter nuclide, d, in the solid
(rock, soil, or sediment) varies with time as a function of
156 CHEMICAL WEATHERING (U-SERIES)

Chemical Weathering (U-Series), Figure 3 Schematic representation of the role of the different components of weathering
products to their U-series isotope composition. U-poor minerals (e.g., quartz) do not contribute to the radioactive disequilibrium of
soils or sediments because of their low U concentration. U-rich minerals that weather readily (e.g., monazite) do not contribute to the
radioactive disequilibrium of soils or sediments because they are entirely dissolved and their entire 238U, 234U, and 230Th isotope
budget is transferred to the soil-forming solution. Insoluble U-rich minerals (e.g., zircon) do not contribute to radioactive
disequilibrium because they are highly resistant to weathering. However, these minerals can contribute to radioactive disequilibrium
if (i) they are highly metamict, which favors release of 238U and its decay products, and (ii) they implant 234U and 230Th to adjacent
minerals as a result of decay. U-rich primary minerals that weather, but survive weathering (e.g., muscovite or apatite), as well as
secondary minerals derived from the incongruent dissolution of these primary minerals provide the dominant control on the U-series
isotope composition of soils or sediments. In addition, authigenic phases (as secondary minerals precipitated from soil pore water)
and organic matter inherit the U-series isotope composition of the solution from which they formed and can also contribute to
the U-series isotope composition of the soil or sediments.

the amount of nuclides gained by radioactive decay of the
parent, p, and the amount lost by their own decay and via
dissolution:
dN d
¼ lp  N p  ld  N d  wd  N d
dt
where Np and Nd are the number of atoms of the parent and
daughter nuclides, respectively; lp and ld are the decay
constants for the parent and daughter nuclides, respec-
tively (in a1); and Wd is a coefficient that represents the
combined loss of the daughter nuclide during mineral dis-
solution and as a result of recoil when applicable (in a1).
For 238U, the first two terms on the right-hand side are
absent because there is no parent, and 238U decay can be
neglected over the timescales considered (<1 Ma).
Scott et al. (1992) proposed a similar weathering model
that also explicitly accounts for the loss via recoil of a
daughter nuclide produced by a decay.
dN d
¼ ð1  f d Þlp  N p  ld  N d  w0d  N d
dt
where fd is the fraction of daughter nuclide lost by recoil.
In this formulation, w0 d is a coefficient that accounts for
the loss of daughter nuclide via mineral dissolution only,
i.e., if fd > 0, wd0 < wd. Note that for non-radiogenic
nuclides (e.g., 238U) or nuclides produced by b decay
(e.g., 231Pa), w0 ¼ w. While 234U is produced by b decay
from 234Pa, it is lost by a recoil of 234Th which rapidly
ð1Þ decays into 234Pa (half-life, T 234Th ¼ 24.1 days), which
rapidly decays into 234U (T 234Pa ¼ 6.09 h).
It has been proposed that in silt-sized sediments
234
U-238U fractionation occurs mainly via recoil and
loss of 234U to the surrounding medium (Maher
et al., 2004; DePaolo et al., 2006). In this case, it is
assumed that w40 ¼ w80 and Eq. (2) can be rearranged to
express the (234U/238U) of the sediment as a function of
time:
234  234 
U U t
 T comm

238 U
¼ 238  el4 þ ð1  f 4 Þ 1  el4
U 0
ð3Þ
where (234U/238U)0 is the initial activity ratio. Tcomm is
termed the comminution age and represents the amount
ð2Þ of time needed for the activity ratio to evolve from its ini-
tial value (generally the source rock in secular equilib-
rium) to the observed value in the sediment. Note that
because this approach neglects the effect of chemical
weathering (mineral dissolution) on U-series isotope
 CHEMICAL WEATHERING (U-SERIES)
fractionation, it is not discussed further in this contribu-
tion. The reader is referred to DePaolo et al. (2012) for a
recent review.
Dequincey et al. (1999, 2002) proposed a model that
allows not only for the loss of nuclide via mineral dissolu-
tion but also for its gain via precipitation of secondary
minerals or dust deposition. For instance, the evolution
of the abundance of a radiogenic nuclide in the solid is
predicted by the equation:
dN d Fd
¼ lp  N p  ld  N d þ  wd  N d ð4Þ
dt ld
where Fd is the activity input of the daughter nuclide per
time unit (in atoms/a2). Note that here, recoil is not explic-
itly accounted for and is lumped into the dissolution coef-
ficient, as in Latham and Schwarcz (1987).
Dosseto et al. (2008b) later modified this expression
and introduced an input coefficient, G (in a1), such that
input and dissolution coefficients could be directly
compared.
dN d
¼ lp  N p  ld  N d þ Gd N d, i  wd  N d ð5Þ
dt
The input coefficient Gd reflects how fast the initial con-
centration Nd,i is increased by nuclide addition. Thus, the
term GdNd,i represents the flux of added nuclides
(in atoms/g/a). Note that the model is sensitive to the ini-
tial isotopic ratios but not to the initial isotopic abundances
(i.e., the initial U concentration, Nd,i, does not impact cal-
culated activity ratios, but the initial activity ratios obvi-
ously do).
In order to better understand the evolution of each
nuclide’s abundance in the solid according to these
models, processes that affect the mobility of each nuclide
are reviewed below.
Mobility of 238U:
• Loss during mineral dissolution, releasing 238U to the
solution (coefficient w8). The ability of 238U to stay in
solution then depends of the redox conditions, as
explained above: in reducing conditions, tetravalent
238
U has low solubility and will precipitate out of the
solution, while in oxidizing conditions (typical of
near-surface environments where mineral dissolution
by weathering takes place), 238U will form the uranyl
ion 238UO22+ which readily complexes with the carbon-
ate ion to create a very stable and highly soluble species.
• Sorption from soil pore water onto mineral surfaces
(coefficient G8). It has been shown that uranium can
be easily adsorbed on micas, clays, iron oxyhydroxides,
and organic matter (Ames et al., 1983a; Ames
et al., 1983b; Ames et al., 1983c; Chabaux
et al., 2003), with a sorption capacity ranking as fol-
lows: hydrous Fe oxides > well-crystallized
goethite >> montmorillonite > kaolinite (Szalay,
1964; Manceau et al., 1992). Because sorption is
enhanced by greater surface areas, this process has been
invoked to account for the increasing radionuclide
157
concentrations with decreasing grain size in soils
(Megumi and Mamuro, 1977).
• Precipitation of authigenic phases (coefficient G8). The
formation of carbonates and Fe-Mn oxyhydroxides can
scavenge significant amounts of U from solution
where U will be either adsorbed onto surfaces (see
above) or incorporated into the crystal structure by iso-
morphic substitution (see review in Chabaux
et al., 2003). When considering bulk soils or stream
sediments, the 238U in these secondary phases can rep-
resent a significant proportion of the 238U budget
(e.g., Plater et al., 1992).
• Dust deposition (coefficient G8). Gain of 238U can also
occur via deposition of mineral aerosols. Pett-Ridge
et al. (2007) attributed U enrichments up to 531 % in
Hawaiian soils older than 150 ka by dust deposition.
Pelt et al. (2013) showed that dust deposition can
account for up to 25 % of the U and Th budget of basal-
tic soils in Cameroon. Note that these are extreme sce-
narios because the parent material (basalt) has a low
U content. The contributions from dust to the soil
U budget would be less significant in granitic and sedi-
mentary environments. Nevertheless, gain of 238U via
dust deposition is accounted for in the model by the
parameter G8.
It is important to note that the numerical models only
account for the gain of 238U without differentiating
between the different processes of sorption, precipitation,
or dust deposition that contribute to that gain.
Mobility of 234U:
• Loss during mineral dissolution (coefficient w4). Con-
gruent dissolution will release 234U at the same rate
as 238U (not taking into account the preferential
leaching mentioned below). In this case, no fraction-
ation between these isotopes is expected, i.e., w4 ¼ w8.
• Direct recoil of 234Th and subsequent decay into 234U
contributes to enhance 234U mobility compared to
238
U. The fraction 234U lost by recoil in Scott et al.’s
model is 1 – f4.
• Preferential leaching of 234U embedded in recoil tracks.
Tracks can provide pathways for migrating solutions to
reach 234U, thereby enhancing the mobility of 234U
compared to undecayed 238U, i.e., w4 > w8.
• Preferential oxidation of 234U compared to
238
U. Computer simulations of the recoiled 234Th
motion have shown that in minerals with a low
U content, there is a high probability for 234U to be
found in the vicinity of oxygen atoms or radicals, mak-
ing it more prone to oxidation to the hexavalent state
resulting in preferential mobilization compared to
tetravalent 238U (Adloff and Roessler, 1991). This will
also contribute to a coefficient w4 greater than w8.
• Sorption of 234U on mineral phases (coefficient G4).
Although untested, it is believed that no fraction-
ation between 234U and 238U occurs during this
process. Thus, the adsorbed U will record the
(234U/238U) ratio of the solution (i.e., G4/ G8), which
158 CHEMICAL WEATHERING (U-SERIES)
in most cases will be greater than 1. This relation is
supported by measurements of (234U/238U) ratios in
the exchangeable fraction leached from organic matter,
carbonates, and crystalline Fe-Mn oxides that were
bound to river sediments: all leachates showed values
within analytical error of 1.3 (Plater et al., 1992).
• Precipitation of authigenic phases (coefficient G4). As
with 238U, 234U can be added to the bulk soil or stream
sediment during the precipitation of secondary phases.
Because no fractionation between 234U and 238U is
believed to occur during secondary mineral precipita-
tion, U in the precipitated phases will record the
(234U/238U) ratio of the solution (in most cases greater
than 1), which will define values of G4/G8.
• Dust deposition (coefficient G4). Similarly to 238U, gain
of 234U can occur via the deposition of mineral aerosol.
If this is a substantive process for U gain, the G4/G8 ratio
represents the (234U/238U) activity ratio of the dust
component.
Considering the processes described above, if sorbed
U and authigenic phases can be removed by sequential
leaching, we should observe (234U/238U) activity ratios
<1 in weathering products. However, this is not always
the case (Lee et al., 2010). A (234U/238U) >1 could be
explained by the Sheng-Kuroda effect (Sheng and
Kuroda, 1986a; Sheng and Kuroda, 1986b). In this case,
U-poor residual minerals can be enriched in 234U by
implantation from U-rich minerals, which are subse-
quently removed by dissolution. Note that this process
excludes zircon as a source for 234U implantation as it is
highly resilient to weathering.
Mobility of 230Th:
• Loss during mineral dissolution (coefficient w0). As
230
Th is produced by alpha decay of 234U, it will be
located in recoil tracks. Thus, as the mineral dissolves,
recoil tracks will be exposed to migrating solutions
allowing release of 230Th. However, Th is highly insol-
uble, and any Th released to the solution is likely to pre-
cipitate back and/or be adsorbed on residual minerals.
For these reasons, in most studies, it is assumed that
w0 << w4 or w8. Nevertheless, significant 230Th solubi-
lization can occur in the presence of organic acids and
colloids and/or at low pH (Langmuir and Herman,
1980; Viers et al., 1997; Chabaux et al., 2003; Dosseto
et al., 2006a; Dosseto et al., 2006b). For instance, sedi-
ments from organic-rich rivers of the Amazon basin are
characterized by (230Th/238U) <1 (Dosseto
et al., 2006a).
• Direct recoil of 230Th during decay of 234U. If decay of
234
U occurs within a few tens of nm of the mineral sur-
face, 230Th can be recoiled out of the mineral into the
surrounding medium (air or water). The fraction of
230
Th lost by recoil is 1 – f0. However, as explained
above, because of Th low solubility and its high reactiv-
ity for mineral surfaces (especially Fe-Mn
oxyhydroxides), it is likely that 230Th will be rapidly
re-adsorbed onto mineral surfaces.
• Sorption of 230Th from the solution onto minerals or
incorporation into secondary minerals (coefficient G0).
Th can sorb onto organic matter, clays, and Fe-Mn
oxyhydroxides (Chabaux et al., 2003), resulting in a
gain of 230Th. Based on (230Th/238U) and (230Th/234U)
activity ratios measured during sequential leaching
experiments of river sediments (Plater et al., 1992), it
appears that U has a higher affinity for organic matter
than Th (thus, G8 or G4 > G0), while the opposite is
observed for Fe-Mn oxides (G0 > G8 or G4).
• Dust deposition (coefficient G0). Gain of 230Th can also
occur via deposition of mineral aerosols. If this is a sub-
stantive process for the gain of U and Th isotopes, then
the G0/G8 ratio reflects the (230Th/238U) activity ratio of
the dust component (expected to be greater than 1).
Hence, based on the processes described above,
(230Th/238U) ratios > or < 1 can be observed in
weathering products. Because (230Th/238U) > 1 are more
commonly observed, it is likely that conditions where
230
Th is insoluble, or where any 230Th lost via recoil is
adsorbed onto Fe-Mn oxides, generally prevail.
A corollary from these observations is that at low disso-
lution rates, and where sorbed nuclides and secondary
minerals have been removed, recoil of 230Th should
become important and (230Th/238U) <1 should be
expected in solids. This is similar to what has been pro-
posed to explain (234U/238U) in waters from New Zealand
(Robinson et al., 2004), where loss of 234U is enhanced in
situations where physical weathering (comminution)
dominates over chemical weathering, thus favoring
recoiled 234U over the bulk loss of U.
The concepts and models presented above are used to
describe how (234U/238U) and (230Th/238U) ratios evolve
from initial to observed sets of values over a period of
time. The initial (234U/238U) and (230Th/238U) are gener-
ally taken as those of the parent rock and assumed to rep-
resent secular equilibrium (i.e., all U-series activity ratios
are equal to 1). Alternatively, when modeling regolith
compositions in a weathering profile, initial ratios can be
those of a deeper regolith sample. The amount of time
required to evolve from initial to observed isotopic com-
positions represents the duration of weathering. In a
weathering profile, this can be equivalent to the age of
the weathering front, termed weathering age, i.e., the time
elapsed since the weathering front was at a given depth
(Dosseto et al., 2008b, 2012). In river sediments, this is
equivalent to the residence time of sediments in the catch-
ment, i.e., the time spent by sediments in the catchment
since production from the bedrock, integrating storage in
weathering profiles and transport in the river (Chabaux
et al., 2008; Dosseto et al., 2008a; Vigier and Bourdon,
2011).
The set of parameters that best describe the observed
(234U/238U) and (230Th/238U) ratios is sought for, gener-
ally by inverse modeling. The resulting parameters
derived from the models are the duration of weathering
and the set of coefficients for each nuclide (depending
 CHEMICAL WEATHERING (U-SERIES)
on the models: fi, wi, and/or Gi, where i is the nuclide
considered).
To solve these models, several important assumptions
are required:
1. The coefficients for nuclide loss, w, and gain, G, are
constant over the duration of weathering. This assump-
tion is commonly made in studies that model chemical
weathering (e.g., Chamberlain et al., 2005; Ferrier and
Kirchner, 2008) and laboratory experiments that sug-
gest constant w values for a given mineral (White and
Brantley, 2003).
2. In Scott et al.’s (1992) model, the fraction of recoiled
nuclides (coefficient f) is constant over the duration
of weathering.
3. Nuclide loss and gain operate on similar timescales.
4. Because of the large number of unknowns, several
samples are required, and it is assumed that their activ-
ity ratios differ only as a result of different durations of
weathering. Hence, each coefficient for each nuclide is
assumed to be constant across samples.
These assumptions are difficult to verify because
(1) conditions for mineral dissolution and precipitation
are variable with time, especially over the timescales typ-
ically modeled (103–106 years). It is generally assumed
that any actual variations will average out over such long
timescales. (2) The fraction of recoiled nuclide is depen-
dent on the surface properties of the material, which may
vary with time (e.g., as the result of mineral dissolution).
(3) The onset of nuclide loss (weathering) could take place
significantly earlier than nuclide gain (e.g., illuviation).
Nevertheless, even in the early stages of weathering, sec-
ondary minerals that contain significant amounts of
U are observed (Guthrie and Kleeman, 1986; Pelt
et al., 2008). (4) Constant nuclide gain and loss coeffi-
cients between samples are difficult to verify because con-
ditions of mineral dissolution and precipitation can be
variable even over small length scales. Despite these
strong caveats, previous studies have yielded unprece-
dented constraints on rates of soil formation and sediment
transport (see Chabaux et al., 2008; Dosseto et al., 2008a;
Vigier and Bourdon, 2011 for recent reviews).
Current investigations, controversies, and gaps in
knowledge
This section presents a review of studies using U-series
isotopes to constrain timescales of chemical weathering
in weathering profiles, rinds, and soil chronosequences.
For a review of applications to fluvial sediments, the
reader is referred to the reviews by Dosseto et al. (2008a)
and DePaolo et al. (2012).
Weathering profiles
When discussing weathering profiles, we must first define
a few terms. Here, the soil is defined as the top part of the
weathering profile, which is mobile as it creeps downhill
(colluvial transport). Soil production occurs by physical
159
and chemical weathering of the underlying saprolite or
bedrock, often with living organisms playing a preponder-
ant role. The saprolite is defined as weathered material
that still retains some structural aspects of the bedrock.
It is produced at the weathering front by physical and
chemical weathering of the bedrock. The regolith is the
ensemble of saprolite and soil, which together constitutes
the weathering profile. Thus, saprolite production, rego-
lith production, and weathering front advance refer to the
same process.
The development of weathering profiles is the result of
a complex interaction between tectonics, climatic, and
biotic processes. The thickness and nature of these profiles
are an expression of landscape evolution in response to
these various mechanisms. The topsoil is of particular
importance because it is the basic resource required for
sustaining nearly all biological ecosystems including agri-
culture and human civilizations. Despite this importance,
little is known on how fast weathering profiles develop.
In contrast, their rate of denudation is well constrained,
even at millennial timescales (using cosmogenic isotopes;
Brown et al., 1995, 1998). Cosmogenic isotopes can also
be used to determine soil production rates (Heimsath
et al., 1997; Larsen et al., 2014); however, this requires
the assumption that the soil thickness is in steady state,
which is seldom verifiable. This approach, by definition,
renders it impossible to evaluate soil evolution, i.e., imbal-
ances between production and denudation. Indeed, even if
soil thickness has reached a steady state, how was it
achieved? The soil must have been thickening before
reaching steady state. Furthermore, soil production is only
one aspect of the development of a weathering profile,
with production of saprolite or regolith being the other,
even more important aspect. It has been proposed that cos-
mogenic isotopes could also be used to infer saprolite pro-
duction rates (Dixon et al., 2009); however, this
suggestion also requires the assumption that the entire
weathering profile thickness is in steady state, which is
not easily tested.
U-series isotopes offer an alternative to quantify soil
and saprolite production rates. Using the model of
Dequincey et al. (2002), it is possible to model the amount
of time that elapsed to produce the observed U-series iso-
tope composition since departure from secular equilib-
rium. This is equivalent to the time elapsed since the
onset of chemical weathering and is termed weathering
age (Figure 4). By using the weathering age of a given
saprolite sample in the profile and its height above the
weathering front (i.e., the difference between its depth
and the depth of the weathering front), we can infer an
average saprolite (or regolith) production rate. This repre-
sents the average rate at which the weathering front has
propagated between the two depths considered
(Figure 4). Similarly, considering a soil sample and using
the composition of the underlying saprolite as the initial
conditions, we can use the time required to evolve from
the U-series isotope composition of the saprolite to that
of the soil sample and the height of the soil sample above
160 CHEMICAL WEATHERING (U-SERIES)

Chemical Weathering (U-Series), Figure 4 Conceptual definitions of the weathering age and the saprolite/regolith production rate.
If one considers the downward propagation of the weathering front into bedrock (lower dashed line), the weathering age, tsap, of a
given saprolite sample (red dot) is the amount of time required for the weathering front to migrate over the distance hsap, which is the
height of the sample above the current weathering front. The saprolite or regolith production rate esap (in mm/ka), averaged over
the distance hsap, can then be calculated by dividing hsap (in mm) by tsap (in ka) (Modified from Dosseto et al. (2008b)).

the soil-saprolite boundary to infer an average soil produc-
tion rate (Figure 5).
This approach has been applied to weathering profiles
developed over granitic lithologies in tectonically passive,
tropical/equatorial regions of south America (Mathieu
et al., 1995) and western Africa (Boulad et al., 1977;
Dequincey et al., 1999, 2002). In Cameroon, Boulad
et al. (1977) estimated regolith production rates of
70 mm/ka (Figure 6 and Table 2). In Burkina Faso, while
no regolith production rate was estimated, Dequincey
et al. (2002) determined weathering ages of
400–500 ka. In the Amazon, Mathieu et al. (1995) cal-
culated that it took 300 ka to develop 15 m of regolith
in the Amazon. This implies an average regolith produc-
tion rate of 50 mm/ka (Table 2). These production rates
are surprisingly high considering that (i) erosion rates in
these areas are most likely <10 mm/ka and
(ii) weathering rates are expected to be slow because the
weathering front is several tens of meters deep and likely
to be isolated from meteoric waters (Gaillardet
et al., 1995, 1997, 1999). However, it is important to note
that regolith production rates inferred from U-series iso-
topes are averaged over a timescale that represents the
weathering age. Thus while modern weathering rates are
most likely slow, they were probably much higher in the
past when the weathering front was closer to the surface
and subject to higher rainfall and temperatures.
A U-series isotope study was also conducted in Puerto
Rico (Chabaux et al., 2013), where the weathering profile
is also developed over a granitic parent rock under tropical
conditions, but erosion rates are much higher than in west-
ern Africa or Brazil (43  15 mm/ka (Brown et al., 1995);
33  8 mm/ka (Riebe et al., 2003)). Calculated regolith
production rates were 45  12 mm/ka (Chabaux
et al., 2013) (Table 2). This implies that, at the scale of
the weathering profile, the loss of regolith is largely bal-
anced by its production from the bedrock. A recent
U-series study of stream sediments in the same area shows
that a balance between rates of regolith production and
erosion is also true at the catchment scale (Dosseto
et al., 2014). These results are in agreement with previous
work (White et al., 1998; Riebe et al., 2003; Turner
et al., 2003a; Pett-Ridge et al., 2009a, b; Ferrier
et al., 2010). Furthermore, similar regolith production
rates over granitic parent rock in different areas suggest
that production is relatively insensitive to erosion rates
and is mostly controlled by the composition of the parent
rock and/or climatic conditions.
Dosseto et al. (2012) suggested that the composition
of the parent rock is a more important control on rego-
lith production rates than climatic conditions, because
they calculated production rates of 334  46 mm/ka
for a weathering profile developed over a volcaniclastic
sedimentary rock in the same region (and thus very sim-
ilar climatic conditions) as the granitic profile of
Chabaux et al. (2013) (Figure 6 and Table 2). These
rates are faster than those for the granitic profile by a
factor of 4.
 CHEMICAL WEATHERING (U-SERIES) 161

Chemical Weathering (U-Series), Figure 5 Conceptual definitions of the soil weathering age and production rate. If one considers a
given soil sample (yellow dot), the soil weathering age, tsoil, is the amount of time required for the soil-saprolite boundary (dashed line)
to migrate downward over the distance h. The distance h represents the height of the soil sample above the current soil-saprolite
boundary. Two cases are considered: in case 1, bioturbation is neglected, and the soil production rate e1, averaged over the
distance h, is the ratio of h over tsoil. In case 2, bioturbation is considered, and the soil production rate e2, averaged over the entire soil
thickness H, is the ratio of H over tsoil. In the latter case, the soil thickness must be in steady state in order to obtain a meaningful
production rate (Modified from Dosseto et al. (2008b)).

Chemical Weathering (U-Series), Table 1 Parameters used for U-series isotope models

Parameter Description Units

N8, N4, N0 Concentrations of 238U, 234U, and 230Th, respectively Atoms g1
l8, l4, l0 Decay constants for 238U, 234U, and 230Th, respectively a1
w8, w4, w0 Dissolution coefficients for 238U, 234U, and 230Th, respectively a1
G8, G4, G0 Gain coefficients for 238U, 234U, and 230Th, respectively a1
f4, f0 Fraction of recoiled 234Th (and thus 234U) and 230Th, respectively Unitless

Studies have also been conducted in temperate regions
for weathering profiles developed over granitic lithologies
in southeastern Australia (Dosseto et al., 2008b; Suresh
et al., 2013) and shale in northeastern USA
(Ma et al., 2010, 2012). In southeastern Australia, the
region studied is characterized by a retreating escarpment
separating a coastal plain to the east from highland pla-
teaus (elevation 900 m) to the west. Weathering profiles
were studied in two different geomorphic contexts: on the
escarpment (Dosseto et al., 2008b) and on the plateau
(Suresh et al., 2013). Sampling strategies differed between
the two studies in that Dosseto et al. (2008b) sampled pro-
files across the hillslope, while Suresh et al. (2013) sam-
pled profiles on ridgetops so that colluvial transport
could be neglected.
On the escarpment, saprolite weathering ages ranged
from 0.55 to 6.2 Ma. Values increase downhill with dis-
tance from the ridge. These estimates represent the amount
of time required to develop 25 m of saprolite and result
in saprolite production rates ranging between 4 and
46 mm/ka (Dosseto et al., 2008b) (Table 2). These values
are similar to local estimates of denudation rates
162 CHEMICAL WEATHERING (U-SERIES)

Chemical Weathering (U-Series), Figure 6 Compilation of U-series-isotope-derived regolith (closed rectangles) and soil (open
rectangles) production rates (in mm/ka).

Chemical Weathering (U-Series), Table 2 Compilation of regolith and soil production rates from U-series isotope studies

Regolith (or saprolite) Soil production

Location Parent rock production rate (mm/ka) rate (mm/ka) Reference

Weathering profiles
Cameroon Granite 70 n/d Boulad et al. (1977)
Burkina Faso Granodiorite 30–40 n/d Dequincey et al. (2002)
Amazon (Brazil) Granite 50 n/d Mathieu et al. (1995)
Icacos (Puerto Rico) Granodiorite 45  12 n/d Chabaux et al. (2013)
Bisley (Puerto Rico) Volcaniclastic 334  46 n/d Dosseto et al. (2012)
sedimentary rock
Nunnock (Australia) Granodiorite 4–46 12–77 Dosseto et al. (2008b)
Frogs Hollow (Australia) Granite n/d 10–24 Suresh et al. (2013)
Shale Hills (PA, USA) Shale n/d 17–50 Ma et al. (2010, 2013)

Weathering rinds Rind formation rate

(mm/ka)
Costa Rica Basalt 0.5  0.2 Pelt et al. (2008)
Guadeloupe Basalt 0.18–0.24 Ma et al. (2012)
Icacos (Puerto Rico) Granodiorite 52  13 Chabaux et al. (2013)

(9–68 mm/ka; Heimsath et al., 2000), thus demonstrating
that the entire weathering profile is in steady state. This sug-
gests that, over the 0.55–6.2 Ma of saprolite development,
an equilibrium has been reached between erosion and the
conversion of granitic bedrock into saprolite. Similarly, soil
weathering ages range from 6 to 38 ka representing the
amount of time needed to develop 0.5–1 m of soil from
the underlying saprolite. Soil production rates calculated
from U-series range from 12 to 77 mm/ka (Table 2). Here
again, it is similar to denudation rates and demonstrates that
the soil thickness is in steady state too. Thus, at this site, the
three major interfaces that constitute the weathering profile
(atmosphere-soil, soil-saprolite, and saprolite-rock) are
progressing downward at similar rates. These results are
surprising as the escarpment is the most dynamic part of
the landscape in this region.
On the plateau, calculated soil weathering ages are up
to 30 ka. This represents the amount of time required to
develop 30–60 cm of soil from the saprolite. Soil produc-
tion rates show a much narrower range than at the
 CHEMICAL WEATHERING (U-SERIES)
escarpment, between 10 and 24 mm/ka (Suresh et al., 2013)
(Table 2). Here too, production rates are similar to denuda-
tion rates inferred from cosmogenic isotopes, effectively
demonstrating that soil thickness has also reached a steady
state.
At the Susquehanna/Shale Hills Critical Zone Observa-
tory in northeastern USA, Ma et al. (2010, 2013) quanti-
fied soil weathering ages and production rates from pits
along two opposing hillslopes, developed over a
Palaeozoic shale. On the north-facing hillslope, soil
weathering ages vary between 7 and 40 ka, representing
the amount of time to develop 30–70 cm of soil. This
amounts to soil production rates of 17–45 mm/ka
(Ma et al., 2010) (Table 2). On the south-facing hillslope,
soil weathering ages vary between 12 and 16 ka
(to develop 60–80 cm of soil). Inferred soil production
rates are 40–50 mm/ka (Ma et al., 2013) (Table 2). These
rates are similar to the range observed in granitic environ-
ments. This suggests that the composition of the parent
rock has little effect on the rate of soil production. How-
ever, in the case of Shale Hills, the soil is directly devel-
oped over the parent rock, while in granitic
environments it is developed over a saprolite. It is thus dif-
ficult to use results from the Shale Hills site to assess the
role of parent rock composition since pedogenetic pro-
cesses are quite different from those related to weathering
profiles developed over granitic or volcaniclastic sedi-
mentary parent rock.
Soil chronosequences
Soil chronosequences are defined as a group of soils that
differ in age but have similar parent materials and have
formed under similar conditions of climate, vegetation,
and geomorphic position (Jenny, 1941). They are particu-
larly useful for U-series isotopic studies if the soil’s max-
imum weathering age is known as they can then be used to
calibrate the U-series isotope weathering models.
Pett-Ridge et al. (2007) have studied a chronosequence
of soils developed in Hawaiian tephras with ages ranging
from 0.3 to 4,100 ka. They showed that the U budget of
the soil is largely dominated by dust deposition at older
sites and that Fe oxides are the main mineral phases carry-
ing U in these soils. Pelt et al. (2013) reached similar con-
clusions on the role of dust deposition on the U budget of
soils in a study of Holocene basaltic flows in Cameroon.
These results are a consequence of rapid dissolution of pri-
mary phases in basaltic parental material under wet condi-
tions (average annual rainfall: 2,500 mm/a). By 20 ka,
they are completely exhausted (Crews et al., 1995). In
basaltic parent material, U is mostly contained in the
groundmass which dissolves very readily, leaving the soil
U budget easily dominated by dust deposition and Fe
oxides. Where U is contained in primary minerals that
are more resistant to chemical weathering (e.g., biotite
and apatite in granitic rocks, illite in sedimentary rocks),
the importance of dust deposition and the role of
solution-derived secondary phases would be lessened.
163
The U-series isotope composition of soils developed
along a chronosequence has been modeled in a recent
study by Keech et al. (2013). The parent material is com-
posed of glacial outwash sediments with deposition ages
ranging from 40 ka to 3 Ma. For each soil profile, they
averaged the U-series isotope composition of samples col-
lected at different depths and attempted to use the model
described in Eq. (1) to reproduce the average activity
ratios, knowing the maximum weathering age of each soil
profile (i.e., the deposition age). In doing so, they assumed
that the known age of the parent material represents an
average weathering age for the profile. While it is unclear
whether an average activity ratio for a given soil profile
has any significance in terms of pedogenic processes, the
average weathering age of a soil profile should be a value
between 0 and the age of the parent material. The authors
used the model described by Eq. (1) but could not repro-
duce the observed average values for (234U/238U) and
(230Th/238U), which showed a decrease with increasing
age. To explain this discrepancy, the authors proposed a
two-stage model where labile U is mobilized first,
followed by the mobilization of U in crystal lattices during
mineral dissolution. This two-stage model assumes that
the parent material was in secular equilibrium prior to soil
formation. However, in this case, the parent material is an
unconsolidated sediment and is likely to have already
undergone significant weathering. This pre-deposition
weathering episode would have imparted a disequilibrium
to the parent material which would have decayed away
only over >300 ka (assuming closed system, which is
unlikely during the deposition of an alluvial sediment).
Therefore, the greater disequilibrium observed in younger
soils may simply be the result of pre-deposition radioac-
tive disequilibria (with 234U and 230Th excesses over
238
U) which had not yet decayed away. Thus, there is no
need for a two-stage model. Without knowing the
U-series isotopic composition of the parent material
(which was not sampled), it is not possible to derive reli-
able time information from this work.
Andersen et al. (2013) have studied the U-series isoto-
pic compositions of soils from a chronosequence formed
by glacier retreat in Scotland. The parent material is com-
posed of alluvial sediments deposited in fluvial terraces
with ages between 0.1 and 13 ka. At this location, the
authors observed that, on average, the (234U/238U) activity
ratio decreased with increasing deposition age, starting
with values >1, similarly to what was observed in Keech
et al. (2013), but (230Th/238U) increased with deposition
age, from values initially <1. The model described in
Eq. (1) was used to determine 238U dissolution coeffi-
cients for 5.6-ka- and 13-ka-old soils, assuming that the
soil weathering age was the same as the deposition age.
However, here too the parent material was assumed to be
in secular equilibrium. This is unlikely to be the case
because soils were developed on fluvial sediments, which
were unlikely to be at secular equilibrium at the onset of
soil formation (as observed in all fluvial sediments; see
the review in Dosseto et al. (2008a)). The decrease in
164 CHEMICAL WEATHERING (U-SERIES)
(234U/238U) with soil age could reflect loss of 234U via
recoil or more likely (as pointed out by the authors) via
preferential leaching. The initial (230Th/238U) <1 in the
parent material could be explained as the sediment is
likely to be derived from organic-rich hillslopes. The pres-
ence of organic acids is known to mobilize Th (Porcelli
et al., 1997; Viers et al., 1997; Andersson et al., 1998;
Porcelli et al., 2001; Chabaux et al., 2003), and organic-
rich soil horizons can exhibit (230Th/238U) <1 (Dosseto
et al., 2008b; Ma et al., 2010; Chabaux et al., 2013; Suresh
et al., 2013). The increase in (230Th/238U) could be
explained by the subsequent loss of U during soil
development.
While the study of soil chronosequences is an exciting
endeavor with a great potential to improve our understand-
ing of soil processes and the use of U-series isotopes, the
different aspects of U-series models need to be carefully
considered. For instance, initial conditions (i.e., the compo-
sition of the parent material) must be adequately assessed in
order to apply the appropriate model parameters.
Weathering rinds
In the early stages of weathering, the parent rock can
sometimes develop rinds of weathered material. The mea-
surement of U-series isotopes on a depth profile across the
pristine core and the outer weathering rinds can be used to
infer how fast the weathering front propagates into the
rock. As shown above, it is expected that U is lost prefer-
entially compared to Th and 234U preferentially to 238U, so
increasing (230Th/238U) and decreasing (234U/238U) activ-
ity ratios, respectively, should develop with time. How-
ever, in a study of weathering rinds developed on a
basaltic clast from a 125-ka-old alluvial terrace in Costa
Rica, Pelt et al. (2008) observed the opposite with
(230Th/238U) <1 and (234U/238U) >1 in the most exten-
sively weathered parts of the rinds. This was interpreted
as the result of (i) the conservative behavior of Th and
(ii) external input of U from soil solutions having
(234U/238U) >1, with minimal loss of U during
weathering. The input of U was modeled, resulting in a
rind formation rate of 0.5  0.2 mm/ka (Table 2). In
another study, Ma et al. (2012) also found (230Th/238U)
<1 and (234U/238U) >1 in weathering rinds from a basaltic
clast from Guadeloupe Island. They inferred a rind forma-
tion rate of 0.18–0.24 mm/ka (Table 2). These rates are
three orders of magnitude lower than what has been
inferred considering the scale of the weathering-profile
development for basaltic lithologies (Dosseto
et al., 2012). As explained in Navarre-Sitchler and
Brantley (2007), this apparent discrepancy stems from the
fractal nature of the weathering advance rate, its value being
largely controlled by the length scale over which it is mea-
sured. This is a consequence of the variable spatial resolu-
tions used to measure the surface area of the weathering
front (from the 103 m scale when studying weathering
rinds to the meter scale when investigating weathering pro-
files). Navarre-Sitchler and Brantley (2007) showed that it
is possible to compare weathering advance rates across
length scales by calculating the roughness of the weathering
front at each length scale, using the fractal dimension char-
acteristic of the weathering advance rate (for instance, 2.3
for basaltic weathering).
In a weathering profile developed over a granitic
corestone in Puerto Rico, Chabaux et al. (2013) also quan-
tified the rate at which weathering-rind formation
advances into the corestone and estimated a value of
52  13 mm/ka. Interestingly, this is similar to the sapro-
lite production rate inferred in the same profile (Table 2).
This can be explained by the small length scales of
weathering-rind development studied by Pelt et al. (2008)
and Ma et al. (2012) measured in millimeters, compared to
the weathering rinds studied by Chabaux et al. (2013),
which cover a length scale similar to that of the weathering
profile developed above them.
Implications for landscape evolution
A compilation of calculated regolith and soil production
rates, similar to the one shown in Dosseto et al. (2012),
allows us to make several observations concerning the
development of weathering profiles (Figure 6). Note that
these observations need to be considered with caution,
given the small number of studies published so far. How-
ever, they hint at some aspects of landscape evolution that
are worth investigating.
It has been known for a long time that different rocks
weather at different rates, depending on the susceptibility
to dissolution of the minerals that compose them. This is
illustrated on Figure 6, where in Puerto Rico, regolith pro-
duction over a volcaniclastic sedimentary rock occurs at a
rate more than six times faster than over granodiorite.
However, parent material composition does not appear to
play a strong role for soil production: soils over a shale
are produced at about the same rate as those produced over
a granitic rock (Figure 6). This may imply that soil produc-
tion is mostly driven by physical weathering, via the
mechanical action of plants and animals, rather than by
chemical weathering.
The role of climate (runoff, temperature) on regolith
production is not clear from results shown in Figure 6:
granitic profiles seem to develop at similar rates across a
range of climatic conditions. While it is possible that reg-
olith production rates that are lower than expected in
Africa and the Amazon may be the consequence of thick
weathering profiles shielding the weathering front from
meteoric waters, profiles in Puerto Rico and southeastern
Australia have similar thicknesses and production rates
despite very different climatic conditions.
Erosion rates in lowlands of western Africa and the
Amazon are expected to be <10 mm/ka (Gaillardet
et al., 1997; Wittmann et al., 2010). In contrast, Larsen
(2012) has estimated long-term hillslope erosion rates in
the Rio Icacos basin of Puerto Rico to be 75 mm/ka,
mostly driven by landslides. Despite about an order of
magnitude difference in erosion rates, regolith production
 CHEMICAL WEATHERING (U-SERIES)
rates for these granitic profiles are relatively similar. They
exceed erosion rates in tropical lowlands, while they are
similar to or lower than erosion rates in Puerto Rico. This
could suggest a decoupling between the surface (where
erosion occurs) and the weathering front. The advance rate
of the weathering front seems to be most sensitive to the
composition of the parent material, showing similar values
for granitic profiles across a range of climatic conditions
and erosion rates.
Gaps in knowledge
The models described above are intended to track the
U-series isotope composition of minerals derived from
the parent rock, whether they are residual primary min-
erals or secondary minerals formed during incongruent
dissolution of primary minerals. Existing studies have
focused on bulk material (regolith, soil, sediments). As
explained above, the regolith is a complex mixture of pri-
mary minerals (derived from the parent rock), secondary
minerals precipitated from soil pore water (e.g., carbon-
ates and Fe-Mn oxyhydroxides), allochthonous minerals
(via dust deposition), and organic matter (Figure 3). Each
of these phases is likely to have its own U-series isotope
composition. Thus, the application of the U-series models
is complicated by the contribution of each of these compo-
nents to the isotopic composition of the bulk sample.
Future works should investigate techniques to remove
these phases either physically (e.g., mineral separation;
(Menozzi and Dosseto, 2013; Rihs et al., 2013)) or chem-
ically (e.g., sequential extraction; (Schultz et al., 1998;
Blanco et al., 2004; Martin et al., submitted)). Until then,
we will still be relying on whichever phases control the
U budget of the samples. If the U budget is controlled by
only a single primary phase or a secondary phase derived
from primary minerals, the models may provide reliable
results. If the U budget is controlled by more than one pri-
mary phase or secondary phase derived from primary min-
erals, there might be some noise in the data that make the
models less straightforward to apply. If controlled by a
mixture of primary and associated secondary minerals,
as well as solution-derived secondary minerals,
allochthonous minerals, or organic matter, it is likely that
the models will yield less than straightforward informa-
tion. Finally, if the U budget of the samples is controlled
by solute-derived secondary minerals, allochthonous min-
erals, or organic matter, the models will provide meaning-
less information in terms of weathering age, as defined
above. Thus, the future of this technique depends on our
ability to understand the U-series contributions from the
various components involved and how they impact the
overall compositions from which the models are
developed.
Conclusions
U-series isotopes in weathering products (weathering
rinds, regolith, sediments) allow us to place time con-
straints on weathering processes. These then can be used
165
to determine the rate of formation of weathering profiles
(regolith production rates). Results from available stud-
ies, when combined with other tools like cosmogenic iso-
topes, provide unprecedented insights on the development
of weathering profiles and landscape evolution. Results
from these studies show that climate seems to have a sec-
ondary role on regolith production rates. Instead, the pri-
mary control on regolith production appears to be the
composition of the parent material. Under tropical cli-
mates, for instance, regolith can be produced 6 times
faster over a volcaniclastic sedimentary rock than over a
granodiorite parent. Conversely, soil production seems to
be relatively insensitive to the parent material composi-
tion: under temperate climates, soil is produced at the
same rates over shale and granodiorite. Finally, regolith
production rates are similar for sites that show a range of
erosion rates. This suggests a decoupling between the land
surface and the weathering front. In this case, the topogra-
phy of the weathering front would be determined by the
lithology, while the surface topography is determined by
denudation in response to climatic and tectonic
conditions.
Several studies have emphasized the role of dust depo-
sition on weathering profile development and in particular
on the uranium budget of the soil. While this is taken into
account in existing models of U-series isotopic evolution,
further work is needed to improve the robustness of the
information derived from these models. In order to
achieve this objective, innovative chemical and/or physi-
cal protocols for sample preparation are important avenues
to explore.
Finally, soil chronosequences represent fascinating nat-
ural laboratories to test the application of U-series isotope
methods and to further improve the quality of information
that U-series models provide. However, these studies need
to be carefully undertaken, with a full understanding of the
assumptions required by each different model.
With these considerations in mind, it is clear that, com-
bined with other geochronological and geochemical tech-
niques, U-series isotopes supply us with an ability to better
understand the evolution of the Earth’s surface. As sum-
marized in this entry, more than 10 years of studies on
the topic illustrate the unique prospects that this technique
offers.
Bibliography
Adams, J. A. S., Osmond, J. K., and Rogers, J. J. W., 1959. The geo-
chemistry of thorium and uranium. Physics and Chemistry of the
Earth, 3, 298–348.
Adloff, J. P., and Roessler, K., 1991. Recoil and transmutation
effects in the migration behaviour of actinides. Radiochimica
Acta, 52–53, 269–274.
Ames, L. L., McGarrah, J. E., and Walker, B. A., 1983a. Sorption of
trace constituents from aqueous solutions onto secondary min-
erals. I Uranium. Clays and Clay Minerals, 31, 321–334.
Ames, L. L., McGarrah, J. E., and Walker, B. A., 1983b. Sorption of
uranium and radium by biotite, muscovite and phlogopite. Clays
and Clay Minerals, 31, 343–351.
166 CHEMICAL WEATHERING (U-SERIES)
Ames, L. L., McGarrah, J. E., Walker, B. A., and Salter, P. F.,
1983c. Uranium and radium sorption on amorphous ferric
oxyhydroxide. Chemical Geology, 40, 135–148.
Andersen, M. B., Erel, Y., and Bourdon, B., 2009. Experimental
evidence for 234U–238U fractionation during granite weathering
with implications for 234U/238U in natural waters. Geochimica
et Cosmochimica Acta, 73, 4124–4141.
Andersen, M. B., Vance, D., Keech, A., Rickli, J., and Hudson, G.,
2013. Estimating U fluxes in a high-latitude, boreal post-glacial
setting using U-series isotopes in soils and rivers. Chemical
Geology, 354, 22–32.
Andersson, P. S., Porcelli, D., Wasserburg, G. J., and Ingri, J., 1998.
Particle transport of 234U-238U in the Kalix River and the Baltic
Sea. Geochimica et Cosmochimica Acta, 62, 385–392.
Balan, E., Neuville, D. R., Trocellier, P., Fritsch, E., Muller, J. P.,
and Calas, G., 2001. Metamictization and chemical durability
of detrital zircon. American Mineralogist, 86, 1025–1033.
Blanco, P., Tome, F. V., and Lozano, J. C., 2004. Sequential extrac-
tion for radionuclide fractionation in soil samples: a comparative
study. Applied Radiation and Isotopes, 61, 345–350.
Borovec, Z., 1981. The adsorption of uranyl species by fine clay.
Chemical Geology, 32, 45–58.
Boulad, A. P., Muller, J. P., and Bocquier, G., 1977. Essai de déter-
mination de l’âge et de la vitesse d’altération d’un sol
ferrallitique camerounais à l’aide de la méthode du déséquilibre
radioactif uranium–thorium. Sciences Géologiques Bulletin,
30, 175–188.
Bourdon, B., Joron, J. L., and Allègre, C. J., 1999. A method for
231
Pa analysis by thermal ionization mass spectrometry in sili-
cate rocks. Chemical Geology, 157, 147–151.
Bourdon, B., Henderson, G. M., Lundstrom, C. C., and Turner, S. P.,
2003. Uranium-Series Geochemistry. Washington, DC: Mineral-
ogical Society of America. Reviews of Mineralogy and Geo-
chemistry, Vol. 52, 656 p.
Brown, E. T., Stallard, R. F., Larsen, M. C., Raisbeck, G. M., and
Yiou, F., 1995. Denudation rates determined from the accumula-
tion of in-situ produced 10Be in the Luquillo Experimental Forest,
Puerto Rico. Earth and Planetary Science Letters, 129, 193–202.
Brown, E. T., Stallard, R. F., Larsen, M. C., Bourles, D. L.,
Raisbeck, G. M., and Yiou, F., 1998. Determination of
predevelopment denudation rates of an agricultural watershed
(Cayaguas River, Puerto Rico) using in-situ-produced 10Be in
river-borne quartz. Earth and Planetary Science Letters, 160,
723–728.
Chabaux, F., Othman, D. B., and Birck, J. L., 1994. A new Ra-Ba
chromatographic separation and its application to Ra mass-
spectrometric measurement in volcanic rocks. Chemical Geol-
ogy, 114, 191–197.
Chabaux, F., Riotte, J., and Dequincey, O., 2003. U-Th-Ra fraction-
ation during weathering and river transport. Uranium-series geo-
chemistry. In Bourdon, B., Henderson, G. M., Lundstrom, C. C.,
and Turner, S. P. (eds.), Uranium-Series Geochemistry. Wash-
ington, DC: Mineralogical Society of America. Reviews of Min-
eralogy and Geochemistry, Vol. 52, pp. 533–576.
Chabaux, F., Bourdon, B., and Riotte, J., 2008. U-Series Geochem-
istry in Weathering Profiles, River Waters and Lakes. In
Krishnaswami, S., and Cochran, J. K. (eds.), U/Th Series Radio-
nuclides in Aquatic Systems. Radioactivity in the Environment,
Vol. 13, pp. 49–104.
Chabaux, F., Blaes, E., Stille, P., di Chiara Roupert, R., Pelt, E.,
Dosseto, A., Ma, L., Buss, H. L., and Brantley, S. L., 2013. Reg-
olith formation rate from U-series nuclides: implications from
the study of a spheroidal weathering profile in the Rio Icacos
watershed (Puerto Rico). Geochimica et Cosmochimica Acta,
100, 73–95.
Chamberlain, C. P., Waldbauer, J. R., and Jacobson, A. D., 2005.
Strontium, hydrothermal systems and steady-state chemical
weathering in active mountain belts. Earth and Planetary
Science Letters, 238, 351–366.
Cherdyntsev, V. V., 1969. Uranium-234. Moskva: Atomizdat.
English translation by Schmorak, J., 1971. Jerusalem: Israel Pro-
gram for Scientific Translations.
Cherdyntsev, V. V., Malyshev, V. I., Sokolova, Z. A., Kazachevskii,
I. V., and Borisov, I. V., 1964. Isotopic composition of uranium
and thorium in the zone of hypergenesis. Investigation of the
peat bog matter. Geokhimiya (U.S.S.R.). English translation
see. Geochemistry (USSR), 5, 399–403.
Cowart, J. B., and Osmond, J. K., 1977. Uranium isotopes in
groundwater: their use in prospecting for sandstone-type ura-
nium deposits. Journal of Geochemical Exploration, 8,
365–379.
Crews, T. E., Kitayama, K., Fownes, J., Riley, R. H., Herbert, D. A.,
Mueller-Dombois, D., and Vitousek, P. M., 1995. Changes in
soil phosphorus fractions and ecosystem dynamics across a long
chronosequence in Hawaii. Ecology, 76, 1407–1424.
DePaolo, D. J., Maher, K., Christensen, J. N., and McManus, J.,
2006. Sediment transport time measured with U-series isotopes:
results from ODP North Atlantic drift site 984. Earth and Plan-
etary Science Letters, 248, 394–410.
DePaolo, D. J., Lee, V. E., Christensen, J. N., and Maher, K., 2012.
Uranium comminution ages: sediment transport and deposition
time scales. Comptes Rendus Geoscience, 344, 678–687.
Dequincey, O., Chabaux, F., Clauer, N., Liewig, N., and Muller, J.-
P., 1999. Dating of weathering profiles by radioactive
disequilibria: contribution of the study of authigenic mineral
fractions. Comptes Rendus de l’Academie des Sciences – Series
IIA – Earth and Planetary Science, 328, 679–685.
Dequincey, O., Chabaux, F., Clauer, N., Sigmarsson, O., Liewig, N.,
and Leprun, J.-C., 2002. Chemical mobilizations in laterites: evi-
dence from trace elements and 238U-234U-230Th disequilibria.
Geochimica et Cosmochimica Acta, 66, 1197–1210.
Dixon, J. L., Heimsath, A. M., and Amundson, R., 2009. The criti-
cal role of climate and saprolite weathering in landscape evolu-
tion. Earth Surface Processes and Landforms, 34, 1507–1521.
Dosseto, A., Bourdon, B., Gaillardet, J., Allègre, C. J., and Filizola,
N., 2006a. Timescale and conditions of chemical weathering
under tropical climate: study of the Amazon basin with
U-series. Geochimica et Cosmochimica Acta, 70, 71–89.
Dosseto, A., Turner, S. P., and Douglas, G. B., 2006b. Uranium-
series isotopes in colloids and suspended sediments: timescale
for sediment production and transport in the Murray-Darling
River system. Earth and Planetary Science Letters, 246,
418–431.
Dosseto, A., Bourdon, B., and Turner, S. P., 2008a. Uranium-series
isotopes in river materials: insights into the timescales of erosion
and sediment transport. Earth and Planetary Science Letters,
265, 1–17.
Dosseto, A., Turner, S. P., and Chappell, J., 2008b. The evolution of
weathering profiles through time: new insights from uranium-
series isotopes. Earth and Planetary Science Letters, 274,
359–371.
Dosseto, A., Turner, S. P., and van Orman, J. A., 2010. Timescales of
Magmatic Processes: From Core to Atmosphere. Oxford: Wiley-
Blackwell. ISBN 1444348264.
Dosseto, A., Buss, H. L., and Suresh, P. O., 2012. Rapid regolith for-
mation over volcanic bedrock and implications for landscape evo-
lution. Earth and Planetary Science Letters, 337–338, 47–55.
Dosseto, A., Buss, H. L., and Chabaux, F., 2014. Age and
weathering rate of sediments in small catchments: the role of hill-
slope erosion. Geochimica et Cosmochimica Acta, 132,
238–258.
Dran, J. C., Langevin, Y., and Petit, J. C., 1988. Uranium isotopic
disequilibrium: reappraisal of the alpha-recoil effect (abst.).
Chemical Geology, 70, 126.
 CHEMICAL WEATHERING (U-SERIES)
Ferrier, K. L., and Kirchner, J. W., 2008. Effects of physical erosion
on chemical denudation rates: a numerical modeling study of
soil-mantled hillslopes. Earth and Planetary Science Letters,
272, 591–599.
Ferrier, K. L., Kirchner, J. W., Riebe, C. S., and Finkel, R. C., 2010.
Mineral-specific chemical weathering rates over millennial time-
scales: measurements at Rio Icacos, Puerto Rico. Chemical
Geology, 277, 101–114.
Fleischer, R. L., 1980. Isotopic disequilibrium of uranium: alpha-
recoil damage and preferential solution effects. Science, 207,
979–981.
Fleischer, R. L., 1982a. Alpha-recoil damage and solution effects in
minerals: uranium isotopic disequilibrium and radon release.
Geochimica et Cosmochimica Acta, 46, 2191–2201.
Fleischer, R. L., 1982b. Nature of alpha-recoil damage: evidence
from preferential solution effects. Nuclear Tracks and Radiation
Measurements, 6, 35–42.
Fleischer, R. L., 1983. Theory of alpha recoil effects on radon
release and isotopic disequilibrium. Geochimica et
Cosmochimica Acta, 47, 779–784.
Fleischer, R. L., and Raabe, O. G., 1978a. On the mechanism of
‘dissolution’ in liquids of PuO2 by alpha decay. Health Physics,
35, 545–548.
Fleischer, R. L., and Raabe, O. G., 1978b. Recoiling alpha-emitting
nuclei. Mechanisms for uranium-series disequilibrium.
Geochimica et Cosmochimica Acta, 42, 973–978.
Gaillardet, J., Dupré, B., and Allègre, C. J., 1995. A global mass
budget applied to the Congo Basin rivers: erosion rates and con-
tinental crust composition. Geochimica et Cosmochimica Acta,
59, 3469–3485.
Gaillardet, J., Dupré, B., Allègre, C. J., and Négrel, P., 1997. Chem-
ical and physical denudation in the Amazon River Basin. Chem-
ical Geology, 142, 141–173.
Gaillardet, J., Dupré, B., Louvat, P., and Allègre, C. J., 1999. Global
silicate weathering and CO2 consumption rates deduced from the
chemistry of large rivers. Chemical Geology, 159, 3–30.
Guthrie, V. A., and Kleeman, J. D., 1986. Changing uranium distri-
butions during weathering of granite. Chemical Geology, 54,
113–126.
Hansen, R. O., and Huntington, G. L., 1969. Thorium movements in
morainal soils of the High Sierra, California. Soil Science, 108,
257–265.
Hansen, R. O., and Stout, P. R., 1968. Isotopic distribution of ura-
nium and thorium in soils. Soil Science, 105, 44–50.
Harmon, R. S., and Rosholt, J., 1982. Igneous Rocks. In Ivanovich,
M., and Harmon, R. S. (eds.), Uranium-Series Disequilibrium:
Application to Environmental Problems. Oxford: Oxford Sci-
ences Publications, pp. 145–166.
Harvey, B. G., 1962. Introduction to Nuclear Physics and Chemis-
try. New Jersey: Prentice Hall Inc., p. 382.
Hashimoto, T., Aoyagi, Y., Kudo, H., and Sotobayashi, T., 1985.
Range calculation of alpha-recoil atoms in some minerals using
LSS-theory. Journal of Radioanalytical and Nuclear Chemistry,
90, 415–438.
Heimsath, A. M., Dietrich, W. E., Nishiizumi, K., and Finkel, R. C.,
1997. The soil production function and landscape equilibrium.
Nature, 388, 358–361.
Heimsath, A. M., Chappell, J., Dietrich, W. E., Nishiizumi, K., and
Finkel, R. C., 2000. Soil production on a retreating escarpment in
southeastern Australia. Geology, 28, 787–790.
Horwitz, E. P., 1996. Extraction chromatography of actinides and
selected fission products: principles and achievement of selectiv-
ity. In International Workshop on the Application of Extraction
Chromatography in Radionuclide Measurement, IRMM, Geel,
Belgium, pp. 27–37.
Horwitz, E. P., Chiarizia, R., Dietz, M. L., and Diamond, H., 1993.
Separation and preconcentration of actinides from media by
167
extraction chromatography. Analytica Chimica Acta, 281,
361–372.
Hsi, D. C.-k., and Langmuir, D., 1985. Adsorption of uranyl onto
ferric oxyhydroxides: application of the surface complexation
site-binding model. Geochimica et Cosmochimica Acta, 49,
1931–1941.
Hussain, N., and Lal, D., 1986. Preferential solution of 234U from
recoil tracks and 234U/238U radioactive disequilibrium in natural
waters. Proceedings of the Indian Academy of Sciences – Earth
and Planetary Sciences, 95, 245–263.
Jenny, H., 1941. Factors of Soil Formation. New York: McGraw-
Hill.
Keech, A. R., West, A. J., Pett-Ridge, J. C., and Henderson, G. M.,
2013. Evaluating U-series tools for weathering rate and duration
on a soil sequence of known ages. Earth and Planetary Science
Letters, 374, 24–35.
Kigoshi, K., 1971. Alpha-recoil thorium-234: dissolution into water
and the uranium-234/uranium-238 disequilibrium in nature. Sci-
ence, 173, 47–48.
Krishnaswami, S., Graustein, W., Turekian, K., and Dowd, J., 1982.
Radium, thorium and radioactive lead isotopes in groundwaters:
application to the in-situ determination of adsorption-desorption
rate constants and retardation factors. Water Resources
Research, 18, 1633–1675.
Langmuir, D., and Herman, J. S., 1980. The mobility of thorium in
natural waters at low temperatures. Geochimica et
Cosmochimica Acta, 44, 1753–1766.
Larsen, M. C., 2012. Landslides and sediment budgets in four
Watersheds in Eastern Puerto Rico. In Murphy, S. F., and
Stallard, R. F. (eds.), Water Quality and Landscape Processes
of Four Watersheds in Eastern Puerto Rico. U.S. Geological
Survey Professional Paper 1789-F, pp.157–178.
Larsen, I. J., Almond, P. C., Eger, A., Stone, J. O., Montgomery,
D. R., and Malcolm, B., 2014. Rapid soil production and
weathering in the Western Alps, New Zealand. Science, 343,
637–640.
Latham, A. G., and Schwarcz, H. P., 1987. On the possibility of
determining rates of removal of uranium from crystalline igne-
ous rocks using U-series disequilibria—1: a U-leach model,
and its applicability to whole-rock data. Applied Geochemistry,
2, 55–65.
Lee, V., DePaolo, D. J., and Christensen, J. N., 2010. Uranium-
series comminution ages of continental sediments: case study
of a Pleistocene alluvial fan. Earth and Planetary Science
Letters, 296, 244–254.
Lenhart, J. J., and Honeyman, B. D., 1999. Uranium (VI) sorption to
hematite in the presence of humic acid. Geochimica et
Cosmochimica Acta, 63, 2891–2901.
Linhard, J., and Scharff, M., 1961. Energy dissipation by ions in the
kev region. Physical Review, 124, 128–130.
Lively, R. S., Harmon, R. S., Levinson, A. A., and Bland, C. J.,
1979. Disequilibrium in the 238Uranium series in samples from
Yeelirrie, Western Australia. Journal of Geochemical Explora-
tion, 12, 57–65.
Lundstrom, C. C., 2003. Uranium-series disequilibria in Mid-Ocean
Ridge basalts: observations and models of basalt genesis. In
Bourdon, B., Henderson, G. M., Lundstrom, C. C., and Turner,
S. P. (eds.), Uranium-Series Geochemistry. Washington, DC:
Mineralogical Society of America. Reviews of Mineralogy and
Geochemistry, Vol. 52, pp. 175–214.
Ma, L., Chabaux, F., Pelt, E., Blaes, E., Jin, L., and Brantley, S. L.,
2010. Regolith production rates calculated with uranium-series
isotopes at Susquehanna/Shale Hills Critical Zone Observatory.
Earth and Planetary Science Letters, 297, 211–255.
Ma, L., Chabaux, F., Pelt, E., Granet, M., Sak, P. B., Gaillardet, J.,
Lebedeva, M., and Brantley, S. L., 2012. The effect of curvature
on weathering rind formation: evidence from Uranium-series
168 CHEMICAL WEATHERING (U-SERIES)
isotopes in basaltic andesite weathering clasts in Guadeloupe.
Geochimica et Cosmochimica Acta, 80, 92–107.
Ma, L., Chabaux, F., West, N., Kirby, E., Jin, L., and Brantley, S. L.,
2013. Regolith production and transport in the Susquehanna
Shale Hills Critical Zone Observatory, Part 1: insights from
U-series isotopes. Journal of Geophysical Research F: Earth
Surface, 118, 722–740.
Maher, K., DePaolo, D. J., and Lin, J. C. F., 2004. Rates of silicate
dissolution in deep-sea sediment: in situ measurement using
U-234/U-238 of pore fluids. Geochimica et Cosmochimica Acta,
68, 4629–4648.
Manceau, A., Charlet, L., Boisset, M. C., Didier, B., and Spadini, L.,
1992. Sorption and speciation of heavy metals on hydrous Fe
and Mn oxides. From microscopic to macroscopic. Applied Clay
Science, 7, 201–223.
Mangini, A., Sonntag, C., Bertsch, G., and Müller, E., 1979. Evi-
dence for a higher natural uranium content in world rivers.
Nature, 278, 337–339.
Martin, A., Dosseto, A., and Kinsley, L. P. J., (under revision). Eval-
uating sample pre-treatment procedures for the uranium-series
surface area dating method. Chemical Geology.
Martin, J. M., Nijampurkar, V., and Salvadori, F., 1978. Uranium
and Thorium isotope behavior in estuarine systems. In Goldberg,
E. D. (ed.), Biogeochemistry of Estuarine Sediments. Paris:
Unesco Press, pp. 111–126.
Mathieu, D., Bernat, M., and Nahon, D., 1995. Short-lived U and Th
isotope distribution in a tropical laterite derived from granite
(Pitinga river basin, Amazonia, Brazil): application to assess-
ment of weathering rate. Earth and Planetary Science Letters,
136, 703–714.
Megumi, K., and Mamuro, T., 1977. Concentration of uranium
series nuclides in soil particles in relation to their size. Journal
of Geophysical Research, 82, 353–356.
Menozzi, D., and Dosseto, A., 2013. U-series isotope composition
of primary minerals to determine regolith production rates on
three different lithologies (abst.). In Proceedings of the 2013
Goldschmidt Conference, Florence, Italy, August 25–30, 2013.
Abstract accessed October 2014 at; http://goldschmidt.info/
2013/abstracts/finalPDFs/1741.pdf
Moreira-Nordemann, L. M., 1977. Etude de la vitesse d’altération
des roches au moyen de l’uranium utilisé comme traceur naturel.
Application à deux bassins du Nord Est du Brésil Ph.D.,
Université Paris VI.
Moreira-Nordemann, L. M., 1980. Use of 234U/238U disequilibrium
in measuring chemical weathering rate of rocks. Geochimica et
Cosmochimica Acta, 44, 103–108.
Navarre-Sitchler, A., and Brantley, S. L., 2007. Basalt weathering
across scales. Earth and Planetary Science Letters, 261, 321–334.
Osmond, J. K., and Cowart, J. B., 1976. The theory and uses of nat-
ural uranium isotopic variations in hydrology. Atomic Energy
Reviews, 14, 621–679.
Osmond, J. K., and Cowart, J. B., 1982. Groundwater. In Ivanovich,
M., and Harmon, R. S. (eds.), Uranium-Series Disequilibrium:
Application to Environmental Problems. Oxford: Oxford Sci-
ences Publications, pp. 202–245.
Osmond, J. K., Cowart, J. B., and Ivanovich, M., 1983. Uranium
isotopic disequilibrium in ground water as an indicator of anom-
alies. Journal of Applied Radiation and Isotopes, 34, 283–308.
Pelt, E., Chabaux, F., Innocent, C., Navarre-Sitchler, A. K., Sak,
P. B., and Brantley, S. L., 2008. Uranium-thorium chronometry
of weathering rinds: rock alteration rate and paleo-isotopic
record of weathering fluids. Earth and Planetary Science Let-
ters, 276, 98–105.
Pelt, E., Chabaux, F., Stille, P., Innocent, C., Ghaleb, B., Gérard, M.,
and Guntzer, F., 2013. Atmospheric dust contribution to the bud-
get of U-series nuclides in soils from the Mount Cameroon vol-
cano. Chemical Geology, 341, 147–157.
Pett-Ridge, J. C., Monastra, V. M., Derry, L. A., and
Chadwick, O. A., 2007. Importance of atmospheric inputs and
Fe-oxides in controlling soil uranium budgets and behavior
along a Hawaiian chronosequence. Chemical Geology, 244,
691–707.
Pett-Ridge, J. C., Derry, L. A., and Barrows, J. K., 2009a. Ca/Sr and
87Sr/86Sr ratios as tracers of Ca and Sr cycling in the Rio Icacos
watershed, Luquillo Mountains, Puerto Rico. Chemical Geol-
ogy, 267, 32–45.
Pett-Ridge, J. C., Derry, L. A., and Kurtz, A. C., 2009b. Sr isotopes
as a tracer of weathering processes and dust inputs in a tropical
granitoid watershed, Luquillo Mountains, Puerto Rico.
Geochimica et Cosmochimica Acta, 73, 25–43.
Plater, A. J., Ivanovich, M., and Dugdale, R. E., 1992. Uranium series
disequilibrium in river sediments and waters: the significance of
anomalous activity ratios. Applied Geochemistry, 7, 101–110.
Pliler, R., and Adams, J. A. S., 1962. The distribution of thorium
and uranium in a Pennsylvanian weathering profile. Geochimica
et Cosmochimica Acta, 26, 1137–1146.
Porcelli, D., Andersson, P. S., Wasserburg, G. J., Ingri, J., and
Baskaran, M., 1997. The importance of colloids and mires for
the transport of uranium isotopes through the Kalix River water-
shed and Baltic Sea. Geochimica et Cosmochimica Acta, 61,
4095–4113.
Porcelli, D., Andersson, P. S., Baskaran, M., and Wasserburg, G. J.,
2001. Transport of U- and Th-series nuclides in a Baltic Shield
watershed and the Baltic Sea. Geochimica et Cosmochimica
Acta, 65, 2439–2459.
Riebe, C. S., Kirchner, J. W., and Finkel, R. C., 2003. Long-term
rates of chemical weathering and physical erosion from cosmo-
genic nuclides and geochemical mass balance. Geochimica et
Cosmochimica Acta, 67, 4411–4427.
Rihs, S., Prunier, J., Thien, B., Lemarchand, D., Pierret, M.-C.,
and Chabaux, F., 2011. Using short-lived nuclides of the
U- and Th-series to probe the kinetics of colloid migration in
forested soils. Geochimica et Cosmochimica Acta, 75,
7707–7724.
Rihs, S., Gontier, A., Turpault, M. P., Lemarchand, D., Voinot, A.,
and Chabau, F., 2013. Insight into Biotite Weathering Rate Using
U-Series Isotopes (abst.). In Proceedings of the 2013
Goldschmidt Conference. Florence, Italy, August 25–30, 2013.
Abstract accessed October 2014 at; http://goldschmidt.info/
2013/abstracts/originalPDFs/5107.pdf
Robinson, L. F., Henderson, G. M., Hall, L., and Matthews, I., 2004.
Climatic control of riverine and seawater uranium-isotope ratios.
Science, 305, 851–854.
Rosholt, J. N., 1982. Mobilization and weathering. In Ivanovich,
M., and Harmon, R. S. (eds.), Uranium-Series Disequilibrium:
Application to Environmental Problems. Oxford: Oxford Sci-
ences Publications, pp. 167–180.
Rosholt, J., Doe, B., and Tatsumoto, M., 1966. Evolution of the
isotopic composition of uranium and thorium in soil profiles.
Geological Society of America Bulletin, 77, 987–1004.
Schultz, M. K., Burnett, W. C., and Inn, G. W., 1998. Evaluation of a
sequential extraction method for determining actinide fraction-
ation in soils and sediments. Journal of Environmental Radioac-
tivity, 40, 155–174.
Scott, M. R., 1968. Thorium and uranium concentrations and iso-
tope ratios in river sediments. Earth and Planetary Science Let-
ters, 4, 245–252.
Scott, R. D., MacKenzie, A. B., and Alexander, W. R., 1992. The
interpretation of 238U-234U-230Th-226Ra disequilibria produced
by rock-water interactions. Journal of Geochemical Exploration,
45, 323–343.
Sheng, Z. Z., and Kuroda, P. K., 1984. Alpha-recoil effects of ura-
nium isotopes in radioactive minerals. Radiochimica Acta, 37,
93–98.
 CHERT
Sheng, Z. Z., and Kuroda, P. K., 1986a. Further studies on the sep-
aration of acid residues with extremely high 234U/238U ratios
from a Colorado Carnotite. Radiochimica Acta, 40, 95–102.
Sheng, Z. Z., and Kuroda, P. K., 1986b. Isotopic fractionation of
uranium: extremely high enrichments of 234U in the acid-
residues of a Colorado Carnotite. Radiochimica Acta, 39,
131–138.
Shirvington, P. J., 1983. Fixation of radionuclides in the 238U decay
series in the vicinity of mineralized zones: 1. The Austatom Ura-
nium Prospect, Northern Territory, Australia. Geochimica et
Cosmochimica Acta, 47, 403–412.
Sun, H., and Furbish, D. J., 1995. Moisture content effect on radon
emanation in porous media. Journal of Contaminant Hydrology,
18, 239–255.
Suresh, P. O., Dosseto, A., Hesse, P. P., and Handley, H. K., 2011.
Time Constraints on Soil Evolution from Uranium-Series Iso-
topes in the South-Eastern Australian Highlands: Evidences
for a Coupling Between Soil Erosion and Production (abst.).
2011 Fall Meeting, American Geophysical Union, San
Francisco, CA, December 5–9, Abstract #EP41B-0604.
Suresh, P. O., Dosseto, A., Hesse, P. P., and Handley, H. K.,
2013. Soil formation rates determined from uranium-series
isotope disequilibria in soil profiles from southeastern
Australian Highlands. Earth and Planetary Science Letters,
379, 26–37.
Syromyatnikov, N. G., and Ivanova, E. I., 1968. A radiologic study
of the granites of the Ortau massif, central Kazakhstan. Geo-
chemistry International, 5, 299–305.
Syromyatnikov, N. G., Ibrayev, R. A., and Mukashev, F. A., 1967.
Interpretation of uranium anomalies in arid regions with the aid
of the U234/U238 ratio. Geokhimiya, 7, 482–496.
Szalay, A., 1964. Cation exchange properties of humic acids and
their importance in the geochemical enrichment of UO2++ and
other cations. Geochimica et Cosmochimica Acta, 28,
1605–1614.
Tieh, T. T., and Ledger, E. B., 1981. Fission track study of uranium
in two granites of central texas. Contributions to Mineralogy and
Petrology, 76, 12–16.
Titayeva, N. A., and Veksler, T. I., 1977. The state of radioactive
equilibrium in the uranium and thorium series as an indicator
of migration of radioactive elements and active interaction
between phases under natural conditions. Geochemistry Interna-
tional, 14, 99–107.
Turner, S., van Calsteren, P., Vigier, N., and Thomas, L., 2001.
Determination of thorium and uranium isotope ratios in
low-concentration geological materials using a fixed multi-
collector-ICP-MS. Journal of Analytical Atomic Spectrometry,
16, 612–615.
Turner, B. F., Stallard, R. F., and Brantley, S. L., 2003a. Investiga-
tion of in situ weathering of quartz diorite bedrock in the Rio
Icacos basin, Luquillo Experimental Forest, Puerto Rico. Chem-
ical Geology, 202, 313–341.
Turner, S., Bourdon, B., and Gill, J., 2003b. Insights into magma
genesis at convergent margins from U-series isotopes. In
Bourdon, B., Henderson, G. M., Lundstrom, C. C., and Turner,
S. P. (eds.), Uranium-Series Geochemistry. Washington, DC:
Mineralogical Society of America. Reviews of Mineralogy and
Geochemistry, Vol. 52, pp. 255–315.
Viers, J., Dupre, B., Polve, M., Schott, J., Dandurand, J.-L., and
Braun, J.-J., 1997. Chemical weathering in the drainage basin
of a tropical watershed (Nsimi-Zoetele site, Cameroon); compar-
ison between organic-poor and organic-rich waters. Chemical
Geology, 140, 181–206.
Vigier, N., and Bourdon, B., 2011. Constraining rates of chemical
and physical erosion using U-series radionuclides. In Baskaran,
M. (ed.), Handbook of Environmental Isotope Geochemistry.
Berlin/Heidelberg: Springer, pp. 553–571.
169
Vigier, N., Bourdon, B., Lewin, É., Dupré, B., Turner, S., van
Calsteren, P., Subramanian, V., and Allègre, C. J., 2005. Mobility
of U-series nuclides during basalt weathering: an example of the
Deccan Traps (India). Chemical Geology, 219, 69–91.
Waite, T. D., Davis, J. A., Payne, T. E., Waychunas, G. A., and Xu,
N., 1994. Uranium(VI) adsorption to ferrihydrite: application of
a surface complexation model. Geochimica et Cosmochimica
Acta, 58, 5465–5478.
White, A. F., and Brantley, S. L., 2003. The effect of time on the
weathering of silicate minerals: why do weathering rates differ
in the laboratory and field? Chemical Geology, 202, 479–506.
White, A. F., Blum, A. E., Schulz, M. S., Vivit, D. V., Stonestrom,
D. A., Larsen, M., Murphy, S. F., and Eberl, D., 1998. Chemical
weathering in a tropical watershed, Luquillo Mountains, Puerto
Rico: I. Long-term versus short-term weathering fluxes.
Geochimica et Cosmochimica Acta, 62, 209–226.
Wittmann, H., von Blanckenburg, F., Maurice, L., Guyot, J. L.,
Filizola, N., and Kubik, P. W., 2010. Sediment production and
delivery in the Amazon River basin quantified by in situ-
produced cosmogenic nuclides and recent river loads. Geologi-
cal Society of America Bulletin, 123, 934–950.
Cross-references
Aquifer Characteristics (U-Series)
Carbonates, Pedogenic (U-Series)
Mass Spectrometry
Paleosol
Radiation and Radioactivity
Radiation Defect
Terrestrial Cosmogenic Nuclide Dating
U-Series Dating
CHERT
L. Paul Knauth
School of Earth and Space Exploration, Arizona State
University, Tempe, AZ, USA
Definition
Cherts are nodular masses and beds of microcrystalline
quartz that replace carbonate sediment or opaline biologic
oozes of diatoms or radiolarians during burial. The
replacement can occur during early to deep burial, so chert
is younger than the time of host sediment deposition by
variable amounts not easily determined. The quartz itself
contains no radionuclides that can be used for age-dating
the time of crystallization. Detrital impurity grains may
contain datable material, but the age is that of the source
area and not of the sediment. The value of cherts in age-
dating relates to the unsurpassed preservation that silicifi-
cation affords fossils.
Silicification involves a slow micron-by-micron
replacement process. Silicified fossils better survive the
alteration effects of subsequent burial, uplift, and
weathering. Acid dissolution of carbonate frees isolated
examples, and these probably represent the majority of
invertebrate fossils used to establish the paleontological
record and to age-date sedimentary rocks. Ages
170 CHROMATOGRAPHY
established from the fossils relate to the time of original
deposition even though silicification is secondary.
Organic remains in nodular cherts can be preserved down
even to the cellular level. Microfossils in such cherts pro-
vide much of the record of the Precambrian life, but their
evolutionary history is still insufficiently established to
be widely useful for age-dating.
In Cenozoic and Paleozoic bedded cherts, microquartz
in the tiny volumes originally occupied by radiolarian tests
resist dissolution in HF and NaOH relative to the matrix
microquartz. Gentle etching reveals the radiolarian mor-
phology so well that biostratigraphic ages can be
established with surprising precision. Radiolarian age-
dating of bedded chert sequences was a breakthrough in
unraveling the structural geology and history of the great
folded mountain sequences in the Alps and Western
United States (e.g., De Wever et al., 2001). Tertiary diato-
maceous oozes are abundant and incompletely
transformed into microquartz. Fossil ages can be deter-
mined from these usually without acid etching.
The 18O/16O ratio of cherts is highly variable but gener-
ally decreases in progressively older cherts (Knauth,
2005). The explanation remains controversial but is likely
due to past higher climatic temperatures. The highest
values for Archean examples are lower than the lowest
values for younger examples, so the 18O/16O ratio is
a crude age indicator of cherts older than 2.5 billion years.
Dissolved argon in the silicifying fluid can be trapped
within tiny fluid inclusions in the microquartz that precip-
itates from it. Since the isotopic composition of atmo-
spheric Ar evolves with time, appropriate samples might
yield approximate age dates. Analyses of 3.5 billion-
year-old hydrothermal quartz have now better constrained
atmospheric Ar isotopic evolution, so this method holds
considerable promise (Pujol et al., 2013).
While the geochemical and radiometric age-dating of
cherts remains limited, biostratigraphic ages of the
enclosed fossils remain among the most accurate achieved
for the Phanerozoic sedimentary rocks. The exquisite
microfossil record in older cherts, the “amber of the Pre-
cambrian,” offers hope for similar utility extending back
in time to the earliest sediments.
Bibliography
De Wever, P., Dumitrica, P., Caulet, J. P., Nigrini, C., and Caridroit,
M., 2001. Radiolarians in the Sedimentary Record. Amsterdam:
Gordon and Breach Scientific Publishers, p. 533.
Knauth, L. P., 2005. Temperature and salinity history of the
Precambrian ocean: implications for the course of microbial evo-
lution. Paleogeography, Paleoclimatology, Paleoecology, 219,
53–69.
Pujol, M., Marty, B., Rurgess, R., and Turner, G., 2013. Argon iso-
topic composition of Archaean atmosphere probes early Earth
geodynamics. Nature, 498, 87–90.
Cross-references
Biostratigraphy
Hominid Evolution Timescale
CHROMATOGRAPHY
John F. Wehmiller
Department of Geological Sciences, University of
Delaware, Newark, DE, USA
Synonyms
Method for chemical separation
Definition
Chromatography is a general term describing chemical
procedures for the separation of mixtures of elements or
compounds. Chromatographic methods are used for both
preparation and analysis of samples. A fundamental prin-
ciple of chromatography is that components of a mixture
will have different distribution or partition coefficient
between a mobile and stationary phase. Distribution coef-
ficients are a measure of a compound “preference” for
either the mobile or the stationary phase. A solution
containing compounds to be separated is usually dissolved
in the mobile phase, which then is allowed to flow through
a stationary phase. The dissolved components elute at dif-
ferent times in this flow system, the times being deter-
mined by the distribution coefficients – those that are
preferentially retained by the stationary phase appear later
in the chromatogram, which is the graphical output of the
chromatographic separation.
Preparative chromatography is part of the purification
of a sample prior to more advanced instrumental analysis.
Many early advances in chromatography involved the
extraction and separation of natural pigments found on
biological samples and sediments. Examples of prepara-
tive chromatographic techniques used on geochronology
include the separation of individual compounds (amino
acids, lignins, chlorophyll) for radiocarbon analysis (Hatte
et al., 2008; Marom et al., 2012) and separation of ele-
ments from solute mixtures for isotope ratio geochronol-
ogy (Sanna et al., 2010; Isnard et al., 2012; Souders
et al., 2012; Rotenberg et al., 2012). Analytical methods
that follow these preparative steps often involve high-res-
olution, sensitive instrumental analytical chromatographic
methods.
Analytical chromatographic procedures in geochronol-
ogy include high-performance liquid chromatography
(HPLC), gas chromatography (GC), size-exclusion chro-
matography, and various combinations of liquid or gas
chromatography with instrumental mass spectrometric
analysis. HPLC methods involve a liquid containing the
mixture to be analyzed flowing under high pressure
through a small diameter column containing the stationary
phase. A range of HPLC methods exist, each tailored to
the specific analysis to be conducted. Some involve ion-
exchange chromatography (IEC), where the stationary
phase is charged and interacts with solute ions as they flow
through the column. Adjusting the pH of the solute during
the analytical run will change the affinity of the solutes for
the stationary phase. IEC is a common tool in amino acid
 CLAYS AND GLAUCONITES (K–Ar/Ar–Ar)
racemization geochronology. Other forms of HPLC rely
upon different polarities of the stationary and mobile
phases, reverse phase chromatography (RPC – where the
mobile phase is more polar than the stationary phase)
now being a common procedure for analysis of amino acid
enantiomers (D/L values)(Kaufman and Manley, 1998;
Penkman et al., 2013). GC methods involve a gas as the
mobile phase and a thin liquid coating on a narrow-bore
capillary column as the stationary phase. The mixture to
be analyzed is vaporized in the injection port of a gas chro-
matograph, being carried through the liquid column by
a carrier gas. Elution order is determined by the affinity
of compounds for the solid phase. Flow rate and overall
elution times are influenced by carrier gas pressure and
column temperature. GC methods have also been tradi-
tional in amino acid racemization geochronology
(Wehmiller and Miller, 2000; Wehmiller, 2013). Size-
exclusion chromatography is a preparative chromato-
graphic method useful for separation of compounds based
on their molecular weight, an important step in the further
analysis of individual compounds or mixtures found in
geological samples. Mass spectrometric analyses often
are focused on isotope ratio determinations (often with
geochronological applications) and many examples exist
where gas- or liquid-chromatographic systems are
coupled with mass spectrometric systems to separate com-
plex mixtures prior to the final mass spectrometric
analysis.
Bibliography
Hatte, C., Hodgins, G., Jull, A. J. T., Bishop, B., and Tesson, B.,
2008. Marine chronology based on 14C dating on diatom pro-
teins. Marine Chemistry, 109, 143–151.
Isnard, H., Gourgiotis, A., Mialle, S., Bourgeois, M., Vio, L.,
Nonell, A., and Chartier, F., 2012. Coupling between separative
techniques and inductively coupled plasma mass spectrometry
for lanthanides separation and on-line isotope ratio measure-
ments. Mineralogical Magazine, 76, 1883.
Kaufman, D. S., and Manley, W. F., 1998. A new procedure for
determining DL amino acid ratios in fossils using reverse phase
liquid chromatography. Quaternary Science Reviews (Quater-
nary Geochronology), 17, 987–1000.
Marom, A., McCullagh, J. S. O., Higham, T. F. G., Sinitsyn, A. A.,
and Hedges, R. E. M., 2012. Single amino acid radiocarbon dat-
ing of Upper Paleolithic modern humans. Proceedings National
Academy of Sciences, 109, 6878–6871.
Penkman, K. E. H., Preece, R. C., Bridgland, D. R., Keen, D. H.,
Meijer, T., Parfitt, S. A., White, T. S., and Collins, M. J., 2013.
An aminostratigraphy for the British Quaternary based on
Bithynia opercula. Quaternary Science Reviews, 61, 111–134.
Rotenberg, E., Davis, D. W., Amelin, Y., Ghosh, S., and Bergquist,
B. A., 2012. Determination of the decay-constant of 87Rb by lab-
oratory accumulation of 87Sr. Geochimica et Cosmochimica
Acta, 85, 41–57.
Sanna, L., Saez, F., Simonsen, S., Constantin, S., Calaforra, J., Forti,
P., and Lauritzen, S., 2010. Uranium-series dating of gypsum
speleothems; methodology and examples. International Journal
of Speleology, 39, 35–46.
Souders, A. K., and Sylvester, P. J., 2012. Use of MC-ICPMS for
laser ablation U/Pb geochronology of baddeleyite. Mineralogi-
cal Magazine, 76, 2397.
171
Wehmiller, J. F., 2013. Interlaboratory comparison of amino acid
enantomeric ratios in Pleistocene fossils. Quaternary Geochro-
nology, 16, 173–182.
Wehmiller, J. F., and Miller, G. H., 2000. Aminostratigraphic dating
methods in Quaternary and geology. In Noller, J. S., Sowers,
J. M., and Lettis, W. R. (eds.), Quaternary Geochronology,
Methods and Applications. Washington, DC: American Geo-
physical Union ReferenceShelf, Vol. 4, pp. 187–222.
Web links
http://www.iupac.org/publications/pac/1993/pdf/6504  0819.pdf
http://ocw.mit.edu/resources/res-5-0001-digital-lab-techniques-manual-
spring-2007/videos/# 3, 4, and 10
http://www.rpi.edu/dept/chem-eng/Biotech-Environ/CHROMO/
chromintro.html
http://www.justchromatography.com/
Cross-references
Amino Acid Racemization Dating
Radiocarbon Dating
Rubidium–Strontium Dating, Hydrothermal Events
U-Series Dating
CLAYS AND GLAUCONITES (K–Ar/Ar–Ar)
Horst Zwingmann
CSIRO Earth Science and Resource Engineering,
Australian Resources Research Centre, Perth,
WA, Australia
School of Earth and Environment, The University of
Western Australia, Crawley, WA, Australia
Department of Applied Geology, Curtin University,
Bentley, WA, Australia
Institut de Recherche sur les Archéomatériaux, UMR
5060 CNRS – Université de Bordeaux, Centre de
Recherche en Physique Appliquée à l’Archéologie
(CRP2A), Maison de l’archéologie, Pessac, France
Introduction
Both conventional K–Ar and 40Ar/39Ar dating methods
can be applied to clay minerals. Clay minerals are sheet
silicates characterized by their small grain size (<2 mm),
thus requiring specialized separation and analytical
methods for isotopic dating. Detailed accounts of the con-
ventional K–Ar and 40Ar/39Ar dating technique have been
described by Dalrymple and Lanphere (1969), Faure
(1986), Dickin (1995), McDougall and Harrison (1999),
and Kelley (2002). The ages of clay minerals can be deter-
mined by measuring the amount of the argon isotope 40Ar
relative to the potassium content. 40Ar is produced by the
radioactive decay of the potassium isotope 40K. Minerals
generally contain negligible amounts of argon when they
are formed, although small amounts of atmospheric argon
may adhere to samples, which can be corrected for by using
the known atmospheric 40Ar/39Ar ratio of 295.5 (Steiger
and Jäger, 1977). Thus, by measuring the ratio of 40Ar to
172 CLAYS AND GLAUCONITES (K–Ar/Ar–Ar)
40
K, and knowing the decay rate of 40K, it is possible to cal-
culate the time since the clay mineral formed.
What are clays?
Clay minerals are phyllosilicates; a comprehensive sum-
mary of their properties and genesis has been compiled
by Srodón and Eberl (1984), Velde (1992), and Meunier
(2005). Two clay minerals groups, illite and glauconite,
can be dated using isotopic methods. Illite is a general term
for a 10 angstrom (Å) potassium dioctahedral mica-like clay
mineral common in sedimentary rocks such as shales and
sandstones. Glauconite is an iron potassium phyllosilicate
mineral of characteristic green color that occurs in sandstone
and carbonate-rich sedimentary rocks (Velde, 2004).
As a generalization, illite and glauconite group clays
have a dioctahedral mineral structure: K(Mg,Fe,Al)2(Si,
Al)4O10, with illite containing K < 0.9 (Al), and glauco-
nite characterized by the presence of Fe with K <0.9
(Al, Fe, Mg).
Clay minerals occur in different geological environments
as both detrital and authigenic phases. One of the main chal-
lenges in clay geochronology is the separation of, and differ-
entiation between, authigenic and detrital mineral phases
that is necessary to obtain meaningful geological ages.
Why date clays?
Clay minerals are widespread in a variety of geological
environments and their occurrence manifests numerous
important processes throughout Earth history. Due to their
capacity to swell and incorporate water, clay minerals are
considered important in the evolution of life (i.e.,
Hashizume, 2012). They form significant economic reser-
voir rocks for water and energy resources (Nadeau, 2011).
Glauconite is a common authigenic sedimentary mineral
and has been used to constrain the geological timescale
where biostratigraphic control is limited (Velde, 2004).
Clay dating by K–Ar and 40Ar/39Ar methods:
a history and literature review
Clauer and Chaudhuri (1995) and Meunier and Velde
(2004) summarized the impact, use, and limits of different
isotope dating and tracing systems including conventional
K–Ar and 40Ar/39Ar dating of sedimentary systems. Arm-
strong (1991) and Clauer et al. (2012) summarized the sci-
entific history and technical developments of the
conventional K–Ar and 40Ar/39Ar method applied to illite
clay minerals. Fundamental nanoparticles of illite with
sizes of <0.02 mm or <20 nm can be dated successfully
(Clauer et al., 1997).
The concept of glauconite K–Ar dating was described
by Odin et al. (1982). Smith et al. (1993, 1998) report
40
Ar/39Ar dating of single grain glauconite minerals using
micro-encapsulation to address 39Ar recoil. Numerous
glauconite dating studies are reported in the literature
(e.g., Clauer et al., 2005; Derkowski et al., 2009). How-
ever, some controversy exists over the merits of K–Ar
and 40Ar/39Ar glauconite dating compared to zircons
separated from the same stratigraphic units and dated by
U–Pb geochronology (Selby, 2009).
K–Ar and 40Ar/39Ar dating of clays: basic principles
The K–Ar and 40Ar/39Ar methods allow dating of clay
minerals formed since the formation of the solar system
(Dalrymple, 1991) until a few thousand years before pre-
sent (Renne et al., 1997), because of the relatively short
half-life of about 1,250 Ma ago (Ma) of the 40K isotope.
Special care is required in dating young geological sam-
ples due to atmospheric 40Ar/39Ar contamination and error
magnification (McDougall and Harrison, 1999). Ages are
reported to international timescales such as Gradstein
et al. (2012).
The age of clay minerals determined by the K–Ar or
40
Ar/39Ar method is valid when the following assump-
tions are satisfied:
1. The clay mineral must behave as a closed system. The
system should not have lost any radiogenic 40Ar pro-
duced by decay of 40K, which requires a closed
system relative to radiogenic 40Ar and K since clay
formation.
2. The radiogenic 40Ar should only be produced by decay
of 40K. No 40Ar should have been incorporated into the
mineral after its formation.
3. The atmospheric 40Ar should be adjusted to the
present-day 40Ar/39Ar value of 295.5.
Dating authigenic clay mineral crystallization involves
a fundamentally different approach from dating detrital
clays (Meunier et al., 2004). Theoretical aspects that need
to be considered when trying to date authigenic clay min-
erals include: (1) contamination effects, i.e., mixing with
other K-bearing phases, (2) Ar closure, i.e., loss or capture
of radiogenic 40Ar during crystallization, (3) the duration
of K accumulation, and (4) initial clay mineralogy (illite,
illite–smectite).
There is a fundamental difference between the classical
mineral cooling model, which is characterized by the Ar
“closure temperature”, considered to be of short duration
relative to the mineral “life time,” and that pertaining to
clays (Faure and Mensing, 2005). Dating clay minerals
involves the assumption that a newly formed authigenic
clay mineral can be dated by isotopic techniques indepen-
dent of the concept of a closure temperature. This crystal-
lization process can encompass different timescales. If the
time interval over which an authigenic mineral has crystal-
lized is shorter than the analytical precision of the method,
described in 1 or 2 sigma values, an actual “age” is
recorded. However, in some geological case studies, min-
eral authigenesis and crystallization can extend over
extensive time periods, larger than the analytical uncer-
tainties, and an “integrated age” is recorded (Meunier
and Velde, 2004). Smectite–illite-type minerals growing
at any given temperature, labeled as the “Ar retention tem-
perature” (70
C), will start the K–Ar radiometric chro-
nometer. Instead of a simple cooling model, the K–Ar
 CLAYS AND GLAUCONITES (K–Ar/Ar–Ar)
dating of clays may be subjected to further recrystalliza-
tion episodes when the temperature exceeds the crystalli-
zation temperature (e.g., Clauer and Chaudhuri, 1995,
1998; Meunier et al., 2004).
Ar diffusion effects have been investigated for fine-
grained clay minerals. The general theory of Ar diffusion
was summarized by Crank (1975) and Watson and Baxter
(2007) and its geological applications to the K–Ar and
40
Ar/39Ar methods have been outlined by McDougall
and Harrison (1999). Huon et al. (1993) described an Ar
diffusion model using clay particle morphology as the
shape of fine-grained clay minerals will influence diffu-
sion. Models employing cylindrical, planar, and spherical
shapes have been applied in investigating or constraining
the influence of thermal overprints on the preserved ages
in clay minerals. Moreover, only limited Ar diffusion
parameters for illite (muscovite) have been published
(Wijbrans and McDougall, 1986; Harrison et al., 2009).
Clay dating methods
K-bearing clay minerals such as illite and glauconite can
be dated by the conventional “spiked” K–Ar, the
“unspiked” K–Ar, and 40Ar/39Ar methods.
Classical spiked K–Ar clay dating
For dating K-bearing rocks or minerals, it is necessary to
determine the concentrations of 40K and of radiogenic
40
Ar on two aliquots of the same sample. Since two splits
of any sample are required, a critical assumption of homo-
geneity of the sampled clays may be violated. Due to the
hygroscopic nature of clays, special care has to be taken
in the preparation of both K and Ar sample splits. For Ar
analysis by noble gas mass spectrometry, sample splits
are loaded into clean Al, Cu, or Mo foil, weighed and
preheated to 80
C for several hours to remove moisture,
and reweighed using a high-precision balance. The mea-
sured dry weight is used in the K–Ar age calculation.
For both K and Ar analytical procedures, international
mineral standards and blank analyses are performed for
quality control purposes. Farley et al. (2013a) developed
a new double-spiked K–Ar dating method using a solid
39
Ar and 41K spike for K and Ar isotope measurements.
This method uses a single mass spectrometer and double
spiking requires a single sample aliquot, enabling in situ
age determination on other planet surfaces such as Mars
(Farley et al., 2013b).
Unspiked K–Ar clay dating
The theory and analytical procedures of the unspiked K–
Ar method have been summarized by Cassignol and Gillot
(1982) and Sudo et al. (1996). The unspiked K–Ar
method is able to date geologically very young basalt sam-
ples (several kyr) with high atmospheric 40Ar contamina-
tion. With the use of this procedure on clays, the initial
40
Ar/36Ar is calculated from the present atmospheric
38
Ar/36Ar isotope ratio assuming mass-dependent isotopic
fractionation during mineral formation.
173
40
Ar/39Ar clay dating
The basic principles, applications, and analytical methods
of dating clay minerals by 40Ar/39Ar dating are summa-
rized by McDougall and Harrison (1999) and Clauer
et al. (2012) and highlight technical challenges related to
the micron-scale grain size of clay minerals. Numerous
studies have documented the 39Ar recoil phenomenon,
resulting from irradiation of the samples by fast neutrons,
needed to transmute 39K into 39Ar. This recoil effect,
which can cause problems of isotopic loss at the surface
of small clay crystals, represents a major disadvantage
for clay 40Ar/39Ar dating, as a precise 39Ar measurement
is required to calculate indirectly the amount of
K present in the mineral phases. Turner and Cadogan
(1974) and Kelley (2002) calculated the distances of
39
Ar recoil during irradiation to be a mean of 0.08 mm,
refined by Onstott et al. (1995) and Villa (1997). Foland
et al. (1992) developed a sample encapsulation method,
which allows capture of 39Ar released during irradiation
to correct 40Ar/39Ar ages of illite and glauconite (Smith
et al., 1993; Onstott et al., 1995; Smith et al., 1998). Clauer
et al. (2012) suggested that the encapsulation method
requires a standardization process to allow comparison
of data obtained in different laboratories. 40Ar/39Ar clay
dating is controversial as shown by the literature, mainly
due to different interpretations of the 39Ar recoil effect
(Dong et al., 1995; Yun et al., 2010; Clauer et al., 2011,
2012). In order to minimize 39Ar recoil losses, special
sample pre-treatments such as pre-heating under vacuum,
compaction under vacuum, neutron irradiation in vacuum,
and/or using a different neutron source (e.g., deuterium–
deuterium) might offer new analytical pathways in reduc-
ing 39Ar recoil effects (Kapusta et al., 1997; Renne
et al., 2005).
Applications of clay dating
Clay minerals occur in different geological settings and
are associated with soil formation, erosion and sedimenta-
tion processes, and diagenesis processes that occur during
burial from surface locations (Velde, 1992; Meunier
and Velde, 2004). The K–Ar and 40Ar/39Ar dating of dia-
genetic events in sedimentary systems faces methodolog-
ical challenges that are discussed within four main
applications.
Deep burial clay mineral authigenesis in sandstones
Dating of illite offers the prospect of establishing the abso-
lute timing of diagenetic events such as heating or fluid
flow events within a sedimentary basin (Pevear, 1999).
The radiogenic isotope systematics of sedimentary
rocks is complex due to the intimate mixture of minerals
of different origins such as detrital phases possibly from
a variety of sources and authigenic minerals. It is therefore
often difficult to unambiguously interpret measured ages.
Special sample preparation techniques involving freeze–
thaw disaggregation to avoid over-crushing and extensive
size separation to reduce the proportion of detrital phases
174 CLAYS AND GLAUCONITES (K–Ar/Ar–Ar)
have been developed (Liewig et al., 1987). Progressive
size reduction to submicron size fractions (<0.1 mm or
finer) increases the proportion of authigenic clay phases
in the clay component and minimizes contamination.
The most reliable isotopic ages for authigenic clay min-
erals are obtained from the finest size fractions. The valid-
ity and importance of the assumptions involved in K–Ar
or 40Ar/39Ar dating of authigenic illite must be carefully
addressed, and the sample material characterized using
a wide range of methods such as x-ray diffraction
(XRD), scanning electron microscopy (SEM), transmis-
sion electron microscopy (TEM), and particle
granulometry. Authigenic illite formation is dependent
on several parameters including: temperature and pressure
ranges and formation fluid chemistry (Hamilton
et al., 1992; Wilkinson and Haszeldine, 2002). Variables
that control the rate of illite growth are temperature, pres-
sure, concentration of reactants (K, Si), rate of supply of
reactants, pore fluid velocity, rate of diffusion or active
fluid flow, time, and the presence of an aqueous pore fluid.
Comprehensive summaries concerning authigenic illite
dating can be found in Clauer and Chaudhuri (1995),
Clauer et al. (2012), Hamilton et al. (1989, 1992), Worden
and Morad (2003), Meunier and Velde (2004), and
Meunier et al. (2004).
Smectite to illite transition during burial diagenesis
in shales
Since Aronson and Hower’s (1976) pioneering study
focusing on burial diagenesis in the US Gulf coast, numer-
ous studies have investigated other basins (Girard
et al., 1989; Clauer and Chaudhuri, 1995). The implica-
tions for dating of the illite to smectite transition due to
burial diagenesis have been summarized in Meunier and
Velde (2004) and Frey and Robinson (1999). Illite ages
are interpreted as maximum ages because the neoformed
illite captured in the finest separated size fraction is
formed at the ends of filamentous grains, representing
the most recently grown illite, due to Ostwald ripening
processes (see Eberl and Środoń, 1988; Eberl
et al., 1990). Illite grain-size fractions of neoformed illite
are mixtures of illite particles formed at different times
during growth, and this growth history is usually investi-
gated by dating a range of different grain-size fractions
(0.02 to 2–6 mm). Hunziker et al. (1986) and Pevear
(1999) discussed the influence of detrital mineral phases
on authigenic illite formation in detail for sedimentary sys-
tems. To resolve the influence of detrital contamination,
a regression method was proposed (Pevear, 1999), in
which an extrapolated diagenetic age should represent
the mean integrated age of the time interval over which
illite crystallizes. This method addresses the contamina-
tion effect due to intimate authigenic–detrital clay mineral
phase mixing at very small grain sizes (<2 mm). In a differ-
ent approach focusing on the detrital clay mineral
end-member, van Laningham and Mark (2011) reported
40
Ar/39Ar standard mineral mixture step heating data and
proposed a fine-grained bulk sediment provenance tool
that can be applied to date clay minerals.
Glauconite dating for stratigraphic age control
Glauconite formation takes place during deposition of
sediment at the water–sediment interface. Glauconite is
therefore an attractive target for isotopic dating of deposi-
tional ages of sedimentary rocks (e.g., Odin et al., 1982).
Dating of K-bearing clay minerals using the isotopic sys-
tems offers the prospect of establishing the absolute
timing of sedimentation. To ensure that the dated separates
of authigenic glauconite provide meaningful stratigraphic
ages, and to avoid any age bias by detrital contribution,
Odin (1975) suggested that only glauconites with a K con-
centration above 5.4 wt% should be used for isotopic
dating.
Deformation events: brittle fault and gouge zone
dating
Displacement on discrete fault planes often results in the
development of fault gouge composed of crushed rock
fragments and authigenic illite clay minerals. Numerous
studies have highlighted the potential for determining the
absolute timing of brittle fault history using isotopic dat-
ing techniques including K–Ar and 40Ar/39Ar dating
(Vrolijk and van der Pluijm, 1999; van der Pluijm
et al., 2001; Zwingmann et al., 2004; Solum et al., 2005;
Haines and van der Pluijm, 2008; Zwingmann
et al., 2010; Viola et al., 2013). The understanding of fault
processes and specifically the timing and extent of fault
gouge formation are important for several reasons:
(1) regional correlation of shallow fault activity that is of
critical importance in neotectonic studies, (2) for hydro-
carbon exploration as faults may act as either a conduit
zone or a seal for fluids and/or hydrocarbons, (3) CO2
sequestration and possible CO2 diffusion along fault
zones, (4) civil engineering and the evaluation of earth-
quake hazards, and (5) in assessing the suitability of sites
for waste storage such as nuclear waste.
Fault gouge illite grain-size fractions of authigenic ori-
gin are mixtures of illite particles formed at different times
during growth, and this growth history is also investigated
by dating a range of different grain-size fractions.
40
Ar/39Ar dating of coarse-grained mica is a valuable tool
in dating fault processes (Sherlock et al., 2003). However,
as outlined, the dating of fine-grained micron size illite
mineral fractions is complicated due to 39Ar recoil and
the consequent corrections. The inverse relationship
between grain size and amount of contamination can
therefore make dating of illite fault gouges problematic:
the finest grain sizes are normally least contaminated
but are influenced by 39Ar recoil and therefore not
suitable for 40Ar/39Ar dating without encapsulation.
Fine-grained illite can easily be dated by traditional K–
Ar and can distinguish authigenic and detrital illite clay
mineral phases.
 CLAYS AND GLAUCONITES (K–Ar/Ar–Ar)
Summary and conclusions
The conventional K–Ar and 40Ar/39Ar dating methods
have both advantages and limitations:
1. K–Ar dating requires relatively large samples
(ca. 50 mg) which can introduce inaccuracy with any
sample heterogeneity as two sample aliquots are
required for K and Ar analyses.
2. A major obstacle for 40Ar/39Ar dating is 39Ar recoil
resulting in potential loss of 39Ar, which occurs during
irradiation, mostly in the clay mineral size range
below <2 mm.
3. 39Ar recoil can be measured using an encapsulation
technique that needs to be technically evaluated and
standardized.
4. 40Ar/39Ar dating and the conventional K–Ar method
are valuable tools for clay dating. Both methods have
distinct applications in the field of clay geochronology.
Bibliography
Armstrong, R. L., 1991. A brief history of geochronometry and radio-
genic isotopic studies. In Haeman, L., and Ludden, J. N. (eds.),
Applications of Radiogenic Isotope Systems to Problems in Geol-
ogy. Toronto: Mineral Association Canada, Vol. 19, pp. 1–26.
Aronson, J. L., and Hower, L., 1976. Mechanism of burial metamor-
phism of argillaceous sediments. 2. Radiogenic argon evidence.
Geological Society of America Bulletin, 87, 738–744.
Cassignol, C., and Gillot, P. Y., 1982. Range and effectiveness of
unspiked potassium-argon dating: experimental ground-work
and applications. In Odin, G. S. (ed.), Numerical Dating in Stra-
tigraphy. Chichester, Part 1: Wiley, pp. 159–179.
Clauer, N., and Chaudhuri, S., 1995. Clays and Crustal Cycles. Hei-
delberg/New York: Springer. 359 p.
Clauer, N., Środoń, J., Francú, J., and Šucha, V., 1997. K-Ar dating
of illite fundamental particles separated from illite/smectite. Clay
Minerals, 32, 181–196.
Clauer, N., Huggett, J. M., and Hillier, S., 2005. How reliable is the
K-Ar glauconite chronometer? A case study of Eocene sedi-
ments from the Isle of Wight. Clay Minerals, 40, 167–176.
Clauer, N., Jourdan, F., and Zwingmann, H., 2011. Dating petro-
leum emplacement by illite 40Ar/39Ar laser stepwise heating:
a comment. American Association of Petroleum Geologists Bul-
letin, 95(12), 2107–2111, doi:10.1306/03241110115.
Clauer, N., Zwingmann, H., Liewig, N., and Wendling, R., 2012.
Comparative 40Ar/39Ar and K-Ar dating of illite-type clay min-
erals. Earth Science Reviews, 115, 76–96.
Crank, J., 1975. The Mathematics of Diffusion, 2nd edn. Oxford:
Oxford University Press.
Dalrymple, G. B., 1991. The Age of the Earth. Stanford: Stanford
University Press.
Dalrymple, G. B., and Lanphere, M. A., 1969. Potassium-Argon
Dating. San Francisco: W.H. Freeman. 258 p.
Dickin, A. P., 1995. Radiogenic Isotope Geology. Cambridge: Cam-
bridge University Press. 490 p.
Dong, H., Hall, C. M., Peacor, D. R., and Halliday, A. N., 1995.
Mechanisms of argon retention in layers revealed by laser
40
Ar-39Ar dating. Science, 267, 355–359.
Eberl, D. D., and Środoń, J., 1988. Ostwald ripening and
interparticle-diffraction effects for illite crystals. American Min-
eralogist, 73, 1335–1345.
Eberl, D. D., Środoń, J., Kralik, M., Taylor, B. E., and Peterman,
Z. E., 1990. Ostwald ripening of clays and metamorphic min-
erals. Science, 248, 474–477.
175
Farley, K. A., Hurowitz, J. A., Asimov, P. D., Jacobson, N. S., and
Cartwright, J. A., 2013a. A double-spike method for K–Ar mea-
surement: a technique for high precision in situ dating on Mars
and other planetary surfaces. Geochimica et Cosmochimica
Acta, 110, 1–12.
Farley, K. A., Malespin, C., Mahaffy, P., Grotzinger, J. P.,
Vasconcelos, P. M., Milliken, R. E., Malin, M., Edgett, K. S.,
Pavlov, A., Hurowitz, J. A., Grant, J. A., Miller, H. B., Arvidson,
R., Beegle, L., Calef, F., Conrad, P. G., Dietrich, W. E.,
Eigenbrode, J., Gellert, R., Gupta, S., Hamilton, V., Hassler,
D. M., Lewis, K. W., McLennan, S. M., Ming, D., Navarro-
González, R., Schwenzer, S. P., Steele, A., Stolper, E. M., Sum-
ner, Y. D., Vaniman, D., Vasavada, A., Williford, K., Wimmer-
Schweingruber, R. F., and 21 the MSL Science Team, 2013b. In
situ radiometric and exposure age dating of the martian surface.
Science, 343(6169), 1247166, doi:10.1126/science.1247166.
Faure, G., 1986. Principles of Isotope Geology. New York: Wiley.
589 p.
Faure, G., and Mensing, T. M., 2005. Isotopes – Principles and
Applications, 3rd edn. Hoboken, NJ: Wiley. 897p.
Foland, K. A., Hubacher, F. A., and Arehartet, G. B., 1992.
40
Ar/39Ar dating of very fine-grained samples: an
encapsulated-vial procedure to overcome the problem of 39Ar
recoil loss. Chemical Geology (Isotope Geoscience Section),
102, 269–276.
Frey, M., and Robinson, D., 1999. Low-Grade Metamorphism.
Oxford: Blackwell Science. 313 p.
Girard, J. P., Savin, S. M., and Aronson, J., 1989. Diagenesis of the
lower cretaceous arkoses of the Angola margin; petrologic, K/Ar
dating and 18O/16O evidence. Journal of Sedimentary Research,
59(4), 519–538.
Gradstein, F., Ogg, J., Schmitz, M. D., and Ogg, G., 2012. The Geo-
logic Time Scale 2012. Oxford: Elsevier.
Haines, S. H., and van der Pluijm, B. A., 2008. Clay quantification
and Ar-Ar dating of synthetic and natural gouge: application to
the Miocene Sierra Mazatan detachment fault, Sonora, Mexico.
Journal of Structural Geology, 30, 525–538.
Hamilton, P. J., Kelley, S., and Fallick, A. E., 1989. K-Ar dating of
illite in hydrocarbon reservoirs. Clays Clay Minerals, 24,
215–231.
Hamilton, P. J., Giles, M. R., and Ainsworth, P., 1992. K-Ar dating
of illites Brent Group reservoirs: a regional perspective. In Mor-
ton, A. C., Haszeldine, R. S., Giles, M. R., and Brown, S. (eds.),
Geology of the Brent Group. London: Geological Society of
London, Vol. 61, pp. 377–400. Special Publications.
Harrison, T. M., Celerier, J., Amos, B., Aikman, A. B., Hermann, J.,
and Heizler, M. T., 2009. Diffusion of Ar in muscovite.
Geochimica et Cosmochimica Acta, 73, 1039–105,
doi:10.1016/j.gca.2008.09.038.
Hashizume, H., 2012. Role of clay minerals in chemical evolution
and the origins of life. In: Valakova, M., and Martynkova,
G. S. (eds.), Clay Minerals in Nature – Their Characterization,
Modification and Application, ISBN 978-953-51-0738-5,
http://dx.doi.org/10.5772/50172.
Hunziker, J. C., Frey, M., Clauer, N., Dallmeyer, R., Friedrichsen,
H., Flehmig, W., Hochstrasser, K., Roggwiler, P., and
Schwander, H., 1986. The evolution of illite to muscovite: min-
eralogical and isotopic data from the Glarus Alps, Switzerland.
Contributions to Mineralogy and Petrology, 92, 157–180.
Huon, S., Cornee, J. J., Pique, A., Rais, N., Clauer, N., Liewig, N.,
and Zayane, R., 1993. Mise en évidence au Maroc d’événements
thermiques d’âge triascico-liasique à l’ouverture de l’Atlantique.
Bulletin de la Societe Geologique de France, 164, 165–176.
Kapusta, Y., Steinitz, G., Akkerman, A., Sandler, A., Kotlarsky, P.,
and Nägler, A., 1997. Monitoring the deficit of 39Ar in irradiated
clay fractions and glauconites: modeling and analytical proce-
dure. Geochimica et Cosmochimica Acta, 61, 4671–4678.
176 CLAYS AND GLAUCONITES (K–Ar/Ar–Ar)
Kelley, S., 2002. K-Ar and Ar-Ar dating. Reviews in Mineralogy
and Geochemistry, 47(1), 785–818.
Liewig, N., Clauer, N., and Sommer, F., 1987. Rb-Sr and K-Ar dat-
ing of clay diagenesis in Jurassic sandstone oil reservoirs, North
Sea. American Association of Petroleum Geologists Bulletin, 71,
1467–1474.
Mark, D. F., Stuart, F. M., and de Podesta, M., 2011. New high-
precision measurements of the isotopic composition of atmo-
spheric argon. Geochimica et Cosmochimica Acta, 75,
7494–7501.
McDougall, I., and Harrison, T. M., 1999. Geochronology and
Thermochronology by the 40Ar/39Ar Method, 2nd edn. Oxford:
Oxford University Press.
Meunier, A., 2005. Clays. Berlin: Springer. 472 p.
Meunier, A., and Velde, B., 2004. Illite: Origins, Evolution and
Metamorphism. Berlin: Springer. 286 p.
Meunier, A., Velde, B., and Zalba, P., 2004. Illite K–Ar dating and
crystal growth processes in diagenetic environments: a critical
review. Terra Nova, 16, 296–304.
Nadeau, P. H., 2011. Earth’s energy “Golden Zone”: A synthesis
from mineralogical research. Clay Minerals, 46, 1–24.
Odin, G. S., 1975. De Glauconarium Constitutione, Aetateque:
Ph.D. thesis, University of Paris-Sud, 280 p.
Odin, G. S., et al., 1982. Interlaboratory standards for dating pur-
poses. In Odin, G. S. (ed.), Numerical Dating in Stratigraphy.
Chichester, Part 1: Wiley, pp. 123–148.
Onstott, T. C., Miller, M. L., Ewing, R. C., Arnold, G. W., and
Walsh, D. S., 1995. Recoil refinements: implications for the
40Ar/39Ar dating technique. Geochimica et Cosmochimica
Acta, 59, 1821–1834.
Pevear, D. R., 1999. Illite and hydrocarbon exploration.
Proceedings of the National Academy of Sciences of the
United States of America, 96, 3440–3446, doi:10.1073/
pnas.96.7.3440.
Renne, P. R., Sharp, W. D., Deino, A. L., Orsi, G., and Civetta, L.,
1997. 40Ar/39Ar dating into the historical realm: calibration
against Pliny the Younger. Science, 277, 1279–1280.
Renne, P. R., Knight, K. B., Nomade, S., Leung, K., and Lou, T.,
2005. Application of deuteron-deuteron (D-D) fusion neutrons
to 40Ar/39Ar geochronology. Applied Radiation Isotopes, 62,
25–32, doi:10.1016/j.apradiso.2004.06.004.
Selby, D., 2009. U-Pb zircon geochronology of the Aptian/Albian
boundary implies that the GL-O international glauconite stan-
dard is anomalously young. Cretaceous Research, 30,
1263–1267.
Sherlock, S. C., Kelley, S. P., Zalasiewicz, J., Schofield, D., Evans,
J., Merriman, R. J., and Kemp, S. J., 2003. Absolute dating of
low-temperature deformation: strain-fringe dating by Ar-Ar
laser microprobe. Geology, 31, 219–222.
Smith, P. E., Evensen, N. M., and York, D., 1993. First successful
40
Ar/39Ar dating of glauconites: argon recoil in single grain
cryptocrystalline material. Geology, 21, 41–44.
Smith, P. E., Evensen, N. M., York, D., and Odin, G., 1998. Single-
grain 40Ar/39Ar ages of glauconies: implications for the geologic
time scale and global sea level variations. Science, 279,
1517–1519, doi:10.1126/science.279.5356.1517.
Solum, J., van der Pluijm, B., and Peacor, D. R., 2005. Neocrystal-
lization, fabrics and age of clay minerals from an exposure of the
Moab fault zone, Utah. Journal of Structural Geology, 27,
1563–1576.
Srodón, J., and Eberl, D. D., 1984. Illite. In Bailey, S. W. (ed.),
Reviews in Mineralogy. Washington, DC: Mineralogical Society
of America, Vol. 13, pp. 495–544.
Steiger, R. H., and Jäger, E., 1977. Subcommission on geochronol-
ogy: convention on the use of decay constants in geo-and
cosmochronology. Earth and Planetary Science Letters, 36,
359–362.
Sudo, M., Tagami, T., Sato, K., Hasebe, N., and Nishimura, S.,
1996. Calibration of a new Ar analytical system for the K-Ar
dating method and analytical results of K-Ar age known
samples. Memoirs of the Faculty of Science, Kyoto University,
Series of Geology and Mineralogy, 58, 21–40.
Turner, G., and Cadogan, P. H., 1974. Possible effects of 39Ar recoil
in 40Ar/39Ar dating. Geochimica et Cosmochimica Acta, 5,
1601–1615.
van der Pluijm, B. A., Hall, C. M., Vrolijk, P. J., Pevear, D. R., and
Covey, M. C., 2001. The dating of shallow faults in the Earth’s
crust. Nature, 412, 172–175.
van der Pluijm, B. A., Vrolijk, P. J., Pevear, D. R., Hall, C. M., and
Solum, J., 2006. Fault dating in the Canadian rocky mountains:
evidence for late Cretaceous and early Eocene orogenic pulses.
Geology, 34, 837–840, doi:10.1130/G22610.1.
van Laningham, S., and Mark, D. F., 2011. Step heating of
40
Ar/39Ar standard mineral mixtures: investigation of a fine-
grained bulk sediment provenance tool. Geochimica et
Cosmochimica Acta, 75, 2324–2335.
Velde, B., 1992. Introduction to Clay Minerals: Chemistry, Origins,
Uses, and Environmental Significance. New York: Chapman &
Hall, p. 198.
Velde, B., 2004. Green clay minerals. In: Holland, H. D., and
Turekian, K. K. (eds.), Treatise on Geochemistry 7. Sediments,
Diagnosis, and Sedimentary Rocks. Amsterdam: Elsevier,
Vol. 7, pp. 309–24.
Villa, I. M., 1997. Direct determination of 39Ar recoil distance.
Geochimica et Cosmochimica Acta, 61, 689–691.
Viola, G., Zwingmann, H., Mattila, J., and Käpyaho, A., 2013. K/Ar
illite constraints on the Proterozoic formation and reactivation
history of a brittle fault in Fennoscandia. Terra Nova, 25(3),
236–244, doi:10.1111/ter.12031.
Vrolijk, P., and van der Pluijm, B. A., 1999. Clay gouge. Journal of
Structural Geology, 21, 1039–1048.
Watson, E., and Baxter, E. F., 2007. Diffusion in solid-Earth sys-
tems. Earth and Planetary Science Letters, 253, 307–327.
Wijbrans, J. R., and McDougall, I., 1986. 40Ar/39Ar dating of
white micas from an alpine high-pressure metamorphic belt on
Naxos. Contributions to Mineralogy and Petrology, 93,
187–194.
Wilkinson, M., and Haszeldine, S., 2002. Fibrous illite in oilfield
sandstones – a nucleation kinetic theory of growth. Terra Nova,
14(1), 56–60.
Worden, R., and Morad, S., 2003. Clay Mineral Cements in
Sandstones. Oxford: Blackwell. Special Publication
Number 34 of the International Association of Sedimentologists.
509 p.
Yun, J. B., Shi, H. S., Zhu, J. Z., Zhao, L. H., and Qiu, H. N., 2010.
Dating petroleum emplacement by illite 40Ar/39Ar laser stepwise
heating. American Association of Petroleum Geologists Bulletin,
94, 759–771.
Zwingmann, H., Offler, R., Wilson, T., and Cox, S., 2004. K-Ar
dating of fault gouge in the northern Sydney basin,
Australia – implications for the breakup of Gondwana. Journal
of Structural Geology, 26, 2285–2295.
Zwingmann, H., Mancktelow, N., Antognini, M., and Lucchini, R.,
2010. Dating of shallow faults – new constraints from the
AlpTransit tunnel site (Switzerland). Geology, 38, 487–490,
doi:10.1130/G30785.1.
Cross-references
Ar–Ar and K–Ar Dating
Fault Zone (Thermochronology)
Minerals (40Ar–39Ar)
 CONTINENTAL DRIFT (PALEOMAGNETISM)
CONTINENTAL DRIFT (PALEOMAGNETISM)
Trond H. Torsvik, Pavel V. Doubrovine and
Mathew Domeier
Centre for Earth Evolution and Dynamics (CEED),
University of Oslo, Oslo, Norway
Definition
Paleomagnetism is the study of the Earth’s ancient mag-
netic field through the record of remanent magnetism pre-
served in rocks. The directions of remanent magnetization
are used to deduce the position of the Earth’s magnetic
pole relative to the study location at the time when this
magnetization was acquired. By studying magnetizations
of varying age from a single lithospheric plate, one can
construct a path of apparent polar wandering (APWP) that
tracks the motion of that plate relative to the geographic
pole. Awell-defined APWP can serve as a geochronological
tool, i.e., for dating magnetizations of unknown age through
a comparison of their directions with those expected from
the reference APWP. Paleomagnetism can be used to date
any geologic event that engenders the acquisition of rema-
nent magnetization, including formation of igneous and sed-
imentary rocks, deposition of ore minerals, episodes of
deformation, and other remagnetization processes.
Introduction
The development and broad acceptance of the theory of
plate tectonics in the 1960s (e.g., McKenzie and Parker,
1967) was the culmination of decades of scrutiny and
intense debate among the Earth science community that
ultimately witnessed the establishment of a mobilistic par-
adigm for the Earth’s surface, which had first been intro-
duced by Wegener (1912) in the form of “continental
drift.” Paleomagnetism was a key to this revolution, as it
provided unambiguous evidence both for the “drift” of
continents (Runcorn, 1956) and for seafloor spreading
(Hess, 1962). The latter was validated by the observation
of marine magnetic anomalies, which revealed symmetric,
age-dependent stripes of remanent magnetization carried
by seafloor that was created at mid-ocean ridges during
periods of opposite geomagnetic polarity (Vine and Mat-
thews, 1963). While the concept of seafloor spreading pro-
vided a qualitative explanation for the mechanism of
continental drift, identifications of marine magnetic anoma-
lies allowed quantitative estimates of displacements
between continents through geologic time, paving the way
for the development of the modern theory of plate tectonics.
The geomagnetic field averaged over a sufficiently long
time interval (tens to hundreds of thousands of years) is
closely approximated by the field of a geocentric dipole
aligned parallel to the Earth’s rotation axis (Merrill et al.,
1996; McElhinny and McFadden, 2000). The dipolar
geometry of the time-averaged field allows paleomagne-
tists to estimate positions of geographic poles relative to
the studied locations by analyzing directions of remanent
magnetization, provided that these magnetizations were
177
acquired over a time interval that was sufficiently long to
average out short-term deviations of the geomagnetic field
from that of a geocentric axial dipole (paleosecular varia-
tion). The poles obtained through measurements of mag-
netic remanence are commonly referred to as
“paleomagnetic poles.” Due to the motion of lithospheric
plates relative to the Earth’s spin axis, the position of the
paleomagnetic pole in the reference frame of a specific
plate varies with age. To an observer on the plate who is
not aware of plate motion, it would appear that the spin
axis slowly changes its orientation through time, so that
the geographic pole moves along a unique path that we
refer to as the “apparent polar wander path” (APWP).
The motion reflected by APWPs may be due to the “drift”
of individual plates, the rotation of the entire planet rela-
tive to the spin axis (true polar wander, TPW), or
a combination of these two processes.
Creer et al. (1954) were the first to publish an APWP
for “Europe” based on paleomagnetic poles from Britain.
The paleomagnetic poles (plotted as south poles in Fig-
ure 1a) differed markedly from the present-day pole and
left Creer et al. (1954) with two possible explanations:
either the pole itself had moved (TPW) or Britain/Europe
(Figure 1b) had moved relative to the pole (“continental
drift”). Initially, they interpreted their results as “a slow
change in the axis of rotation of the earth with respect to
its surface,” in other words, TPW. However, two years
later, Runcorn (1956) published an APWP for North
America and crucially observed that it was separated from
the APWP of Europe by 30
longitude, thus demonstrat-
ing that the continents must have moved relative to each
other. Specifically, Runcorn (1956) attributed the separa-
tion of the APWPs to the “drift” between North America
and Europe corresponding to the opening of the Atlantic
Ocean. This interpretation was later corroborated by the
analysis of marine magnetic anomalies (Vine and Mat-
thews, 1963).
It is now well established that tectonic plates (which
may include both continental and oceanic lithosphere)
are constantly moving relative to each other, although at
times they may coalesce or fragment, and their boundaries
are otherwise frequently modified. APWPs from the indi-
vidual continents attest to their often independent history
of motion but also reveal an incessant, but commonly sub-
ordinate component of TPW. After more than a half-
century of paleomagnetic research, the APWPs of most
continents can be traced back to the dawn of the Phanero-
zoic, although they are generally much better defined for
younger time. Consequently, when and where these
APWPs are well constrained, it is now possible to invert
the traditional workflow of paleomagnetic studies and
use the APWP as a geochronological tool for dating mag-
netizations of unknown age.
Constructions of reference APWPs
APWPs constructed as paths of apparent motion for the
northern or southern geographic pole conveniently
178 CONTINENTAL DRIFT (PALEOMAGNETISM)

Continental Drift (Paleomagnetism), Figure 1 (a) The first Phanerozoic polar wander path for “Europe” (Creer et al., 1954) based on
six paleomagnetic poles from Britain. These poles – here shown as south poles – differ from the present-day South Pole, and this
was originally interpreted as a change in the Earth’s rotation axis through time, i.e., true polar wander (TPW). In this plot, the continent
(British Isles) stays fixed, but the polar axis is left to wander and following a path that is dubbed apparent polar wander path
(APWP). “Apparent” means that the wandering may not be real and that it could be the result of “continental drift” as in (b).
(b) The location of the British Isles at different times (longitude is arbitrary) according to the poles in (a), and here assuming a moving
continent with a fixed polar axis. This was originally coined “continental drift.” Note that Scotland and England were actually
separated by the Iapetus Ocean before the Silurian (Figure 2a). This was not known in 1954 and thus the British Isles cannot be
plotted as a coherent unit in the Cambrian.

summarize paleomagnetic data for continents, terranes,
and other tectonic blocks. In order to define an APWP,
paleomagnetic poles of various ages are plotted on
a stereonet and then fitted with a model path of polar
motion, either with the use of spherical splines or by
applying a running mean technique. With the spherical
spline method (Jupp and Kent, 1987), a spline curve
constrained to lie on the surface of the sphere is fitted to
the paleomagnetic poles providing a regression model that
smoothly approximates the positions of the poles at their
respective ages (Figure 2b–c). The paleomagnetic poles
that serve as the input data for the regression are assigned
statistical weights that are inversely proportional to the
precision of the poles (A95) or the paleomagnetic quality
factor Q (Van der Voo, 1990). In Figure 2b–c, the input
poles were weighted by the value of 7/Q, which causes
the spline path to pass closer to high-quality poles (7 is
the highest possible value of Q factor).
In the running mean method, the position on the APWP
corresponding to a specific age is calculated as a Fisherian
mean (Fisher, 1953) of paleomagnetic poles whose ages
fall within a selected time window (e.g., 20 Ma) centered
on that age. For example, the 160 Ma pole in Figure 3a
averages all Laurussian paleomagnetic poles within the
150–170 Ma age window. In practice, the mean poles
are commonly calculated at age increments equal to half
the width of the averaging window (e.g., 10 Ma incre-
ments for the 20 Ma window in Figure 3a). An advantage
of the running mean technique is that the pole uncertainty,
commonly expressed as a circle of 95 % confidence (A95),
can be estimated for each mean pole through Fisherian sta-
tistics (Figures 3–5). Both the spline method and the run-
ning mean technique effectively average out random
biases of individual paleomagnetic poles and allow the
basic pattern of APW to be determined.
APWPs have changed dramatically over the course of
the past six decades, in part due to a much greater pool
of paleomagnetic data. For example, in contrast to the ini-
tial compilation of merely five Phanerozoic poles from
Britain (Figure 1a), the most recent Phanerozoic APWP
for Europe (Figure 2c) is based on several hundred paleo-
magnetic poles. APWP refinements have also been made
through the implementation of more sophisticated analyt-
ical methods and better instrumentation. Important recent
developments include novel techniques to identify and
correct for the inclination shallowing errors in data from
clastic sedimentary rocks, which are now recognized to
be commonly biased (Torsvik et al., 2012). An improved
understanding of paleogeography has also driven notable
changes in how we group paleomagnetic data.
 CONTINENTAL DRIFT (PALEOMAGNETISM) 179

Continental Drift (Paleomagnetism), Figure 2 (a) Reconstruction of Laurussia at 420 Ma. Avalonia collided with Baltica at 450 Ma
and later Baltica/Avalonia collided with Laurentia (including North America, Greenland, Scotland, and Northern Ireland) at 430 Ma
along the Iapetus Suture. The locations of paleomagnetically dated fault rocks in Western Norway (WN) and mineral deposits in
Ireland (I) are shown as open white circles. The Irish mineral deposits are located near the Iapetus Suture and about 50 km north of
the Variscan Front in Figure 3b. (b) APWP for Laurentia. Moderately smoothed spherical spline path shown in 10 Ma intervals from
the Early Cambrian (530 Ma) to the present. (c) APWP for Baltica/Stable Europe from Early Ordovician to the present. Input poles
for both Laurentia and Baltica/Europe are shown with 95 % confidence ovals (yellow shading) and detrital sedimentary input
poles have been corrected for potential I-errors (After Torsvik et al., 2012).

For example, the “European” path in Figure 2c is Avalonia (which was a separate terrane in the Ordovician),
constructed only with paleomagnetic poles from Baltica and progressively younger poles from Europe derived
before 430 Ma, but after this time we also include poles from rocks formed after their accretion to Laurussia. The
from Scotland (originally a part of Laurentia, Figure 2a), Laurentian path in Figure 2b is constructed from North
180 CONTINENTAL DRIFT (PALEOMAGNETISM)

Continental Drift (Paleomagnetism), Figure 3 (a) Laurussian (Baltica/Stable Europe and Laurentia rotated to European coordinates)
running mean path shown with A95 circles (yellow shading). The APWP is corrected for potential inclination shallowing in detrital
sedimentary rocks (Torsvik et al., 2012). The APWP is shown in 10-Ma intervals from the Late Carboniferous (320 Ma) to the Cretaceous
(120 Ma), and each mean pole is based on a sliding time window of 20 Ma. The APWP is compared with paleomagnetic poles derived
from two generations of fault rocks in Western Norway (Torsvik et al., 1992) and Zn–Pb and Ba deposits in Ireland (Symons et al., 2007;
Pannalal et al., 2008a; Pannalal et al., 2008b). These poles are shown with 95 % confidence ovals (dp/dm). (b) The Zn–Pb/Ba and the green
breccia poles in (a) dates to around 310–290 Ma; they can potentially be linked to the waning stages of Hercynian orogenic activity.

American and Greenland poles; the latter have been
rotated to North America after adjusting for the Cenozoic
opening of the Labrador Sea and Baffin Bay.
The relative positions between North America, Green-
land, and Baltica/Europe are known reasonably well after
the Silurian, and paleomagnetic poles from these
continents can be combined into a Laurussian APW path
(Figures 3 and 4b, Torsvik et al., 2012). In the following
section, we will use this APWP as a reference path to illus-
trate the essentials of the paleomagnetic dating technique
with examples of age estimates from fault rocks in West-
ern Norway, Zn–Pb deposits in Ireland, intrusions in
southern Illinois (USA), and remagnetized carbonate
rocks and mineral deposits in North America.
Examples of paleomagnetic age determinations
Example 1: Fault rocks in Western Norway
After the Silurian Caledonide collision of Baltica and
Laurentia (Figure 2a), post-orogenic collapse and
 CONTINENTAL DRIFT (PALEOMAGNETISM) 181

Continental Drift (Paleomagnetism), Figure 4 (a) Laurentia running mean path with A95 circles (yellow shading) from the Late
Carboniferous (320 Ma) to the Triassic (220 Ma) shown together with remagnetized carbonate rock poles from North America (Van
der Voo and Torsvik, 2012) and two Zn–Pb mineralization poles (blue squared is the pole from Eastern Tennessee, Symons and
Stratakos, 2002; green squared is the pole from Wisconsin, Pannalal et al., 2004). Remagnetized carbonate poles (black squares) are
shown without 95 % confidence ovals for clarity. (b) Similar to (a) but the North American poles are compared with the Laurussia
APWP (North American frame). (c) Geographic locations of sampling sites poles in (a–b) but plotted for sites with pole latitude <40
S
(presumed oldest remagnetizations, i.e., Late Carboniferous when compared with the Laurentia APWP), 40–50
S (Permian), and
>50
S (Late Permian-Triassic). Yellow shading denotes regions of remagnetization and mineralization. TE Tennessee, WI Wisconsin,
OK Oklahoma, MO Missouri, AL Alabama, AZ Arizona, PA Pennsylvania, MI Michigan, NY New York.

extensional shearing was accommodated along the
Nordfjord-Sogn Detachment (NSD) in Western Norway
(e.g., Andersen et al., 1991). The NSD and Devonian
mylonites are cut by younger brittle faults, most spectacu-
larly exposed in outcrops on the island of Atløy (Torsvik
et al., 1992; Eide et al., 1997) where a nearly flat-lying
fault-breccia zone comprises a green network breccia
crosscut by a red breccia. The magnetic mineralogy of
the green breccia is dominated by magnetite, and magne-
tization components yielded a mean pole position that
was originally assigned a Late Permian age
(250–260 Ma) by Torsvik et al. (1992). The green breccia
pole plots slightly off the APWP for Laurussia, but is
within the 95 % confidence circle of the 310 Ma reference
pole (Figure 3a). Considering that the acquisition of fault-
related remanence in the green breccia has likely occurred
over a time interval that does not fully average the secular
variation of the geomagnetic field (Torsvik et al., 1992),
the uncertainty of the pole location may be substantially
larger than that implied by its nominal 95 % confidence
region (shown by a green ellipse in Figure 3a). Hence,
the post-Devonian fault rejuvenation along the NSD can
only be loosely constrained to Late Carboniferous
(310 Ma) to Early Permian (280 Ma) time. The red breccia
pole overlaps with the 140 Ma mean pole for Laurussia
(Figure 3a), suggesting Early Cretaceous fault movements
182 CONTINENTAL DRIFT (PALEOMAGNETISM)
corresponding to the formation of the red breccia along the
NSD.
Brittle fault rocks are notoriously difficult to date with
conventional isotopic methods and 40Ar/39Ar whole-rock
spectra from the breccias defined several age groups that
were interpreted to correspond to separate thermal epi-
sodes (Eide et al., 1997). However, the oldest whole-rock
ages from the green breccia, 296.2  2.6 Ma (8 steps,
47 % of total 39Ark gas) and 296.4  2.8 Ma (4 steps,
58.4 % of total 39Ark gas), are in agreement with the age
suggested by paleomagnetic data (280–310 Ma). The
youngest 40Ar/39Ar ages range from 96  3 Ma to 163
 4 Ma and may reflect argon loss related to low-
temperature hematite precipitation during the formation
of the red breccia at 140 Ma (Figure 3a).
Example 2: Zn–Pb deposits in Ireland
The Lisheen Zn–Pb deposit in Ireland is one of the major
Lower Carboniferous carbonate-hosted base metal
deposits, but the timing of ore mineralization and its gen-
esis have been debated due to the lack of absolute ages.
Paleomagnetic results from twelve Zn–Pb mineralized
sites and eight host-rock sites (Pannalal et al., 2008a) yield
a combined pole that is indistinguishable from the 300 and
290 Ma mean poles for Laurussia (pole I1 in Figure 3a).
A paleomagnetic pole (Symons et al., 2007) from the
nearby Magcobar Ba deposit (pole I3 in Figure 3a) is vir-
tually identical to the Lisheen pole and a pole from the
Galmoy Zn–Pb deposit plots slightly off the
300–290 Ma poles (pole I2 in Figure 3a). These poles
postdate peak Variscan folding and deformation (Fig-
ure 3b), and Pannalal et al. (2008b) argued for an epige-
netic model in which mineralization occurred during
cooling from the Variscan thermal episode. These Irish
Zn–Pb and Ba deposits were previously paleomagneti-
cally dated through a comparison with an APWP
published in Torsvik et al. (1996) that suggested
a 269–290 Ma range for the age of mineralization. We
now favor an Early Permian age of approximately
295 Ma based on our updated APWP (Figure 3a).
Example 3: Remagnetization and mineralization of
carbonate rocks in North America
A well-documented remagnetization event is known to
have widely affected Paleozoic carbonate rocks in North
America. The cause of this far-reaching event remains
enigmatic, but the age of the remagnetization can be
deduced by a comparison against reference APWPs. This
was most recently done by Van der Voo and Torsvik
(2012), who also illustrated how the reference APWP for
Laurentia had evolved over the past three decades.
Figure 4a shows a comparison of the remagnetization
poles listed by Van der Voo and Torsvik (2012, Table 1)
with the most recent APWP for Laurentia (Torsvik et al.,
2012) that suggests that the remagnetization events date
from the Late Carboniferous (310 Ma) to the Early
Triassic (230 Ma). This range of ages is also manifested
when the poles are compared with the Laurussian APWP
(Figure 4b), but the majority of the remagnetization poles
plot slightly west of the main Laurussian path. This occurs
because the Permian poles from Europe differ slightly
from those of North America, and the inclusion of the
European poles pulls the combined Laurussian APWP
eastward. The reasons for this discrepancy are unclear.
Because the North American and European poles of both
older and younger age are in good agreement, it is unlikely
that the discrepancy originates from erroneous relative fits
between Laurentia and Baltica/Stable Europe. Nonethe-
less, this example demonstrates the importance of consid-
ering the differences among reference paths, as well as
their limitations.
Remagnetization sites are located mostly within
1,000 km of the Alleghanian/Quachitan Sutures, but
remagnetization has also been observed in sites as far as
3,000 km away (Figure 4c), highlighting the broad scale
of this protracted, continent-wide event. Spatial trends in
the timing of remagnetization are not entirely clear
(Figure 4c), but as pointed out in earlier compilations,
the southern states of Alabama, Georgia, and Tennessee
near the Appalachian Suture appear to be affected earlier
than those of the Central Appalachians (Miller and Kent,
1988; Van der Voo and Torsvik, 2012). Late Carbonifer-
ous (310–300 Ma) remagnetization and Zn–Pb mineral-
ization events were confined to an area from Alabama,
Georgia, and Tennessee to Colorado. By contrast, Permian
remagnetizations are much more widespread – from New-
foundland to Nevada – whereas Late Permian to Early Tri-
assic remagnetizations are confined to the Central
Appalachians (e.g., New York and Michigan) and the
southwestern parts of North America (e.g., Colorado and
Arizona). The cause of this pattern of remagnetization
and its relation to the emplacement of Mississippi Val-
ley-type deposits (e.g., Zn–Pb deposits) is still mystifying,
but a commonly preferred hypothesis invokes tectonically
expelled fluids driven from the Alleghanian orogenic belt
(Oliver, 1986).
Example 4: Intrusions in Southern Illinois (USA)
In the south end of the Illinois Basin, near the Illinois-
Kentucky border (USA), a series of dikes, sills, and
diatreme breccia bodies are observed to invade Devo-
nian-Carboniferous sedimentary rocks on and around
a structural dome (Hicks Dome) that is inferred to be
cogenetic with the intrusive rocks. Due to extensive sur-
face weathering and alteration, the intrusive rocks have
proven difficult to classify and interpret. Consequently,
many outstanding questions remain regarding their rela-
tionship to each other and to the greater tectonic frame-
work of the region. Initial attempts to isotopically date
the intrusive episode(s) ended in ambiguity, leading Reyn-
olds et al. (1997) to employ paleomagnetism to date the
rocks. From an intrusive breccia and a lamprophyric sill,
 CONTINENTAL DRIFT (PALEOMAGNETISM)
Reynolds et al. (1997) isolated a remanent magnetization
carried predominantly by high-temperature-oxidized
titanomagnetite, as witnessed by microscopic observa-
tions of magnetite-ilmenite intergrowths that suggest
a primary igneous origin of the mineral. Matching the
resulting paleomagnetic pole with the APWP of Laurentia,
Reynolds et al. (1997) concluded that the magnetization
was acquired in late Paleozoic-early Mesozoic time during
the emplacement of the intrusions. A comparison of the
results of a recent paleomagnetic study that revisited these
intrusive rocks (Domeier et al., 2011) with the updated
APWP for Laurussia indeed reveals a close correspon-
dence between the paleomagnetic pole and the
240–260 Ma mean poles of the APWP (Figure 5a).
Uncertainties of paleomagnetic age estimates
In the examples of the previous section, we restricted our
discussion to qualitative age assignments through
a visual comparison of paleomagnetic poles derived from
magnetizations of unknown age with a reference APWP.
In most cases, however, it is desirable to obtain a formal
age estimate and specify its uncertainty. In this section,
we will present a simple statistical approach that can be
used to estimate the “best-fitting” age and its confidence
interval.
The accuracy of ages estimated from paleomagnetic
data depends on several factors, and certain conditions
must be satisfied for paleomagnetic age estimates to be
meaningful. It is critical that the magnetizations used to
calculate a pole of unknown age have sampled
a sufficiently long time interval to average out
paleosecular variation (e.g., at least tens to hundreds of
thousands of years) and that tectonic tilts or rotations
postdating the acquisition of magnetization have been
structurally corrected. When either of these two conditions
is not met, the calculated pole may be significantly biased,
resulting in an erroneous age estimate. In cases when it is
suspected that the paleosecular variation has not been
averaged out (e.g., as in the fault breccias from Western
Norway discussed in the previous section) or that there
may have been unaccounted structural tilts, the most con-
servative approach is to conclude that whereas the pole
provides some indication for a particular age, the age can-
not be constrained with certainty.
Although the reference APWP can vary depending on
the construction method, the differences between the
spline-fitted and running mean paths that are defined using
a large number of high-quality paleomagnetic poles are
usually not significant (e.g., Torsvik et al., 2012). It is
most common to construct running mean APWPs at
10-Ma intervals with a sliding time window of 20 Ma
(e.g., Figures 3 and 4). Irrespective of whether the refer-
ence APWP is a spline or a running mean path, the best
estimate of the unknown age of the pole is the age
corresponding to a position on the APWP, for which the
great circle distance (GCD) between the pole and APWP
183
is minimal. Because the reference APWPs are usually tab-
ulated at rather large age increments (10 Ma in most
cases), we recommend resampling the reference APWP
at smaller steps (e.g., 1 Ma) by interpolation between the
consecutive poles assuming a constant rate of polar
motion, calculating GCDs between the pole and APWP
at these increments, and defining the age corresponding
to the minimum GCD value.
The confidence region for the estimated age can be
defined by testing whether the dated pole is statistically
distinct from the reference APWP at a specified level of
significance (e.g., 5 % to constrain a 95 % confidence
interval). In spline models, the uncertainty of the APWP
is not known, and it is not possible to rigorously define
the confidence region. If the spline curve intersects the
95 % confidence circle (A95) of the pole, the range of ages
corresponding to the APWP segment within the circle may
serve as a “minimal” confidence region. Without incorpo-
rating the uncertainty of the APWP, it clearly underesti-
mates the interval of acceptable ages.
In the running mean APWPs, the uncertainty of
a mean pole (A95) depends on the directional dispersion
(Fisher precision parameter K) and the number (N) of
the individual poles within the averaging window that
have been used to define the mean. The directional dis-
persion reflects both the random deviations of the indi-
vidual poles from the “true” mean and the differences
of their ages from the mean pole age. If the ages of indi-
vidual poles are evenly distributed within the averaging
window (which is normally the case), the mean age is
expected to be nearly identical to the median age of the
window. Hence, in our simplified approach, we will not
consider the age uncertainties of the mean poles sepa-
rately and will treat each pole of a running mean APWP
as an “unbiased estimate” of the pole corresponding
exactly to the nominal age (median age of the averaging
window).
The range of acceptable ages can be defined by compar-
ing the pole of unknown age and the reference poles from
the APWP and deciding whether the differences are sig-
nificant at the 95 % confidence level. First-order conclu-
sions can be made through a visual comparison of
confidence regions. If the A95 circles of two poles do not
overlap, it can be concluded that the poles are distinct at
the 95 % confidence level. When one of the poles is within
the A95 circle of the other, the difference between them is
not significant. In the remaining cases, when the confi-
dence regions overlap, but neither of the two poles is
within the A95 of the other, the significance of the observed
difference can be assessed through a formal test for
a common mean (McFadden and Lowes, 1981).
Suppose that the pole of unknown age was estimated
from a dataset comprising N1 individual measurements
(VGPs) that yielded a resultant vector R1 (sum of unit vec-
tors corresponding to the individual VGPs) and the refer-
ence pole is based on averaging N2 paleomagnetic poles,
with a resultant vector R2. Assuming Fisherian-distributed
184 CONTINENTAL DRIFT (PALEOMAGNETISM)

Continental Drift (Paleomagnetism), Figure 5 (a, b) Laurussia running mean path with A95 circles (yellow shading) from the Late
Carboniferous (320 Ma) to the Triassic (220 Ma) shown together with a paleomagnetic pole from intrusions in southern Illinois
(red square with the pink A95 circle) from Domeier et al. (2011) and Zn–Pb mineralization poles from Tennessee (TE) and Wisconsin (WI)
(black and green square with light blue and light green A95 circles; same as in Figure 4, but the dp/dm ovals were recalculated into
A95 circles using the formulations of Cox (1970)). The running mean path is shown in two versions: (a) time window ¼ 20 Ma and
(b) time window ¼ 10 Ma. (c, d) Great circle distance and the probability of the McFadden and Lowes (1981) test for the comparisons
of the poles with the APWPs ((c) 20-Ma window, (d) 10-Ma window). Red curves correspond to the pole from the southern Illinois
intrusions, blue curves to the Tennessee Zn–Pb mineralization pole, and green curves to the Wisconsin Zn–Pb mineralization
pole. Thick dashed vertical lines show the best estimates of ages; thin dashed lines indicate their nominal 95 % confidence limits.
Shaded bands show the conservative confidence intervals (pink shading for the southern Illinois pole, light blue shading for the
Tennessee pole, and light green shading for the Wisconsin pole).
 CONTINENTAL DRIFT (PALEOMAGNETISM)
data, the precision parameters of the underlying distribu-
tions of VGPs and poles can be estimated as
N1  1 N2  1
K1 ¼ , K2 ¼ ð1Þ
N 1  R1 N 2  R2
where R1 and R2 are the lengths of the respective resultant
vectors. In cases when the R values have not been
presented in original studies (which is common for the
running mean APWP compilations), they can be com-
puted from the Fisher (1953) equation for A95:
"  #
N R 1 1=ðN 1Þ
cos A95 ¼ 1  1 ð2Þ
R 0:05
The statistics f for testing the null hypothesis that the
two directional datasets (VGPs and poles averaged by
the running mean) can be drawn from Fisherian distribu-
tions sharing a common mean is calculated using the
equation
K2 h i
ðR1 þ R2 Þ2  R2c
K1
f ¼ ðN 1 þ N 2  2Þ   
K2 K2
2 ðN 1  R1 Þ þ ðN 2  R2 Þ R1 þ R2
K1 K1
ð3Þ
where Rc is the length of the resultant vector for the com-
bined dataset (Rc ¼ R1+ R2). McFadden and Lowes
(1981) showed that this statistic is approximately distrib-
uted as the F distribution with 2 and 2(N1 + N2  2)
degrees of freedom, and the probability of observing the
test statistic as large as f or larger is
 2N 1 N 2
f for the Tennessee pole and 305 Ma (302–311 Ma nominal
PðF
f Þ ¼ þ1 ð4Þ confidence interval, 300–315 Ma conservative confidence
N1 þ N2  2
The probability P(F
f) corresponds to the signifi- The examples considered here illustrate the sensitivity
cance level for retaining the null hypothesis. Hence, if it
is smaller than 0.05, we can conclude that the poles are
distinct at the 5 % significance level. Otherwise (P
using the APWPs with the 20-Ma and 10-Ma averaging
0.05), the poles should be considered indistinguishable
at that level of significance.
Considering that the ages of the reference poles that are
not distinguishable from the pole of unknown age at the
95 % confidence level should be within the confidence
interval for the estimated age, and the ages of the reference
poles that are distinct from the pole should be outside this
interval, the confidence limits can be defined by the range
of ages for which the probability values calculated using
Eqs. (3) and (4) are equal or greater than 5 %. In practice,
we recommend performing the McFadden and Lowes
(1981) test for all poles of the reference APWP, plotting
the calculated probability as a function of age, defining
the age interval for which the probability curve is above
the 5 % value, and using this interval as a nominal 95 %
confidence region. A more conservative estimate can be
obtained by using the ages of the two closest reference
185
poles that significantly differ from the dated pole (i.e., fail
the test) as the upper and lower confidence limits. The
“conservative region” may in fact be preferable because
of the simplifying assumptions for the treatment of pole
uncertainties discussed above and the approximate nature
of the test itself (see McFadden and Lowes, 1981, for
discussion).
To illustrate this technique, we will now derive
formal age estimates and their confidence limits for
some examples discussed in the previous section
(examples 3 and 4). In Figure 5a and b, we show the
Laurussia APWPs calculated at 5-Ma increments with
two different time windows, 20 and 10 Ma. The pole from
the southern Illinois intrusions (example 4) plots directly
on the APWPs, and the analysis of GCD values suggests
an age of 244 Ma for both comparisons (Figure 5c and d).
The probability values for the McFadden and Lowes
(1981) test (Eqs. 3 and 4) are plotted as solid red curves
in Figure 5c and d, suggesting a 236–267 Ma confidence
region for the age estimated using the APWP with the
20-Ma window (Figure 5c) and a 233–270 Ma region
for the estimate obtained with the 10-Ma-window APWP
(Figure 5d). The conservative estimates are 235–270 Ma
and 230–270 Ma, respectively. Using similar analysis
for the Tennessee and Wisconsin Zn–Pb mineralization
poles (example 3) and the 20-Ma-window APWP (Fig-
ure 5a and c), we arrive to the age estimates of 314 Ma
(308–320 Ma nominal confidence interval, 305–320 Ma
conservative confidence interval) and 308 Ma
(307–313 Ma nominal confidence interval, 305–315 Ma
conservative confidence interval), respectively. With the
use of the 10-Ma-window APWP (Figure 5d), the esti-
mates become 312 Ma (306–319 Ma nominal confidence
interval, 305–320 Ma conservative confidence interval)
interval) for the Wisconsin pole.
of the estimated ages and their precision to the choice of
a reference APWP. In all examples, the ages estimated
windows do not differ significantly (the differences are
below the uncertainty limits). The reference poles in the
10-Ma-window APWP have larger uncertainties (A95),
producing slightly wider confidence intervals for the esti-
mated ages. This result shows that that there is no real
advantage in using a reference path with an averaging
window shorter than 20 Ma for obtaining a better
constrained age (smaller confidence region). Hence, we
recommend using APWPs constructed with a 20-Ma
window for most applications of paleomagnetic dating
technique.
Our examples also illustrate that the precision of paleo-
magnetic age estimates critically depends on the rate of
polar motion (spacing of reference poles along the
APWP). A 40-Ma-long confidence interval for the esti-
mated age of the southern Illinois intrusions can be tracked
back to a virtual standstill between 245 and 265 Ma in
186 CONTINENTAL DRIFT (PALEOMAGNETISM)
both APWPs, whereas higher rates of polar motion over
the 300–320 Ma period produce much tighter confidence
intervals for the Zn–Pb mineralization poles from eastern
Tennessee and Wisconsin (10–15 Ma). Thus, the temporal
resolution of the reference APWP imposes a fundamental
limitation on the precision of ages estimated using paleo-
magnetic data.
Summary and conclusions
APWPs have proven invaluable as a means to succinctly
communicate plate motions, as well as a convenient tool
by which to study them. Well-defined APWPs may be
used as a geochronological tool, capable of revealing the
unknown age of a magnetization according to the position
of the paleomagnetic pole derived from it on a reference
APWP. The paleomagnetic dating method can be used to
deduce the age of any geologic event which produces an
attendant remanent magnetization, assuming there is
a well-characterized APWP corresponding to the age and
area (continent/terrane) of the event. We have presented
several case studies which demonstrate the application of
this method to the dating of intrusive rocks, faulting, ore
mineralization, and regional remagnetization. We have
also discussed a simple statistical technique that can be
used to derive a best-fit estimate of the unknown age and
constrain its 95 % confidence interval. It is important to
note that random but persistent short-term variations in
the geomagnetic field (paleosecular variation), uncer-
tainties of the reference APWP, and its temporal resolution
(rate of polar motion) impose fundamental limitations on
the geochronological resolution of this method. Even the
most ideal paleomagnetic result would normally yield
a paleomagnetic pole with an intrinsic uncertainty of
A95  5
, which would roughly correspond to an age error
of approximately  11 Ma if the host continent had a N-S
drift of 5 cm/year and its APWP was perfectly defined.
A slower plate speed and/or a poorly defined APWP will
see an increase in the age uncertainty. Finally, a word of
caution: paleomagnetic results dated by this technique
cannot subsequently be used in the construction of an
APWP unless their age is verified independently, as the
estimated age of the magnetization is derived from the
path itself.
Bibliography
Andersen, T. B., Jamtveit, B., Dewey, J. F., and Swensson, E., 1991.
Subduction and eduction of continental crust: major mechanisms
during continent-continent collision and orogenic extensional
collapse, a model based on the south Norwegian Caledonides.
Terra Nova, 3, 303–310.
Cox, A., 1970. Latitude dependence of the angular dispersion of the
geomagnetic field. Geophysical Journal of the Royal Astronom-
ical Society, 20, 253–269.
Creer, K. M., Irving, E., and Runcorn, S. K., 1954. The direction of
the geomagnetic field in remote epochs in Great Britain. Journal
of Geomagnetism and Geoelectricity, 250, 164–168.
Domeier, M., Van der Voo, R., and Denny, F. B., 2011. Widespread
inclination shallowing in Permian and Triassic paleomagnetic
data from Laurentia: support from new paleomagnetic data from
Middle Permian shallow intrusions in southern Illinois (USA)
and virtual geomagnetic pole distributions. Tectonophysics,
511, 38–52.
Eide, E. A., Torsvik, T. H., and Andersen, T. B., 1997. Absolute dat-
ing of fault breccias: late Palaeozoic and early Cretaceous fault
reactivation in Western Norway. Terra Nova, 9, 135–139.
Fisher, R., 1953. Dispersion on a sphere. Proceedings of the Royal
Society of London. Series A: Mathematical and Physical Sci-
ences, 217(1130), 295–305.
Hess, H. H., 1962. History of ocean basins. In Engle, E. A. J. (ed.),
Petrologic Studies. New York: Geological Society of America.
Jupp, P. E., and Kent, J. T., 1987. Fitting smooth paths to spherical
data. Applied Statistics, 36, 34–36.
McElhinny, M. W., and McFadden, P. L., 2000. Paleomagnetism:
Continents and Oceans. San Diego: Academic, 386 pp.
McFadden, P. L., and Lowes, F. J., 1981. The discrimination of
mean directions drawn from Fisher distribution. Geophysical
Journal of the Royal Astronomical Society, 67, 19–33.
McKenzie, D. P., and Parker, R. L., 1967. The North Pacific: an
example of tectonics on a sphere. Nature, 216, 1276–1280.
Merrill, R. T., McElhinny, M. W., and McFadden, P. L., 1996. The
Magnetic Field of the Earth: Paleomagnetism, the Core, and
the Deep Mantle. San Diego: Academic, 531 pp.
Miller, J. D., and Kent, D. V., 1988. Regional trends in the timing of
Alleghenian remagnetization in the Appalachians. Geology, 16,
588–591.
Oliver, J., 1986. Fluids expelled tectonically from orogenic belts:
their role in hydrocarbon migration and other geologic phenom-
ena. Geology, 14, 99–102.
Pannalal, S. J., Symons, D. T. A., and Sangster, D. F., 2004. Paleo-
magnetic dating of upper Mississippi valley zinc-lead
mineralisation, WI, USA. Journal of Applied Geophysics, 56,
135–153.
Pannalal, S. J., Symons, D. T. A., and Sangster, D. F.,
2008a. Paleomagnetic evidence for an early Permian Age of
the Lisheen Zn-Pb deposit, Ireland. Economic Geology, 103,
1641–1655.
Pannalal, S. J., Symons, D. T. A., and Sangster, D. F., 2008b. Paleo-
magnetic evidence of a Variscan age for the epigenetic Galmoy
zinc-lead deposit, Ireland. Terra Nova, 20, 385–393.
Reynolds, R. L., Goldhaber, M. B., and Snee, L. W., 1997. Paleo-
magnetic and 40Ar/39Ar results from the Grant intrusive breccia
and comparison to the Permian Downeys Bluff sill - evidence for
Permian igneous activity at Hicks Dome, southern Illinois Basin.
U. S. Geological Survey Bulletin 2094-G, 16 pp.
Runcorn, S. K., 1956. Palaeomagnetic comparisons between
Europe and North America. Proceedings of the Geological Asso-
ciation of Canada, 8, 77–85.
Symons, D. T. A., and Stratakos, K. K., 2002. Paleomagnetic dating
of Alleghanian orogenesis and mineralisation in the Mascot-
Jefferson City zinc district of East Tennessee, USA.
Tectonophysics, 348, 51–72.
Symons, D. T. A., Pannalal, S. J., Kawasaki, K., Sangster, D. F., and
Stanley, G. A., 2007. Paleomagnetic age of the Magcobar Ba
deposit, Silvermines, Ireland. In Andrew, C. J., et al. (eds.), Min-
eral Exploration and Research: Digging Deeper. Dublin: Irish
Association for Economic Geology, pp. 377–380.
Torsvik, T. H., Sturt, B. A., Swensson, E., Andersen, T. B., and
Dewey, J. F., 1992. Palaeomagnetic dating of fault rocks: evi-
dence for Permian and Mesozoic movements and brittle defor-
mation along the extensional Dalsfjord Fault, western Norway.
Geophysical Journal International, 109, 565–580.
Torsvik, T. H., Smethurst, M. T., Meert, J. G., Van der Voo, R.,
McKerrow, W. S., Brasier, M. D., Sturt, B. A., and Walderhaug,
H. J., 1996. Continental break-up and collision in the
Neoproterozoic and Paleozoic—a tale of Baltica and Laurentia.
Earth Science Reviews, 40, 229–258.
 CORALS (SCLEROCHRONOLOGY)
Torsvik, T. H., Van der Voo, R., Preeden, U., Mac Niocaill, C.,
Steinberger, B., Doubrovine, P. V., van Hinsbergen, D. J. J.,
Domeier, M., Gaina, C., Tovher, E., Meert, J. G., McCausland,
P. J., and Cocks, L. R. M., 2012. Phanerozoic polar wander,
paleogeography and dynamics. Earth Science Reviews, 114,
325–368.
Van der Voo, R., 1990. Phanerozoic paleomagnetic poles from
Europe and North America and comparisons with continental
reconstructions. Reviews of Geophysics, 28, 167–206.
Van der Voo, R., and Torsvik, T. H., 2012. The history of
remagnetization of sedimentary rocks: deceptions, develop-
ments, discoveries. Geological Society, London, Special Publi-
cations, 371, 23–53 doi: 10.1144/SP371.2.
Vine, F. J., and Matthews, D. H., 1963. Magnetic anomalies over
oceanic ridges. Nature, 199, 947–949.
Wegener, A., 1912. Die Entstehung der Kontinente. Petermann’s
Mittelungen aus Justus Perthes’ Geographischer Anstalt, 58,
185–195, 253–256, 305–309.
Wegener, A., 1915. Die Entstehung der Kontinente und Ozeane.
Brunswick: Vieweg.
CORALS (SCLEROCHRONOLOGY)
Kristine L. DeLong
Department of Geography and Anthropology, Lousiana
State University, Baton Rouge, LA, USA
Synonyms
Coral chronometer; Related to dendrochronology
Definition
Sclerochronology: The study of the incremental layers
contained in the hard parts or skeletons of organisms that
form at regular intervals during the organism’s life span.
Sclerochronology is derived from the Greek words sklero
meaning “hard,” khronos meaning “time,” and logos
meaning “science of.” Sclerochronology is related to den-
drochronology or the study of annual growth rings in
trees.
Coral: Corals are in the class Anthozoa of the phylum
Cnidaria and include scleractinian or “stony corals,”
which form colonies of many individual polyps with an
external skeleton or exoskeleton of calcium carbonate
(CaCO3), and non-stony corals or soft corals. Some stony
corals contain incremental layers in their exoskeleton.
Coral-based sclerochronological research is focused on
stony corals yet recent research reveals that some soft
coral species contain incremental layers (see Roark et al.,
2006).
Introduction
The general term of sclerochronology includes a diverse
array of marine and terrestrial organisms, both extinct
and extant, that produce hard structures with regular time
intervals related to the organism’s growth and/or environ-
ment. Such organisms include corals, molluscs, fish oto-
liths or “ear bones,” coralline algae, sclerosponges,
187
brachiopods, and bones of vertebrate animals with
research expanding to other organisms (see Arnold and
Jones, 2007; Schöne, 2010). This description is focused
on coral-based sclerochronology, in particular
scleractinian corals, with descriptions of mollusc-based
sclerochronology appearing in another entry.
The exoskeleton of scleractinian corals is composed of
aragonite, a mineral form of calcium carbonate, deposited
by the living coral polyps within the surface of the exo-
skeleton. The individual coral polyps calcify new skeleton
thecal “walls” and dissepiment “floors,” lifting up as new
skeleton is created, similar to building a skyscraper. As the
coral deposits its skeleton, it produces a high-density band
and a low-density band each year while simultaneously
recording environmental signals, such as water tempera-
ture and chemistry, in its skeleton (e.g., d13C, d18O, and
Sr/Ca). Coral colonies have different morphologies
including branching, columnar, and boulder-shaped or
massive corals, for which annual density bands are found
in massive and columnar colonies. Massive coral colonies,
which grow up to 8 m in height (Veron, 1986; Isdale et al.,
1998; Corrège, 2006; Brown et al., 2009), can live for
many centuries providing a continuous annual chronology
preserved in its exoskeleton. Coral sclerochronology is
used for establishing time for a variety of purposes includ-
ing paleoceanographic and paleoclimatic reconstructions,
investigations of environmental change, refining radio-
metric dating methods, assessing sea-level variations,
and documenting earthquakes.
History
Research on coral reefs began in earnest with Charles
Darwin’s hypothesis of reef formation with the passage
of time (Darwin, 1842). Early work with the branching
coral Acropora palmata in the Bahamas notes regular
undulations on the surface of the broad fronds (Whitfield,
1898) shown empirically as annual in nature (Vaughan,
1915). A Chinese paleontologist, Ting-Ying H. Ma
(1934), notes that Paleozoic and recent corals contain
internal skeletal structures with varying densities. The
report of Wells (1963) examines diurnal growth lines in
the epitheca of Devonian corals, which he interprets as
evidence for more days in a Devonian year compared to
present.
Despite these early studies, the annual nature of density
variations within coral skeletons is not confirmed until the
study of Knutson et al. (1972) examines corals from the
nuclear test site at Eniwetok Atoll. That study uses autora-
diography to image radioactive layers in the coral colony
by placing X-ray film in contact with a thin radial cross
section for over one month. Next, they couple autoradiog-
raphy with X-radiography, which reveals alternating high-
and low-density bands in the cross section, and then count
the number of density band couplets between the radioac-
tive bands. They found the number of bands equals the
number of years between nuclear weapons tests, thus
confirming the annual nature of high-low-density bands
188 CORALS (SCLEROCHRONOLOGY)
in corals. Knutson et al. (1972) suggest combining the
“coral internal calendar” with isotope thermometry in
corals (Weber and Woodhead, 1972) will allow for the
“construction of coral-ring chronologies” of past oceanic
variability, similar to dendrochronology. In June 1973,
the papers presented at the Second International Sympo-
sium on Coral Reefs include publications related to
sclerochronology (Buddemeier, 1974; Macintyre and
Smith, 1974; Moore and Krishnaswami, 1974). The use
of term “sclerochronology” first appears in two 1974 pub-
lications (Buddemeier et al., 1974; Macintyre and Smith,
1974), yet this term appears to originate from this confer-
ence. Subsequent studies confirm the annual nature of the
density couplets for Atlantic and Indo-Pacific coral spe-
cies by radiometric analysis, staining, and direct measure-
ment (Buddemeier, 1974; Buddemeier et al., 1974; Dodge
and Thomson, 1974; Macintyre and Smith, 1974; Moore
and Krishnaswami, 1974) as well as revealing diurnal
increments (Barnes, 1970) and monthly or lunar cycles
(Buddemeier, 1974).
Sclerochronological methods
Samples for sclerochronological studies include cores,
3–10 cm in diameter, extracted from large colonies using
hydraulic or pneumatic drills or whole colonies if small
enough to recover by divers or by a boat-mounted winch.
The cores or colonies are sectioned along the primary
growth axis to thin slabs of constant thickness of approx-
imately 0.5 cm. The slabs are exposed to X-rays to pro-
duce X-radiographs (i.e., X-ray images) using either film
or digital X-ray systems. X-radiographs are negative
images exhibiting density variations from light high-
density areas to dark low-density areas in the coral slab
for which alternating bands of high and low density repre-
sent a single year’s growth. Years are assigned by
counting annual density couplets from a known date or
collection date from the top of the slab following the den-
sity bands and growth patterns. A floating chronology
(e.g., year 1, year 2, etc.) is utilized for samples without
a known collection date such as dead or fossil corals.
Two parameters of coral growth are measured from
X-radiographs: linear extension per year (cm/year) and
skeletal density (g/cm3), both of which vary indepen-
dently on a seasonal basis. Relative calcification rate
(g/cm3/year) is the product of linear extension and density.
Annual linear extension (cm) is physically measured from
the top of one high-density band to the top of the next
high-density band using calipers. X-ray and gamma densi-
tometers provide density measurements from coral slabs
(Buddemeier, 1974; Dodge and Thomson, 1974; Chalker
et al., 1985; Chalker and Barnes, 1990). For digital
X-radiographs, imaging software such as Coral XDS
(www.nova.edu/ocean/coralxds) (Helmle et al., 2002),
Image J (sb.info.nih.gov/ij), and Adobe Photoshop©
(www.photoshop.com) are used to determine linear exten-
sion and relative density. Linear extension can be deter-
mined from the winter extremes in seasonal coral
chemical variations (e.g., d18O or Sr/Ca) using the dis-
tance (i.e., cm from core top) between these samples. This
method is particularly useful for comparing chemical var-
iations and extension rates, but precision is limited to the
chemical sampling resolution.
The study of Dodge and Thomson (1974) is the first to
apply methodologies from dendrochronology to build
a master chronology using density variations from multi-
ple coral colonies in Bermuda. Further work with
a larger network of colonies reveals long-term covariance
with temperature and air pressure records (Dodge and
Vaisnys, 1975). The report of Hudson et al. (1976) demon-
strates that stress bands in Montastraea annularis formed
by cold events can be used as stratigraphic markers for
cross-dating between live and dead colonies from the
same reef. Some corals skeletons contain luminescent
lines under ultra-violet light from soil-derived humic acids
that are interpreted as provies of coastal precipitation and
runoff (Boto and Isdale, 1985; Omata et al., 2006; Grove
et al., 2010). Researchers utilize these lines to cross-date
coral colonies to produce runoff histories for individual
rivers and regional scale precipitation reconstructions that
span multiple centuries (Isdale et al., 1998; Hendy et al.,
2003; Nyberg et al., 2007; Lough, 2011). Cross-dated
luminescence corals are coupled with chemical records
to extend these records back in time (Tudhope et al.,
1996; Hendy et al., 2002). The study of Cobb et al.
(2003) uses high-precision uranium-thorium dating
(5–10 years) and correlation between coral d18O varia-
tions to generate long overlapping coral records
reconstructing El Niño-Southern Oscillation variability
(Cobb et al., 2013). A relatively new application is
cross-dating coral Sr/Ca variations, a temperature proxy,
between cores from the same Porites lutea colony to pro-
duce a 350-year-long sea surface temperature (SST)
reconstruction (DeLong et al., 2012). Those authors con-
firm their cross-dated Sr/Ca chronology with high-
precision (1 year) 230Th dating (Shen et al., 2008),
which allows them to assess sources of chronology error
(DeLong et al., 2013). Further application of this cross-
dating method with live and dead Siderastrea siderea
corals has produced a 274-year-long SST reconstruction
for the Gulf of Mexico (DeLong et al., in review).
Applications of coral sclerochronology
Growth in corals varies with environmental factors such as
temperature, light, and nutrients; thus, environmental var-
iations are recorded in the sclerochronology. Initial studies
at Eniwetok suggest that density bands in the corals form
in response to seasonal variations in cloud cover or precip-
itation (Knutson et al., 1972; Buddemeier et al., 1974).
Conversely, the report of Weber et al. (1975), with
a wide array of corals in locality and genera, finds seasonal
temperature variations drive coral growth rates. Further
research with coral growth records continues to produce
conflicting results (see reviews of Buddemeier and Kinzie,
1976; Barnes and Lough, 1996); consequently, these
 CORALS (SCLEROCHRONOLOGY)
records were relegated to chronology development for
coral chemical studies in the early 1990s, yet recent inter-
est in ocean acidification has revived interest in coral
growth histories (see reviews of Lough, 2008; Lough
and Cooper, 2011).
Soon after the confirmation of annual increments in
corals, researchers started sampling coral skeletons across
the density bands to produce time series of chemical vari-
ability. These early studies coupled sclerochronology with
coral chemistry including stable isotopes (d13C and d18O)
(Goreau, 1977; Emiliani et al., 1978; Nozaki et al., 1978;
Fairbanks and Dodge, 1979), trace elements (Fe, Mg,
Na, and Sr) (Goreau, 1977; Oomori et al., 1982; Schneider
and Smith, 1982), and radiocarbon (Druffel, 1980). These
coral chemical variations and others represent water
chemistry histories (e.g., B, Ba/Ca, 14C, Mn/Ca, P/Ca,
and Pb/Ca) as well as serve as proxies for water tempera-
ture (d18O, Mg/Ca, Sr/Ca, and U/Ca) and ocean circula-
tion (e.g., 14C, Ba, and Cd) (see reviews of Shen, 1993;
Druffel, 1997). Advances in sampling methodologies
(e.g., up to weekly) and analytical improvements (i.e.,
reducing the amount of sample needed and improved ana-
lytical precision) have resulted in rapidly expanding coral
chemical research compared to growth-rate studies
(Lough and Cooper, 2011). Coral sclerochronology-
chemistry studies include numerous high-resolution,
multi-century long records (see www.ncdc.noaa.gov/
paleo/corals) that provide valuable information to the
fields of paleoclimatology, ocean circulation, and environ-
mental change (see reviews of Swart, 1983; Druffel, 1997;
Gagan et al., 2000; Cole, 2003; Felis and Patzold, 2004;
Corrège, 2006; Grottoli and Eakin, 2007; Lough, 2010).
The sclerochronology determined by counting the
annual bands in corals allows for comparisons with radio-
metric dating methods (e.g., U-Th) to improve analytical
methods (e.g., Shen et al., 2008, 2012; Clark et al.,
2012). Radioisotopes are incorporated in the coral exo-
skeleton during calcification (e.g., 14C, 228Ra, and 210Pb)
or by decay after incorporation (e.g., 230Th and 231Pa).
Uranium-thorium(U-Th) dating of carbonates provides
ages with high precision (3 years) with relatively small
sample sizes (Edwards et al., 1987). The study of Shen
et al. (2008) refines U-Th dating for corals by measuring
141 coral samples each extracted from a single density
band of known age, determined by counting annual bands.
Those authors demonstrate that corals with an age less
than 100 years can be 230Th-dated with a precision of
1 year. Further analytical refinement resulted in reduced
sample size for U-Th analysis (10–50 mg of coral carbon-
ate) with a precision of less than 1 year (Shen et al.,
2012). Such high-precision U-Th dates allow for the
cross-dating of dead and fossil corals to extend the inclu-
sive records further in time (e.g., DeLong et al., in review).
Microatolls are massive corals in very shallow water
with a dead flat or concave top resembling an atoll island.
As these corals grow out of the water, the exposed surface
dies while the submerge surface continues to live; thus,
these corals serve as an indicator of sea level (Smithers
189
and Woodroffe, 2000). Counting the annual bands in the
coral’s X-radiograph reveals the date the coral surface
emerged from the water. Using microatolls as tide gauges,
the study of Taylor et al. (1987) documents the date of four
earthquake events in Vanuatu, an island in the southwest
Pacific, that uplifted the corals during earthquakes. These
coral earthquake records provide dates of past earthquakes
in remote areas that lack such records, as well as informa-
tion on vertical uplifting rates and seismic movements
(Buskirk et al., 1982; Taylor et al., 1982). Other
researchers utilize microatolls to study twentieth-century
sea-level variations (e.g., Smithers and Woodroffe, 2001)
and Holocene sea-level changes (e.g., Yu et al., 2009;
Woodroffe et al., 2012). Investigations with microatolls
include corals from the tropical Pacific, Indian, and Atlan-
tic oceans (see review of Buddemeier and Taylor, 2000).
Summary
Coral sclerochronology started with the discovery of
annual density bands within the coral skeleton. These
bands are visible in X-radiographs and dates are assigned
by counting bands from the collection date. The primary
use of coral sclerochronology is assigning time to syn-
chronous chemical records store in the skeletons of these
long-lived organisms. These time series of coral chemical
variations are used for paleoclimate reconstructions,
investigations of environmental change, and refining
radiometric dating methods. Additionally, annual density
bands in corals are utilized to investigate changes in sea
level and earthquakes. The field of coral sclerochronology
is continually advancing with new applications being
developed to decipher the hidden messages stored in the
skeletons of corals.
Bibliography
Arnold, B., and Jones, D. S., 2007. Program & abstract book. In
Arnold, B., and Jones, D. S. (ed.), 1st International
Sclerochronology Conference, Saint Petersburg, p. 118.
Barnes, D. J., 1970. Coral skeletons: an explanation of their growth
and structure. Science, 170, 1305–1308.
Barnes, D. J., and Lough, J. M., 1996. Coral skeletons: storage and
recovery of environmental information. Global Change Biology,
2, 569–582.
Boto, K., and Isdale, P., 1985. Fluorescent bands in massive corals
result from terrestrial fulvic acid inputs to nearshore zone.
Nature, 315, 396–397.
Brown, D., Basch, L., Barshis, D., Forsman, Z., Fenner, D., and
Goldberg, J., 2009. American Samoa’s island of giants: massive
Porites colonies at Ta’u island. Coral Reefs, 28, 735.
Buddemeier, R. W., 1974. Environmental controls over annual and
lunar monthly cycles in hermatypic coral calcification. In Cam-
eron, A. M., Cambell, B. M., Cribb, A. B., Endean, R., Jell,
J. S., Jones, O. A., Mather, P., and Talbot, F. H. (eds.), Proceed-
ings of the Second International Coral Reef Symposium. Bris-
bane: Great Barrier Reef Committee, Vol. 2, pp. 259–267.
Buddemeier, R. W., and Kinzie, R. A., 1976. Coral growth. Ocean-
ography and Marine Biology Annual Review, 14, 183–225.
Buddemeier, R. W., and Taylor, F. W., 2000. Sclerochronology. In
Quaternary Geochronology. Washington, DC: American Geo-
physical Union, pp. 25–40.
190 CORALS (SCLEROCHRONOLOGY)
Buddemeier, R. W., Maragos, J. E., and Knutson, D. W., 1974.
Radiographic studies of reef coral exoskeletons: rates and pat-
terns of coral growth. Journal of Experimental Marine Biology
and Ecology, 14, 179–199.
Buskirk, R. E., Taylor, F. W., O’Brien, W. P., Maillet, P., and Gilpin,
L., 1982. Seasonal growth patterns and mortality of corals in the
New Hebrides (Vanuatu). In Gomez, E. D., Birkeland, C. E.,
Buddemeier, R. W., Johannes, R. E., Marsh, J. A., and Tsuda,
R. T. (eds.), Proceedings of the 4th International Coral Reef
Symposium. Manila: Marine Science Center, University of the
Philippines, Vol. 2, pp. 197–200.
Chalker, B. E., and Barnes, D. J., 1990. Gamma densitometry for the
measurement of skeletal density. Coral Reefs, 9, 11–13.
Chalker, B., Barnes, D., and Isdale, P., 1985. Calibration of x-ray
densitometry for the measurement of coral skeletal density.
Coral Reefs, 4, 95–100.
Clark, T. R., Zhao, J.-X., Feng, Y.-X., Done, T. J., Jupiter, S., Lough,
J., and Pandolfi, J. M., 2012. Spatial variability of initial
230
Th/232Th in modern Porites from the inshore region of the
Great Barrier Reef. Geochimica et Cosmochimica Acta, 78,
99–118.
Cobb, K. M., Charles, C. D., Cheng, H., and Edwards, R. L., 2003.
El Niño/Southern Oscillation and tropical Pacific climate during
the last millennium. Nature, 424, 271–276.
Cobb, K. M., Westphal, N., Sayani, H. R., Watson, J. T., Di
Lorenzo, E., Cheng, H., Edwards, R. L., and Charles, C. D.,
2013. Highly variable El Niño–Southern Oscillation throughout
the Holocene. Science, 339, 67–70.
Cole, J. E., 2003. Holocene coral records: windows on tropical cli-
mate variability. In McKay, A., Battarbee, R., Birks, J., and
Oldfield, F. (eds.), Global Change in the Holocene. London:
Arnold, pp. 168–184.
Corrège, T., 2006. Sea surface temperature and salinity reconstruc-
tion from coral geochemical tracers. Palaeogeography,
Palaeoclimatology, Palaeoecology, 232, 408–428.
Darwin, C., 1842. The structure and distribution of coral reefs
smith. London: Elder and Co.
DeLong, K. L., Maupin, C. R., Flannery, J. A., Quinn, T. M., Shen,
C.–C., and Lin, K., In Revision. A Reconstruction of sea surface
temperature variability in the south eastern Gulf of Mexico from
1734–2008 CE using cross-dated Sr/Ca records from the coral
Siderastrea siderea, Paleoceanography, 2013PA002524.
DeLong, K. L., Quinn, T. M., Taylor, F. W., Lin, K., and Shen, C.–.C.,
2012. Sea surface temperature variability in the southwest tropi-
cal Pacific since AD 1649. Nature Climate Change, 2, 799–804.
DeLong, K. L., Quinn, T. M., Taylor, F. W., Lin, K., and Shen, C.–.C.,
2013. Improving coral-base paleoclimate reconstructions by rep-
licating 350 years of coral Sr/Ca variations. Palaeogeography,
Palaeoclimatology, Palaeoecology, 373, 6–24.
Dodge, R. E., and Thomson, J., 1974. The natural radiochemical
and growth records in contemporary hermatypic corals from
the Atlantic and Caribbean. Earth and Planetary Science Letters,
23, 313–322.
Dodge, R. E., and Vaisnys, J. R., 1975. Hermatypic coral growth
banding as environmental recorder. Nature, 258, 706–707.
Druffel, E. R., 1980. Radiocarbon in annual coral rings of Belize
and Florida. Radiocarbon, 22, 363–371.
Druffel, E. R. M., 1997. Geochemistry of corals: proxies of past
ocean chemistry, ocean circulation, and climate. Proceedings of
the National Academy of Sciences of the United States of Amer-
ica, 94, 8354–8361.
Edwards, L. R., Chen, J. H., and Wasserburg, G. J., 1987.
238
U–234U–230Th–232Th systematics and the precise measure-
ment of time over the past 500,000 years. Earth and Planetary
Science Letters, 81, 175–192.
Emiliani, C., Hudson, J. H., Shinn, E. A., and George, R. Y., 1978.
Oxygen and carbon isotopic growth record in a reef coral from
the Florida Keys and a deep-sea coral from Blake Plateau. Sci-
ence, 202, 627–629.
Fairbanks, R. G., and Dodge, R. E., 1979. Annual periodicity of the
18 16
O/ O and 13C/12C ratios in the coral Montastrea annularis.
Geochimica et Cosmochimica Acta, 43, 1009–1020.
Felis, T., and Patzold, J., 2004. Climate reconstructions from annu-
ally banded corals. In Shiyomi, M. (ed.), Global Environmental
Change in the Ocean and on Land. Tokyo: Terrapub,
pp. 205–227.
Gagan, M. K., Ayliffe, L. K., Beck, J. W., Cole, J. E., Druffel,
E. R. M., Dunbar, R. B., and Schrag, D. P., 2000. New views
of tropical paleoclimates from corals. Quaternary Science
Reviews, 19, 45–64.
Goreau, T. J., 1977. Coral skeletal chemistry – physiological and
environmental regulation of stable isotopes and trace-metals in
Montastrea Annularis. Proceedings of the Royal Society of Lon-
don Series B-Biological Sciences, 196, 291–315.
Grottoli, A., and Eakin, M., 2007. A review of modern coral d18O
and D14C proxy records. Earth-Science Reviews, 81, 67–91.
Grove, C., Nagtegaal, R., Zinke, J., Scheufen, T., Koster, B.,
Kasper, S., McCulloch, M., van den Bergh, G., and Brummer,
G., 2010. River runoff reconstructions from novel spectral lumi-
nescence scanning of massive coral skeletons. Coral Reefs, 29,
579–591.
Helmle, K. P., Kohler, K. E., and Dodge, R. E., 2002. Relative opti-
cal densitometry and the coral x-radiograph densitometry sys-
tem: CoralXDS. In Brooks, S., Spencer, T., Teleki, K., and
Taylor, M. (eds.), European Meeting of the International Society
for Reef Studies (ISRS). Cambridge, UK: University of
Cambridge.
Hendy, E. J., Gagan, M. K., Alibert, C. A., McCulloch, M. T.,
Lough, J. M., and Isdale, P. J., 2002. Abrupt decrease in tropical
Pacific sea surface salinity at end of Little Ice Age. Science, 295,
1511–1514.
Hendy, E. J., Gagan, M. K., and Lough, J. M., 2003. Chronological
control of coral records using luminescent lines and evidence for
non-stationary ENSO teleconnections in northeast Australia.
The Holocene, 13, 187–199.
Hudson, J. H., Shinn, E. A., Halley, R. B., and Lidz, B., 1976.
Sclerochronology: a tool for interpreting past environments.
Geology, 4, 361–364.
Isdale, P. J., Stewart, B. J., Tickle, K. S., and Lough, J. M., 1998.
Palaeohydrological variation in a tropical river catchment:
a reconstruction using fluorescent bands in corals of the Great
Barrier Reef, Australia. The Holocene, 8, 1.
Knutson, D., Buddemeier, R., and Smith, S., 1972. Coral chronom-
eters: seasonal growth bands in reef corals. Science, 177,
270–272.
Lough, J. M., 2008. Coral calcification from skeletal records
revisited. Marine Ecology: Progress Series, 373, 257–264.
Lough, J. M., 2010. Climate records from corals. Wiley Interdisci-
plinary Reviews: Climate Change, 1, 318–331.
Lough, J., 2011. Measured coral luminescence as a freshwater
proxy: comparison with visual indices and a potential age arte-
fact. Coral Reefs, 30, 169–182.
Lough, J. M., and Cooper, T. F., 2011. New insights from coral
growth band studies in an era of rapid environmental change.
Earth-Science Reviews, 108, 170–184.
Ma, T.-Y. H., 1934. On the seasonal growth change of growth in the
reef coral Favia speciosa (Dana). Proceedings of the Imperial
Academy, Tokyo, 10, 353–356.
Macintyre, I. G., and Smith, S. V., 1974. X-radiographic studies of
skeletal development in coral colonies. In Cameron, A. M.,
Cambell, B. M., Cribb, A. B., Endean, R., Jell, J. S., Jones,
O. A., Mather, P., and Talbot, F. H. (eds.), Proceedings of the
Second International Coral Reef Symposium. Brisbane: The
Great Barrier Reef Committee, Vol. 2, pp. 277–287.
 CRUSTAL SULFIDE MINERALS (Re-Os)
Moore, W. S., and Krishnaswami, S., 1974. Correlation of
x-radiographs revealed banding in corals with radiometric
growth rates. In Cameron, A. M., Cambell, B. M., Cribb,
A. B., Endean, R., Jell, J. S., Jones, O. A., Mather, P., and Talbot,
F. H. (eds.), Proceedings of the Second International Coral Reef
Symposium. Brisbane: Great Barrier Reef Committee, Vol. 2,
pp. 269–276.
Nozaki, Y., Rye, D. M., Turekian, K. K., and Dodge, R. E., 1978.
A 200 year record of carbon-13 and carbon-14 variations in
a Bermuda coral. Geophysical Research Letters, 5, 825–828.
Nyberg, J., Malmgren, B. A., Winter, A., Jury, M. R., Kilbourne,
K. H., and Quinn, T. M., 2007. Low Atlantic hurricane activity
in the 1970s and 1980s compared to the past 270 years. Nature,
447, 698–701.
Omata, T., Suzuki, A., Kawahata, H., Nojima, S., Minoshima, K.,
and Hata, A., 2006. Oxygen and carbon stable isotope systemat-
ics in Porites coral near its latitudinal limit: the coral response to
low-thermal temperature stress. Global and Planetary Change,
53, 137–146.
Oomori, T., Kaneshima, K., and Nakamura, Y., 1982. Seasonal var-
iation of minor elements in coral skeletons. Galaxea, 1, 77–86.
Roark, E. B., Guilderson, T. P., Dunbar, R. B., and Ingram, B. L.,
2006. Radiocarbon-based ages and growth rates of Hawaiian
deep-sea corals. Marine Ecology: Progress Series, 327, 1–14.
Schneider, R. C., and Smith, S. V., 1982. Skeletal Sr content and
density in Porites spp. in relation to environmental factors.
Marine Biology, 66, 121.
Schöne, B. R., 2010. Program and abstracts. In Schöne, B. R. (ed.),
2nd International Sclerochronology Conference, Mainz, p. 118.
Shen, G. T., 1993. Reconstruction of El Niño history from reef
corals. Bulletin de l’Institut Francais d’Etudes Andines, 22,
125–158.
Shen, C.–. C., Li, K.-S., Sieh, K., Natawidjaja, D., Cheng, H.,
Wang, X., Edwards, R. L., Lam, D. D., Hsieh, Y.-T., Fan, T.-
Y., Meltzner, A. J., Taylor, F. W., Quinn, T. M., Chiang, H.-W.,
and Kilbourne, K. H., 2008. Variation of initial 230Th/232Th
and limits of high precision U–Th dating of shallow-water
corals. Geochimica et Cosmochimica Acta, 72, 4201–4223.
Shen, C.–. C., Wu, C.–. C., Cheng, H., Lawrence, E. R., Hsieh, Y.-T.,
Gallet, S., Chang, C.–. C., Li, T.-Y., Lam, D. D., Kano, A.,
Hori, M., and Spötl, C., 2012. High-precision and high-
resolution carbonate 230Th dating by MC-ICP-MS with SEM
protocols. Geochimica et Cosmochimica Acta, 99, 71–86.
Smithers, S. G., and Woodroffe, C. D., 2000. Microatolls as sea-
level indicators on a mid-ocean atoll. Marine Geology, 168,
61–78.
Smithers, S. G., and Woodroffe, C. D., 2001. Coral microatolls and
20th century sea level in the eastern Indian Ocean. Earth and
Planetary Science Letters, 191, 173–184.
Swart, P. K., 1983. Carbon and oxygen isotope fractionation in
scleractinian corals: a review. Earth-Science Reviews, 19, 51–80.
Taylor, F. W., Jouannic, C., Gilpin, L., and Bloom, A. L., 1982.
Coral colonies as monitors of change in relative level of the land
and sea: applications to vertical tectonism. In Gomez, E. D.,
Birkeland, C. E., Buddemeier, R. W., Johannes, R. E., Marsh,
J. A., and Tsuda, R. T. (eds.), Proceedings of the 4th Interna-
tional Coral Reef Symposium. Manila: Marine Science Center,
University of the Philippines, Vol. 1, pp. 484–492.
Taylor, F. W., Frohlich, C., Lecolle, J., and Strecker, M., 1987. Anal-
ysis of partially emerged corals and reef terraces in the central
Vanuatu Arc: comparison of contemporary coseismic and
nonseismic with quaternary vertical movements. Journal of
Geophysical Research: Solid Earth, 92, 4905–4933.
Tudhope, A. W., Lea, D. W., Shimmield, G. B., Chilcott, C. P., and
Head, S., 1996. Monsoon climate and Arabian sea coastal
upwelling recorded in massive corals from southern Oman.
Palaios, 11, 347–361.
191
Vaughan, T. W., 1915. The geologic significance of the growth-rate
of the Floridian and Bahaman shoal-water corals. Journal of the
Washington Academy of Sciences, 5, 591–600.
Veron, J. E. N., 1986. Corals of Australia and the Indo-Pacific. Aus-
tralia: North Ryde NSW, University of Hawaii Press.
Weber, J. N., and Woodhead, P. M., 1972. Temperature dependence
of oxygen-18 concentration in reef coral carbonates. Journal of
Geophysical Research, 77, 463–473.
Weber, J. N., White, E. W., and Weber, P. H., 1975. Correlation of
density banding in reef coral skeletons with environmental
parameters: the basis for interpretation of chronological records
preserved in the corolla of corals. Paleobiology, 1, 137–149.
Wells, J. W., 1963. Coral growth and geochronometry. Nature, 197,
948–950.
Whitfield, R. P., 1898. Notice of a remarkable specimen of the West
India coral Madrepora palmata. Bulletin of the American
Museum of Natural History, 10(19), 463–464.
Woodroffe, C. D., McGregor, H. V., Lambeck, K., Smithers, S. G.,
and Fink, D., 2012. Mid-Pacific microatolls record sea-level sta-
bility over the past 5000 year. Geology, 40, 951–954.
Yu, K.-F., Zhao, J.-X., Done, T., and Chen, T.-G., 2009. Microatoll
record for large century-scale sea-level fluctuations in the mid-
Holocene. Quaternary Research, 71, 354–360.
Cross-references
Bivalve Sclerochronology
Luminescence, Biogenic Carbonates
Luminescence, Coastal Sediments
Sea Level Change (U-Series)
CRUSTAL SULFIDE MINERALS (Re-Os)
Robert A. Creaser
Department of Earth and Atmospheric Sciences,
University of Alberta, Edmonton, AB, Canada
Definition
The rhenium-osmium (187Re-187Os) radioactive isotope
system can be used to determine absolute age dates for
some sulfide minerals formed in a variety of crustal geo-
logic environments, including those associated with igne-
ous, sedimentary, and metamorphic processes. Sulfide
minerals that have yielded reliable Re-Os age dates
include molybdenite (MoS2), pyrite (FeS2), and arsenopy-
rite (FeAsS), although many other sulfide minerals have
potential for yielding Re-Os age dates. This entry deals
with Re-Os geochronology of crustal sulfide minerals typ-
ically formed from aqueous fluids, either hydrothermal or
diagenetic, and does not consider orthomagmatic sulfide
minerals.
Introduction
Following the discovery of the natural radioactivity of
187
Re by negatron decay (Naldrett and Libby, 1948), some
of the earliest applications of 187Re-187Os geochronology
involved the crustal sulfide mineral molybdenite
(MoS2 – Hirt et al., 1963; Luck and Allègre, 1982). Dur-
ing these early years, the analytical methods available
192 CRUSTAL SULFIDE MINERALS (Re-Os)
were such that only molybdenite, a mineral that contains
parts-per-million (ppm) level Re abundances, was realisti-
cally the sole crustal sulfide mineral amenable for Re-Os
isotopic analysis. Following major analytical break-
throughs in mass spectrometry (Creaser et al., 1991;
Volkening et al., 1991), sample digestion, and spike-
sample equilibration methods (Shirey and Walker, 1995;
Cohen and Waters, 1996; Birck et al., 1997), other crustal
sulfide minerals such as pyrite, typically containing parts-
per-billion (ppb) level of Re and parts-per-trillion (ppt)
level of Os, became amenable for isotopic analysis, with
potential application as geochronometers. Given the wide-
spread geological occurrence of crustal sulfide minerals
such as pyrite, development and application of the Re-
Os method for these minerals is an important recent devel-
opment in isotopic dating, particularly for research areas
such as ore deposit geology, in which the mineralogy
can be dominated by sulfide minerals and for which
obtaining accurate and precise ages dates for mineral for-
mation can be problematic (Richards and Noble, 1998).
Methodology: analytical protocol
The majority of modern analyses for Re-Os in crustal sul-
fide minerals involve mineral separation and purification,
spiking with isotopically enriched tracer solution(s) and
chemical digestion in inverse aqua regia using Carius tube
methodology (Shirey and Walker, 1995). Chemical purifi-
cation of Re and Os typically uses organic solvent extrac-
tion, ion-exchange, and microdistillation methods (Cohen
and Waters, 1996; Birck et al., 1997) and is followed by
isotope dilution negative thermal ionization mass spec-
trometry for isotopic analysis of both Os and Re (NTIMS –
Creaser et al., 1991; Volkening et al., 1991). Sample prep-
aration – crushing and milling of rock to purify sulfide
minerals – uses nonferrous materials such as ceramic
and agate, to avoid contamination by Re and Os present
in steel. The choice of isotopically enriched tracer depends
on the nature of Os present in the sulfide mineral. For
example, molybdenite rarely contains measurable “com-
mon Os” – Os incorporated at the time of formation –
and consists entirely of radiogenic 187Os. In this case,
a “double spike” of 188Os + 190Os is preferred, combined
with 185Re (e.g., Markey et al., 2003) to permit accurate
mass bias correction during isotopic analysis and to mon-
itor possible common Os in the sample. In some cases,
crustal sulfide minerals such as arsenopyrite also contain
little or no common Os (Morelli et al., 2005), and the
“mixed double-spike” method is also used. In these cases,
uncertainties in the quantification of 188Os can be large, as
it is dominated by blank 188Os, so uncertainties in
187
Re/188Os and 187Os/188Os are highly correlated on
a traditional isochron plot. In this case, the correlation
coefficient rho (r) must be used for isochron regression
analysis (Morelli et al., 2005) or a modified plot used that
eliminates 188Os (Stein et al., 2000). Many crustal sulfide
minerals like pyrite typically do contain trace amounts of
common Os, and a mixed 185Re + 190Os spike is typically
used for quantification of Os and Re abundances in these
cases. Re and Os abundances can be very low in many
cases (Ootes et al., 2011), and laboratory blank levels for
Re and Os must be extremely low (<1 pg) and monitored
for blank 187Os/188Os value. Age quantification for sam-
ples with no common Os uses the simplified equation:

t ¼ ln 187 Os=187 Re þ 1 =l
from which a single age is calculated based on the mea-
sured 187Re and 187Os abundances, whereas for samples
with common Os, the isochron approach using multiple
co-genetic samples is applied. The value of the 187Re
decay constant (l) of 1.666e11.a1 (Smoliar et al.,
1996) is used. Although a relatively new method, Re-Os
geochronology of crustal sulfide minerals has progressed
to the point of establishing reference age materials for
inter-laboratory comparison (Markey et al., 2007).
Re-Os geochronology of molybdenite
Molybdenite (MoS2) is ideal for geochronology using the
187
Re-187Os isotope system, as it is enriched in Re and is
usually devoid of common Os, so nearly all 187Os is radio-
genic, derived from the decay of 187Re. Although some
early studies concluded Re-Os age dating of molybdenite
to be inaccurate or easily disturbed (Luck and Allègre,
1982; McCandless et al., 1993; Suzuki et al., 2001), Stein
et al. (2001) found that this likely relates to internal spatial
decoupling of radiogenic 187Os from Re within molybde-
nite grains and that accurate age dates could be obtained if
sufficiently large sample sizes were analyzed. Investiga-
tion of Re-Os in molybdenite at the micron scale using
laser ablation ICP-MS methods confirmed these findings,
by producing highly inaccurate ages of very small sam-
pled volumes (Stein et al., 2003; Kosler et al., 2003; Selby
and Creaser, 2004). Since this limitation was discovered
and suitable analysis methods adopted, the Re-Os
geochronometer in molybdenite has proven to be highly
robust, showing excellent cross-calibration to U-Pb zircon
ages for igneous-associated molybdenite deposits back to
Archean time (Selby et al., 2007). Bingen and Stein
(2003) found that the Re-Os system in molybdenite has
a thermal resetting temperature (closure or blocking tem-
perature) well above 800
C. Molybdenite Re-Os geo-
chronology has developed into an invaluable dating
method in metallic ore deposit research, as it is
a common accessory mineral in porphyry Cu-(Mo-Au)
deposits, pegmatite and rare metal deposits, iron oxide-
Cu-Au-U (IOCG-type) deposits, and intrusion-related
Au deposits. It provides direct age control for the forma-
tion of sulfide minerals in a variety of ore-forming systems
to evaluate the general tectonomagmatic affinity of ores
(Chiaradia et al., 2009; Mao et al., 2003), to test the igne-
ous affiliation of ores (Reid et al., 2013; Selby et al.,
2002), and to investigate the temporal connections
between mineralization and volcanism (Rosera et al.,
2013). The precision of Re-Os molybdenite ages is typi-
cally 0.4 % (2s, e.g., Stein et al., 2001), equating to
 CRUSTAL SULFIDE MINERALS (Re-Os)
age uncertainties of 5 to 10 Ma in Archean time for indi-
vidual molybdenite age dates and thus not comparable to
the precision attained by U-Pb dating of minerals like zir-
con (e.g., Piercey et al., 2008). However, in geologically
young ore systems, Re-Os age dates of molybdenite can
have age precision of <50 Ka, providing temporal control
within the history large mineralized intrusive complexes
(e.g., Deckart et al., 2013). Although rare, multiple age
components have been described within molybdenite
crystals (Aleinikoff et al., 2012). Molybdenite paragenesis
is mostly related to hydrothermal ore deposition, but it has
also been found to form under high-grade metamorphic
conditions associated with partial melting reactions and
leucosome formation (Stein, 2006).
Re-Os geochronology of pyrite and arsenopyrite
Following the major advancements in the analysis of Re-
Os isotopes made in the 1990s, analysis of crustal sulfide
minerals other than molybdenite became feasible, with
several early papers investigating Re-Os geochronology
of pyrite (Stein et al., 1998; Mathur et al., 1999; Stein
et al., 2000; Arne et al., 2001). These studies found Re
levels in pyrite ranging up to a few tens of ppb, in some
cases with little common Os present (Stein et al., 2000)
enabling direct Re-Os age dating of a very common ore
sulfide mineral in massive sulfide and gold vein deposits.
Since then, the Re-Os geochronometer in pyrite has been
used widely in ore deposit research to date the timing of
vein versus massive ore formation in sedimentary
exhalative-type deposits (SedEx; Morelli et al., 2004), to
connect orogenic gold deposits with deep crustal meta-
morphism (Ootes et al., 2011), and to link polymetallic
deposits to magmatism (Zhu et al., 2013). In terms of ther-
mal resetting, Brenan et al. (2000) found that Os diffusion
in pyrite is very slow, leading to estimates of 500
C for
10 Ma to affect the cores of 500 mm pyrite crystals, and
Morelli and Creaser (2007) studied metamorphosed sul-
fide ores to estimate Re-Os resetting temperatures above
600
C for pyrite. Beyond ore deposit research,
synsedimentary to early digenetic pyrite from carbona-
ceous shales has been used to provide age constraints
within Proterozoic sedimentary basins (Hannah et al.,
2004).
Arsenopyrite (FeAsS) is typically restricted to hydro-
thermal vein and contact metamorphism occurrences, but
As is an important pathfinder element for gold ores (Nude
et al., 2012) as arsenopyrite is a host for many refractory
gold deposits. Thus, the ability to precisely date arsenopy-
rite with Re-Os is an invaluable tool for gold deposit
research. This was first reported by Arne et al. (2001)
who used the method to confirm an Ordovician age for
the historical Bendigo gold reefs in Australia. Arne et al.
(2001) found arsenopyrite to have Re contents above
10 ppb, with little common Os and high Re/Os ratios, thus
ideal for geochronology. Morelli et al. (2005) presented
a detailed study of arsenopyrite geochronology from the
Meguma terrane, Canada, where they identified two
193
distinct ages of mineralization at 380 Ma and 408 Ma.
Within the deposits studied, some arsenopyrite contained
no common Os, and the ages determined from those sam-
ples using the “mixed double-spike” approach were iden-
tical to other samples from which the age was determined
using the conventional isochron approach. The Re-Os
method in arsenopyrite has been used to establish the
age of giant gold ore deposits worldwide (Yu et al.,
2005; Morelli et al., 2007; Morelli et al., 2010) and to eval-
uate models for their origin. An important cross-
calibration to the U-Pb zircon geochronometer was
presented by Scherstén et al. (2012) who reported Re-Os
ages from arsenopyrite of 2,636  23 Ma and U-Pb ages
from metamorphic zircon of 2,633  6 Ma, both of which
formed during the same episode of metamorphism at that
time. This implies that the closure temperature for Re-Os
in arsenopyrite is, like molybdenite, high and well above
500
C (e.g., Morelli et al., 2010). Davies et al. (2010)
found that Re-Os ages recorded by arsenopyrite (1,768
 11 Ma) overlaps with U-Pb ages recorded by monazite
(1,781  4 Ma) in a metamorphosed Proterozoic gold
deposit, again implying a high closure temperature for
Re-Os in arsenopyrite (>600
C) that is demonstrably
higher than the Ar-Ar mica closure temperature
(350
C) which records cooling ages of ca. 1,720 Ma
from the deposit studied.
Re-Os geochronology of other crustal sulfide
minerals
Several copper sulfide minerals including chalcopyrite
(CuFeS2) and bornite (Cu5FeS4) have been investigated
for Re-Os geochronology. A detailed study by Selby
et al. (2009) found extremely high Re contents in both
minerals up to 5,000 ppb, with little common Os, in sedi-
ment-hosted Cu deposits from Alaska, and bornite, chal-
copyrite, and pyrite yielded a precise Re-Os age that
resolved possible origin models for the deposit. Extremely
high Re contents also characterized bornite and chalcopy-
rite from sandstone-hosted Cu deposits of Dzhezkazgan,
Kazakhstan (Box et al., 2013). In this study, bornite and
chalcopyrite yielded similar ages from one deposit but dis-
parate ages from another. Zhu and Sun (2013) used Re-Os
geochronology of chalcopyrite to provide the age of for-
mation of IOCG deposits from China and again found
Re levels above 1,000 ppb and low common Os contents.
Thus, the copper sulfide minerals hold significant promise
for Re-Os geochronology, and both Selby et al. (2009) and
Zhu and Sun (2013) report retention of Re-Os ages in
these minerals despite low-grade metamorphic overprints
for each deposit.
Hydrothermal pyrrhotite (Fe(1-x)Sx) was analyzed by
Morelli et al. (2010) from the Homestake Au deposit, from
which a precise 1,736 Ma Re-Os arsenopyrite age was
produced. In comparison to arsenopyrite, the pyrrhotite
Re-Os data were highly scatted and did not yield useful
age information. Brenan et al. (2000) measured Os diffu-
sion in pyrrhotite and found it to be much faster than
194 CRUSTAL SULFIDE MINERALS (Re-Os)
pyrite, yielding a likely closure temperature for pyrrhotite
of 300–400
C. The Re-Os pyrrhotite results of Morelli
et al. (2010) fit broadly with these experimental findings.
Similarly, sphalerite (ZnS) was found to yield scattered
Re-Os data from veins studied by Morelli et al. (2004)
from which pyrite yielded a precise Re-Os age.
Few minerals contain Re as major component of their
structure, but rheniite (ReS2) has been reported in modern
volcanic centers and remarkably has yielded isotopic Re-
Os ages as young as 79  11 a (Tessalina et al., 2008).
Summary
The common occurrence of sulfide minerals in metallic ore
deposits (e.g., those of Cu, Au, Zn, and Pb) makes Re-Os
dating of these minerals a key method in ore deposit
research, particularly given the difficulty of directly dating
ore minerals with other isotopic methods. It is relatively
a new field of geochronology, enabled by significant
advancements in analysis methods, that is becoming widely
applied. Many crustal sulfide minerals (molybdenite, pyrite,
arsenopyrite, chalcopyrite, bornite) have been found to pro-
vide reliable Re-Os ages, whereas some do not (pyrrhotite,
sphalerite).
Bibliography
Aleinikoff, J. N., Creaser, R. A., Lowers, H. A., Magee, C., and
Grauch, R. I., 2012. Multiple age components in individual
molybdenite grains. Chemical Geology, 300–301, 55–60.
Arne, D. C., Bierlein, F. P., Morgan, J. W., and Stein, H. J., 2001.
Re-Os dating of sulfides associated with gold mineralization in
central Victoria, Australia. Economic Geology, 96, 1455–1459.
Bingen, B., and Stein, H. J., 2003. Molybdenite Re-Os dating of
biotite dehydration melting in the Rogaland high-temperature
granulites, S. Norway. Earth and Planetary Science Letters,
208, 181–195.
Birck J.-L., Roy Barman M., and Capmas F., 1997. Re-Os isotopic
measurements at the femtomole level in natural samples.
Geostandards Newsletter, 20, 19–27.
Box, S. E., Syusyura, B., Seltmann, R., Creaser, R. A.,
Dolgopolova, A., and Zientek, M. L., 2013. Dzhezkazgan and
associated sandstone copper deposits of the Chu-Sarysu basin,
central Kazakhstan. Society of Economic Geologists Special
Publication, 16, 303–328.
Brenan, J. M., Cherniak, D. J., and Rose, L. A., 2000. Diffusion of
osmium in pyrrhotite and pyrite: implications for closure of the
Re-Os isotopic system. Earth and Planetary Science Letters,
180, 399–413.
Chiaradia, M., Vallance, J., Fontboté, L., Stein, H., Schaltegger, U.,
Coder, J., Richards, J., Villeneuve, M., and Gendall, I., 2009.
U-Pb, Re-Os, and 40Ar/39Ar geochronology of the Nambija
Au-skarn and Pangui porphyry Cu deposits, Ecuador: implica-
tions for the Jurassic metallogenic belt of the northern Andes.
Mineralium Deposita, 44, 371–387.
Cohen, A. S., and Waters, F. G., 1996. Separation of osmium from
geological materials by solvent extraction for analysis by ther-
mal ionization mass spectrometry. Analytica Chimica Acta,
332, 269–275.
Creaser, R. A., Papanastassiou, D. A., and Wasserburg, G. J., 1991.
Negative thermal ion mass spectrometry of osmium, rhenium
and iridium. Geochimica et Cosmochimica Acta, 55, 397–401.
Davies, T., Richards, J. P., Creaser, R. A., Heaman, L. M., Chacko,
T., Simonetti, A., Williamson, J., and McDonald, D. W., 2010.
Paleoproterozoic age relationships in the Three Bluffs archean
iron formation-hosted gold deposit, Committee Bay greenstone
belt, Nunavut, Canada. Exploration and Mining Geology, 19,
55–80.
Deckart, K., Clark, A. H., Cuadra, P., and Fanning, M., 2013.
Refinement of the time-space evolution of the giant Mio-
Pliocene Río Blanco-Los Bronces porphyry Cu–Mo cluster,
Central Chile: new U–Pb (SHRIMP II) and Re–Os geochronol-
ogy and 40Ar/39Ar thermochronology data. Mineralium
Deposita, 48, 57–79.
Hannah, J. L., Bekker, A., Stein, H. J., Markey, R. J., and Holland,
H. D., 2004. Primitive Os and 2316 Ma age for shale: implica-
tions for Paleoproterozoic glacial events and the rise of atmo-
spheric oxygen. Earth and Planetary Science Letters, 225,
43–52.
Hirt, B., Herr, W., and Hoffmeister, W., 1963. Age determination by
the rhenium-osmium method. International Atomic Energy
Agency, Radioactive Dating, 35–44.
Kosler, J., Simonetti, A., Sylvester, P., Cox, R., Tubrett, M. N., and
Wilton, D., 2003. Laser ablation ICP-MS measurements of
Re/Os in molybdenites and implications for Re-Os geochronol-
ogy. Canadian Mineralogist, 41, 307–320.
Luck, J. M., and Allègre, C. J., 1982. The study of molybdenites
through the 187Re-187Os chronometer. Earth and Planetary
Science Letters, 61, 291–296.
Mao, J., Du, A., Seltmann, R., and Yu, J., 2003. Re–Os ages for the
Shameika porphyry Mo deposit and the Lipovy Log rare metal peg-
matite, central Urals, Russia. Mineralium Deposita, 38, 251–257.
Markey, R., Hannah, J. L., Morgan, J. W., and Stein, H. J., 2003.
A double spike for osmium analysis of highly radiogenic sam-
ples. Chemical Geology, 200, 395–406.
Markey, R., Stein, H. J., Hannah, J. L., Selby, D., and Creaser, R. A.,
2007. Standardizing Re-Os geochronology: a new molybdenite
Reference Material (Henderson, USA) and the stoichiometry of
Os salts. Chemical Geology, 244, 74–87.
Mathur, R., Ruiz, J., and Tornos, F., 1999. Age and sources of the
ore at Tharsis and Rio Tinto, Iberian pyrite belt, from Re-Os iso-
topes. Mineralium Deposita, 34, 790–793.
McCandless, T. E., Ruiz, J., and Campbell, A. R., 1993. Rhenium
behavior in molybdenite in hypogene and near-surface environ-
ments: implications for Re-Os geochronometry. Geochimica et
Cosmochimica Acta, 57, 889–905.
Morelli, R. M., and Creaser, R. A., 2007. Testing the robustness of
low-level Re-Os sulfide geochronometers: an example from
metamorphosed VMS ores from the Trans Hudson Orogen, Can-
ada. Geological Society of America Abstracts with Programs,
39, 276.
Morelli, R. M., Creaser, R. A., Selby, D., Kelley, K. D., Leach,
D. L., and King, A. R., 2004. Re-Os sulfide geochronology from
the Red Dog sediment-hosted Zn-Pb Ag deposit, Alaska. Eco-
nomic Geology, 99, 1569–1576.
Morelli, R. M., Creaser, R. A., Selby, D., Kontak, D. J., and Horne,
R. J., 2005. Rhenium-Osmium arsenopyrite geochronology of
Meguma Group gold deposits, Meguma Terrane, Nova Scotia,
Canada: evidence for multiple gold mineralizing events. Eco-
nomic Geology, 100, 1229–1242.
Morelli, R. M., Creaser, R. A., Seltmann, R., Stuart, F. M., Selby,
D., and Graupner, T., 2007. Age and source constraints for the
giant Muruntau gold deposit, Uzbekistan, from coupled Re-Os-
He isotopes in arsenopyrite. Geology, 35, 795–798.
Morelli, R. M., Bell, C. C., Creaser, R. A., and Simonetti, A., 2010.
Constraints on the genesis of gold mineralization at the
Homestake Gold Deposit, Black Hills, South Dakota, from rhe-
nium-osmium sulfide geochronology. Mineralium Deposita,
45, 461–480.
Naldrett, S. N., and Libby, W. F., 1948. Natural radioactivity of rhe-
nium. Physical Review, 73, 487–493.
 CRUSTAL SULFIDE MINERALS (Re-Os)
Nude, P., Asigri, J., Yidana, S., Arhin, E., Foli, G., and Kutu, J.,
2012. Identifying pathfinder elements for gold in multi-element
soil geochemical data from the Wa-Lawra belt, northwest Ghana:
a multivariate statistical approach. International Journal of
Geosciences, 3, 62–70.
Ootes, L., Morelli, R. M., Lentz, D. R., Falck, H., Creaser, R. A.,
and Davis, W. J., 2011. The timing of Yellowknife gold mineral-
ization: a relationship with crustal anatexis? Economic Geology,
106, 713–720.
Piercey, S. J., Chaloux, E. C., Peloquin, A. S., Hamilton, M. A., and
Creaser, R. A., 2008. Synvolcanic and younger rocks from the
Blake River Group: implications for regional metallogenesis.
Economic Geology, 103, 1243–1268.
Reid, A., Smith, R. N., Baker, T., Jagodzinski, E. A., Selby, D.,
Gregory, C. J., and Skirrow, R. G., 2013. Re-Os dating of molyb-
denite within hematite breccias from the Vulcan Cu-Au prospect,
Olympic Cu-Au province, South Australia. Economic Geology,
108, 883–894.
Richards, J.P., and Noble, S.R., 1998. HYPERLINK “http://easweb.
eas.ualberta.ca/page/publications/?id=1910” Application of
radiogenic isotope systems to the timing and origin of
hydrothermal processes. In: Richards, J. P., and Larson, P.B.,
(eds.), Reviews in Economic Geology, 10, 195–233.
Rosera, J. M., Coleman, D. S., and Stein, H. J., 2013. Re-evaluating
genetic models for porphyry Mo mineralization at Questa, New
Mexico: implications for ore deposition following silicic ignimbrite
eruption. Geochemistry, Geophysics, Geosystems, 14, 787–805.
Scherstén, A., Creaser, R. A., van Gool, J. A. M., Næraa, T., Szilas,
K., and Østergaard, C., 2012. Re-Os and U-Pb constraints on
gold mineralisation events in the Meso- to Neoarchaean Storø
greenstone belt, Storø, southern West Greenland. Precambrian
Research, 203, 149–162.
Selby, D., and Creaser, R. A., 2004. Macroscale NTIMS and micro-
scale LA-MC-ICP-MS Re-Os isotopic analysis of molybdenite:
testing spatial restrictions for reliable Re-Os age determinations,
and implications for the decoupling of Re and Os within molyb-
denite. Geochimica et Cosmochimica Acta, 68, 3897–3908.
Selby, D., Creaser, R. A., Hart, C. J. R., Rombach, C. S., Thompson,
J. F. H., Smith, M. T., Bakke, A. A., and Goldfarb, R. J., 2002.
Absolute timing of sulfide and gold mineralization:
a comparison of Re-Os molybdenite and Ar-Ar mica methods
from the Tintina Gold Belt, Alaska. Geology, 30, 791–794.
Selby, D., Creaser, R. A., Stein, H. J., Markey, R. J., and Hannah,
J. L., 2007. Assessment of the 187Re decay constant by cross cal-
ibration of the 187Re-187Os molybdenite and U-Pb zircon chro-
nometers. Geochimica et Cosmochimica Acta, 71, 1999–2013.
Selby, D., Kelley, K. D., Hitzman, M. W., and Zieg, J., 2009. Re-Os
sulfide (bornite, chalcopyrite, and pyrite) systematics of the car-
bonate-hosted copper deposits at Ruby Creek, southern Brooks
Range, Alaska. Economic Geology, 104, 437–444.
195
Shirey, S. B., and Walker, R. J., 1995. Carius tube digestion for low
blank rhenium-osmium analysis. Analytical Chemistry, 67,
2136–2141.
Smoliar, M. I., Walker, R. J., and Morgan, J. W., 1996. Re-Os ages
of Group IIA, IIIA, IVA and IVB iron meteorites. Science, 271,
1099–1102.
Stein, H. J., 2006. Low-rhenium molybdenite by metamorphism in
northern Sweden: recognition, genesis, and global implications.
Lithos, 87, 300–327.
Stein H. J., Scherstén A., Hannah J. L., and Markey R. J., 2003.
Sub-grain scale decoupling of Re and 187Os and assessment of
laser ablation ICPMS spot dating in molybdenite. Geochimica
et Cosmochimica Acta, 67, 3673–3686.
Stein, H. J., Sundblad, K., Markey, R. J., Morgan, J. W., and
Motuza, G., 1998. Re-Os ages for Archean molybdenite
and pyrite, Kuittila-Kivisuo, Finland and Proterozoic molybde-
nite, Kabeliai, Lithuania: testing the chronometer in
a metamorphic and metasomatic setting. Mineralium Deposita,
33, 329–345.
Stein, H. J., Morgan, J. W., and Scherstén, A., 2000. Re-Os dating of
low-level highly radiogenic (LLHR) sulfides: the Harnäs gold
deposit, southwest Sweden, records continental-scale tectonic
events. Economic Geology, 95, 1657–1671.
Stein, H. J., Markey, R. J., Morgan, J. W., Hannah, J. L.,
and Scherstén, A., 2001. The remarkable Re-Os chronometer
in molybdenite: how and why it works. Terra Nova, 13,
479–486.
Suzuki, K., Feely, M., and O’Reilly, C., 2001. Disturbance of the
Re-Os chronometer of molybdenites from the late-Caledonian
Galway Granite, Ireland, by hydrothermal fluid circulation. Geo-
chemical Journal, 35, 29–35.
Tessalina, S., Yudovskaya, M. A., Chaplygin, I. V., Birck, J. L., and
Capmas, F., 2008. Sources of unique rhenium enrichment in
fumaroles and sulphides at Kudryavy volcano. Geochimica et
Cosmochimica Acta, 72, 889–909.
Völkening, J., Walczyk, T., and Heumann, K. G., 1991. Osmium
isotope ratio determinations by negative thermal ion mass spec-
trometry. International Journal of Mass Spectrometry and Ion
Processes, 105, 147–159.
Yu, G., Yang, G., Chen, J., Qu, W., Du, A., and He, W., 2005. Re-Os
dating of gold-bearing arsenopyrite of the Maoling gold deposit,
Liaoning Province, Northeast China and its geological signifi-
cance. Chinese Science Bulletin, 50, 1509–1514.
Zhu, Z., and Sun, Y., 2013. Direct Re-Os dating of chalcopyrite
from the Lala IOCG deposit in the Kangdian copper belt, China.
Economic Geology, 108, 871–882.
Zhu, M.-T., Zhang, L.-C., Wu, G., Jin, X.-D., Xiang, P., and
Li, W.-J., 2013. Zircon U–Pb and pyrite Re–Os age constraints
on pyrite mineralization in the Yinjiagou deposit, China. Inter-
national Geology Review, 55(13), 1616–1625.
 D
DENDROCHRONOLOGY, DWELLINGS
Stephen E. Nash
Denver Museum of Nature & Science, Denver, CO, USA
Synonyms
Tree-ring dating
Definition
Dendrochronology is literally “the study of tree time” and
is the science of assigning calendar-year dates to the
growth rings of trees. It is based on the fundamental prin-
ciple of cross dating (Figure 1), which is classically
defined as “the procedure of matching ring width
variations. . . among trees that have grown in nearby areas,
allowing the identification of the exact year in which each
ring formed” (Fritts, 1976, p. 534). Cross dating “utilizes
the presence and absence of [ring] synchrony from differ-
ent cores and trees to identify the growth rings that may be
misinterpreted” (Fritts and Swetnam, 1989, p. 121). Three
aspects warrant emphasis. First, tree-ring dating requires
ring pattern matching, not ring counting. Second, sample
sizes must be large, so that tree-growth variability in
a given region is well understood. Third, one begins by
studying living trees in a given area, cross dating their ring
series internally, and working back in time to successively
older specimens that are usually found as dead snags on
the landscape or as construction beams in ancient dwell-
ings (Nash, 1999).
Methods
The process of dating an archaeological tree-ring speci-
men requires, first and foremost, creation of a smooth
surface on which all the growth rings are clearly visible.
This smoothing is achieved using standard wood-working
equipment and progressively finer-grit sandpaper
J.W. Rink, J.W. Thompson (eds.), Encyclopedia of Scientific Dating Methods, DOI 10.1007/978-94-007-6304-3,
© Springer Science+Business Media Dordrecht 2015
(up to American National Standards Institute 400 grit;
20.6–23.6 mm or thousandths of a millimeter), such that
individual cell walls on the specimen can be seen. Once
a smooth, clear surface is prepared, the dendrochronolo-
gist can measure individual ring widths using specialized
hardware (like a Velmex™ measuring stage, precise to
0.001 mm, or similar equipment) interfaced with
a computer running specialized software (like
MeasureJ2X™ software or similar programs). Once all
rings have been accurately and precisely measured,
open-source COFECHA software is used to compare the
ring-width measurement series to previously developed
local or regional tree-ring chronologies in order to suggest
dating possibilities, which must then be checked against
the original wood specimen. If all goes well, calendar
dates may then be assigned to each growth ring on that
specimen. In places like the American Southwest, where
tree growth is often quite sensitive to precipitation, cross
dating can be performed by hand, as it was originally
developed by Andrew Ellicott Douglass in the early
1900s (Douglass, 1929; Nash, 1999).
In the Douglass method, the dendrochronologist pre-
pares a smooth surface as above but then creates
a summary graphic representation of the specimen’s
ring-width variability known as a skeleton plot, in which
long lines are written on graph paper to indicate relatively
narrow rings, a “B” indicates a relatively large ring, and in
which other unique attributes (e.g., wide or narrow early-
wood or latewood bands) are noted that may assist in dat-
ing (Figure 2; see Stokes and Smiley, 1968). The skeleton
plot is then visually compared to the previously developed
local or regional master tree-growth chronology. When the
ring-width pattern summarized on the skeleton plot
matches the local or regional growth pattern on the master
chronology and all missing, double, or locally absent rings
on the specimen have been identified, an accurate and pre-
cise date can be assigned to all rings on that specimen.
198 DENDROCHRONOLOGY, DWELLINGS

Dendrochronology, Dwellings, Figure 1 Cross dating, the basic principle of tree-ring dating, illustrated. The landscape across the
American Southwest has living trees, dead but exposed snags and remnant trees, and archaeological wood preserved in existing
structures and cliff dwellings. A dendrochronologist begins by analyzing long ring growth patterns in cores from living trees
(represented by “A” in the top portion of the figure). He/she then analyzes long ring series in successively older specimens from dead
snags and remnant trees (represented by “B” in the top portion of the figure), taking care to make sure that the ring series overlap by
at least 50 years. Finally, he/she analyzes ring series in archaeological wood (or charcoal) specimens (represented by “C” above), again
making sure that the ring series overlap by at least 50 years, to develop a long master chronology. Once a master tree-ring
chronology (represented by A + B + C above; in reality chronologies are based on dozens of specimens) has been developed, newly
collected archaeological and historic tree-ring specimens (represented by specimens A through J in the lower portion of the figure)
may be analyzed and dated against this master ring-width chronology (Courtesy of the Laboratory of Tree-Ring Research).

From a dendrochronological perspective, all cross-
dated and verified tree-ring dates are created equal; there
is no statistical uncertainty associated with cross-dated
tree-ring dates, which are the most precise chronometric
data available to archaeologists in the absence of historic
records (Jacoby, 2000). Unfortunately, and for a variety
of reasons including chaotic or complacent ring growth,
poor preservation, and other factors, many tree-ring spec-
imens simply cannot be dated, either quantitatively or by
the Douglass method. In this situation, dendrochronolo-
gists should not succumb when archaeologists ask for
a “likely date” for a specimen, nor should they use inde-
pendent archaeological data to propose a probable date
for an undated specimen (Baillie, 1995). Once a cross-
dated tree-ring date is determined, the interpretation of
that date within its archaeological context becomes the
archaeologist’s responsibility (Dean, 1978). From this
perspective, all tree-ring dates are not created equal, and
an extensive body of theory has been developed to guide
the proper interpretation of archaeological tree-ring dates
(Bannister, 1962; Dean, 1978; Hillam, 1987, 1998).
Tree-ring dating is most commonly used in the
North American Southwest, where ponderosa pine
(Pinus ponderosa), Douglas fir (Pseudotsuga menziesii),
pinyon pine (Pinus edulis), and juniper (Juniperus
scopulorum) are sensitive to variations in precipitation,
are well preserved in the archaeological record, and were
commonly used by indigenous populations. Dated sites
include cliff dwellings like those at Mesa Verde National
Park, standing architecture at Chaco Canyon National His-
torical Park, and pit houses across much of the region. In
the southeastern United States, researchers focus on dating
baldcypress (Taxodium distichum), juniper (Juniperus
sp.), white oak (Quercus alba), and other species. Dated
sites include standing historic structures and a very small
number of archaeological sites, which remain poorly dated
because wood specimens are not well preserved in wet
environments and because historic deforestation in the
region makes it difficult to build long chronologies going
back in time. In Western Europe and the Mediterranean,
dating efforts focus on oak (Quercus sp.), juniper
(Juniperus sp.), and other species, which are used to date
churches, dwellings, ship wrecks, and a wide variety of
other archaeological and historic features (Nash, 2002).
In theory, tree-ring dating is a relatively straightforward
process; in practice it can be astonishingly difficult.
 DENDROCHRONOLOGY, DWELLINGS 199

Dendrochronology, Dwellings, Figure 2 A skeleton plot for a wood specimen collected from a cliff dwelling in the American
Southwest. Especially long lines on the graph represent especially narrow rings on the specimen, which are caused by especially dry
years in the tree’s existence. The date for the last ring present on the specimen is A.D. 1273; the “vv G incomp.” suffix describes
noteworthy specimen attributes that modify the outside date. The “vv” indicates that it is a noncutting date and that there is no way
of estimating how far it is from an actual date-of-death of the tree. The “G” indicates, however, that galleries from wood-boring
beetles are present on the outside of the specimen; the “incomp.” indicates that the outer growth ring is incomplete, suggesting that
the tree was cut down during its growth season, usually late spring or early summer (Courtesy of the Laboratory of Tree-Ring
Research).

It requires rigorous sample collection and preparation,
methodical attention to detail, and deep knowledge of
tree-growth characteristics and wood attributes across vast
regions. For additional information and resources on the
method, theory, and results of tree-ring dating, see Henri
Grissino-Mayer’s excellent Science of Tree-Rings website
(http://web.utk.edu/~grissino/; accessed 14 August 2013).
Origins
Andrew Ellicott Douglass, the father of dendrochronol-
ogy, introduced tree-ring dating in “Talkative Tree-Rings
and the Tales They Tell,” published in the December,
1929, issue of National Geographic magazine (Douglass,
1929). He offered tree-ring dates for dozens of archaeo-
logical sites in the American Southwest. As a result,
archaeologists were forced to revise their interpretation
of southwestern prehistory, for they had grossly
overestimated the age of the ruins (Nash, 1999). Nearly
as important, however, was Douglass’ inference that
a “Great Drought” had occurred across the American
Southwest from A.D. 1276 until 1299 and that it had
affected pre-Columbian populations living in the area,
thus, also began the field of dendroclimatology.
Types of information
The analytic and interpretive utility of tree-ring dating
stems from the fact that it can inform archaeological
research in three broad arenas: chronometric, environmen-
tal, and behavioral (Dean, 1996a). The chronometric
aspect of tree-ring dating has the longest history and is
the most commonly known – tree rings can be used to date
objects, features, dwellings, and sites. By extension,
tree-ring dates can be used to date abstract archaeological
entities such as periods, stages, phases, and styles. The
environmental aspect of modern tree-ring dating today
enjoys the most worldwide application, as tree rings can
be used to reconstruct numerous environmental variables,
including temperature, precipitation, stream flow, drought
severity, fire frequency and intensity, insect infestation,
atmospheric circulation patterns, and others. The behav-
ioral aspect of tree-ring dating allows archaeologists to
make inferences regarding wood-use practices, trade,
and other economic variables (Dean, 1996b; Towner and
Clary, 2001; Towner et al., 2009; Towner and Creaseman,
2010; Windes, 2010).
Cutting dates, noncutting dates, and date clusters
The difference between cutting dates and noncutting dates
is important, as is the concept of date clustering. Cutting
dates are assigned to cross-dated wood or charcoal speci-
mens that possess evidence that the last ring present on
the specimen was the last ring grown by the tree before
it died. Noncutting dates are assigned to cross-dated spec-
imens if there is no evidence indicating that the last ring
present on the specimen is the last one grown before the
tree died. Note that cutting and noncutting dates are not
mutually exclusive – a noncutting date may actually
record the date of a tree’s death, but there happens to be
no discernable, definitive criteria (e.g., bark, beetle galler-
ies, patinated surfaces) present on the specimen that allow
200 DENDROCHRONOLOGY, DWELLINGS
the “cutting date” designation. Noncutting dates are there-
fore potentially biased estimates of beam procurement and
use events because they underestimate the actual date-of-
death of the tree (Ahlstrom, 1997).
The archaeological interpretation of noncutting dates is
complicated because there is no way of knowing, a priori,
how many rings are missing from the outside portion of
the specimen. The vagaries of tree growth, site formation
processes, specimen preservation, and wood-use behavior
ensure that only a fraction of tree-ring specimens submit-
ted for analysis date and, of those that do, most yield
noncutting dates because the outer portion of a beam is
more likely to be lost due to decay, insect infestation, or
anthropogenic removal. Despite recent efforts to alleviate
the interpretive difficulties associated with noncutting
dates, they remain the most recalcitrant of tree-ring dates
for archaeological purposes (Nash, 1997).
Another concept useful for the interpretation of tree-
ring dates is date clustering (Ahlstrom, 1985; Baillie,
1995). If a number of tree-ring dates from a given site or
provenience cluster in one to three calendar years, one
can infer that some construction event occurred and has
been dated by the tree rings. For the archaeologist inter-
ested in determining an accurate and precise date of con-
struction for a given room or site, the ideal date
distribution curve would consist of a vertical line of x
number of cutting dates in a given year. Because of activ-
ities such as beam stockpiling, manipulation, reuse, and
structure repair, actual date distribution curves are often
skewed left, with long tails, as noncutting dates dominate
the distribution (Ahlstrom, 1997, p. 343; Ahlstrom and
Smiley, 1998, p. 150).
Cutting date estimation
Dendrochronologists have attempted to estimate cutting
dates for specimens that yield only noncutting dates, the
logic being that if the number of missing rings can be esti-
mated with some reasonable degree of probability, one can
then simply add the estimated number of rings to the
existing outer ring date to obtain an estimated cutting date.
In general, there are three techniques for estimating cut-
ting dates (Hughes et al., 1981). The first is to determine
some statistical relationship between the number of sap-
wood rings, and therefore the number of rings that might
be lost, and various tree-growth parameters. The second
is merely to assume an immutable number of sapwood
rings. The third is to calculate the mean and standard devi-
ations of the number of sapwood rings from frequency dis-
tributions of samples assumed to be from the same
statistical populations.
Nash (1997) used heartwood and sapwood data from
living-tree cores to calculate regression equations that
predict the number of sapwood rings on a specimen based
on the number of heartwood rings. Using case studies
from the well-dated protohistoric, historic, and modern
dwellings at Walpi, Arizona, he concluded that because
of the statistical uncertainty associated with estimated
cutting dates, they are best considered in the aggregate,
in terms of the site’s entire date distribution curve.
Cutting-dating estimation is much more common in
Europe than in North America. Empirical research has
now been extensive enough to document a decreasing
west–east gradient in the number of sapwood rings found
in European oaks: In Ireland, mature oaks tend to have
32
9 sapwood rings; German oaks tend to have 19
8
sapwood rings, while Finnish oaks have 14
3 (Baillie
et al., 1985; see Pilcher, 1987).
Recent research
Nash (2002) and Towner (2002) published their perspec-
tives on the status of American archaeological tree-ring
dating at the turn of the millennium; Baillie (2002) and
Kuniholm (2002) did so for Europe and the Mediterra-
nean, respectively. Since then, the bulk of published
research on the dendrochronology of ancient dwellings
has occurred in the southern and eastern United States,
in Western Europe, and, to a lesser degree, in the greater
Mediterranean area. In a stunning recent development,
the full implications of which will take years to realize,
Morales et al. (2013) dated pre-Hispanic chullpas (burial
towers and storage chambers) in the Andean Altiplano,
marking the first time that reliable tree-ring dates have
been obtained for archaeological structures in South
America.
Grissino-Mayer (2009; see also references therein) and
colleagues led the effort to date historic structures in the
eastern and southern United States, publishing a special
issue of the journal Tree-Ring Research in the process.
The effort to use tree rings to date historic structures often
begins with a disagreement about the purported age of the
structures, which often stems from a lack of concordance
between documentary evidence and oral histories. Den-
drochronological specialists are then called in to provide
a complementary dataset. Much to the chagrin of local
heritage specialists, however, tree-ring dates can under-
mine previously held ownership histories for individual
structures. How, for example, can the John Sevier Cabin
have been occupied by Tennessee Governor Sevier, who
died in 1815, if tree-ring dates indicate that the cabin
was built sometime between the summer of 1835 and the
early spring of 1836 (Slayton et al., 2009, p. 35; for addi-
tional examples see also Grissino-Mayer and van de
Gevel, 2007; Towner, 2008)?
In other instances, tree-ring dates confirm, or at the least
do not contradict, previously held interpretations about
settlement origins and histories, including a slave cabin
at The Hermitage, President Andrew Jackson’s estate
(Lewis et al., 2009), an antebellum plantation house in
Arkansas (Therrell and Stahle, 2012), an antebellum
house in Georgia (Wight and Grissino-Mayer, 2004),
and structures in southeastern Nova Scotia (Robichaud
and Laroque, 2008), among many others.
In Western Europe (and also in the American South-
west), much archaeological tree-ring dating is driven by
 DENDROCHRONOLOGY, DWELLINGS
heritage resource management; tree-ring dates and inter-
pretive results are therefore often published in proprietary
reports. English Heritage has, however, made available
more than 100 tree-ring dating reports (search “tree-ring
dating” and “dendrochronological analysis” at http://
research.english-heritage.org.uk/; accessed 21 May
2013) on dwellings, churches, and other structures
throughout the United Kingdom. Hoffsummer (2009; see
also Lambert, 2009; Trenard, 2009; Haneca and Debonne,
2012) discussed the issues complicating tree-ring dating
of European structures in an edited volume on the typol-
ogy and development of roof frames in Belgium and
northern France from the eleventh to the nineteenth centu-
ries. Tegel et al. (2012) used tree rings to date early Neo-
lithic water wells, which they describe as the earliest
wood architecture but is more accurately described as the
earliest tree-ring-dated wood architecture. Further to the
east, Bernabei and Bontadi (2012) used tree-ring dating
to identify previously undocumented, or poorly
documented, construction activity at the Church of the
Nativity in Bethlehem. They dated previously undocu-
mented repair events between A.D. 1000 and the early
1200s and conclusively identified the source of wood,
far afield in the Republic of Venice, for a massive mid-fif-
teenth-century repair event. The Aegean Dendrochronol-
ogy Project, formerly based at Cornell University, has
moved to the Laboratory of Tree-Ring Research at the
University of Arizona in Tucson, where it continues
a long tradition of dating ancient dwellings in the greater
Mediterranean region.
Summary
Nearly a century after Douglass (1929) demonstrated that
tree rings could be used to date ancient dwellings, the
interpretation of archaeological tree-ring dates has
become more sophisticated with respect to understanding
prehistoric wood-use behavior and the discrepancies that
sometimes occur between historic documentation, oral
history, and tree-ring data. Arguably even more impres-
sive, however, is the fact that tree-ring dating is now rou-
tinely applied in areas like eastern and southern North
America, in which scholars once believed tree-ring dating
would not be successful. The expansion of tree-ring dating
to South America and the tropics has the potential to be
revolutionary, if sites with well-preserved wood can be
found in quantity.
Bibliography
Ahlstrom, R. V. N., 1985. The Interpretation of Archaeological
Tree-Ring Dates. PhD Dissertation, University of Arizona. Ann
Arbor, University Microfilms.
Ahlstrom, R. V. N., 1997. Sources of variation in the southwestern
archaeological tree-ring record. Kiva, 62, 321–348.
Ahlstrom, R. V. N., and Smiley, F., 1998. Archaeological Chronom-
etry: Radiocarbon and Tree-Ring Models and Applications from
Black Mesa, Arizona. Carbondale: Southern Illinois University
Center for Archaeological Investigations Occasional Paper 16.
Baillie, M. G. L., 1995. A Slice Through Time: Dendrochronology
and Precision Dating. London: B.T. Batsford.
201
Baillie, M. G. L., 2002. Future of dendrochronology with respect to
archaeology. Dendrochronologia, 20, 69–85.
Baillie, M. G. L., Hillam, J., Briffa, K., and Brown, D. M., 1985. Re-
dating the English art-historical tree-ring chronologies. Nature,
315, 317–319.
Bannister, B., 1962. The interpretation of tree-ring dates. American
Antiquity, 27, 508–514.
Bernabei, M., and Bontadi, J., 2012. Dendrochronological analysis
of the timber structure of the church of the nativity of Bethlehem.
Journal of Cultural Heritage. http://dx.doi.org/10.1016/j.
culher.2012.10.002; accessed 21 May 2013.
Dean, J. S., 1978. Independent dating in archaeological analysis. In
Schiffer, M. B. (ed.), Advances in Archaeological Method and
Theory 1. New York: Academic, pp. 223–255.
Dean, J. S., 1996a. Dendrochronology and the study of human
behavior. In Dean, J. S., Meko, D. M., and Swetnam, T. W.
(eds.), Tree-Rings, Environment, and Humanity: Proceedings
of the International Conference, Tucson, 1994. Tucson: Radio-
carbon, Department of Geosciences, University of Arizona,
pp. 461–469.
Dean, J. S., 1996b. Behavioral sources of error in archaeological
tree-ring dating: Navajo and Pueblo wood use. In Dean, J. S.,
Meko, D. M., and Swetnam, T. W. (eds.), Tree-Rings, Environ-
ment, and Humanity: Proceedings of the International Confer-
ence, Tucson, 1994. Tucson: Radiocarbon, Department of
Geosciences, University of Arizona, pp. 497–502.
Douglass, A. E., 1929. The secret of the southwest solved by talka-
tive tree-rings. National Geographic, 56, 736–770.
Fritts, H. C., 1976. Tree-Rings and Climate. New York: Academic.
Fritts, H. C., and Swetnam, T. W., 1989. Dendroecology: a tool for
evaluating variations in past and present forest environments.
Advances in Ecological Research, 19, 111–188.
Grissino-Mayer, H. D., 2009. Preface: an introduction to the
dendroarchaeology of the southeastern United States. Tree-Ring
Research, 65, 5–10.
Grissino-Mayer, H. D., and van de Gevel, S. L., 2007. Tell-tale trees:
historical dendroarchaeology of log structures at Rocky Mount,
Piney Flats, Tennessee. Historical Archaeology, 41, 32–49.
Haneca, K., and Debonne, V., 2012. Precise tree-ring dating of
building activities despite the absence of bark: a case-study of
Medieval Church Roofs in Damme, Belgium.
Dendrochronologia, 30, 23–34.
Hillam, J., 1987. Problems of dating and interpreting results from
archaeological timbers. In Ward, R. G. W. (ed.), Applications
of Tree-Ring Studies: Current Research in Dendrochronology
and Related Subjects. Oxford: British Archaeological Report.
BAR International Series, Vol. 333, pp. 141–155.
Hillam, J., 1998. Dendrochronology: Guidelines on Producing and
Interpreting Dendrochronological Dates. London: Ancient
Monuments Laboratory.
Hoffsummer, P., 2009. Critical aspects of dendrochronology. In
Hoffsummer, P. (ed.), Roof Frames from the 11th to the 19th
Century. Architectura Medii Aevi, III. Turnhout: Brepols,
pp. 99–104.
Hughes, M. K., Milson, S. J., and Leggett, P. A., 1981. Sapwood
estimates in the interpretation of tree-ring dates. Journal of
Archaeological Science, 8, 381–390.
Jacoby, G. C., 2000. Dendrochronology. In Noller, J. S., Sowers,
J. M., and Lettis, W. R. (eds.), Quaternary Geochronology:
Methods and Applications. Washington, DC: American Geo-
physical Union, pp. 11–20.
Kuniholm, P. I., 2002. Archaeological dendrochronology.
Dendrochronologia, 20, 63–68.
Lambert, G.-N., 2009. Dendrochronological reliability of timber
roof carpentry: reliability and precision. In Hoffsummer, P.
(ed.), Roof Frames from the 11th to the 19th Century.
Architectura Medii Aevi, III. Turnhout: Brepols, pp. 85–97.
202 DENDROCHRONOLOGY, ENTOMOLOGY
Lewis, D. B., Nelson, W. L., Grissino-Mayer, H. D., Cook, E. R.,
and Jones, R. D., 2009. Dendrochronological dating of eastern
red cedar (Juniperus Virginia L.) logs from Alfred’s Cabin,
The Hermitage, Home of President Andrew Jackson. Tree-Ring
Research, 65, 47–55.
Morales, M. S., Nielsen, A. E., and Villalba, R., 2013. First
dendroarchaeological dates of prehistoric contexts in South
America: Chullpas in the Central Andes. Journal of Archaeolog-
ical Science, 40, 2393–2401.
Nash, S. E., 1997. A cutting-date estimation technique for
ponderosa pine and Douglas fir wood specimens. American
Antiquity, 62, 260–272.
Nash, S. E., 1999. Time, Trees, and Prehistory: Archaeological
Tree-Ring Dating and the Development of North American
Archaeology 1914–1950. Salt Lake City: The University of Utah
Press.
Nash, S. E., 2002. Archaeological tree-ring dating at the millen-
nium. Journal of Archaeological Research, 10, 243–275.
Nash, S. E. (ed.), 2009. Readings in Chronometric Analysis: Selec-
tions from American Antiquity and Latin American Antiquity,
1935–2006. Washington, DC: The Society for American
Archaeology.
Pilcher, J. R., 1987. A 700-year old chronology from northern
France. In Ward, R. G. W. (ed.), Applications of Tree-Ring Stud-
ies: Current Research in Dendrochronology and Related Sub-
jects. Oxford: British Archaeological Report. BAR
International Series, Vol. 333, pp. 127–139.
Robichaud, A., and Laroque, C. P., 2008. Dendroarchaeology
in Southwestern Nova Scotia and the Construction of a
Regional red Spruce Chronology. Tree-Ring Research, 64,
17–25.
Slayton, J. D., Stevens, M. R., Grissino-Mayer, H. D., Cook, E. R.,
and Jones, R. D., 2009. The historical dendroarchaeology of two
log structures at the Marble Springs Historic site, Knox County,
Tennessee, USA. Tree-Ring Research, 65, 23–36.
Stokes, M. A., and Smiley, T. L., 1968. An Introduction to Tree-Ring
Dating. Chicago: University of Chicago Press.
Tegel, W., Elburg, R., Hakelburg, D., Stäubel, H., and Büntgen, U.,
2012. Early neolithic water wells reveals the world’s oldest
wood architecture. PLoS One, 7(12), http://www.plosone.org/
article/info:doi/10.1371/journal.pone.0051374; accessed 21
May 2013.
Therrell, M. D., and Stahle, D. W., 2012. Tree-ring dating of an
Arkansas Antebellum Plantation House. Tree-Ring Bulletin, 68,
59–67.
Towner, R. H., 2002. Archaeological dendrochronology in the
Southwestern United States. Evolutionary Anthropology, 11,
68–84.
Towner, R. H., 2008. The Navajo depopulation of Dinetah. Journal
of Anthropology Research, 64, 511–527.
Towner, R. H., and Clary, M. R., 2001. Historical
dendroarchaeology in central Colorado: lessons from the key-
stone area. Southwestern Lore, 67, 8–31.
Towner, R. H., and Creaseman, P. P., 2010. Historical
dendroarchaeology in the El Malpais area: lessons from the Sav-
age Homestead. Historical Archaeology, 44, 8–27.
Towner, R. H., Salzer, M. W., Parks, J. A., and Baralow, K. R., 2009.
Assessing the importance of past human behavior in dendroarch-
aeological research: examples from Range Creek Canyon, Utah,
USA. Tree-Ring Research, 65, 117–127.
Trenard, Y., 2009. The basics of dendrochronology. In Hoffsummer,
P. (ed.), Roof Frames from the 11th to the 19th Century.
Architectura Medii Aevi, III. Turnhout: Brepols, pp. 75–83.
Ward, R. G. W., 1987. Applications of Tree-Ring Studies: Current
Research in Dendrochronology and Related Subjects. Oxford:
British Archaeological Report. BAR International Series, Vol.
333, pp. 127–129.
Wight, G. D., and Grissino-Mayer, H. D., 2004. Dendrochronolog-
ical dating of an antebellum period house, Forsyth county, Geor-
gia, USA. Tree-Ring Research, 60, 91–99.
Windes, T. C., 2010. Dendrochronology and structural wood use at
Pueblo del Arroyo, Chaco Canyon, New Mexico. Journal of
Field Archaeology, 35, 78–98.
Cross-references
Dendrochronology, Progress
Historical Development of Dating Methods
Radiocarbon Dating
DENDROCHRONOLOGY, ENTOMOLOGY
Barry Cooke
Government of Canada, Northern Forestry Centre,
Canadian Forest Service, Natural Resources Canada,
Edmonton, AB, Canada
Definition
Dendroentomology, a subfield of dendrochronology, docu-
ments past occurrence of forest insect outbreaks, and pro-
vides an understanding of insect population dynamics,
including duration of outbreaks, interval between outbreaks
and spread. (Speer, 2010)
Dendroentomology is the study of spatiotemporal pat-
terns of herbivory caused by forest insects, based on infer-
ences from (1) the analysis of annually resolved tree-ring
data and (2) relatively robust species associations between
forest insects and the host-tree tissues that they tend to pre-
fer to consume. See also the definition in Text Box 1.
Introduction
As illustrated in Text Box 1, dendroentomology is related
to dendrochronology – which is the use of annually
resolved tree rings to date growth anomalies (narrow
rings, frost rings, discolored rings, etc.) within woody
samples (stems, roots, branches) and to cross-date such
anomalies across samples. Dendroentomology is
a subfield of dendroecology – which is the use of annually
resolved tree rings to accurately date (Stokes and Smiley,
1996), and to study, growth anomalies related to ecologi-
cal factors (including site and species characteristics) that
affect tree growth rates. It is distinct from, but not
unrelated to, dendroclimatology – which is the use of
annually resolved tree rings to date and to study growth
anomalies related to climatic events of the past, especially
as pertaining to positive and negative anomalies in tem-
perature, precipitation, radiation, etc., which influence tree
growth rates (Fritts, 1976).
For foliage-grazing insects, such as the defoliating Lep-
idoptera (caterpillars) or Hymenoptera (sawflies), it is
often possible to study negative growth anomalies in
extant trees (or coarse woody debris) that are not attribut-
able to climate and to attribute them to a foliage-grazing
 DENDROCHRONOLOGY, ENTOMOLOGY 203

insect that does not kill its host tree. Because tree growth Dendrochronology, Entomology, Table 1 Tree-defoliating
rings, when properly prepared, are annually resolved, it insects which have been a target of dendroentomological
is often possible to date such impacts with relative accu- reconstruction of past outbreaks
racy, to the year in which the impact occurred. Host-tree Defoliator insect Geographic
species species region Authority
Text Box 1: Definitions from Grissino-Mayer’s Douglas fir Western spruce USA Swetnam and
(2014) “principles of dendrochronology” budworm Lynch (1993)
Henri Grissino-Mayer (2014) provides a definition of Eastern Larch sawfly Canada Jardon et al. (1994)
larch
dendrochronology that is broadly inclusive: European Larch budmoth Switzerland Weber (1997)
Dendrochronology: The science that uses tree larch
rings dated to their exact year of formation to ana- Ponderosa Pandora moth USA Speer et al. (2001)
lyze temporal and spatial patterns of processes in pine
the physical and cultural sciences. Trembling Forest tent Canada Cooke and Roland
such that dendroentomology is considered as one of aspen caterpillar (2007)
White Eastern spruce Canada Boulanger
many subfields of dendrochronology: spruce budworm et al. (2012)
Subfields of Dendrochronology Douglas fir Western spruce Canada Alfaro et al. (2014)
• Dendroarchaeology: Dating of Archaeological budworm
dwellings.
• Dendroclimatology: Developing a record of past
climate.
• Dendrogeomorphology: Dating land movements
such as landslides in the past. between insect species and tree species is typically not
• Dendrohydrology: Creating a record of past water strictly one-to-one, but rather several-to-several, where
availability and flooding. there are several potential herbivore species that might
• Dendroglaciology: Dating past movements of grow to significant numbers on a given tree species, and
glaciers. (2) insect populations, when they rise, do not always grow
• Dendrovolcanology: Dating the past eruptions of to devastating levels, and light herbivory will be indistin-
volcanoes. guishable from background variations that are expected
• Dendrochemistry: Using tree rings as a monitor of to result from numerous ancillary agents, especially mois-
the chemical makeup of the soil. ture or temperature limitations occurring during the grow-
• Dendroecology: Recording ecological processes ing season.
such as tree-line movement, insect outbreaks, or Frequently, the study of neighboring tree species that are
movement of invasive tree species. typically not defoliated by the target insect species of inter-
• Dendropyrochronology: Dating the past occur- est will be studied as a “control.” If this control species is not
rence of forest fires. attacked by either the target insect or other insects and if,
• Dendroentomology: The use of tree rings to recon- further, its responses to climatic fluctuations mirror those
struct past population levels of insects. of the primary host-tree species that is attacked, then the
• Dendromastecology: The use of tree rings to recon- two species will tend to share a common dendroclimatic
struct fruiting events in trees signal that may be removed by subtraction, thus enhancing
the target signal of interest to the dendroentomologist.
Uncertainty in the reconstruction of entomological his-
Applications tory includes imprecision in dating accuracy and impreci-
Some well-known examples of applications in dendroen- sion in impact calibration. Dating accuracy within
tomology are provided in Table 1. a sample is limited by the fact that radial tree growth
For phloem-feeding, fungus-vectoring bark beetles responses often lag a year and more behind the herbivory
(Coleoptera: Scolytidae), which tend to kill their hosts in event. Dating accuracy across samples is limited by the
order to successfully reproduce, herbivory is inferred not fact that while climatic influences may be highly corre-
from negative anomalies in annual radial growth, but pos- lated across samples, entomological disturbances may or
itive anomalies among surviving trees that did not experi- may not be as highly correlated. For example,
ence lethal attack (Campbell et al., 2007). a spreading epidemic will result in spatial and temporal
The attributes of tree rings that are used to infer past lags in growth reductions occurring across samples. This
entomological disturbances may include radial ring width, hinders the otherwise straightforward procedure of accu-
the ratio of latewood to earlywood, wood density, chemi- rately cross-dating samples by “wiggle matching” the
cal profiles, and characteristic mis-colorations associated high-frequency variations in ring attributes over time
with altered wood chemistry and/or cell structure. (Stokes and Smiley, 1996).
The reconstruction of herbivory events though time is Estimating the proportion of sample trees affected dur-
necessarily an imprecise science, as (1) the association ing an insect outbreak is sensitive to assumptions about
204 DENDROCHRONOLOGY, FIRE REGIMES
the relationship between insect population density, defoli-
ation levels, and the impact of defoliation on the ring attri-
butes of interest, and in many cases these assumptions are
not verified. Outbreak reconstructions therefore typically
carry an unknown level of imprecision. A final, irreduc-
ible source of imprecision and bias is the fact that histori-
cal reconstructions tend to make use of living samples,
which are the survivors of previous disturbance events.
As the most intensely attacked trees are selectively
removed from the extant population available for sam-
pling, the result is an unstably biased reconstruction sig-
nal, with an unknown level of bias that grows over time
with each disturbance event that removes trees from the
sampling population (Cooke et al., 2003).
When insect populations are extremely high, the mor-
tality that occurs, through the creation of gaps in the forest
canopy, gives rise to the establishment of a new cohort of
seedlings and saplings. Indeed, the mortality that may
occur during heavy defoliation events leads to characteris-
tic patterns of cohort development which may be used to
supplement inferences about the impact of past outbreaks.
The reconstruction of past herbivory based on tree rings
is aided by, but distinct from, the study of insect
paleomacrofossils, which are fossilized remnants of insect
body parts (e.g., heavily scleritized mouthparts, head cap-
sules, or wings) or frass (i.e., excrement), which are often
taxonomically unique identifiers to the species level.
A thick deposit of macrofossils of a particular insect spe-
cies that is found in an ancient lake or bog may be
carbon-dated (within a limited range of accuracy), thus
providing corroborating evidence for inferences drawn
from tree-ring analysis (Bhiry and Filion, 1996; Simard
et al., 2006; Brunelle et al., 2008).
Bibliography
Alfaro, R. I., Berg, J., and Axelson, J., 2014. Periodicity of western
spruce budworm in Southern British Columbia, Canada. Forest
Ecology and Management, 315, 72–79.
Bhiry, N., and Filion, L., 1996. Mid-Holocene hemlock decline in
eastern North America linked with phytophagous insect activity.
Quaternary Research, 45, 312–320.
Boulanger, Y., Arseneault, D., Morin, H., Jardon, Y., Bertrand, P.,
and Dagneau, C., 2012. Dendrochronological reconstruction of
spruce budworm (Choristoneura fumiferana) outbreaks in
southern Quebec for the last 400 years. Canadian Journal of
Forest Research, 42, 1264–1276.
Brunelle, A., Rehfeldt, G. E., Bentz, B., and Munson, S., 2008.
Holocene records of Dendroctonus bark beetles in high elevation
pine forests of Idaho and Montana, USA. Forest Ecology and
Management, 255, 836–846.
Campbell, E. M., Alfaro, R. I., and Hawkes, B., 2007. Spatial distri-
bution of mountain pine beetle outbreaks in relation to climate
and stand characteristics: a dendroecological analysis. Journal
of Integrative Plant Biology, 49, 168–178.
Cooke, B. J., and Roland, J., 2007. Trembling aspen responses to
drought and defoliation by forest tent caterpillar and reconstruc-
tion of recent outbreaks in Ontario. Canadian Journal of Forest
Research, 37, 1586–1598.
Cooke, B. J., Miller, W. E., and Roland, J., 2003. Survivorship bias
in tree-ring reconstructions of forest tent caterpillar outbreaks
using trembling aspen. Tree Ring Research, 59, 29–36.
Fritts, H. C., 1976. Tree Rings and Climate. Caldwell: The Black-
burn Press.
Grissino-Mayer, H. 2014. Principles of Dendrochronology. http://
web.utk.edu/grissino/principles.htm#def
Jardon, Y., Filion, L., and Cloutier, C., 1994. Tree-ring evidence for
endemicity of the larch sawfly in North America. Canadian
Journal of Forest Research, 24, 742–747.
Simard, I., Morin, H., and Lavoie, C., 2006. A millennial-scale
reconstruction of spruce budworm abundance in Saguenay, Qué-
bec, Canada. The Holocene, 16, 31–37.
Speer, J. H., 2010. Fundamentals of Tree Ring Research. Tucson,
AZ: The University of Arizona Press, 333 p. https://www.
uapress.arizona.edu/Books/bid2216.htm
Speer, J. H., Swetnam, T. W., Wickman, B. E., and Youngblood, A.,
2001. Changes in Pandora Moth outbreak dynamics during the
past 622 years. Ecology, 82, 679–697. http://www.jstor.org/sta-
ble/2680188
Stokes, M. A., and Smiley, T. L., 1996. An Introduction to Tree-Ring
Dating. Arizona: University of Arizona.
Swetnam, T. W., and Lynch, A. M., 1993. Multicentury, Regional-
Scale Patterns of Western Spruce Budworm Outbreaks. Ecolog-
ical Monographs, 63, 399–424.
Weber, U., 1997. Dendroecological reconstruction and interpretation of
larch budmoth (Zeiraphera diniana) outbreaks in two central alpine
valleys of Switzerland from 1470–1990. Trees, 11, 277–290.
Cross-references
Dendrochronology, Progress
DENDROCHRONOLOGY, FIRE REGIMES
Peter M. Brown
Rocky Mountain Tree-Ring Research, Inc., Fort Collins,
CO, USA
Definition
Use of tree-ring data and methods to reconstruct past fire
timing, fire regimes, and fire effects on individuals, com-
munities, and ecosystems.
Introduction
Fire directly or indirectly affects woody plants in
many ways, some of which will leave evidence in age or
growth patterns in individual trees or community structure
that can be cross-dated using dendrochronological
methods. This evidence can be used to reconstruct past
fire dates, fire regimes, and fire effects on individuals,
communities, and ecosystems (“pyrodendroecology”).
Fire regimes are defined as the combination of fire fre-
quency, severity, size, seasonality, and relationships with
forcing factors such as climate or changes in human land
use.
Fire evidence in trees and community structure
A common type of fire evidence is from severe fire that
kills all or most trees in an area which opens up space
for new trees to establish. Even-aged forest structure is
often used as evidence of past lethal fire, also referred to
as “stand-replacing” fire. Such evidence provides
 DENDROCHRONOLOGY, FIRE REGIMES 205

Dendrochronology, Fire Regimes, Figure 1 (Left) A Pinus ponderosa stump with a fire-created “catface” facing to the right. The
catface was formed from repeated fire scars (visible as distinct ridges inside the catface). Bark and sapwood have eroded from the
outside of the stump, and only the heartwood is left. The stump was cut with a crosscut saw, placing the date of cutting likely in the
late nineteenth or early twentieth century before widespread use of motorized chainsaws. (Right) Cross-section sample cut with
a chainsaw from a fire-scarred P. ponderosa snag (standing dead tree). Dates of fire scars are marked on the section. The tree died in
1916 from a bark beetle attack as evidenced by blue stain in the small remnant of sapwood visible at the top of the photograph. Most
of the sapwood has decayed since the tree died, limiting the amount of time that a death date could be established (Photograph
credits: P. M. Brown).

a minimum date for the fire (i.e., the fire occurred some-
time before the innermost dates of the oldest trees), and,
by sampling multiple stands in an area, estimates of the
fire size can be made based on broader-scale patterns of
tree ages. Conversely, multiaged structure is considered
to be evidence of less severe fires or other patchy mortality
and recruitment events. Note that a dating limitation to age
structure studies is that rarely are total tree ages able to be
reconstructed. Tree ages are usually derived from incre-
ment cores or cross sections collected at a height above
and several years after the date of seedling germination.
Often cores or sections for age structure studies are col-
lected low on the tree bole (e.g., 10–30 cm above ground
level), and some studies have applied a correction factor to
estimate germination dates from sample-height pith dates.
However, regardless of these methods, generally age
structure data are not accurate to exact years of tree
germination.
Another commonly used line of evidence of past fire is
from fire scars (Figure 1). Fire scars are areas of cambial
mortality resulting mainly from lower intensity fires that
burn through surface fuels such as leaf litter, grasses and
other herbaceous plants, or shrubs. If the cambium is
killed all the way around the tree circumference, the tree
is girdled and dies. However, if only a portion of the cam-
bium is killed, a fire scar is formed and the tree continues
to live. The formation of a fire scar is a function of both
bark thickness (thicker bark helps to insulate the
cambium) and the intensity of the fire (a certain tempera-
ture is necessary to kill the cambium through the bark).
After scar formation, circumferential growth will con-
tinue from the margins of the dead cambial zone, often
completely regrowing over the scar unless the area is
burned again in a subsequent fire (Smith and Sutherland,
2001). After a tree has been scarred once, rapidly grow-
ing woundwood on the scar margins tends to have thinner
bark making it more susceptible to subsequent fires.
Because of this susceptibility, long sequences of multiple
fire scars are often found on trees in forested ecosystems
that experienced frequent, episodic, low intensity fires,
such as many dry Pinus ecosystems around the world
(Falk et al., 2011; Figure 1 right).
Fire scars can usually be dated to their exact year of
formation within a cross-dated ring series. Fire frequency
is derived from sequences of composited fire-scar
records found on multiple trees in a stand or study area.
Seasonality of burning also can be estimated from the
position of the scar within the annual ring and knowledge
of the general seasonal timing of cambial growth for
a species and location. The presence of a fire scar on
a tree implies that the fire that formed it was relatively
low intensity and nonlethal, at least at the scale of the
individual tree. By compiling dates of fire scars from
multiple trees or plots across a landscape, the relative
sizes and severities of past fires can be estimated with
a great deal of accuracy (Farris et al., 2010). Increasingly,
206 DENDROCHRONOLOGY, FIRE REGIMES
fire history studies use both tree age structure and fire-
scar data to develop more complete estimates of the fre-
quency and patterns of lethal and nonlethal burning
across a study area (Brown et al., 2008). Furthermore,
the accuracy of annual fire dates provided through den-
drochronological cross-dating permits comparison of fire
timing between sites, mountain ranges, and regions and
with independently derived tree-ring reconstructions of
climate variables (Swetnam and Brown, 2010). Impor-
tant insights into historical fire climatology have been
made owing to the accuracy of fire dates provided
through regional networks of fire-scar records (Falk
et al., 2011).
Other evidence of fires recorded by trees includes tree
death dates and tree growth responses. Trees killed by fire
can be dated by cross-dating with living trees. However,
sapwood typically does not last very long after a tree is
dead (even on standing dead trees, sapwood lasts at most
a few decades; Figure 1 left), which limits the length of
time that this line of evidence can be used to reconstruct
fire-caused mortality. Growth responses include abrupt
growth suppressions or releases. For example, a growth
suppression may occur after a fire as a result of severe
scorching of a tree canopy. A growth release may occur
as a result of neighboring trees being killed with
a subsequent loss of competitive pressure on the surviving
tree. Other responses include traumatic resin ducts or other
ring features that indicate injury to the tree (Brown and
Swetnam, 1994).
A note of caution is warranted here, in that many other
environmental factors may leave similar evidence in trees
as those resulting from the effects of fire. For example,
lighting and other injury events may partially kill the cam-
bium and leave a scar in the ring series. The approach used
by most studies has been to composite potential fire evi-
dence from multiple trees in a stand or study area and
depend on synchronous patterns as indications of fire
timing and effects. For example, lightning will result in
a scar on only a single tree in any given year, whereas fire
scars will be recorded on several trees in the same year due
to the spreading nature of fire within a stand of trees. Con-
versely, not all fires that burn at the base of a tree will leave
a fire scar or other fire-caused evidence, and again
compositing evidence from multiple trees is a usual step
to compile more complete spatiotemporal fire histories
(e.g., Farris et al., 2010). Dendrochronological cross-
dating is, of course, crucial in these studies to provide
absolute calendrical dates for fire evidence to assess syn-
chronous or asynchronous patterns within and between
trees, sites, or regions.
Application of dendrochronological fire regime
studies
Studies using pyrodendroecological data have generally
had three broad and often overlapping goals: (1) to docu-
ment and characterize past fire effects on ecosystem
processes, structure, and function across spatiotemporal
scales; (2) to understand how fire regimes have been
affected by forcing factors such as temporal or spatial
changes in climate, human land use, and vegetation (fuel)
structure; and (3) to develop historical information used in
forest and ecosystem management. Pyrodendroecological
data provide understanding of fire as an ecological process
over periods longer than those available from typical eco-
logical studies or even instrumental or observational
records. For example, a major focus of many
pyrodendroecological studies has been to examine lon-
ger-term (multi-decadal to multi-centennial) variation in
fire climatology (Swetnam and Brown, 2010; Falk et al.,
2011). Fire history and other dendroecological data also
provide “historical ranges of variability” in disturbance
regimes and forest structure for ecological restoration
efforts intended to restore ecological processes in
degraded, damaged, or mismanaged forested ecosystems
(Brown et al., 2008). This is especially the case in western
North America where fire suppression efforts over the past
century have led to profound changes in forest and fuel
structure and fire regimes in many “fire-adapted” dry coni-
fer ecosystems (e.g., Friederici, 2003).
Conclusion
Dendrochronological methods are useful for dating,
reconstructing, and interpreting past fires, fire regimes,
and fire effects on individuals, communities, and ecosys-
tems. Multiple lines of evidence of fire and its effects are
recorded in tree-ring series, and many of these can be
dated to the exact year and even season of formation.
Pyrodendroecological data are useful not only for dating
of fires but, more importantly, for understanding the role
of fire in ecosystem structure and function; for under-
standing the role of climate, human land use, and vegeta-
tion structure on variations in fire regimes; and for
current and future management of ecosystems around
the world.
Bibliography
Brown, P. M., and Swetnam, T. W., 1994. A cross-dated fire history
from a stand of coast redwood near Redwood National Park,
California. Canadian Journal of Forest Research, 24, 21–31.
Brown, P. M., Wienk, C. L., and Symstad, A. J., 2008. Fire and for-
est history at Mount Rushmore. Ecological Applications, 18,
1984–1999.
Falk, D. A., Heyerdahl, E. K., Brown, P. M., Farris, C., Fulé, P. Z.,
McKenzie, D., Swetnam, T. W., Taylor, A. H., and Van Horne,
M. L., 2011. Multicentury spatial dynamics of western North
American forest fires from fire-scar networks. Frontiers in Ecol-
ogy and the Environment, 8, 446–454.
Farris, C. A., Baisan, C. H., Falk, D. A., Yool, S. R., and Swetnam,
T. W., 2010. Spatial and temporal corroboration of a fire-scar-
based fire history in a frequently burned ponderosa pine forest.
Ecological Applications, 20, 1598–1614.
Friederici, P. (ed.), 2003. Ecological Restoration of Southwestern
Ponderosa Pine Forests. Washington, DC: Island Press.
 DENDROCHRONOLOGY, PROGRESS
Smith, K. T., and Sutherland, E. K., 2001. Terminology and biology
of fire scars in selected central hardwoods. Tree-Ring Research,
57, 141–147.
Swetnam, T. W., and Brown, P. M., 2010. Climatic inferences from
dendroecological reconstructions. In Hughes, M. K., Diaz, H. F.,
and Swetnam, T. W. (eds.), Dendroclimatology: Progress and
Prospects. Dordrecht: Springer. Developments in Paleoenvir-
onmental Research, Vol. 11, pp. 263–295.
Cross-references
Dendrochronology, Progress
DENDROCHRONOLOGY, PROGRESS
Laia Andreu-Hayles and Caroline Leland
Tree-Ring Laboratory, Lamont-Doherty Earth
Observatory of Columbia University, Palisades, NY, USA
Definition
Dendrochronology derives from the Greek words
“dendron” (tree) and “chronology” (study of time), and it
allows for the precise dating of annual tree rings based
on the analysis of tree-ring patterns among several trees.
Dendrochronology combined with other disciplines has
allowed for substantial progress in environmental studies
(see Cross-References).
Introduction
During recent years, dendrochronologists have faced the
challenge of getting more information from tree rings,
pushing its own limits to understand a wider range of cli-
matic/environmental factors, to obtain seasonal resolu-
tion, and to reconstruct larger spatial scales and longer
temporal time spans of climate variability. This is impor-
tant because instrumental records cover only a limited
time period and are insufficient for characterizing
decadal and centennial climatic features. Different strate-
gies are being explored to address these challenges. First,
different standardization approaches have been designed
to retain lower frequencies of climate variability,
a frequent pitfall of traditional standardization methods
used in tree-ring research. Low and medium frequencies
allow us to assess long-term variations in climate and to
determine the magnitude and range of climatic conditions
through time. This information provides a context to
investigate, for instance, whether the recent anthropo-
genic upward temperature trend is unusual compared to
other persistent warm periods during the past, such as
one occurring during Medieval Times. Second, several
technological advances have allowed for measuring
other tree-ring parameters in addition to the more classi-
cal methods (i.e., the width and the maximum density
of the rings). Stable isotope analyses that include geo-
chemistry and mass spectrometry, as well as wood
207
anatomy analyses using high-resolution imaging, are
becoming some of the most popular approaches in the
search for different kinds of environmental information.
These and other related topics are discussed in the sec-
tions below.
Standardization methods
Tree-ring width series often consist of different compo-
nents, such as age-size growth trends, environmental sig-
nals (e.g., climate variability), growth patterns due to
disturbance, and noise. The objective of the study defines
which component is considered signal or noise. There-
fore, retaining the signal of interest (or removing the
noise) in the tree-ring data determines the standardization
method to be chosen. Standardization is the process
through which we convert individual tree-ring measure-
ments into dimensionless indices to create a site chronol-
ogy. This step is used to first make all the trees equally
contribute to the mean of the site chronology and second
to reduce or remove variability in tree-ring series that is
not of interest for the overall research question. For
instance, the removal of any nonclimatic variability in
tree-ring measurements (e.g., age-size trend) is necessary
to retain as much climatic information as possible, and it
is one of the greatest challenges in dendroclimatology.
Proper removal of nonclimatic trends might suggest that
the resulting indices represent an ideally pure climatic
signal. The concept of standardization is fairly simple in
theory, but the practice has proven to be much more com-
plex. Some new methods designed to preserve as much
climatic information as possible are discussed in the
following.
The traditional methods of standardization have been
outlined and thoroughly discussed in several publications
(e.g., Fritts, 1976; Cook and Kairiukstis, 1990). Perhaps
the most basic of these methods is removal of the
age-size growth trend through fitting deterministic sloping
lines or negative exponential curves to the measurement
series (Fritts, 1976). However, several adaptations and
improvements to this method have been applied within
the last few decades, such as standardization methods for
closed-canopy forests or trees with potential disturbance
signals using stochastic (i.e., flexible) curves such as the
cubic smoothing spline (Cook, 1985; Cook and Peters,
1997). Additionally, attempts have been made to reduce
variance biases from the conventional “ratio” method of
calculating tree-ring indices through the “residuals”
method in conjunction with power transformation (Cook
and Peters, 1997; Helama et al., 2004). Please refer to an
earlier version of the encyclopedia for more specific infor-
mation on the classical standardization techniques
(Buckley, 2009). All standardization methods have
their own advantages and disadvantages, and the
method(s) used for a particular study must consider the
trees and the environment in which they are located. How-
ever, as a common feature, all these traditional methods
208 DENDROCHRONOLOGY, PROGRESS
are based on data-adaptive curve-fitting approaches that
limit the preservation of medium- and low-frequency var-
iability. The extent of longer-term variability removed
depends on the chosen detrending method, though the
longest timescales are restricted by the “segment length
curse” (Cook et al., 1995) to periods below the age of
the trees. Thus, further progress in standardization
and chronology building has been towards retaining as
much low- and medium-frequency climatic information
as possible while mitigating (1) the noise inherent in
tree-ring series and (2) biases and trend distortion from
statistical procedures. We will elaborate on some of the
more recent developments that have been especially
important in dendroclimatology: the regional curve stan-
dardization (RCS), signal-free standardization methods,
and the combined step and trend (CST) intervention
approach.
Regional curve standardization
Regional curve standardization (RCS) was popularized in
dendroclimatology primarily by Briffa et al. (1992), and
its primary goal was to recover more low-frequency vari-
ability in tree-ring chronologies for understanding changes
in climate from a long-term perspective. This method aligns
individual series by age and then compares each ring width
with a regional curve representing an “expectation of
growth” to produce indices. For more specifics on compu-
tation, please refer to Briffa and Melvin (2011). The RCS
method can recover more low-frequency information than
other curve-fitting methods of standardization because indi-
ces from the latter are forced to have an equal mean 1995),
whereas RCS allows for varying index series means
through time (Briffa et al., 1992; Briffa and Melvin,
2011). There are several important limitations and potential
biases to consider when using RCS to produce a site
chronology: the “trend-in-signal,” the “differing-
contemporaneous-growth-rate,” and the “modern-sample”
bias. Briffa and Melvin (2011) describe these biases in
detail. In short, the “trend-in-signal” problem occurs when
climate signals nearing or extending beyond the full length
of the chronology remain in the RCS mean, thereby under
or (over) estimating the final chronology at the beginning
and end. This bias can be reduced, in part, by having many
overlapping series. The “modern-sample” bias, which is
related to the “contemporaneous-growth-rate” problem,
should also be seriously considered before using RCS. This
bias occurs as a result of the relationship between growth
rate and tree longevity. For example, if the number of
fast-growing young trees is proportionally greater than
slow-growing old trees in a set of all living trees, there will
be a positive trend bias in the final chronology. Despite
these biases, this standardization method is very powerful
for retaining low-frequency variability beyond other
methods if the input tree-ring data is appropriate. The
RCS performs the best when there are (1) pith-offset esti-
mates in the case where pith dates are not achieved, (2) a
sufficient number of samples (Melvin, 2004), and (3) many
overlapping series of subfossil wood to account for issues
with the “trend-in-signal” and “modern-sample” biases.
Signal-free method
The signal-free (SF) method introduced by Melvin (2004)
and Melvin and Briffa (2008) seeks to reduce the effect of
trend distortion that has been recognized using other stan-
dardization methods (both traditional and RCS). Trend
distortion occurs when variance in tree-ring measurements
related to climate is removed during the process of remov-
ing age-size growth trends with data-adaptive curve fitting
(Melvin, 2004). This method produces a series of signal-
free measurements, which are void of the common
(climate) signal, and develops detrending curves based
on the age-size growth-related trend of the signal-free
measurements; these detrending curves are applied to the
original measurements, thereby producing new indices,
and the entire process is iterated as many times as needed
until convergence is attained and a final chronology (with
reduced trend distortion) is produced. In the effort to infer
past climate while preserving low-frequency variability,
recent studies have implemented and adopted the SF
method (Anchukaitis et al., 2013) or even combined SF
with RCS (Björklund et al., 2012). The signal-free method
in conjunction with RCS might retain more low-frequency
information while also helping to mitigate some of the
potential biases imparted on final chronologies from
RCS (Briffa and Melvin, 2011).
Combined step and trend (CST) intervention
approach (by Daniel Druckenbrod)
In closed-canopy forests, scientists have long observed
that understory trees may experience a rapid increase in
growth with removal of the canopy above them (e.g.,
Marshall, 1927). This growth increase is typically
referred to as a release from suppression (Oliver and
Larson, 1996) as the tree experiences greater light avail-
ability, which in turn provides greater growth. Light
availability constitutes an important nonclimatic forcing
of tree growth in these forests and is one example
of a larger class of disturbance events (e.g., ice storms,
insect defoliations, etc.) included within a common con-
ceptual model of tree growth (Graybill, 1982; Cook,
1987). Identifying and removing disturbance forcings
has been challenging for limiting noise in standardizing
tree rings. Time series analysis (Box and Jenkins, 1970)
provides a statistical framework for quantifying the
behavior of systems through time, particularly
autocorrelated series like tree rings, and has led
to advances in climate reconstruction (e.g., Cook and
Kairiukstis, 1990; Fritts, 1976). Downing and
McLaughlin (1990) demonstrated that time series analy-
sis with intervention detection could also be applied
to identify disturbance events (i.e., interventions) as step
outliers in time series when examining red spruce
decline; however, few studies have utilized this approach
(see Druckenbrod, 2005). Druckenbrod et al. (2013)
 DENDROCHRONOLOGY, PROGRESS
introduced a new combined step and trend (CST) outlier
approach that is capable of identifying disturbance events
in time series and removing their impact on the subse-
quent growth rates of trees. After power transforming,
removing the age-size growth trend, and estimating the
autoregressive order of a tree-ring time series, this
method iteratively searches for outliers in the means of
residuals with a running window specified by the user.
Disturbance events causing growth suppressions can also
be detected similar to Downing and McLaughlin (1990).
While this method is new, it also shows potential for use
in standardizing tree rings, possibly along with other
methods such as RCS and signal-free, since it shares
a common framework in time series analysis.
Pseudoproxy experiments
Pseudoproxy experiments (e.g., Mann and Rutherford,
2002) have been recently conducted by Anchukaitis
et al. (2013) to evaluate the appropriateness of a set of
detrending approaches to minimize the loss of
low-frequency variability and trend distortion in real data
when reconstructing the target climatic variable. The first
step is the generation of a set of artificial pseudoproxies
created by a combination of known pseudoclimate signals
and growth curves with random noise. The second step is
the standardization of these pseudoproxies with different
methods (e.g., data-adaptive curves, RCS, or signal-free).
Finally, comparisons between the known pseudoclimate
and the pseudoclimate estimated with the pseudoproxies
standardized in different ways can be achieved. This
method allows for detecting potential artifacts directly
derived from the chosen detrending method.
Tree-ring parameters
Stable isotopes
The application of stable isotopes in tree-ring research
continues to be an emerging field (Gagen et al., 2011),
but one with great promise for yielding a better under-
standing of climatic controls and physiological processes
driving growth patterns of forests. Significant relation-
ships between well-replicated, absolute-dated tree-ring
isotopic chronologies and environmental variables have
been well demonstrated in paleoclimatic research
(Robertson et al., 2008). Stable isotopic ratios can poten-
tially record climatic variability at all temporal frequen-
cies, and statistical age/geometry detrending is not
required (McCarroll and Loader, 2004; Gagen
et al., 2007; Young et al., 2011; but see Esper
et al., 2010). This is a big advantage considering the issues
associated with detrending methods (as described above)
for generating tree-ring chronologies based on the classi-
cal parameters (ring width and maximum latewood den-
sity). Indeed, stable isotopic signatures in tree rings can
complement the traditional metrics used in dendrocli-
matology; they can provide a broader spatial signal, pre-
serve low frequencies of climate variability, and reflect
different climatic variables (Gagen et al., 2011) for
209
different seasons. Stable isotopes in tree rings are strongly
related to climate and atmospheric dynamics and, thus, are
powerful paleoclimatic proxies that in some cases record
stronger climatic signals than the classical parameter of
ring width (Andreu et al., 2008; Gagen et al., 2011).
R denotes the ratio of the heavier isotope (e.g.,13C) to the
lighter (e.g.,12C), and it is expressed in delta (d) notation in
parts per thousand (%) with reference to a standard material
of known isotopic ratio:
 
Rsample
dsample ¼  1  1, 000
Rstandard
The chosen standard for d13C was a subfossil belemnite
from the Pee Dee formation of South Carolina, which was
exhausted and replaced by Vienna Pee Dee Belemnite or
VPDB (Coplen, 1995). The d18O signatures are in relation
to the Standard Mean Ocean Water (SMOW), now
replaced by Vienna-SMOW or V-SMOW, and less often,
the Standard Light Antarctic Precipitation (SLAP) can
also be used (IAEA, 1995). The hydrogen isotopes (the
lighter 1H, and the heavier 2H or D, also called deuterium)
are expressed as the D/H ratio (dD) and use the same ref-
erence materials as d18O ratios (IAEA, 1995). Carbon
and oxygen stable isotopes are the most frequently used
in tree-ring research, and occasionally hydrogen. Here
we primarily discuss carbon, but we direct readers to
McCarroll and Loader (2004) for a more extended discus-
sion of oxygen and hydrogen.
Stable carbon isotopes can supply distinct types of
physiological information including intrinsic water-use
efficiency or intercellular CO2 concentrations to study
physiological responses of trees to environmental changes
(e.g., Feng, 1998; Saurer et al., 2004; Waterhouse et al.,
2004; Peñuelas et al., 2010; Andreu-Hayles et al., 2011).
The physiological basis of d13C isotopic signatures in tree
rings involves several processes and assumptions. Carbon
isotope discrimination against 13C (D) is calculated
expressing the difference in isotopic composition between
air (d13Cair) and plant organic matter (d13Ctree) in %
(Eq. 1). For C3 plants, the relationship between D and leaf
gas exchange can be described by the Farquhar et al. (1982,
1989) model in Eq. (2).
d13 Cair  d13 Ctree
D¼ ð1Þ
1 þ d13 Ctree =1, 000
ci
D  d13 Cair  d13 Ctree ¼ a þ ðb  aÞ ð2Þ
ca
The terms ci and ca are the needle intercellular spaces
and ambient concentrations of CO2 (mmol mol1), respec-
tively. Isotopic fractionation is based on two main steps:
diffusion (4.4%) and carboxylation by the photosynthetic
rubisco enzyme (~27%; Farquhar and Richards, 1984).
First, air diffuses through stomata into the intercellular
air spaces towards the carboxylation sites. The CO2
with the lighter carbon isotope diffuses more easily.
Thus, internal air is depleted in 13C relative to ambient
210 DENDROCHRONOLOGY, PROGRESS
air. Second, CO2 is used by the rubisco and discrimination
against 13CO2 is taking place, since biological processes
tend to use 12C in preference to 13C.
Knowing D, ci/ca, and ci, the intrinsic water-use effi-
ciency (iWUE) can be estimated. D can be calculated
using extrapolations of d13Catm mainly derived from Ant-
arctic ice-core records (Francey et al., 1999) and direct
measurements, summarized by McCarroll and Loader
(2004). The ci can be determined using the data of the
ca from Robertson et al. (2001) until 1993 and from
Mauna Loa from 1994 to 2007 (composite series
published in McCarroll et al. (2009)). The iWUE is the
ratio of the fluxes of net photosynthesis A and conduc-
tance for water vapor gH2O (Ehleringer et al., 1993;
Feng, 1999):
A 1
iWUE ¼ ¼ ðca  ci Þ
g H2 O 1:6
While some fractionation processes such as those due
to CO2 diffusion into the stomata and carboxylation (see
above) are constant and consequently insensitive to cli-
mate (McCarroll and Pawellek, 2001), others are deter-
mined by changes in the environment. Stable carbon
isotopes (d13C) in tree rings are excellent recorders of var-
iations in climatic conditions as a result of a balance
between stomatal regulation due to changes in moisture
availability and photosynthetic rates and are a unique
and independent resource of long-term information on
the physiological status of natural forests. Stable oxygen
isotopes (d18O) reflect the source of precipitation and also
climatic information driven by variations in the vapor
pressure deficit that controls stomatal conductance. The
hydrogen isotopes dD are influenced by the same environ-
mental factors as the d18O ratios, and so they are the result
of similar fractionation process.
A downside of using the isotopes approach as com-
pared to traditional morphological measurements is that
stable isotope analyses are relatively more labor intensive
to produce and more expensive, typically limiting the total
number of measurements that can be reasonably
performed. In recent years, however, technical improve-
ments are beginning to successfully circumvent this limi-
tation (e.g., Wieloch et al., 2011). There are several
laboratory protocols for isotopic analysis, and here we
describe some of the most commonly used methods. The
isotopic composition can be measured directly from the
wood, but a-cellulose is commonly extracted in order to
avoid isotope variations due to other wood components
(e.g., lignin or resins). This extraction uses acetic acid,
sodium chlorite, and sodium hydroxide (Loader
et al., 1997; Loader et al., 2003; Rinne et al., 2005) or
nitric acid as described in the Brendel method (Brendel
et al., 2000; Anchukaitis et al., 2008). Several methods
can be applied for obtaining homogeneous cellulose
in relation to its isotopic signature such as mechanical
grinding of wood, cellulose freeze-milling with liquid
nitrogen or an ultrasonic treatment (Laumer et al., 2009).
Pooling can be done with wood from two radii per tree
in order to account for isotopic variability around the cir-
cumference of a given ring (Leavitt, 2010) and to ensure
enough material for the analyses. Although pooling with
wood from different trees is also a well-accepted method
(Leavitt, 1984, 2008) that saves time and resources, if
applied, there is a lack of information from individual trees
that precludes any chance of determining the range of
uncertainty of the resulting isotopic site chronology.
Finally, the cellulose samples need to be converted to
sample gases to be analyzed using isotope ratio
mass-spectrometers (IRMS). This gas conversion can be
done off-line or online as a part of a continuous flow
method that interfaces an elemental analyzer with IRMS.
The implementation of online methods has greatly
improved the efficiency of the entire process, substantially
increasing the possible total number of measurements,
a prior limitation to produce long-term records based on
isotopic ratios. Stable carbon isotopes (13C and 12C) are
measured by combusting small amount of a-cellulose
placed into tin capsules to CO2 in to the elemental analyzer
(EA) coupled with an IRMS, whereas samples are placed
in silver capsules for measuring stable oxygen isotopes
(18O and 16O) using conversion via pyrolysis to CO with
a high temperature conversion/elemental analyzer
(TC/EA). Some recent technical improvements also can
allow measuring both isotopes at the same time using
pyrolysis (Knoller et al., 2005). The measurements of
hydrogen isotopes (1H and 2 H/D), both online and
off-line, involve some complexities that have prevented
progress in technical advancements. As a consequence,
the amount of dD data available is still limited in compar-
ison to d13C or d18O data. Detailed description of mass
spectrometry methods, for each stable isotope ratio, can
be found in the McCarroll and Loader review (2004).
Wood anatomy and high-resolution imaging
The study of specific features within annual tree rings has
also proven to be valuable for understanding and
reconstructing links between tree growth and environmen-
tal conditions, particularly at intra-annual time scales. Mea-
suring wood characteristics, like the width of the earlywood
and latewood for each tree ring, as a source of specific and
intra-year climatic information, has recently advanced (e.g.,
Meko and Baisan, 2001; Griffin et al., 2011). On a finer
microscopic scale, much information can be attained about
past climatic and ecological conditions through careful
observation of specific anatomical features within individ-
ual rings. This investigation of tree-ring anatomy can be
quantified by measuring and counting structural wood com-
ponents. Quantitative wood anatomy (QWA) combines tra-
ditional wood anatomy with dendrochronological
techniques and provides temporal records of what often
used to be only qualitative information. Many wood anato-
mists did not traditionally study these two disciplines in
unison (Schweingruber, 2007; Fonti et al., 2010), but
QWA is currently rising in importance and popularity.
 DENDROCHRONOLOGY, PROGRESS
Recent improvements in image analysis capabilities have
allowed QWA to flourish as a field of research within the
past few decades (Fonti et al., 2010). Measurable parame-
ters for conifers have included tracheid cell area and diam-
eter, cell wall thickness, lumen diameter, and number of
tracheids, which can also be measured separately for early-
wood and latewood (e.g., Vysotskaya and Vaganov, 1989;
Filion and Cournoyer, 1995; Deslauriers and Morin, 2005;
Martin-Benito et al., 2013). For angiosperms, the number
or density of vessels, conductive area, and vessel area, for
example, can be measured (e.g., Woodcock, 1989; Fonti
and Garcia-Gonzalez, 2004; Campelo et al., 2010). Please
refer to Fonti et al. (2010) for a comprehensive listing of sev-
eral wood anatomical studies, including the parameters and
climatic signals identified in each study. Such features as
traumatic resin ducts and frost rings can also record changes
in environmental conditions and can be useful indicators of
extreme events in the past (e.g., LaMarche and Hirschboeck,
1984; Wimmer, 2002; Bollschweiler et al., 2008). Addition-
ally, some wood anatomical studies have identified and
described anatomical features in relation to other classical
dendrochronological parameters, particularly measurements
of density (Yasue et al., 2000; Wang et al., 2002) due to the
relation between the latter and characteristics of the cell
walls. QWA can complement traditional tree-ring parame-
ters as well as deepen and improve our understanding of
linkages between tree growth and the environment
over time.
Conclusions
Methods used in the field of dendrochronology have
advanced significantly over the past few decades. First,
standardization processes designed to obtain as much
low-frequency climatic information as possible have
advanced through different data treatment methods, such
as the RCS and SF methods. Additionally, there are
improvements on identifying and standardizing tree-ring
series with disturbance signals, particularly for closed-
canopy forests. Second, stable isotopes in tree rings
continue to emerge as an additional parameter for under-
standing environmental changes in the past climatic vari-
ability, and advances in cellulose extraction and sample
processing have recently made isotopic analysis more
accessible. In terms of wood anatomy, improvements in
image processing technology have resulted in more poten-
tial for investigating and quantifying anatomical features
in tree rings. All of the aforementioned methodological
advances in dendrochronology illustrate that there has
recently been significant progress in the field, and it indi-
cates that there will likely be more scientific progress in
the years to come.
Bibliography
Anchukaitis, K. J., Evans, M. N., Lange, T., Smith, D. R., Leavitt,
S. W., and Schrag, D. P., 2008. Consequences of a rapid cellulose
extraction technique for oxygen isotope and radiocarbon ana-
lyses. Analytical Chemistry, 80, 2035–2041.
211
Anchukaitis, K. J., D’Arrigo, R. D., and Andreu-Hayles, L., 2013.
Tree-ring-reconstructed summer temperatures from Northwest-
ern North America during the last nine centuries. Journal of Cli-
mate, 26, 3001–3012.
Andreu, L., Planells, O., Gutiérrez, E., Helle, G., and Schleser,
G. H., 2008. Climatic significance of tree-ring width and d13C
in a Spanish pine forest network. Tellus Series B Chemical and
Physical Meteorology, 60, 771–781.
Andreu-Hayles, L., Planells, O., Gutiérrez, E., Muntan, E., Helle,
G., Anchukaitis, K. J., and Schleser, G. H., 2011. Long tree-ring
chronologies reveal 20th century increases in water-use effi-
ciency but no enhancement of tree growth at five Iberian pine
forests. Global Change Biology, 17, 2095–2112.
Björklund, J.A., Gunnarson, B.E., Krusic, P.J., Grudd, H.,
Josefsson, T., Östlund, L., Linderholm, H.W., 2013. Advances
towards improved low-frequency tree-ring reconstructions,
using an updated Pinus sylvestris L. MXD network from the
Scandinavian Mountains. Theoretical and Applied Climatology,
113, 697–710.
Bollschweiler, M., Stoffel, M., Schneuwly, D. M., and Bourqui, K.,
2008. Traumatic resin ducts in Larix decidua stems impacted by
debris flows. Tree Physiology, 28, 255–263.
Box, G. E. P., and Jenkins, G. M., 1970. Time Series Analysis: Fore-
casting and Control. San Francisco: Holden-Day.
Brendel, O., Iannetta, P. P. M., and Stewart, D., 2000. A rapid and
simple method to isolate pure alpha-cellulose. Phytochemical
Analysis, 11, 7–10.
Briffa, K. R., and Melvin, T. M., 2011. A closer look at regional
curve standardization of tree-ring records: justification of the
need, a warning of some pitfalls, and suggested improvements
in its application. In Dendroclimatology. Dordrecht: Springer,
pp. 113–145.
Briffa, K. R., Jones, P. D., Bartholin, T. S., Eckstein, D.,
Schweingruber, F. H., Karlen, W., Zetterberg, P., and Eronen,
M., 1992. Fennoscandian summers from AD 500: temperature
changes on short and long timescales. Climate Dynamics, 7(3),
111–119.
Buckley, B. M., 2009. Dating, dendrochronology. In Encyclopedia
of Paleoclimatology and Ancient Environments. Dordrecht:
Springer, pp. 239–247.
Campelo, F., Nabais, C., Gutiérrez, E., Freitas, H., and García-
Gonzlez, I., 2010. Vessel features of Quercus ilex L. growing
under Mediterranean climate have a better climatic signal than
tree-ring width. Trees, 24, 463–470.
Cook, E. R., 1985. A Time-Series Analysis Approach to Tree-Ring
Standardization. PhD dissertation, Tucson, The University of
Arizona Press, 171 pp.
Cook, E. R., 1987. The decomposition of tree-ring series for envi-
ronmental studies. Tree-Ring Bulletin, 47, 37–59.
Cook, E. R., and Kairiukstis, L. A. (eds.), 1990. Methods of Dendro-
chronology: Applications in the Environmental Science. Boston:
Kluwer Academic Publishers.
Cook, E. R., and Peters, K., 1981. The smoothing spline: a new
approach to standardizing forest interior tree-ring width series
for dendroclimatic studies. Tree-Ring Bulletin, 41, 45–53.
Cook, E. R., and Peters, K., 1997. Calculating unbiased tree-ring
indices for the study of climatic and environmental change.
The Holocene, 7, 361–370.
Cook, E. R., Briffa, K. R., Meko, D. M., Graybill, D. A., and
Funkhouser, G., 1995. The segment length curse in long tree-
ring chronology development for paleoclimatic studies. The
Holocene, 5, 229–237.
Coplen, T. B., 1995. Discontinuance of SMOW and PDB. Nature,
375, 285.
Deslauriers, A., and Morin, H., 2005. Intra-annual tracheid produc-
tion in balsam fir stems and the effect of meteorological vari-
ables. Trees, 19, 402–408.
212 DENDROCHRONOLOGY, PROGRESS
Downing, D. J., and McLaughlin, S. B., 1990. Detecting shifts in
radial growth by use of intervention detection. In Cook, E. R.,
and Kairiukstis, L. A. (eds.), Methods of Dendrochronology. Dor-
drecht: International Institute for Applied Systems Analysis,
pp. 258–270.
Druckenbrod, D. L., 2005. Dendroecological reconstructions of for-
est disturbance history using time-series analysis with interven-
tion detection. Canadian Journal of Forest Research, 35,
868–876.
Druckenbrod, D. L., Pederson, N., Rentch, J., and Cook, E. R.,
2013. A comparison of times series approaches for
dendroecological reconstructions of past canopy disturbance
events. Forest Ecology and Management, 302, 23–33.
Ehleringer, J. R., Hall, A. E., and Farquhar, G. D., 1993. Stable Iso-
topes and Plant Carbon-Water Relations. San Diego:
Academic. 555 pp.
Esper, J., Frank, D. C., Battipaglia, G., Büntgen, U., Holert, C.,
Treydte, K., Siegwolf, R., and Saurer, M., 2010.
Low-frequency noise in d13C and d18O tree ring data: a case
study of Pinus uncinata in the Spanish Pyrenees. Global Biogeo-
chemical Cycles, 24, GB4018.
Farquhar, G. D., and Richards, R. A., 1984. Isotopic composition of
plant carbon correlates with water-use efficiency of wheat geno-
types. Australian Journal of Plant Physiology, 11, 539–552.
Farquhar, G. D., O’Leary, M. H., and Berry, J. A., 1982. On the rela-
tionship between carbon isotope discrimination and the
intercellular carbon dioxide concentration in leaves. Australian
Journal of Plant Physiology, 9, 121–137.
Farquhar, G. D., Ehleringer, F. R., and Hubick, K. T., 1989. Carbon
isotope discrimination and photosynthesis. Annual Review Plant
Physiology and Plant Molecular Biology, 40, 503–537.
Feng, X., 1998. Long-term ci/ca response of trees in western North
America to atmospheric CO2 concentration derived from carbon
isotope chronologies. Oecologia, 117, 19–25.
Feng, X., 1999. Trends in intrinsic water-use efficiency of natural
trees for the past 100–200 years: a response to atmospheric
CO2 concentration. Geochimica et Cosmochimica Acta, 63,
1891–1903.
Filion, L., and Cournoyer, L., 1995. Variation in wood structure of
eastern larch defoliated by the larch sawfly in subarctic Quebec,
Canada. Canadian Journal of Forest Research, 25, 1263–1268.
Fonti, P., and García-Gonzlez, I., 2004. Suitability of chestnut early-
wood vessel chronologies for ecological studies. New
Phytologist, 163, 77–86.
Fonti, P., von Arx, G., García-Gonzlez, I., Eilmann, B., Sass-
Klaassen, U., Gärtner, H., and Eckstein, D., 2010. Studying
global change through investigation of the plastic responses of
xylem anatomy in tree rings. New Phytologist, 185, 42–53.
Francey, R. J., Allison, C. E., Etheridge, D. M., Trudinger, C. M.,
Enting, I. G., Leuenberger, M., Langenfels, R. L., Michel, E.,
and Steele, L. P., 1999. A 1000-year high precision record of
d13C in atmospheric CO2. Tellus Series B Chemical and Physical
Meteorology, 51, 170–193.
Fritts, H. C., 1976. Tree Rings and Climate. New York: Academic.
Gagen, M., McCarroll, D., Loader, N. J., Robertson, I., Jalkanen, R.,
and Anchukaitis, K. J., 2007. Exorcising the ‘egment length
curse’: summer temperature reconstruction since AD 1640 using
non-detrended stable carbon isotope ratios from pine trees in
northern Finland. The Holocene, 17, 435–446.
Gagen, M., McCarroll, D., Loader, N. J., and Robertson, I., 2011.
Stable Isotopes in dendroclimatology: moving beyond ‘Poten-
tial’. In Hughes, M. K., Swetnam, T. W., and Diaz, H. F. (eds.),
Dendroclimatology. Dordrecht: Springer, pp. 147–172.
Graybill, D. A., 1982. Chronology development and analysis. In
Hughes, M. K., Kelly, P. M., Pilcher, J. R., and LaMarche,
V. C., Jr. (eds.), Climate from Tree Rings. Cambridge: Cam-
bridge University Press, pp. 21–28.
Griffin, D., Meko, D. M., Touchan, R., Leavitt, S. W., and
Woodhouse, C. A., 2011. Latewood chronology development
for summer-moisture reconstruction in the US Southwest. Tree-
Ring Research, 67(2), 87–101.
Helama, S., Lindholm, M., Timonen, M., and Eronen, M., 2004.
Detection of climate signal in dendrochronological data analysis:
a comparison of tree-ring standardization methods. Theoretical
and Applied Climatology, 79, 239–254.
International Atomic Energy Agency (IAEA), 1995. TECDOC-825.
Reference and intercomparison materials for stable isotopes of
light elements . In Proceedings of a Consultants Meeting,
Vienna, December 1–3, 1993.
Knoller, K., Boettger, T., Weise, S. M., and Gehre, M., 2005. Car-
bon isotope analyses of cellulose using two different on-line
techniques (elemental analysis and high-temperature pyrolysis) –
a comparison. Rapid Communications in Mass Spectrometry, 19,
343–348.
LaMarche, V. C., and Hirschboeck, K. K., 1984. Frost rings in trees
as records of major volcanic eruptions. Nature, 307(5947),
121–126.
Laumer, W., Andreu, L., Helle, G., Schleser, G. H., Wieloch, T.,
and Wissel, H., 2009. A novel approach for the homogeniza-
tion of cellulose to use micro-amounts for stable isotope
analyses. Rapid Communications in Mass Spectrometry, 23,
1934–1940.
Leavitt, S. W., 2008. Tree-ring isotopic pooling without regard to
mass: no difference from averaging d13C values of each tree.
Chemical Geology, 252, 52–55.
Leavitt, S. W., 2010. Tree-ring C–H–O isotope variability and sam-
pling. Science of The Total Environment, 408, 5244–5253.
Leavitt, S. W., and Long, A., 1984. Sampling strategy for stable
carbon isotope analyses of tree rings in pine. Nature, 311,
145–147.
Loader, N. J., Robertson, I., Barker, A. C., Switsur, V. R., and
Waterhouse, J. S., 1997. An improved technique for the batch
processing of small wholewood samples to a-cellulose. Chemi-
cal Geology, 136, 313–317.
Loader, N. J., Robertson, I., and McCarroll, D., 2003. Comparison
of stable carbon isotope ratios in the whole wood, cellulose
and lignin of oak tree-rings. Palaeogeography, Palaeocli-
matology, Palaeoecology, 196, 395–407.
Mann, M. E., and Rutherford, S., 2002. Climate reconstruction
using ‘pseudoproxies’. Geophysical Research Letters, 29,
1501, doi:10.1029/2001GL014554.
Marshall, R., 1927. The Growth of Hemlock Before and After
Release from Suppression. Cambridge, MA: Harvard University
Press. Harvard Forest Bulletin, Vol. 11.
Martin-Benito, D., Beeckman, H., and Cañellas, I., 2013. Influence
of drought on tree rings and tracheid features of Pinus nigra and
Pinus sylvestris in a mesic Mediterranean forest. European Jour-
nal of Forest Research, 132, 33–45.
McCarroll, D., and Loader, N. J., 2004. Stable isotopes in tree rings.
Quaternary Research Reviews, 23, 771–801.
McCarroll, D., and Pawellek, F., 2001. Stable carbon isotope ratios
of Pinus sylvestris from northern Finland and the potential for
extracting a climate signal from long Fennoscandian chronolo-
gies. The Holocene, 11, 517–526.
McCarroll, D., Gagen, M. H., Loader, N. J., Robertson, I.,
Anchukaitis, K. J., Los, S., Young, G. H. F., Jalkanen, R.,
Kirchhefer, A., and Waterhouse, J. S., 2009. Correction of tree
ring stable carbon isotope chronologies for changes in the carbon
dioxide content of the atmosphere. Geochimica et
Cosmochimica Acta, 73, 1539–1547.
Meko, D. M., and Baisan, C. H., 2001. Pilot study of latewood-
width of conifers as an indicator of variability of summer rainfall
in the North American monsoon region. International Journal of
Climatology, 21(6), 697–708.
 DENDROCHRONOLOGY, SURFICIAL PROCESSES
Melvin, T. M., 2004. Historical Growth Rates and Changing Cli-
matic Sensitivity of Boreal Conifers. Dissertation, University of
East Anglia.
Melvin, T. M., and Briffa, K. R., 2008. A “signal-free” approach to
dendroclimatic standardisation. Dendrochronologia, 26(2),
71–86.
Oliver, C. D., and Larson, B. C., 1996. Forest Stand Dynamics. New
York: Wiley.
Peñuelas, J., Canadell, J. G., and Ogaya, R., 2010. Increased water-
use efficiency during the 20th century did not translate into
enhanced tree growth. Global Ecology and Biogeography, 20,
597–608.
Rinne, K. T., Boettger, T., Loader, N. J., Robertson, I., Switsur,
V. R., and Waterhouse, J. S., 2005. On the purification of
a-cellulose from resinous wood for stable isotope (H, C and O)
analysis. Chemical Geology, 222, 75–82.
Robertson, A., Overpeck, J., Rind, D., Mosley-Thompson, E.,
Zielinski, G., Lean, J., Koch, D., Penner, J., Tegen, I., and Healy,
R., 2001. Hypothesized climate forcing time series for the last
500 years. Journal of Geophysical Research, 106, 14783–14803.
Robertson, I., Leavitt, S. W., Loader, N. J., and Buhay, W., 2008.
Progress in isotope dendroclimatology. Chemical Geology,
252, EX1–EX4.
Saurer, M., Siegwolf, R. T. W., and Schweingruber, F. H., 2004.
Carbon isotope discrimination indicates improving water-use
efficiency of trees in northern Eurasia over the last 100 years.
Global Change Biology, 10, 2109–2120.
Schweingruber, F. H., 2007. Wood Structure and Environment. Ber-
lin/Heidelberg/New York: Springer.
Vysotskaya, L. G., and Vaganov, E. A., 1989. Components of the
variability of radial cell size in tree rings of conifers. IAWA Bul-
letin, 10(4), 417–426.
Wang, L., Payette, S., and Bégin, Y., 2002. Relationships between
anatomical and densitometric characteristics of black spruce
and summer temperature at tree line in northern
Quebec. Canadian Journal of Forest Research, 32, 477–486.
Waterhouse, J. S., Switsur, V. R., Barker, A. C., Carter, A. H. C.,
Hemming, D. L., Loader, N. J., and Robertson, I., 2004. North-
ern European trees show a progressively diminishing response
to increasing atmospheric carbon dioxide concentrations. Qua-
ternary Science Reviews, 23, 803–810.
Wieloch, T., Helle, G., Heinrich, I., Voigt, M., and Schyma, P., 2011.
A novel device for batch-wise isolation of a-cellulose from
small-amount wholewood samples. Dendrochronologia, 29,
115–117.
Wimmer, R., 2002. Wood anatomical features in tree-rings as indi-
cators of environmental change. Dendrochronologia, 20, 21–36.
Woodcock, D. W., 1989. Climate sensitivity of wood-anatomical
features in a ring-porous oak (Quercus macrocarpa). Canadian
Journal of Forest Research, 19, 639–644.
Yasue, K., Funada, R., Kobayashi, O., and Ohtani, J., 2000. The
effects of tracheid dimensions on variations in maximum density
of Picea glehnii and relationships to climatic factors. Trees, 14,
223–229.
Young, G. H. F., Demmler, J. C., Gunnarson, B. E., Kirchhefer,
A. J., Loader, N. J., and McCarroll, D., 2011. Age trends in tree
ring growth and isotopic archives: a case study of Pinus
sylvestris L. from northwestern Norway. Global Biogeochemical
Cycles, 25, GB2020.
Cross-references
Dendrochronology, Dwellings
Dendrochronology, Entomology
Dendrochronology, Fire Regimes
Dendrochronology, Surficial Processes
Dendrochronology, Volcanic Eruptions
213
DENDROCHRONOLOGY, SURFICIAL PROCESSES
Markus Stoffel, Juan A. Ballesteros-Cnovas and
Christophe Corona
Institute for Geological Sciences, University of Bern,
Bern, Switzerland
Synonyms
Earth-surface process; Dendrogeomorphology;
Geomorphology
Definitions
Dendrochronology (tree-ring dating): scientific method of
dating based on the analysis of patterns of tree rings, also
known as growth rings. Dendrochronology can date the
time at which gymnosperms and some angiosperms
(woody dicotyledonous) were formed tree rings to the
exact calendar year.
Surficial processes: processes on the earth surface that
shape or have shaped landforms.
Dendrogeomorphology: scientific method to date earth-
surface processes in terms of frequency, magnitude, and
spatial pattern using the growth-ring series of affected
trees.
Introduction
The significant contribution of tree rings to document
earth-surface processes lies in their capacity to both pre-
serve evidence of past geomorphic activity and provide
critical information on their dating with annual or sub-
annual resolution. In many climates, the tree-ring record
may represent the most valuable and precise natural
archive for the reconstruction and understanding of past
and ongoing earth-surface processes during the last sev-
eral hundred years. As many geomorphic processes are
significant natural hazards, understanding their distribu-
tion, timing, and controls provides valuable information
that can assist in the development of mitigation and
defense against these hazards and their effects on society.
This contribution provides an overview of tree-ring-
based reconstructions of surficial processes resulting from
different geomorphic agents. It outlines the impact of
earth-surface processes on tree morphology and wood
anatomy and clarifies the approaches used for tree-ring
reconstruction of past activity of earth-surface processes.
This entry also illustrates the breadth and diverse applica-
tions of contemporary dendrogeomorphology and under-
lines the growing potential to expand these studies,
possibly leading to the establishment of a range of tech-
niques and approaches that may become standard practice
in the analysis of specific hazards.
What surficial processes can be analyzed
with tree rings?
Dendrogeomorphology has been used in the analysis and
reconstruction of a wide range of surficial processes.
214 DENDROCHRONOLOGY, SURFICIAL PROCESSES
Some of the earliest applications of dendrochronological
research to earth-surface processes utilized roots rather
than stem samples to study erosion processes. In the
1960s, LaMarche (1968) used tree roots exposed by sur-
face lowering as an indicator of ground surface at the time
of germination of 5,000-year-old Pinus aristata Engelm.
in the White Mountains, California. Corona et al. (2011)
focus on areal denudation in marly badlands in the French
Alps and illustrate limitations and possible caveats of past
approaches using growth rings of roots to reconstruct ero-
sion rates. Stoffel et al. (2012) reconstructed channel wall
retreat related to flash floods in the Patagonian Andes of
Argentina, and Fantucci (2007) utilized roots to analyze
shore erosion in Central Italy.
After the pioneering studies of Hupp (1984) conducted
on Mount Shasta (California), tree-ring records were used
by Strunk (1991) to reconstruct debris-flow activity in the
Italian Dolomites. More recently, the frequency and spa-
tial patterns (i.e., lateral spread and extent of flows) have
been analyzed in the Swiss Alps (Stoffel et al., 2008;
Bollschweiler and Stoffel, 2010), including
dendrogeomorphic work on frequency-magnitude rela-
tions of debris flows (Stoffel, 2010) and precipitation
thresholds (Schneuwly-Bollschweiler and Stoffel, 2012).
Dendrogeomorphic analyses of landslides were real-
ized primarily in Canada (Bégin and Filion, 1985) and
the French Alps (Braam et al., 1987; Lopez Saez et al.,
2013) to document spatial patterns of activity,
reactivations, and rainfall patterns triggering landslides.
In the first study in the subtropics, Paolini et al. (2005)
used suppression-release patterns and inception of Alnus
acuminate Kunth. trees to date a number of landslides in
Argentina.
The occurrence of snow avalanches has quite fre-
quently been analyzed with tree rings over the last four
decades, beginning with Schaerer’s (1972) publication
on vegetation in avalanche terrain in British Columbia
(Canada). For a long time, tree-ring-based analysis of
snow avalanche activity was almost exclusively used in
North America (Butler, 1979; Germain et al., 2009). Work
in the European Alps started in the early 2000s with
a focus on runout distances and triggering weather condi-
tions (Stoffel et al., 2006a; Corona et al., 2012).
The analysis of growth rings has also been used in
paleohydrology. The first studies linking fluvial processes
and botanical evidence go back to the 1960s (Sigafoos,
1964). Later on, a broad array of flood signatures have
been defined to date paleofloods (St. George and Nielsen,
2000). More recently, scarred trees have been used to esti-
mate flood discharge (Ballesteros et al., 2011) and to ana-
lyze flood frequency distributions (Ballesteros et al.,
2013).
Surprisingly, and despite the potential of
dendrogeomorphic methods, rockfall activity has only
rarely been studied with tree rings (Stoffel, 2006). The
first studies analyzing the seasonality, frequency, and spa-
tial patterns of rockfalls using tree rings were only very
recently realized by Stoffel et al. (2005) and provided
a reliable basis for the analysis of rockfall risks and the
accuracy assessment of rockfall models (Stoffel et al.,
2006b).
Recent comprehensive discussions of dendroglaciology
can be found in Koch (2009). The first, simplest, applica-
tions involve determining the inception date of the oldest
tree growing on a moraine or other surface to provide
a minimum age for the feature or, more rarely, estimates
of the rates of glacier recession or succession. In many
areas around the world, the glacier advances of the last
few centuries were more extensive than earlier Holocene
events, and significant dendroglaciological work has been
done using recently exposed stumps or logs in glacier
forefields to estimate the dates of earlier events or ice-free
periods, providing the tree-ring records can be cross-dated
against reference chronologies. However only kill dates
from in situ materials provide the precise location of the
glacier when the tree was killed, detrital material having
been killed some unknown distance up-glacier cannot
yield such data.
Tree reactions to earth-surface processes
Apart from the site-specific information common to many
trees at any site (e.g., genetic information, age, or climatic
conditions), individual trees also record the effects of
mechanical disturbance caused by external processes.
Trees can be injured, suffer breakage of their crown or
branches, and have their stems tilted, partially buried, or
their roots exposed. Evidence of these events can be
recorded in the tree-ring series of affected trees (Figure 1).
The analysis of geomorphic processes through the study
of growth anomalies in tree rings is called dendrogeo-
morphology (Alestalo, 1971). Dendrogeomorphic
research is normally based on the “process–event–
response” concept as defined by Shroder (1978) where
the “process” is represented by any geomorphic agent,
e.g., debris flows, snow avalanches, or rockfall (Stoffel
et al., 2010). Individual geomorphic “events” that affect
the tree may result in a range of growth “responses.” These
“events” and associated “responses” are illustrated in the
following paragraphs.
Mechanical disturbance and wounding are very
common features in trees affected by geomorphic pro-
cesses. When impact locally destroys the vascular cam-
bium, increment cell formation is disrupted and new cell
formation ceases in the injured segment of the tree
(Figure 1a, b). To minimize rot and insect attacks after
damage, the injured tree will compartmentalize the wound
and start production of chaotic callus tissue (callus
tissue ¼ mass of unorganized parenchyma cells in plant
tissues formed after wounding; Figure 1c) at the edges of
the injury (Shigo, 1984). The extent of wound healing
greatly depends on the annual increment rate, the age of
the tree, and on the size of the scar. After injury, tangential
rows of traumatic resin ducts (or TRD ¼ compact and con-
tinuous rows of resin canals formed as a result of cambium
damage. TRD are an excellent proxy to date mass
 DENDROCHRONOLOGY, SURFICIAL PROCESSES 215

Dendrochronology, Surficial Processes, Figure 1 Injuries in Larix decidua Mill: (a) injured stem (b) cross section with overgrowth
starting from the lateral edges of the injury. (c) Callus tissue as observed in the overgrowing cell layers bordering the injury. (d)
Tangential row of traumatic resin ducts migrating from earlywood toward later portions of the tree ring with increasing distance from
the wound.

movement-induced mechanical impact in trees; Figure 1d)
are produced in the developing secondary xylem of certain
conifer species, e.g., Larix sp., Picea sp., Pseudotsuga sp.,
or Abies sp. (Stoffel, 2008). They extend tangentially and
axially from the injury (Bollschweiler et al., 2008;
Schneuwly et al., 2009). When wounding occurs during
the vegetation period of the tree, resin production will start
a few days after the event and ducts emerge within 3
weeks after the disturbing event. Therefore, when analyz-
ing cross sections, the intra-seasonal position of the first
series of TRD can be used to reconstruct previous events
with monthly precision (Stoffel et al., 2008; Stoffel and
Hitz, 2008). In broad-leaved species, cambial damage will
initiate various anatomical responses, among those
a reduction vessel sizes (Arbellay et al., 2010, 2012).
Inclination of the stem results from the sudden pressure
induced by geomorphic processes directly, by the associ-
ated deposition of material (e.g., avalanche snow, debris-
flow material), or by the slow but ongoing destabilization
of a tree through landslide activity or erosion (Lundström
et al., 2009). Tilted trees (Figure 2a) are common in many
areas affected by geomorphic processes and have there-
fore been used in many dendrogeomorphic studies to date
previous events (Braam et al., 1987). Subsequent growth
in the trunk of a tilted tree will attempt to restore its verti-
cal position, and the reaction will be most clearly visible in
216 DENDROCHRONOLOGY, SURFICIAL PROCESSES
Dendrochronology, Surficial Processes, Figure 2 (a) Tree morphology and (b) cross sections of a tilted Larix decidua Mill.
(c) Increment curves of a Picea abies (L.) Karst. tree tilted by a debris flow in 1922.
that segment of the tree to which the center of gravity has
been moved to through the inclination of the stem axis. In
the tree-ring record, eccentric growth (and related reaction
wood; Figure 2b, c) will be visible after a tilting event and
thus allow accurate dating of the disturbance. In conifer-
ous trees, compression wood will be produced on the
underside of the trunk. Individual rings will be consider-
ably larger and slightly darker in appearance as compared
to the upslope side. The difference in color results from
much thicker and rounded cell walls of early- and late-
wood tracheids (Timell, 1986). Multiple tilting events in
the same stem may be recognized by changes in the
amount, color, or orientation of reaction wood series in
the ring series. In contrast, stem tilting in broad-leaved
trees leads to the formation of tension wood on the upper
side facing the tilting agent. Broad-leaved trees also react
upon tilting with ultrastructural modifications that are only
visible when studied on microsections (Pilate et al., 2004).
In addition to the formation of different types of reaction
wood, trees may also respond with reduced growth after
tilting (Mayer et al., 2010).
Debris flows, floods, landslides, or “dirty” snow ava-
lanches may bury trees by depositing material around their
stem base (Figure 3a). Growth in these trees will normally
be reduced as the supply of water and nutrients will be
temporarily disrupted or at least limited (Figure 3b;
Kogelnig et al., 2013). If stem burial exceeds a certain
threshold, trees will die from a shortage in water and nutri-
ent supply. According to case-study results from the Ital-
ian Dolomites (Strunk, 1991), Picea abies (L.) Karst.
may tolerate a maximum burial depth of 1.6–1.9 m in
environments dominated by fine-grained debris flows
composed of calcareous material. Occasionally, buried
trees produce adventitious roots close to the new ground
surface. As adventitious roots are typically formed in the
first years after burial, the moment of root sprouting can
be used to derive an approximate date for the sedimenta-
tion processes (Strunk, 1991).
Bouncing rocks and boulders, debris in flowing water,
debris flows and lahars, or the windblast of snow ava-
lanches may cause decapitation of trees or the removal
of branches. The loss of the crown or branches is more
common in bigger trees, when stems have lost their flexi-
bility. Apex (apex ¼ uppermost part or growing tip of
a plant) loss has also been observed as a result of rockfall
impacts close to the ground level. In such cases, the sinu-
soidal propagation of shockwaves in the stem results in the
break-off of the crown. This phenomenon has been
described as whiplash or a “hula-hoop” effect (Lundström
et al., 2009). Trees react upon decapitation or branch loss
with distinct radial growth suppression. One or several lat-
eral branches will form a “leader” that replaces the broken
crown, resulting in the tree morphology called “candela-
bra” growth (Butler and Malanson, 1985).
Erosional processes and the (partial) denudation of
roots may generate different growth reactions, both in
the stem and in the exposed roots (Figure 4a). The type
and intensity of the reaction(s) will depend on the nature
of the erosive event, which may be instantaneous or
progressive and gradual. If several roots are completely
denuded during a sudden erosive event (e.g., debris flow,
lahar, flood, or landslide), they are no longer able to fulfill
 DENDROCHRONOLOGY, SURFICIAL PROCESSES
Dendrochronology, Surficial Processes, Figure 3 (a) Sedimentation and subsequent die-off of trees after sedimentation.
(b) Microsection showing an abrupt growth decrease in Castanea sativa Mill. (c) Several levels of adventitious roots in Populus
deltoides Bartr. ex Marsh.
their primary functions and quickly die. The tree subse-
quently suffers from a shortage of water and nutrient supply,
resulting in suppressed tree growth and the formation of
narrow rings in the stem (LaMarche, 1968). In cases where
only part of a root is exposed and its outer end remains in the
ground, the root will continue to grow and fulfill its func-
tions. In the exposed root, however, anatomical changes
will occur and individual growth rings similar to those in
the stem or branches will be formed (Figure 4b, d). The
localization of this change in the tree-ring series may allow
determination of the moment of exposure (Corona et al.,
2011). The continuous slow exposure of roots is usually
caused by gradual processes and relatively low denudation
rates, e.g., by overland flow; the slow opening of cracks in
soils (e.g., soil creep, landslides) and in disintegrated bed-
rock, along rivers, streams, lakes, and oceans (floods, shore
erosion); as well as with faulting activity and displacements
in relation to earthquake activity (Jacoby et al., 1988, 1992).
Provided that the roots are gradually exposed with time, it is
also possible to determine the erosion rates in such cases
(Stoffel et al., 2013a).
Geomorphic processes can eliminate trees along chan-
nels or couloirs (couloir ¼ French for “passage” or “corri-
dor,” used to refer to a narrow gully with a steep gradient in
a mountainous terrain) through uprooting and stem break-
age and leave neighboring trees intact. This phenomenon
can be observed with, e.g., rockfalls, debris flows, lahars,
extreme floods, landslides, or snow avalanches. The
uninjured survivor trees have less competition, more light,
nutrients, and/or water. As a consequence, they may start
to produce larger increment rings. However, observations
indicate that this growth release in survivor trees can be
217
delayed and therefore this reaction cannot always be used
to date past destructive events with yearly precision. Nev-
ertheless, the growth release in survivor trees can corrobo-
rate the dating of events that have been identified from
evidence in other trees at the same site, e.g., from scars,
tilted stems, etc. (Stoffel and Corona, in press).
Many different earth-surface processes can eliminate
surface vegetation including entire forest stands, leaving
no direct dendrogeomorphic evidence. In such cases, ger-
mination ages of trees growing on bare surfaces can be
used to estimate the time of creation of new landforms
and the disturbance event. This approach provides
a minimum age for that surface and has frequently been
used to date landforms or assess the minimum time
elapsed since the last devastating event in snow avalanche
couloirs, debris-flow channels, or floodplains (Sigafoos,
1964; Pierson, 2007). It involves estimating the time
between the exposure of the surface and the germination
of the first surviving seedling on that surface. This “ece-
sis” estimate varies with the environment, substrate, avail-
able seed sources, and several other factors. Problems of
ecesis determination have been extensively discussed in
studies that attempt to date the formation of glacier
moraines (McCarthy and Luckman, 1993) where ecesis
estimates may vary from a few years to several decades.
Field approach and sampling design
The types of damage to trees described in the preceding
section may result from a wide variety of earth-surface
processes. Linkage between the damage and the causative
process depends on a critical evaluation of the sampling
218 DENDROCHRONOLOGY, SURFICIAL PROCESSES

Dendrochronology, Surficial Processes, Figure 4 (a) Exposed roots of Pinus sylvestris L. (b) In Abies alba Mill., larger increment rings
with distinct latewood are formed after sudden exposure (dashed red line). (c) Following sudden exposure, the arrangement of
vessels in Fraxinus excelsior L. changes from diffuse porous to ring porous. (d) In addition to cell changes, tension wood is formed in
this root of Acer pseudoplatanus L.

site. In many cases, investigations are undertaken with
a specific context where the process or processes involved
are known and appropriate sampling methods and sam-
pling design may be applied. These vary considerably
with the nature of the process or investigation and are best
discussed in the specific context of individual processes,
e.g., the sampling network needed to define past earth-
quake activity may be very different from that required
to reconstruct the history of snow avalanches and debris
flows on a large debris cone.
At an individual site the choice of sampling design and
selection of the trees to be sampled will depend on the pur-
pose of the investigation and the processes or hazards
being sampled. The sampled area and number of trees
sampled may be very small and specific (e.g., on a small
landslide) or may involve systematic sampling of
a larger area (e.g., within a large runout zone of a snow
avalanche) or target individual trees damaged at the mar-
gins of an event. As a rule of thumb, processes that tend
to spread considerably will likely leave larger spatial foot-
prints and will thus be visible in a larger number of trees.
In the case of landslides, a sample size of 50 trees seems
sufficient, a slightly larger sample size (100 trees) appears
appropriate for snow avalanches, and at least 150 trees are
needed to characterize debris flows appropriately (Stoffel
et al., 2013b). The sample size in rockfall studies will
 DENDROCHRONOLOGY, SURFICIAL PROCESSES
depend on the size of the study site, stand density, and
mean rock sizes (Trappmann et al., in press). The choice
of the individual trees sampled is based on an inspection
of the stem and on the location of the trees with relation
to the processes studied.
The tree-ring record of growth disturbances created by
past events is analyzed with cross sections, wedges, or
increment cores. Cross sections or wedges are normally
taken at the location of the growth disturbance and pro-
vide an excellent and very complete insight into the tree’s
history. However, in many locations (e.g., protection for-
ests, National Parks), felling may be prohibited or ill-
advised for aesthetic or safety reasons. Most tree-ring
studies thus use cores extracted with an increment borer.
The nature of visible growth defects observed in the
tree’s morphology will influence sampling height, sam-
pling directions, and minimum number of samples to be
taken per tree.
In trees with visible scars, previous geomorphic events
are most easily dated through the destructive sampling of
trees and the preparation of cross sections taken at the
location where the injury is largest. This approach will
facilitate an accurate and intra-seasonal identification of
the onset of callus tissue (and TRD) formation and there-
fore allows a reconstruction of the impacting event with
very high temporal resolution. Alternatively, wedges can
be sawn from the overgrowing callus and an increment
core extracted from the side opposite to the wound
(Arbellay et al., 2010).
Tilted stems are usually analyzed with at least two incre-
ment cores extracted per tree, one in the direction of the
tilting and the other on the opposite side of the trunk. The
reaction wood will be visible on the tilted side in conifer
trees (¼ compression wood) and on the side opposite to
the tilting direction in broad-leaved trees (¼ tension wood).
Individual cores are best extracted at the location where
tilting is strongest based on an outer inspection of the tree
morphology. However, in situations where multiple tilting
events may vary in direction, e.g., in snow avalanche tracks,
additional cores, transverse to the downslope direction,
may be needed to develop a more complete record.
In the case of buried stem bases, the sampling of two
increment cores in opposite directions (ideally from the
upslope and the downslope side) will normally allow accurate
reconstruction of the event that has led to the sedimentation of
material around the stem’s base. It is best to sample these trees
as close to ground level as possible (ca. 20–40 cm) to extract
a maximum number of tree rings and obtain maximum infor-
mation. However, the influence of roots should be avoided.
The analysis of exposed roots is normally based on
cross sections, as they regularly show partially or
completely missing rings. The position of samples needs
to be carefully noted with respect to the ground surface
or the non-eroded parts of the root. These samples are
essential for the understanding and interpretation of con-
tinuous erosion processes and the determination of denu-
dation rates (Stoffel et al., 2013a).
219
The germination of trees on new landforms is best
determined by counting the annual growth rings in incre-
ment cores taken from the root crown level (root
crown ¼ also known as root collar or root neck; part of
a root system from which a stem arises and where vascular
changes between the stem and the roots occur). However,
branches, obstacles, and rot may sometimes require sam-
pling positions higher up on the stem. In these cases, an
age correction factor needs to be added to compensate
for the time a tree takes to grow to sampling height.
A simple height-age correction can be achieved by divid-
ing the tree height by the number of tree rings to get an
average rate for the yearly apical increment.
In some cases a reference tree-ring chronology may
need to be developed from a nearby forest stand to assist
in cross-dating of trees at the sampled site to verify the
age assignment of rings. Adequate sample size varies
between species and also depends on the strength of the
common signal. In general, 20–30 trees is a useful guide
for most species to minimize the potential influence of
geomorphic processes and hidden growth disturbances.
Sample analysis and interpretation
Ring widths of the disturbed samples may also be mea-
sured and the series compared graphically or statistically
with the reference chronology. These tools also allow for
an assessment of cross-dating accuracy between ring-
width series of individual disturbed trees and the reference
chronology. Once all tree-ring series are checked, individ-
ual growth curves may be analyzed visually to identify the
tree’s reactions to geomorphic processes, e.g., the initia-
tion of abrupt growth reduction or releases. In addition,
the microscopic appearance of the cells may be investi-
gated to yield further evidence of disturbance. Other fea-
tures, e.g., callus tissue overgrowing scars or the
presence of TRD formed following cambium damage,
can only be identified through a visual inspection of the
cores and cross sections. All growth reactions identified
in the samples are noted in order to identify disturbance
events. Except for processes with limited volumes (e.g.,
single rocks or boulders involved in rockfalls), one growth
disturbance identified in a single tree is not considered suf-
ficient evidence to identify an event, and therefore the
identification of events needs to be based on quantitative
or semiquantitative thresholds (Stoffel and Bollschweiler,
2008; Stoffel and Corona, in press).
Summary
The initial employment of tree rings in geomorphic studies
was simply as a dating tool and rarely exploited other
environmental information and records of damage
contained within the tree. However, these unique, annu-
ally resolved, tree-ring records preserve valuable archives
of past earth-surface processes on timescales of decades to
centuries. As many of these processes are significant nat-
ural hazards, understanding their distribution, timing,
220 DENDROCHRONOLOGY, SURFICIAL PROCESSES
and controls provides valuable information that can assist
in the prediction, mitigation, and defense against these
hazards and their effects on society. This entry provides
many illustrations of these themes, demonstrating the
application of tree rings to studies of snow avalanches,
rockfalls, landslides, floods, earthquakes, wildfires, and
several other processes. It illustrates the breadth and
diverse applications of contemporary dendrogeo-
morphology and underlines the growing potential to
expand dendrogeomorphic research, possibly leading to
the establishment of a range of techniques and approaches
that may become standard practice in the analysis and
understanding of earth-surface processes and related natu-
ral hazards in the future.
Bibliography
Alestalo, J., 1971. Dendrochronological interpretation of geomor-
phic processes. Fennia, 105, 1–139.
Arbellay, E., Stoffel, M., and Bollschweiler, M., 2010.
Dendrogeomorphic reconstruction of past debris-flow activity
using injured broad-leaved trees. Earth Surface Processes and
Landforms, 35, 399–406.
Arbellay, E., Fonti, P., and Stoffel, M., 2012. Duration and exten-
sion of anatomical changes in wood structure after cambial
injury. Journal of Experimental Botany, 63, 3271–3277.
Ballesteros, J. A., Eguibar, M., Bodoque, J. M., Díez, A., Stoffel,
M., and Gutiérrez, I., 2011. Estimating flash flood discharge in
an ungauged mountain catchment with 2D hydraulic models
and dendrogeomorphic paleostage indicators. Hydrological Pro-
cesses, 25, 970–979.
Ballesteros-Canovas, J. A., Sanchez-Silva, M., Bodoque, J. M., and
Díez-Herrero, A., 2013. An integrated approach to flood risk
management: a case study of Navaluenga (Central Spain). Water
Resources Management, doi:10.1007/s11269-013-0332-1.
Bégin, C., and Filion, L., 1985. Dendrochronological analysis of
a landslide in the Clearwater-Lake area (northern Quebec).
Canadian Journal of Earth Sciences, 22(2), 175–182.
Bégin, C., and Filion, L., 1988. Age of landslides along Grande
Rivière de la Baleine estuary, eastern coast of Hudson Bay, Que-
bec (Canada). Boreas, 17, 289–299.
Bollschweiler, M., and Stoffel, M., 2010. Variations in debris-flow
occurrence in an Alpine catchment – reconstruction and implica-
tions for the future. Global and Planetary Change, 73, 186–192.
Bollschweiler, M., Stoffel, M., Schneuwly, D. M., and Bourqui, K.,
2008. Traumatic resin ducts in Larix decidua stems impacted by
debris flows. Tree Physiology, 28, 255–263.
Braam, R. R., Weiss, E. E. J., and Burrough, P. A., 1987. Spatial and
temporal analysis of mass movement using dendrochronology.
Catena, 14, 573–584.
Butler, D. R., 1979. Snow avalanche path terrain and vegetation, Glacier
National Park, Montana. Arctic and Alpine Research, 11, 17–32.
Butler, D. R., and Malanson, G. P., 1985. A history of high-magnitude
snow avalanches, Southern-Glacier-National-Park, Montana,
USA. Mountain Research and Development, 5, 175–182.
Corona, C., Lopez, J., Rovéra, G., Stoffel, M., Astrade, L., and
Berger, F., 2011. High-resolution, quantitative reconstruction
of erosion rates based on anatomical changes in exposed roots
(Draix, Alpes de Haute-Provence) – critical review of existing
approaches and independent quality control of results. Geomor-
phology, 125(3–1), 433–444.
Corona, C., Lopez Saez, J., Stoffel, M., Rovéra, G., Edouard, J.-P.,
and Berger, F., 2012. Seven centuries of avalanche activity at
Echalp (Queyras massif, southern French Alps) as inferred from
tree rings. The Holocene, doi:10.1177/0959683612460784.
Fantucci, R., 2007. Dendrogeomorphological analysis of shore ero-
sion along Bolsena lake (Central Italy). Dendrochronologia, 24,
69–78.
St. George, S., and Nielsen, E., 2000. Signatures of high-magnitude
19th century floods in Quercus macrocarpa (Michx.) tree rings
along the Red River, Manitoba, Canada. Geology, 28, 899–902.
Germain, D., Filion, L., and Hétu, B., 2009. Snow avalanche regime
and climatic conditions in the Chic-Chocs Range, Eastern Can-
ada. Climatic Change, 92, 141–167.
Hupp, C. R., 1984. Dendrogeomorphic evidence of debris flow fre-
quency and magnitude at Mount Shasta, California. Environ-
mental Geology and Water Science, 6, 121–128.
Jacoby, G. C., Sheppard, P. R., and Sieh, K. E., 1988. Irregular
recurrence of large earthquakes along the San Andreas Fault.
Evidence from trees. Science, 241, 196–199.
Jacoby, G. C., Williams, P. L., and Buckley, B. M., 1992. Tree ring
correlation between prehistoric landslides and abrupt tectonic
events in Seattle, Washington. Science, 258, 1621–1623.
Koch, J., 2009. Improving age estimates for late Holocene glacial
landforms using dendrochronology – some examples from Gar-
ibaldi Provincial Park, British Columbia. Quaternary Geochro-
nology, 4, 130–139.
Kogelnig, B., Stoffel, M., and Schneuwly-Bollschweiler, M., 2013.
Four-dimensional growth response of mature Larix decidua to
stem burial under natural conditions. Trees: Structure and Func-
tion, 27(5), 1217–1223.
LaMarche, V. C., 1968. Geomorphic and dendroecological impacts
of slushflows in central Gaspé Peninsula (Québec, Canada). U.S.
Geological Survey Professional Paper 352-I.
Lopez Saez, J., Corona, C., Stoffel, M., and Berger, F., 2013. Cli-
mate change increases frequency of shallow spring landslides
in the French Alps. Geology, 41, 619–622.
Lundström, T., Jonsson, M. J., Volkwein, A., and Stoffel, M., 2009.
Reactions and energy absorption of trees subject to rockfall:
a detailed assessment using a new experimental method. Tree
Physiology, 29, 345–359.
Mayer, B., Stoffel, M., Bollschweiler, M., Hübl, J., and Rudolf-
Miklau, F., 2010. Frequency and spread of debris floods on fans:
a dendrogeomorphic case-study from a dolomite catchment in
the Austrian Alps. Geomorphology, 118, 199–206.
McCarthy, D. P., and Luckman, B. H., 1993. Estimating ecesis for
tree-ring dating of moraines – a comparative study from the
Canadian Cordillera. Arctic and Alpine Research, 25, 63–68.
Paolini, L., Villalba, R., and Grau, H. R., 2005. Precipitation vari-
ability and landslide occurrence in a subtropical mountain eco-
system of NW Argentina. Dendrochronologia, 22, 175–180.
Pierson, T. C., 2007. Dating young geomorphic surfaces using age
of colonizing Douglas fir in southwestern Washington and north-
western Oregon, USA. Earth Surface Processes and Landforms,
32, 811–831.
Pilate, G., Chabbert, B., Cathala, B., Yoshinaga, A., Leple, J. C.,
Laurans, F., Lapierre, C., and Ruel, K., 2004. Lignification and
tension wood. C. R. Biologies, 327, 889–901.
Schaerer, P. A., 1972. Terrain and vegetation of snow avalanche
sites at Rogers Pass, British Columbia, B.C. Geographical
Series, 14, 215–222.
Schneuwly, D. M., Stoffel, M., and Bollschweiler, M., 2009. For-
mation and spread of callus tissue and tangential rows of resin
ducts in Larix decidua and Picea abies following rockfall
impacts. Tree Physiology, 29, 281–289.
Schneuwly-Bollschweiler, M., and Stoffel, M., 2012. Hydrometeo-
rological triggers of periglacial debris flows – a reconstruction
dating back to 1864. Journal of Geophysical Research: Earth
Surface, 117, F02033.
Shigo, A. L., 1984. Compartmentalization: a conceptual framework
for understanding how tree growth defend themselves. Annual
Review of Phytopathology, 22, 189–214.
 DENDROCHRONOLOGY, VOLCANIC ERUPTIONS
Shroder, J. F., 1978. Dendrogeomorphological analysis of mass move-
ment on Table Cliffs Plateau, Utah. Quaternary Research, 9,
168–185.
Sigafoos, R. S., 1964. Botanical evidence of floods and flood-
plain deposition. US Geological Survey Professional Paper
485-A, p. 35.
Stoffel, M., 2006. A review of studies dealing with tree rings and
rockfall activity: the role of dendrogeomorphology in natural
hazard research. Natural Hazards, 39, 51–70.
Stoffel, M., 2008. Dating past geomorphic processes with
tangential rows of traumatic resin ducts. Dendrochronologia,
26, 53–60.
Stoffel, M., 2010. Magnitude-frequency relationships of debris
flows – a case study based on field surveys and tree-ring records.
Geomorphology, 116, 67–76.
Stoffel, M., and Bollschweiler, M., 2008. Tree-ring analysis in nat-
ural hazards research – an overview. Natural Hazards and Earth
System Sciences, 8, 187–202.
Stoffel, M., Corona, C., in press. Dendroecological dating of geo-
morphic disturbance in trees. Tree-Ring Research, 70(1).
Stoffel, M., and Hitz, O. M., 2008. Rockfall and snow
avalanche impacts leave different anatomical signatures in tree
rings of juvenile Larix decidua. Tree Physiology, 28,
1713–1720.
Stoffel, M., Schneuwly, D., Bollschweiler, M., Lièvre, I., Delaloye,
R., Myint, M., and Monbaron, M., 2005. Analyzing rockfall
activity (1600–2002) in a protection forest – a case study using
dendrogeornorphology. Geomorphology, 68, 224–241.
Stoffel, M., Bollschweiler, M., and Hassler, G. R., 2006a. Differen-
tiating events on a cone influenced by debris-flow and snow ava-
lanche activity – a dendrogeomorphological approach. Earth
Surface Processes and Landforms, 31, 1424–1437.
Stoffel, M., Wehrli, A., Kühne, R., Dorren, L. K. A., Perret, S., and
Kienholz, H., 2006b. Quantifying the protective effect of moun-
tain forests against rockfall using a 3D simulation model. Forest
Ecology and Management, 225, 113–122.
Stoffel, M., Conus, D., Grichting, M. A., Lievre, I., and Maitre, G.,
2008. Unraveling the patterns of late Holocene debris-flow
activity on a cone in the Swiss Alps: chronology, environment
and implications for the future. Global and Planetary Change,
60, 222–234.
Stoffel, M., Bollschweiler, M., Butler, D. R., and Luckman, B. H.,
2010. Tree Rings and Natural Hazards: A State-of-the-Art. Dor-
drecht/Heidelberg/London/New York: Springer. 505.
Stoffel, M., Casteller, A., Luckman, B. H., and Villalba, R., 2012.
Spatio-temporal analysis of channel sidewall erosion in ephem-
eral debris-flow torrents – an example from the Patagonian
Andes. Geology, 40, 247–250.
Stoffel, M., Corona, C., Ballesteros-Cnovas, J. A., Bodoque, J. M.,
2013a. Dating and quantification of erosion processes based on
exposed roots. Earth-Science Review. http://dx.doi.org/
10.1016/j.bbr.2011.03.031
Stoffel, M., Butler, D. R., and Corona, C., 2013b. Mass movements
and tree rings: a guide to dendrogeomorphic field sampling and
dating. Geomorphology, doi:10.1016/j.geomorph.2012.12.017.
Strunk, H., 1991. Frequency distribution of debris flow in the Alps
since the “Little Ice Age”. Zeitschrift für Geomorphologie Sup-
plement, 83, 71–81.
Timell, T. E., 1986. Compression Wood in Gymnosperms. Berlin:
Springer. 2150.
Trappmann, D., Corona, C., Stoffel, M., in press. Rolling stones and
tree rings: a state of research on dendrogeomorphic reconstructions
of rockfall. Progress in Physical Geography, 37(5), 701–716.
Cross-references
Dendrochronology, Progress
221
DENDROCHRONOLOGY, VOLCANIC ERUPTIONS
Franco Biondi
DendroLab, University of Nevada, Reno, Nevada, USA
Synonyms
Dust veil; Extrusive volcanism; Tephra deposition
Definition
Dating of events associated with volcanism that leaves
a permanent marker in tree-ring records. Such events
include air temperature cooling, deposition of ash layers,
lava flows, and magmatic degassing.
Introduction
The geological applications of dendrochronology cover
a wide range of topics, from paleoseismology (Jacoby,
1997) to paleovolcanism and its effects on climate
(Zielinski, 2000). With specific regard to volcanic erup-
tions, tree-ring records have been used in multiple ways
to date events associated with explosive and effusive vol-
canism, including magmatic degassing. Because wood
formation is influenced by environmental factors
(Plomion et al., 2001), tree species that are directly or indi-
rectly impacted by volcanic eruptions can record such
events in their growth layers. Datable volcanic effects on
woody species range from annihilation (death dates) to
rebirth (establishment dates) but also include growth
changes in surviving trees (Yamaguchi, 1993). Such
changes can be abrupt, hence only detectable in single
rings. These individual growth layers can be very small
(“microrings”), down to the point of not being present
(“locally absent rings”) in one or more samples collected
at a site (Biondi et al., 2003). Individual tree rings that
are typically not so small can show anatomical damage
from freezing (“frost rings”) in response to the sudden
reduction of sunlight and its associated air temperature
drop, caused by volcanic dust veils (LaMarche and
Hirschboeck, 1984). Depending on the type of extrusive
materials (tephra deposition, lava flows, emission of sulfu-
ric gases, etc.), surviving trees can present gradual, rather
than abrupt, changes over multiple consecutive years.
Such departures from previous tree-ring patterns can be
positive (increased growth, or “release”) or negative
(decreased growth, or “suppression”). The diversity of
volcanic effects on trees depends on the variety of mate-
rials that can be ejected, as well as on the location, species,
size, and age of the trees, together with other site condi-
tions, from elevation and topography to vegetation fea-
tures. On one hand, such richness of potential
information has generated a long tradition of tree-ring
applications to volcanology. On the other hand, the diffi-
cult task of identifying what, when, and how trees were
impacted by specific past volcanic events has caused mul-
tiple, heated, and prolonged scientific debates.
222 DENDROCHRONOLOGY, VOLCANIC ERUPTIONS
Historical development
Modern dendrochronology, which is the study of past
changes recorded by wood growth, is based on the rigor-
ous application of Crossdating techniques to assign calen-
dar dates to tree rings (Fritts and Swetnam, 1989). When
so defined, this branch of science was established in the
western United States at the beginning of the twentieth
century by Andrew E. Douglass, an astronomer at the Uni-
versity of Arizona in Tucson (Webb, 1983; Nash, 1999).
Tree-ring dating is closely connected with the study of
wood formation, and relevant studies, for example, on
the anatomy of frost rings (Rhoads, 1923), also date back
about a century. On the other hand, while the hypothesis
that exceptionally cold years could be caused by volcanic
eruptions is found in Benjamin Franklin’s (1785) writings,
the global atmospheric radiative effects of volcanic aero-
sols could only be tested with the advent of satellite and
remote sensing technology in the 1980s (Ramanathan,
1988). At about the same time, the Volcanic Explosivity
Index, or VEI (Newhall and Self, 1982), was introduced
to describe the magnitude of past explosive eruptions in
a semiquantitative manner. Impacts of extrusive eruptions
on tree rings at sites located near a volcano were first
described in the 1960s, but sample sizes were small (i.e.,
less than ten trees), ring boundaries were uncertain either
because of poor sample preservation and surfacing
(Druce, 1966) or because of species growth patterns and
anatomical features (Eggler, 1967), and no crossdating
was used. A notable exception was the A.D. 1064–1065
dating of Sunset Crater, Arizona (Smiley, 1958), where
a larger number of samples from archaeological ruins were
investigated and tree-ring dates were assigned by
crossdating. On the other hand, less than ten trees showed
a distinct growth reduction after 1064, and nonvolcanic
disturbances can cause similar effects in ring widths of
the same species in this geographical area (Sheppard
et al., 2005).
A major catalyst for dendro-volcanology came from the
1980 eruption of Mount St. Helens in the Pacific North-
west region of the USA. The blast that occurred on May
18 of that year with an estimated VEI of 5 (the maximum
is 8) removed the top 400 m of the mountain and rose to
24 km into the stratosphere (Brantley and Myers, 2000).
It was the deadliest and most economically devastating
volcanic event in United States history, exceeding
a billion dollars in damage. A total of about 60 people,
including a volcanologist, and more than nine million
cubic meters of valuable timber were among the losses
suffered over the nearly 600 km2 ravaged by the eruption
(McLean and Lockridge, 2000). Several studies were
published shortly thereafter on tree growth responses to
the eruption (Mahler and Fosberg, 1983; Hinckley
et al., 1984) and on dendrochronological dating of previ-
ous events (Yamaguchi, 1983, 1985). The enhanced inter-
est in past volcanic histories, and the realization of
potential long-range temperature effects from volcanic
dust veils, prompted an extensive study of frost rings in
the wood of high-elevation bristlecone pines (LaMarche
and Hirschboeck, 1984). Another anatomical feature,
i.e., the “light rings” caused by reduced latewood forma-
tion in conifers growing at high elevations or latitudes,
was investigated for their potential connection with
reduced temperatures associated with volcanic eruptions
(Filion et al., 1986). Extensive networks of tree-ring sites,
whose analysis was then facilitated by advances in compu-
tational power and numerical processing, became the tar-
get of sophisticated statistical methods to detect changes
in tree-ring-reconstructed climatic variables that could be
linked to volcanic eruptions (Lough and Fritts, 1987).
Yet another method of tree-ring analysis, X-ray densi-
tometry (Polge, 1970) had come of age in those decades
(Schweingruber et al., 1978), and networks of tree-ring
sites sampled in the late 1970s and early 1980s were ana-
lyzed using this new technique (Schweingruber
et al., 1991). Subsequent research to date has continued
to employ the tools elaborated in previous years, from
light rings (Szeicz, 1996) to densitometry (Jones
et al., 1995), expanding the number of sampled sites
(D’Arrigo and Jacoby, 1999) as well as the geographical
(Hantemirov et al., 2004) or temporal (Salzer and Hughes,
2007) range of previous studies, and applying increasingly
complex numerical treatments for identifying volcanic
signals in tree-ring proxy records (Breitenmoser
et al., 2012). As an alternative to the exhaustive filtering
and massaging of large spatiotemporal datasets, other
studies in the past few decades have investigated the use
of chemical markers in tree rings linked to volcanic erup-
tions (Pearson et al., 2009), including magmatic degassing
in the absence of explosive events (Biondi and Fessenden,
1999).
Basic principles and applications
Trees and other woody plants grow by covering them-
selves with a new layer of tissue every year (Larson,
1994). When seen on a transverse section, such wood
layers appear as concentric tree rings. Because tree growth
is influenced by the environment, tree rings are then natu-
ral archives of past environmental conditions. For
instance, less wood is formed near the base of a tree stem
when climate conditions are less favorable, producing
narrower rings (Speer, 2010). If an eruption takes place,
trees growing close enough to the volcano can be scorched
by hot gases or covered with tephra and may either be
killed or survive depending on their species, age and
vigor, microsite conditions, as well as the chemistry and
temperature of the gases and the thickness and coarseness
of the tephra layer (Segura et al., 1994). Surviving trees
typically experience abrupt suppression of radial growth
(Hinckley et al., 1984), including locally absent rings
(Yamaguchi and Hoblitt, 1995). This immediate decline
can be followed by a prolonged period of reduced
radial increment (Segura et al., 1995) or by its opposite,
i.e., greater than normal growth rates, often in connection
 DENDROCHRONOLOGY, VOLCANIC ERUPTIONS
with concurrent stand dynamics and resulting competitive
interactions (Abrams et al., 1999).
Major explosive volcanic eruptions that inject dust and
aerosols into the stratosphere are capable of causing large-
scale surface cooling (Minnis et al., 1993). Distant, large-
scale networks of tree-ring chronologies from
temperature-limited sites can reflect those eruptions in
anatomical xylem features, such as frost damage
(D’Arrigo et al., 2001) or reduced latewood formation
(Jacoby et al., 1999). Freezing damage to coniferous spe-
cies causes the formation of deformed and collapsed
xylem cells depending on the intensity of the frost, the
degree of cambial activity at the time of the temperature
drop, the tree species and size, and possibly other factors
(Day and Peace, 1934). Frost rings have been reproduced
experimentally (Glerum and Farrar, 1966), and their cross-
sectional features are clearly distinguishable on a well-
prepared wood sample at the magnification (10–50)
normally used in tree-ring studies. While crossdating
among a large enough number of samples guarantees
accurate dates, the year assigned to a frost ring may have
some uncertainty if the injury occurs at a ring boundary,
so that it could represent either a late frost in a year or
an early frost in the following year. This situation is more
frequent as ring widths become smaller (<0.1 mm),
but using well-dated frost rings to resolve uncertain
cases leads to potential bias and should be either avoided
or well documented. With regard to “light rings”, the
reduced cell wall thickening in latewood that causes their
name also causes minimal density and can be caused
either by low air temperature or by insect defoliation
(Liang et al., 1997).
Measured annual growth parameters, such as ring width
(Fritts, 1976) or maximum latewood density (Briffa
et al., 1988), are routinely used for dendroclimatic recon-
structions. When wood formation is limited by air temper-
ature, as it is usually the case for trees located at high
elevations or high latitudes (D’Arrigo et al., 2006), den-
drochronological parameters can reflect a volcanic
cooling, thereby providing a record of past eruptions
(Briffa et al., 1998). Chronologies developed from ring
widths present a greater amount of time-series autocorre-
lation (or “persistence”) than chronologies generated from
annual density peaks (called maximum latewood density,
or MLD), which also tend to reflect growing season condi-
tions rather than integrating environmental signals over
multiple years (D’Arrigo et al., 1992). Therefore, it is nor-
mally assumed that the temporal lag between volcanic
cooling and its incorporation in tree-ring proxy records
increases going from frost rings to light rings and from
maximum latewood density to ring width chronologies
(D’Arrigo and Jacoby, 1999).
The main application of dendrochronology to volcanol-
ogy is the definition of impacts on ecosystems and on
human societies, either current, past, or long lost. Besides
providing information on ecological processes in forests,
tree-ring records of volcanic eruptions contribute to
estimating the past frequency of destructive events.
223
This numerical datum can then improve the definition of
potential risks, especially with regard to explosive phe-
nomena occurring at short intervals from one another
(Yamaguchi, 1985). At the same time, because of possible
global impacts of volcanoes on climate (Robock and Mao,
1995) and biogeochemical cycles (Krakauer and
Randerson, 2003), well-dated records of past eruptions
are needed to improve the accuracy of models used to gen-
erate future scenarios linked to anthropogenic emissions
of greenhouse gases (Ramanathan, 1988). As any other
dating tool, tree-ring records of volcanic eruptions can
alter historical interpretations simply by changing the
order of events. While dates are not by themselves reason
for cause-and-effect relationships, they are able to negate
theoretical ones as no effect can precede its hypothesized
cause.
Current controversies and knowledge gaps
Major volcanic eruptions have been dated using other
types of records, from dust layers and chemical properties
in ice cores (Larsen et al., 2008) to archaeological and
written sources (Stothers, 1999). Not all records agree,
and in particular, dates assigned to some major eruptions
using geochemical markers in Greenland ice cores
(Vinther et al., 2006) are currently in disagreement with
tree-ring records (Baillie, 2010). A particularly important
event is the Minoan eruption of Thera, an island in the
Aegean Sea now called Santorini, which took place in
the mid-second century BC. The date of this volcanic time
marker is of critical importance for the stratigraphic and
archaeological synchronization of ancient Eastern Medi-
terranean societies, the cradle of western culture. The
eruption, which left a caldera and ash deposits up to hun-
dreds of meters in size, has been linked with the collapse
of the Minoan civilization on the island of Crete, 110 km
to the south, possibly spurred by gigantic tsunamis. The
controversy has lasted for decades, with historical/archae-
ological dates centered in 1550–1500 BC (Wiener, 2009),
tree-ring records pointing to 1629–1627 BC (Baillie,
2010), ice-core geochemistry focused on 1642–1641 BC
(Vinther et al., 2006), and radiocarbon dates spread over
1683–1611 BC (Manning et al., 2006). Each of these
sources of evidence has strengths and weaknesses. As an
example, radiocarbon dating of an olive branch from the
eruption site provided a date range of 1627–1600 BC
(Friedrich et al., 2006), but it is difficult to determine if
the sample was dead or alive at the time of the eruption
(Wiener, 2010), and wood anatomical features make any
ring analysis problematic for this species (Cherubini
et al., 2013).
The chemical signature of volcanic eruptions has been
proposed as a possible tool for testing the accuracy of
tree-ring dates assigned to individual events (Pearson
et al., 2005). Because elemental concentrations can vary
considerably in the same tree both around the stem and
at various heights in the trunk (Watt et al., 2007), clear evi-
dence needs to be presented that specific chemical signals
224 DENDROCHRONOLOGY, VOLCANIC ERUPTIONS
are replicable across samples from the same tree as well as
from different trees of the same species and site (Kirchner
et al., 2008). When crossdated ring sequences are com-
bined with localized information on geochemical vari-
ables, a high degree of accuracy and confidence can be
placed in the dating (Sheppard et al., 2008). This, for
instance, was the case with regard to the onset of mag-
matic degassing in forest stands of the Sierra Nevada,
which produced abrupt changes in crossdated ring-width
sequences (Biondi and Fessenden, 1999) and in their
radiocarbon concentrations (Cook et al., 2001).
A number of confounding factors hamper the use of
proxy records for dating volcanic eruptions. Several of
these factors can be categorized as false positives or false
negatives (Cleland, 2002). Considering anatomical
markers in tree rings, not all of them are related to major
volcanic eruptions (false positives), and known major vol-
canic eruptions took place without producing such
markers (false negatives). In addition, not all cold sum-
mers over a region are related to volcanic events, and not
all volcanic years produce a cold summer at the majority
of sampled sites in a tree-ring network (D’Arrigo
et al., 2013). Depending on the prevailing climate regime
and ecological requirements of a species, climate-driven
responses to major eruptions can even favor tree growth
rather than depressing it. Such inverse response was pro-
posed for Thera because the potentially cooler summers
generated by the dust veil would have reduced evapotrans-
piration, resulting in 2–3 years of abnormally high growth
in woody species adapted to the normally dry Mediterra-
nean conditions (Kuniholm et al., 1996). These hypothet-
ical responses, albeit reasonable, require further testing,
which is best attained using detailed ecophysiological
and anatomical studies at sub-monthly time scales (e.g.,
Rossi et al., 2013). Finally, when evaluating unusual epi-
sodes within a time series, it is best to employ
a quantitative approach, such as ranking episode features
(duration, magnitude, peak) according to predefined
numerical criteria (e.g., Biondi et al., 2008).
An additional wrinkle in the protracted debate over
tree-ring reconstructions of volcanic eruptions has
recently emerged. Climate modelers have tried to recon-
cile their simulations of regional and global climate
impacts from volcanic eruptions by postulating that tree-
ring chronologies for northern latitudes or from high ele-
vations are all consistently missing, and in more than
one occasion, a year of growth (Mann et al., 2012). The
ensuing vigorous responses (Anchukaitis et al., 2012;
D’Arrigo et al., 2013; Esper et al., 2013) have concen-
trated on choices made when modeling tree-ring forma-
tion and on correlations with instrumental or proxy
records, showing a general disruption of these statistical
relationships when tree-ring chronologies are shifted by
one or more years. The essential question underlying the
debate is the likelihood that collections of tree-ring sam-
ples from a single site, or even from multiple sites in
a region, would fail to include a year’s growth. The fre-
quency of locally absent rings increases in semiarid
environments as a response to drought, but existing tree-
ring collections pooled by latitude, species, or elevation
include a percentage of “missing” rings that does not
exceed 10 % (St. George et al., 2013). A number of other
issues have been raised, starting from how evidence dete-
riorates going further and further back into the past. For
relatively recent events, such as the 1815 explosion of
the Tambora volcano (Stothers, 1984), there is no chance
of a universally missing ring given the large number of
available proxy and instrumental records. For earlier erup-
tions, on the other hand, even their dates become less reli-
able as temporal separation increases (Plummer
et al., 2012).
Historical archives and environmental records of past
events are notoriously open to multiple interpretations,
but so are model outputs and their underlying assump-
tions. From a tree-ring perspective, not all is known about
the environmental conditions required for, and the biolog-
ical mechanisms leading to, wood formation in areas
where long proxy dendroclimatic records have been
developed. Anatomical and ecophysiological studies have
the potential to uncover such mechanisms without com-
plex numerical analyses (Fonti et al., 2010; Rossi
et al., 2012). On the other hand, the impossibility of exper-
imental testing on past events (Biondi, 2013), when com-
bined with lack of self-criticism or doubt, could maintain
the striking tendency to robust bragging that has become
a distinguishing trait of high-profile articles on volcano-
climate-proxy records connections.
Summary
Dendrochronological studies of volcanic eruptions
employ a variety of techniques, from anatomical to statis-
tical, for detecting the signals left by past events in tree-
ring records. When large datasets composed of multiple
sites and species are independently analyzed, the resulting
tree-ring dates assigned to past events are more reliable
than those derived from other proxy records or from
modeling exercises. Dendrochronological dates are exact
when sufficient sample replication exists, but the correct
interpretation of environmental signals, including volca-
nic ones, that are embedded in wood anatomical markers
and growth patterns requires careful consideration of alter-
native, competing hypotheses.
Bibliography
Abrams, M. D., Copenheaver, C. A., Terazawa, K., Umeki, K.,
Takiya, M., and Akashi, N., 1999. A 370-year dendroecological
history of an old-growth Abies-Acer-Quercus forest in Hok-
kaido, northern Japan. Canadian Journal of Forest Research,
29, 1891–1899.
Anchukaitis, K. J., Breitenmoser, P., Briffa, K. R., Buchwal, A.,
Büntgen, U., Cook, E. R., D’Arrigo, R. D., Esper, J., Evans,
M. N., Frank, D. C., Grudd, H., Gunnarson, B. E., Hughes,
M. K., Kirdyanov, A. V., Körner, C., Krusic, P. J., Luckman,
B. H., Melvin, T. M., Salzer, M. W., Shashkin, A. V., Timmreck,
C., Vaganov, E. A., and Wilson, R. J. S., 2012. Tree rings and
volcanic cooling. Nature Geoscience, 5, 836–837.
 DENDROCHRONOLOGY, VOLCANIC ERUPTIONS
Baillie, M. G. L., 2010. Volcanoes, ice-cores and tree-rings: one
story or two? Antiquity, 84, 202–215.
Biondi, F., 2013. The fourth dimension of interdisciplinary model-
ing. Journal of Contemporary Water Resources Education,
152, 42–48.
Biondi, F., and Fessenden, J. E., 1999. Response of lodgepole pine
growth to CO2 degassing at Mammoth Mountain, California.
Ecology, 80, 2420–2426.
Biondi, F., Galindo Estrada, I., Elizalde Torres, A., and Gavilanes
Ruiz, J. C., 2003. Tree growth response to the 1913 eruption of
Volcn de Fuego de Colima, Mexico. Quaternary Research,
59(3), 293–299.
Biondi, F., Kozubowski, T. J., Panorska, A. K., and Saito, L., 2008.
A new stochastic model of episode peak and duration for
eco-hydro-climatic applications. Ecological Modelling, 211,
383–395.
Brantley, S. R., and Myers, B., 2000. Mount St. Helens – From the
1980 Eruption to 2000. Vancouver, WA: United States Geologi-
cal Survey, 2 p.
Breitenmoser, P., Beer, J., Brönnimann, S., Frank, D., Steinhilber, F.,
and Wanner, H., 2012. Solar and volcanic fingerprints in tree-ring
chronologies over the past 2000 years. Palaeogeography,
Palaeoclimatology, Palaeoecology, 313–314, 127–139.
Briffa, K. R., Jones, P. D., and Schweingruber, F. H., 1988. Summer
temperature patterns over Europe: a reconstruction from 1750
A.D. based on maximum latewood density indices of conifers.
Quaternary Research, 30(1), 36–52.
Briffa, K. R., Jones, P. D., Schweingruber, F. H., and Osborn, T. J.,
1998. Influence of volcanic eruptions on northern Hemisphere
summer temperature over the past 600 years. Nature, 393,
450–454.
Cherubini, P., Humbel, T., Beeckman, H., Gärtner, H., Mannes, D.,
Pearson, C. L., Schoch, W., Tognetti, R., and Lev-Yadun, S.,
2013. Olive tree-ring problematic dating: a comparative analysis
on Santorini (Greece). PLoS ONE, 8(1), e54730 (54735 pages).
Cleland, C. E., 2002. Methodological and epistemic differences
between historical science and experimental science. Philosophy
of Science, 69, 474–496.
Cook, A. C., Hainsworth, L. J., Sorey, M. L., Evans, W. C., and
Southon, J. R., 2001. Radiocarbon studies of plant leaves and
tree rings from Mammoth Mountain, CA: a long-term record of
magmatic CO2 release. Chemical Geology, 177, 117–131.
D’Arrigo, R. D., Wilson, R. J. S., and Jacoby, G. C., 2006. On the
long-term context for late twentieth century warming. Journal
of Geophysical Research, 111, 1–12, D03103,
doi:03110.01029/02005JD006352.
D’Arrigo, R. D., and Jacoby, G. C., 1999. Northern north American
tree-ring evidence for regional temperature changes after major
volcanic events. Climatic Change, 41, 1–15.
D’Arrigo, R. D., Free, R. M., and Jacoby, G. C., 1992. Tree-ring
width and maximum latewood density at the North American
tree line: parameters of climatic change. Canadian Journal of
Forest Research, 22(9), 1290–1296.
D’Arrigo, R., Frank, D., Jacoby, G., and Pederson, N., 2001. Spatial
response to major volcanic events in or about AD 536, 934 and
1258: frost rings and other dendrochronological evidence from
Mongolia and northern Siberia: comment on R. B. Stothers,
‘Volcanic dry fogs, climate cooling, and plague pandemics in
Europe and the middle east’ (Climatic Change, 42, 1999). Cli-
matic Change, 49(1–2), 239–246.
D’Arrigo, R., Wilson, R., and Anchukaitis, K. J., 2013. Volcanic
cooling signal in tree ring temperature records for the past mil-
lennium. Journal of Geophysical Research: Atmospheres,
118(16), 9000–9010.
Day, W. R., and Peace, T. R., 1934. The experimental production
and the diagnosis of frost injury on forest trees. Oxford Forestry
Memoirs, 16, 1–60.
225
Druce, A. P., 1966. Tree-ring dating of recent volcanic ash and
lapilli, Mt Egmont. New Zealand Journal of Botany, 4(1), 3–41.
Eggler, W. A., 1967. Influence of volcanic eruptions on xylem
growth patterns. Ecology, 48(4), 644–647.
Esper, J., Büntgen, U., Luterbacher, J., and Krusic, P. J., 2013. Test-
ing the hypothesis of post-volcanic missing rings in temperature
sensitive dendrochronological data. Dendrochronologia, 31(3),
216–222.
Filion, L., Payette, S., Gauthier, L., and Boutin, Y., 1986. Light rings
in subarctic conifers as a dendrochronological tool. Quaternary
Research, 26, 272–279.
Fonti, P., von Arx, G., García-Gonzlez, I., Eilmann, B., Sass-
Klaassen, U., Gärtner, H., and Eckstein, D., 2010. Studying
global change through investigation of the plastic responses of
xylem anatomy in tree rings. New Phytologist, 185, 42–53.
Franklin, B., 1785. Meteorological imaginations and conjectures.
In Sparks, J. (ed.), The Works of Benjamin Franklin. Boston,
MA: Charles Tappan, pp. 455–457.
Friedrich, W. L., Kromer, B., Friedrich, M., Heinemeier, J., Pfeiffer,
T., and Talamo, S., 2006. Santorini eruption radiocarbon dated to
1627–1600 B.C. Science, 312(5773), 548.
Fritts, H. C., 1976. Tree Rings and Climate. London: Academic Press.
Fritts, H. C., and Swetnam, T. W., 1989. Dendroecology: a tool for
evaluating variations in past and present forest environments. In
Begon, M., Fitter, A. H., Ford, E. D., and MacFadyen, A. (eds.),
Advances in Ecological Research. New York: Academic Press,
pp. 111–188.
Glerum, C., and Farrar, J. L., 1966. Frost ring formation in the stems
of some coniferous species. Canadian Journal of Botany, 44(7),
879–886.
Hantemirov, R. M., Gorlanova, L. A., and Shiyatov, S. G., 2004.
Extreme temperature events in summer in northwest Siberia
since AD 742 inferred from tree rings. Palaeogeography,
Palaeoclimatology, Palaeoecology, 209(1–4), 155–164.
Hinckley, T. M., Imoto, H., Lee, K., Lacker, S., Morikawa, Y., Vogt,
K. A., Grier, C. C., Keyes, M. R., Teskey, R. O., and Seymour,
V. A., 1984. Impact of tephra deposition on growth in conifers:
the year of the eruption. Canadian Journal of Forest Research,
14, 731–739.
Jacoby, G. C., 1997. Application of tree ring analysis to
paleoseismology. Reviews of Geophysics, 35(2), 109–124.
Jacoby, G. C., Workman, K. W., and D’Arrigo, R. D., 1999. Laki
eruption of 1783, tree rings, and disaster for northwest Alaska
Inuit. Quaternary Science Reviews, 18, 1365–1371.
Jones, P. D., Briffa, K. R., and Schweingruber, F. H., 1995. Tree-
ring evidence of the widespread effects of explosive volcanic
eruptions. Geophysical Research Letters, 22(11), 1333–1336.
Kirchner, P., Biondi, F., Edwards, R., and McConnell, J. R., 2008.
Variability of trace metal concentrations in Jeffrey pine (Pinus
jeffreyi) tree rings from the Tahoe Basin, California, USA. Jour-
nal of Forest Research, 13(6), 347–356.
Krakauer, N. Y., and Randerson J. T., 2003. Do volcanic eruptions
enhance or diminish net primary production? Evidence from tree
rings. Global Biogeochemical Cycles, 17(4), 1118, article 1129
(1111 pp).
Kuniholm, P. I., Kromer, B., Manning, S. W., Newton, M., Latini,
C. E., and Bruce, M. J., 1996. Anatolian tree rings and the abso-
lute chronology of the eastern Mediterranean, 2220–718
B.C. Nature, 381, 780–783.
LaMarche, V. C., Jr., and Hirschboeck, K. K., 1984. Frost rings in trees
as records of major volcanic eruptions. Nature, 307(5946), 121–126.
Larsen, L. B., Vinther, B. M., Briffa, K. R., Melvin, T. M., Clausen,
H. B., Jones, P. D., Siggaard-Andersen, M. -L., Hammer, C. U.,
Eronen, M., Grudd, H., Gunnarson, B. E., Hantemirov, R. M.,
Naurzbaev, M. M., and Nicolussi, K., 2008. New ice core evi-
dence for a volcanic cause of the A.D. 536 dust veil. Geophysical
Research Letters, 35, article L04708 (04705 pp).
226 DENDROCHRONOLOGY, VOLCANIC ERUPTIONS
Larson, P. R., 1994. The Vascular Cambium: Development and
Structure. Berlin: Springer.
Liang, C., Filion, L., and Cournoyer, L., 1997. Wood structure of
biotically and climatically induced light rings in eastern larch
(Larix laricina). Canadian Journal of Forest Research, 27(10),
1538–1547.
Lough, J. M., and Fritts, H. C., 1987. An assessment of the possible
effects of volcanic eruptions on North American climate using
tree-ring data, 1602 to 1900 A.D. Climatic Change, 10,
219–239.
Mahler, R. L., and Fosberg, M. A., 1983. The influence of Mount
St. Helens volcanic ash on plant growth and nutrient uptake. Soil
Science, 135, 197–201.
Mann, M. E., Fuentes, J. D., and Rutherford, S., 2012. Underestima-
tion of volcanic cooling in tree-ring-based reconstructions of
hemispheric temperatures. Nature Geoscience, 5, 202–205.
Manning, S. W., Ramsey, C. B., Kutschera, W., Higham, T.,
Kromer, B., Steier, P., and Wild, E. M., 2006. Chronology for
the Aegean Late Bronze Age 1700–1400 B.C. Science,
312(5773), 565–569.
McLean, S., and Lockridge, P., 2000. A Teachers Guide to Strato-
volcanoes of the World. Boulder, CO: National Geophysical
Data Center, 64 p.
Minnis, P., Harrison, E. F., Stowe, L. L., Gibson, G. G., Denn, F. M.,
Doelling, D. R., and Smith, W. L., 1993. Radiative climate forc-
ing by the Mount Pinatubo eruption. Science, 259(5100),
1411–1415.
Nash, S. E., 1999. Time, Trees, and Prehistory: Tree-Ring Dating
and the Development of North American Archaeology,
1914–1950. Salt Lake City, UT: The University of Utah Press.
Newhall, C. G., and Self, S., 1982. The volcanic explosivity index
(VEI) an estimate of explosive magnitude for historical volca-
nism. Journal of Geophysical Research: Oceans, 87(C2),
1231–1238.
Pearson, C. L., Manning, S. W., Coleman, M., and Jarvis, K., 2005.
Can tree-ring chemistry reveal absolute dates for past volcanic
eruptions? Journal of Archaeological Science, 32, 1265–1274.
Pearson, C. L., Dale, D. S., Brewer, P. W., Kuniholm, P. I., Lipton, J.,
and Manning, S. W., 2009. Dendrochemical analysis of a tree-ring
growth anomaly associated with the Late Bronze Age eruption of
Thera. Journal of Archaeological Science, 36, 1206–1214.
Plomion, C., Leprovost, G., and Stokes, A., 2001. Wood formation
in trees. Plant Physiology, 127(4), 1513–1523.
Plummer, C. T., Curran, M. A. J., van Ommen, T. D., Rasmussen,
S. O., Moy, A. D., Vance, T. R., Clausen, H. B., Vinther, B. M.,
and Mayewski, P. A., 2012. An independently dated 2000-yr
volcanic record from Law Dome, East Antarctica, including
a new perspective on the dating of the 1450s CE eruption
of Kuwae, Vanuatu. Climate of the Past, 8(6), 1929–1940.
Polge, H., 1970. The use of X-ray densitometric methods in dendro-
chronology. Tree-Ring Bulletin, 30(1–4), 1–10.
Ramanathan, V., 1988. The greenhouse theory of climate change:
a test by an inadvertent global experiment. Science, 240(4850),
293–299.
Rhoads, A. S., 1923. The Formation and Pathological Anatomy of
Frost Rings in Conifers Injured by Late Frosts. Washington,
DC: U.S. Department of Agriculture,15 p.
Robock, A., and Mao, J., 1995. The volcanic signal in surface tem-
perature observations. Journal of Climate, 8(5), 1086–1103.
Rossi, S., Morin, H., and Deslauriers, A., 2012. Causes and
correlations in cambium phenology: towards an integrated
framework of xylogenesis. Journal of Experimental Botany,
63, 2117–2126.
Rossi, S., Anfodillo, T., Čufar, K., Cuny, H. E., Deslauriers, A.,
Fonti, P., Frank, D., Gričar, J., Gruber, A., King, G. M.,
Krause, C., Morin, H., Oberhuber, W., Prislan, P., and Rathgeber,
C. B. K., 2013. A meta-analysis of cambium phenology and
growth: linear and non-linear patterns in conifers of the northern
hemisphere. Annals of Botany, 112(9), 1911–1920.
Salzer, M. W., and Hughes, M. K., 2007. Bristlecone pine tree rings
and volcanic eruptions over the last 5000 yr. Quaternary
Research, 67, 57–68.
Schweingruber, F. H., Fritts, H. C., Bräker, O. U., Drew, L. G., and
Schaer, E., 1978. The X-ray technique as applied to dendrocli-
matology. Tree-Ring Bulletin, 38, 61–91.
Schweingruber, F. H., Briffa, K. R., and Jones, P. D., 1991. Yearly
maps of summer temperatures in Western Europe from
A.D. 1750 to 1975 and Western North America from 1600 to
1982. Vegetatio, 92(1), 5–71.
Segura, G., Brubaker, L. B., Franklin, J. F., Hinckley, T. M.,
Maguire, D. A., and Wright, G., 1994. Recent mortality and
decline in mature Abies amabilis: the interaction between site
factors and tephra deposition from Mount St. Helens. Canadian
Journal of Forest Research, 24, 1112–1122.
Segura, G., Hinckley, T. M., and Oliver, C. D., 1995. Stem growth
responses of declining mature Abies amabilis trees after tephra
deposition from Mount St. Helens. Canadian Journal of Forest
Research, 25, 1493–1502.
Sheppard, P. R., May, E. M., Ort, M. H., Anderson, K. C., and
Elson, M. D., 2005. Dendrochronological responses to the
24 October 1992 tornado at Sunset Crater, northern Arizona.
Canadian Journal of Forest Research, 35(12), 2911–2919.
Sheppard, P. R., Ort, M. H., Anderson, K. C., Elson, M. D., Vzquez-
selem, L., Clemens, A. W., Little, N. C., and Speakman, R. J.,
2008. Multiple dendrochronological signals indicate the Erup-
tion of ParíCutin Volcano, Michoacn, Mexico. Tree-Ring
Research, 64(2), 97–108.
Smiley, T. L., 1958. The geology and dating of Sunset Crater,
Flagstaff, Arizona. In Anderson, R. Y., and Harshberger, J. W.
(eds.), Guidebook of the Black Mesa Basin, Northeastern
Arizona. Socorro, NM: New Mexico Geological Society,
pp. 186–190.
Speer, J. H., 2010. Fundamentals of Tree-Ring Research. Tucson,
AZ: University of Arizona Press.
St. George, S., Ault, T. R., and Torbenson, M. C. A., 2013. The rar-
ity of absent growth rings in northern Hemisphere forests outside
the American Southwest. Geophysical Research Letters, 40(14),
3727–3731.
Stothers, R. B., 1984. The great Tambora eruption in 1815 and its
aftermath. Science, 224(4654), 1191–1198.
Stothers, R., 1999. Volcanic dry fogs, climate cooling, and plague
pandemics in Europe and the Middle East. Climatic Change,
42(4), 713–723.
Szeicz, J. M., 1996. White spruce light rings in Northwestern Can-
ada. Arctic and Alpine Research, 28(2), 184–189.
Vinther, B. M., Clausen, H. B., Johnsen, S. J., Rasmussen, S. O.,
Andersen, K. K., Buchardt, S. L., Dahl-Jensen, D., Seierstad,
I. K., Siggaard-Andersen, M. L., Steffensen, J. P., Svensson,
A., Olsen, J., and Heinemeier, J., 2006. A synchronized
dating of three Greenland ice cores throughout the Holocene.
Journal of Geophysical Research: Atmospheres, 111(D13),
D13102.
Watt, S. F. L., Pyle, D. M., Mather, T. A., Day, J. A., and Aiuppa, A.,
2007. The use of tree-rings and foliage as an archive of
volcanogenic cation deposition. Environmental Pollution, 148,
48–61.
 DENDROCHRONOLOGY, VOLCANIC ERUPTIONS 227

Webb, G. E., 1983. Tree Rings and Telescopes: The Scientific Yamaguchi, D. K., and Hoblitt, R. P., 1995. Tree-ring dating of
Career of A.E. Douglass. Tucson, AZ: The University of pre-1980 volcanic flowage deposits at Mount St. Helens, Wash-
Arizona Press. ington. Geological Society of America Bulletin, 107(9),
Wiener, M. H., 2009. The state of the debate about the date of the 1077–1093.
Theran eruption. In Warburton, D. A. (ed.), Time’s Up! Dating Zielinski, G. A., 2000. Use of paleo-records in determining variabil-
the Minoan eruption of Santorini. Århus, Denmark: The Danish ity within the volcanism–climate system. Quaternary Science
Institute at Athens, pp. 197–206. Reviews, 19(1–5), 417–438.
Wiener, M. H., 2010. A point in time. In Krzyszkowska, O. (ed.),
Cretan Offerings: Studies in Honour of Peter Warren. London:
British School at Athens. British School at Athens Studies,
Vol. 18, pp. 367–394.
Yamaguchi, D. K., 1983. New tree-ring dates for recent eruptions of Cross-references
Mount St. Helens. Quaternary Research, 20, 246–250. Dendrochronology, Dwellings
Yamaguchi, D. K., 1985. Tree-ring evidence for a two-year interval Dendrochronology, Entomology
between recent prehistoric explosive eruptions of Mount Dendrochronology, Fire Regimes
St. Helens. Geology, 13, 554–557. Dendrochronology, Progress
Yamaguchi, D. K., 1993. Old-growth forest development after Dendrochronology, Surficial Processes
Mount St Helens’ 1480 eruption. National Geographic Research Radiocarbon Dating
& Exploration, 9(3), 294–325. Tephrochronology
 E
EARLY LIFE ON EARTH
David Wacey
Centre for Microscopy Characterisation and Analysis,
University of Western Australia, Crawley, Perth, Australia
Definition
The most obvious signs of early life on Earth are in the
form of body fossils. On the early Earth, only the most
primitive of microorganisms lacking any hard parts would
have been present, so only soft parts in the form of cells,
sheaths, and perhaps associated secreted polymeric sub-
stances are likely to be found. Alternatively, signs of life
may be in the form of trace fossils. These are nonbody
remains that in the broadest sense indicate the activity of
an organism (e.g., a microboring) or a community of
organisms (e.g., a stromatolite or microbially induced sed-
imentary structure). A third way early life may be recog-
nized is in the form of chemical fossils. These are traces
of biological activity indicated by specific chemical sig-
nals left in rocks. Chemical fossils may be preserved in
organic matter or in “biominerals” such as pyrite; they
include isotopic variations in carbon, sulfur, nitrogen, or
iron, distinctive ratios of elements, or molecular com-
pounds that may be tied to a particular group of organisms.
Protocols for identifying early life on earth:
where to look
This is not merely a case of looking at Earth’s oldest rocks.
As well as being demonstrably of the correct age, such
rocks must have been deposited in a setting that could via-
bly host life and ideally should have undergone minimal
postdepositional alteration. When these factors are taken
J.W. Rink, J.W. Thompson (eds.), Encyclopedia of Scientific Dating Methods, DOI 10.1007/978-94-007-6304-3,
© Springer Science+Business Media Dordrecht 2015
into consideration, the search for Earth’s earliest life is
focused on two main areas of the world: the Pilbara Craton
of Western Australia and the Barberton greenstone belt of
southern Africa, both containing sedimentary and volca-
nic rocks up to c. 3.5 billion years old (Ga [giga-annum]).
In addition, parts of Greenland contain putative sedimen-
tary rocks approaching 3.8 Ga, but these have been altered
(metamorphosed) at higher temperatures and strain rates
making the identification of life more difficult and
controversial.
Protocols for identifying early life on earth:
establishing antiquity
Proving that a putative life signature is of undoubted early
Archean (i.e., >c. 3 Ga) age is a vital step in testing any
claim for Earth’s earliest life. Hence, a number of antiquity
criteria have been put forward that any life sign should sat-
isfy. These include being able to demonstrate the age and
provenance of the host rocks, that the candidate life sign
is both physically embedded within the host rock (indige-
nous) and not introduced by later fluids (syngenetic), and
that it has a composition (carbonaceous or non-hydrated
mineral) and color (i.e., black to dark brown if carbona-
ceous) consistent with the deposition and alteration
history of the host rock (see Wacey, 2009, for detailed
criteria).
Protocols for identifying early life on earth:
establishing biogenicity
Criteria have also been put forward to minimize the possi-
bility of interpreting a nonbiological structure as biologi-
cal. Thus, candidate microfossils should exhibit the
following: biological morphology that can be related to
extant cells, sheaths, or waste products; more than
230 EARLY LIFE ON EARTH

Early Life on Earth, Table 1 Summary of reported evidence for life in >3 Ga rocks in the Pilbara region of Western Australia

Age Geological unit Morphological evidence Geochemical evidence References

~3 Ga Farrel Quartzite, Microfossils: spheroids, 2.5–80 mm in Bulk carbon isotopes: d C c.
30 % to

13
Sugitani et al.,
Gorge Creek diameter; lens/spindle shapes up to 40 mm in
35 % consistent with biology 2009
Group length and 35 mm in width; threadlike Laser Raman: fossils are made of
filaments <1 mm diameter and up to 100 mm carbon and carbon structure is
in length; potential carbonaceous biofilms consistent with age and biology
(see Figure 1j) Elemental mapping: C, N, and S found
in microfossil walls
~3.24 Ga Kangaroo Caves Microfossils: solid pyritic filaments 0.5–2 mm None Rasmussen,
Formation, Sulfur in diameter and up to 300 mm in length, often 2000
Springs Group intertwined and densely clustered (see
Figure 1h)
Microfossils: filaments and tubes <1 mm wide, Bulk carbon isotopes: Duck et al.,
often bundled in packages 1–5 mm in d13C c.
27 % to
34 % consistent 2007
diameter and up to 100 mm in length; with biology
aggregates of tiny spheres <50–100 nm in
diameter
~3.35 Ga Euro Basalt, Kelly Trace fossils: microtubes 1–5 mm in diameter Elemental mapping: possible Banerjee
Group and up to 150 mm in length, often branched enrichments of C and N within the et al., 2007
and segmented microtubes
Microfossils: spheroids c. 8 mm in diameter; None Schopf and
spheroids c. 21 mm in diameter sometimes Packer,
enclosed by a potential sheath 1987
~3.4 Ga Strelley Pool Microfossils: carbonaceous spheroids (c. 5–25 In situ carbon isotopes: d13C
33 % to Wacey et al.,
Formation mm in diameter), tubes (c. 7–20 mm in
46 % consistent with biology 2011a
diameter), and rare cellular envelopes (up to In situ sulfur isotopes: d34S range of
80 mm in diameter), associated with micron- 18 % and D33S of +1.43 % to
sized pyrite. Spheroids often cluster and
1.65 % indicating sulfur-processing
organize into chains. Microfossils colonize metabolisms
sand grains and some show taphonomic Elemental mapping: nitrogen
degradation (see Figure 1f) enrichment in microfossil walls
Laser Raman: carbon structure
consistent with age and biology
Stromatolites: best examples are conical and Elemental distributions: co-occurrence Allwood
columnar morphologies with trapped and of C, N, and S in detrital carbonaceous et al., 2006
reworked grains (see Figure 1d) grains; rare earth element enrichment
in carbonate laminae relative to chert
laminae is consistent with younger
microbial carbonates
Microfossils: lenticular/spindle shapes (long Bulk carbon isotopes: d13C c.
31 % to Sugitani et al.,
axis c. 35–120 mm), large spheroids (up to
37 % consistent with biology 2010
c. 100 mm diameter), clusters of smaller Laser Raman: shows microfossils are
spheroids, wrinkled and folded carbonaceous carbonaceous and carbon has structure
films, and threadlike filaments consistent with age and biology
Trace fossils: microborings (channels and Elemental mapping: C and Wacey et al.,
spherical to elliptical surface pits, with near N enrichments in some microborings 2011b
constant diameter of 1–5 mm) often Laser Raman: shows carbonaceous
concentrated in clumps along one side of biofilms associated with some
a detrital pyrite grain (see Figure 1e) microborings; carbon has structure
consistent with age and biology
Possible trace fossils: microtubes containing In situ carbon isotopes: mean d13C of Wacey et al.,
a pyrite grain at their termination – known as c.
26 % from carbonaceous 2008
ambient inclusion trails microtube linings
Elemental mapping: enrichments in C,
N, P, and S in some microtubes
~3.45 Ga Panorama Microbial mat with microfossils: spheroids Bulk carbon isotopes: d13C c.
26 % to Westall et al.,
Formation, (<1 mm in diameter), often clustered or in
30 % consistent with biology 2006b
Warrawoona chains; filaments (<1 mm diameter and up to Laser Raman: spectra indicate carbon
Group 40 mm in length); occasional rod shapes; with a structure consistent with age
films interpreted as fossilized extracellular and biology
polymeric substances (see Figure 1c)
 EARLY LIFE ON EARTH 231

Early Life on Earth, Table 1 (Continued )

Age Geological unit Morphological evidence Geochemical evidence References

~3.46 Ga Apex Basalt, Possible microfossils: filaments – eleven 13

Bulk carbon isotopes: d C averages Schopf, 1993
Warrawoona “species” varying in average diameter from
27.7 % consistent with biology
Group 0.5–16.5 mm and 28–89 mm in length Laser Raman: spectra show objects are
made of carbon and structure
consistent with biology
BUT: many are C-, J-, and L-shaped structures BUT: carbon isotope and Raman signals Brasier et al.,
around spherulitic silica crystals; some are also consistent with abiotic 2002
follow ghost rhombic crystal outlines carbonaceous artifacts; some may not
be carbonaceous
~3.49 Ga Dresser Formation, Microfossils: 3 types of filaments: (a) average 1 Bulk and in situ carbon isotopes: d13C Ueno et al.,
Warrawoona mm in diameter, some mutually interwoven c.
30% to
42 % consistent with 2001
Group and branched; (b) segmented, unbranched, biology
average 8.7 mm in diameter; (c) non-
segmented, tubular, average 12.6 mm in
diameter
Stromatolites: coniform, nodular, domical and None Van
wavy-laminated morphotypes (see Figure 1a)
Kranendonk et al., 2008
Microfossils: entire “hollow cells” and Bulk carbon isotopes: d13C c.
32 % to Glickson
fragments such as “cell walls” observed
36 % consistent with biology et al., 2008
under the transmission electron microscope
Possible biominerals: microscopic pyrite (no Bulk and in situ sulfur isotopes: d34S Shen et al.,
biological morphology) range of 22%; D33S of +6 % to 2009

1.2 % consistent with sulfur-
processing biological metabolisms
Methane inclusions: no Carbon isotopes: d13C
36 % to
56 % Ueno et al., 2006
in biological consistent with microbial methanogenesis
fluid morphology

a single step of biology-like processing, perhaps in
the form of associated biominerals, geochemical fraction-
ations of isotopes, specific organic compounds, or
distinctive elemental ratios; a geological and evolutionary
context that is plausible for life; abundance, ideally
occurring in a multicomponent assemblage showing com-
munity behavior; kerogenous carbon composition or com-
posed of microbially mediated minerals such as pyrite;
a largely hollow nature and occupying a restricted biolog-
ical morphospace; evidence for taphonomic degradation
such as collapse of cells, folding of films, or fracturing;
and dissimilarity from known potentially coexisting
nonbiological organic bodies (e.g., self-organizing spher-
ulitic structures).
Similarly, trace fossils such as microbial borings should
exhibit the following: circular to elliptical cross sections
and a restricted range of diameters; preferential exploita-
tion of certain substrates or horizons; enrichments of bio-
logically important elements, such as carbon and/or
nitrogen enriched linings, or biomineral infillings; and
preferred growth orientations. Stromatolites must also ful-
fill a set of criteria in an attempt to exclude abiotic mimics
such as soft sediment deformation, later structural defor-
mation, and chemical precipitation of stromatolite-like
crusts (see Wacey, 2009, for detailed criteria).
Protocols for identifying early life on earth:
techniques
Fulfilling the criteria outlined above requires the use of
a wide range of techniques that permit investigation from
the kilometer to nanometer scale. Geological mapping at
scales from kilometers to meters, supported by mapping
of petrographic thin sections (c. 30–100 mm-thick rock
slices), is required to show that candidate life structures
are truly syngenetic and ancient. This contextural and pet-
rographic mapping is also the first stage in falsifying the
null hypothesis of a nonbiological origin for a given
object.
Obtaining evidence for a uniquely biological morphol-
ogy requires in situ imaging and mapping of morphospace
to distinguish the fields of biological and nonbiological
morphology and to compare these with self-organizing
structures. This is commonly undertaken under the optical
microscope using petrographic thin sections but with
recent technological advances may also extend to investi-
gation using a scanning electron microscope (SEM), trans-
mission electron microscope (TEM), lasers, or X-rays.
One recent advance is the ability to extract site-specific
ultrathin (c.100 nm) wafers from a petrographic thin sec-
tion or rock chip using a focused ion beam (FIB), allowing
the morphology of candidate life signs to be studied at
232 EARLY LIFE ON EARTH

Early Life on Earth, Table 2 Summary of reported evidence for life in >3 Ga rocks in the Barberton region of South Africa

Age Geological unit Morphological evidence Geochemical evidence References

13
~3.22 Ga Moodies Group Microbially induced sedimentary structures: Bulk carbon isotopes: d C around Noffke et al.,
wrinkle and roll-up structures; aligned grains in
20 % consistent with biology 2006
a matrix of carbonaceous laminae (see Figure 1i)
~3.25 Ga Fig Tree Group Stromatolites: laterally linked domes, pseudo- None Byerly et al.,
columns, and crinkly stratiform varieties 1986
Putative microfossils: spheroidal 1–4 mm in None Knoll and
diameter; possible internal contents; possible Barghoorn,
binary division 1977
~3.3–3.4 Kromberg Microbial mat: carbonaceous laminations with Bulk carbon isotopes: d C of
25 % to Tice and
13

Ga Formation, portions “ripped up” and/or folded over; some
35 % consistent with biology Lowe,
Onverwacht filaments preserved (see Figure 1g) 2004
Group Carbonaceous grains: possible eroded fragments Bulk carbon isotopes: d C of
25 % to Tice and
13

of a microbial mat
35 % consistent with biology Lowe,

2004
Putative microfossils: a selection of <2 mm Bulk carbon isotopes: d13C of
27 % Westall et al.,
spheroidal and rodlike morphologies, often consistent with biology 2001
clustered or paired
Putative microfossils: tubular and solid None Walsh, 1992
carbonaceous and pyritic filaments up to 2.5 mm
diameter and 200 mm long, plus large spindle
shapes
~3.45 Ga Hoogenoeg Trace fossils: segmented microtubes ~ 4 mm in I n situsulfur isotopes: d34S of
40 % to McLoughlin,
Formation, diameter and up to 200 mm in length (see
3% from pyrite in microtubes et al. 2012
Onverwacht Figure 1b), plus clustered spheres indicates sulfur-based metabolism
Group Bulk carbon isotopes (carbonate): d13C Furnes et al.,
of
16 % to +4 % 2004
Putative microfossils: solid filaments up to 2.6 mm None Walsh and
diameter and 200 mm long Lowe,
1985
Putative microfossils: sausage-shaped, up to 3.8 None Westall et al.,
mm long and 1.1 mm wide 2006a
Putative microbial mat: crinkly carbonaceous None Walsh, 1992
laminations; some bundled, broken, or folded
over; carbonaceous grains
Microbial mat: comprising parallel and interwoven Bulk carbon isotopes: d13C of
23 % Westall et al.,
filaments; mat was flexible consistent with biology 2006a

nanometer-scale spatial resolution in the TEM. FIB com-
bined with SEM also permits 3D modeling and recon-
struction of candidate microfossils at nanometer-scale
resolution. Laser Raman microspectroscopy and X-ray
tomography also permit visualization of candidate micro-
fossils in 3D albeit at poorer spatial resolution.
Finally, obtaining geochemical evidence for metabolic
cycling requires micro- to nanoscale in situ mapping and
analysis. Here, potential biological isotope fractionations
can be detected using secondary ion mass spectrometry
(SIMS); high sensitivity and high spatial resolution ele-
mental mapping can be performed using NanoSIMS,
TEM, or synchrotron. The structure of carbon or more
complex organic material can be investigated using laser
Raman microspectroscopy, TEM, or synchrotron, and
the oxidation and bonding states of some biologically
important elements such as Fe and S can also be resolved
using TEM or synchrotron (see Wacey, 2009, 2012, for
details on techniques).
Examples of earth’s earliest life
Reported examples of > 3 Ga life on Earth are summa-
rized in Tables 1 and 2, plus Figure 1.
Summary
Possible signs of life have been reported from eleven
separate geological formations and from at least eight dif-
ferent lithologies in the Pilbara and Barberton regions,
spanning almost the entire exposed 3–3.5 Ga stratigraphy.
Evidence for life includes stromatolites, microfossils,
trace fossils, biominerals, and isotopic signatures and
comes from environments as diverse as a carbonate plat-
form, sandy beach, and deep-sea hydrothermal black
smoker. Putative life signals, in the form of isotopic signa-
tures, have also been reported from rocks as old as
c. 3.8 Ga in Greenland but these are particularly
controversial.
 EARLY LIFE ON EARTH 233

Early Life on Earth, Figure 1 Examples of early life on Earth. (a) Wavy-laminated and domical stromatolites from the c. 3.49 Ga
Dresser Formation, Australia (Penknife for scale; photo courtesy of M. van Kranendonk). (b) Microtubular structures interpreted as
endolithic trace fossils from the outer margin of a pillow basalt from the c. 3.45 Ga Hoogenoeg Formation, South Africa (Photo
courtesy of N. McLoughlin and reproduced under the fair usage policy of the Geological Society of America). (c) Filaments (f), rods (r),
coccoids (c), and possible extracellular polymeric substances from the c. 3.45 Ga Panorama Formation, Australia (Photo reproduced
from Westall et al., 2006b, under the fair usage policy of the Geological Society of America). (d) Complex stromatolite from the
c. 3.4 Ga Strelley Pool Formation, Australia (Photo by author). (e) Microbial trace fossils within pyrite from the c. 3.4 Ga Strelley Pool
Formation (Modified from Wacey et al., 2011b). (f) Tubular microfossils from the c. 3.4 Ga Strelley Pool Formation (Photo by author).
(g) Carbonaceous grains and folded microbial mat from the c. 3.4 Ga Kromberg Formation, South Africa (Photo by author).
(h) Putative filamentous microbes preserved in pyrite from the c. 3.24 Ga Kangaroo Caves Formation, Australia (Photo by author).
(i) Microbially induced sedimentary structures (MISS) from the c. 3.22 Ga Moodies Group, South Africa (Camera lens cap for scale;
photo courtesy of N. Noffke). (j) Folded and torn biofilm from the c. 3.0 Ga Farrel Quartzite, Australia (Photo courtesy of K. Sugitani).

Bibliography Byerly, G. R., Lowe, D. L., and Walsh, M. M., 1986. Stromatolites
from the 3,300–3,500-Myr Swaziland Supergroup, Barberton
Allwood, A. C., Walter, M. R., Kamber, B. S., Marshall, C. P., and
Mountain Land, South Africa. Nature, 319, 489–491.
Burch, I. W., 2006. Stromatolite reef from the Early Archaean
Duck, L. J., Glikson, M., Golding, S. D., and Webb, R. E., 2007.
era of Australia. Nature, 441, 714–718.
Microbial remains and other carbonaceous forms from the 3.24
Banerjee, N. R., Simonetti, A., Furnes, H., Muehlenbachs, K.,
Ga Sulphur Springs black smoker deposit, Western Australia.
Staudigel, H., Heaman, L., and Van Kranendonk, M. J.,
Precambrian Research, 154, 205–220.
2007. Direct dating of microbial ichnofossils. Geology, 35,
Furnes, H., Banerjee, N. R., Muehlenbachs, K., Staudigel, H., and
487–490.
de Wit, M., 2004. Early life recorded in Archean pillow lavas.
Brasier, M. D., Green, O. R., Jephcoat, A. P., Kleppe, A. K., Van
Science, 304, 578–581.
Kranendonk, M. J., Lindsay, J. F., Steele, A., and Grassineau,
Glickson, M., Duck, L. J., Golding, S. D., Hofmann, A., Bolhar, R.,
N. V., 2002. Questioning the evidence for Earth’s oldest fossils.
Webb, R., Baiano, J. C. F., and Sly, L. I., 2008. Microbial
Nature, 416, 76–81.
234 ELECTRON SPIN RESONANCE (ESR) DATING OF CORAL
remains in some earliest Earth rocks: comparison with a potential
modern analogue. Precambrian Research, 164, 187–200.
Knoll, A. H., and Barghoorn, E. S., 1977. Archean microfossils
showing cell division from the Swaziland System of South
Africa. Science, 198, 396–398.
McLoughlin, N., Grosch, E. G., Kilburn, M. R., and Wacey, D.,
2012. Sulfur isotope evidence for a Paleoarchean sub-seafloor
biosphere, Barberton, South Africa. Geology, 40, 1031–1034.
Noffke, N., Eriksson, K. A., Hazen, R. M., and Simpson, E. L.,
2006. A new window into Early Archean life: Microbial mats
in Earth’s oldest siliciclastic tidal deposits (3.2 Ga Moodies
Group, South Africa). Geology, 34, 253–256.
Rasmussen, B., 2000. Filamentous microfossils in a 3235-million-
year-old volcanogenic massive sulphide deposit. Nature, 405,
676–679.
Schopf, J. W., 1993. Microfossils of the Early Archaean Apex
Chert: new evidence for the antiquity of life. Science, 260,
640–646.
Schopf, J. W., and Packer, B. M., 1987. Early Archaean (3.3 Billion
to 3.5 Billion-year-old) microfossils from Warrawoona Group,
Australia. Science, 237, 70–73.
Shen, Y., Farquhar, J., Masterson, A., Kaufman, A. J., and Buick,
R., 2009. Evaluating the role of microbial sulfate reduction in
the early Archean using quadruple isotope systematics. Earth
and Planetary Science Letters, 279, 383–391.
Sugitani, K., Grey, K., Nagaoka, T., Mimura, K., and Walter, M. R.,
2009. Taxonomy and biogenicity of Archaean spheroidal micro-
fossils (ca. 3.0 Ga) from the Mount Goldsworthy-Mount Grant
area in the northeastern Pilbara Craton, Western Australia.
Precambrian Research, 173, 50–59.
Sugitani, K., Lepot, K., Nagaoka, T., Mimura, K., Van Kranendonk,
M., Oehler, D. Z., and Walter, M. R., 2010. Biogenicity of mor-
phologically diverse carbonaceous microstructures from the ca.
3400 Ma Strelley Pool Formation, in the Pilbara Craton, Western
Australia. Astrobiology, 10, 899–920.
Tice, M. M., and Lowe, D. R., 2004. Photosynthetic microbial mats
in the 3,416-Myr-old ocean. Nature, 431, 549–552.
Ueno, Y., Isozaki, Y., Yurimoto, H., and Maruyama, S., 2001.
Carbon isotopic signatures of individual Archean microfossils
(?) from Western Australia. International Geology Review, 43,
196–212.
Ueno, Y., Yamada, K., Yoshida, N., Maruyama, S., and Isozaki, Y.,
2006. Evidence from fluid inclusions for microbial
methanogenesis in the early Archaean era. Nature, 440,
516–519.
Van Kranendonk, M. J., Phillipot, P., Lepot, K., Bodorkos, S., and
Pirajno, F., 2008. Geological Setting of Earth’s oldest fossils in
the ca. 3.5Ga Dresser Formation, Pilbara Craton, Western
Australia. Precambrian Research, 167, 93–124.
Wacey, D., 2009. Early Life on Earth: A Practical Guide.
Dordrecht: Springer. 285p.
Wacey, D., 2012. In situ, high spatial resolution techniques in
the search for the origin of life. In Seckbach, J. (ed.), Cellular
Origin, Life in Extreme Habitats and Astrobiology, 22,
Dordrecht: Springer pp. 391–411.
Wacey, D., Kilburn, M. R., McLoughlin, N., Parnell, J., Stoakes,
C. A., and Brasier, M. D., 2008. Use of NanoSIMS to investigate
early life on Earth: ambient inclusion trails in a c. 3400 Ma sand-
stone. Journal of the Geological Society of London, 165, 43–53.
Wacey, D., Kilburn, M. R., Saunders, M., Cliff, J., and Brasier,
M. D., 2011a. Microfossils of sulfur metabolizing cells in
3.4 billion year old rocks of Western Australia. Nature Geosci-
ence, 4(10), 698–702.
Wacey, D., Saunders, M., Brasier, M. D., and Kilburn, M. R., 2011b.
Earliest microbially mediated pyrite oxidation in 3.4 billion-
year-old sediments. Earth and Planetary Science Letters, 301,
393–402.
Walsh, M. M., 1992. Microfossils and possible microfossils from
the early Archean Onverwacht Group, Barberton Mountain
Land, South Africa. Precambrian Research, 54, 271–293.
Walsh, M. M., and Lowe, D. L., 1985. Filamentous microfossils
from the 3,500 Myr-old Onverwacht Group, Barberton Moun-
tain Land, South Africa. Nature, 314, 530–532.
Westall, F., de Wit, M. J., Dann, J., van der Gaast, S., de Ronde,
C. E. J., and Gerneke, D., 2001. Early Archean fossil bacteria
and biofilms in hydrothermally-influenced sediments from the
Barberton greenstone belt, South Africa. Precambrian Research,
106, 93–116.
Westall, F., de Ronde, C. E. J., Southam, G., Grassineau, N., Colas,
M., Cockell, C., and Lammer, H., 2006a. Implications of
a 3.472-3.333 Gyr-old subaerial microbial mat from the Barber-
ton greenstone belt, South Africa for the UV environmental con-
ditions on the early Earth. Philosophical Transactions of the
Royal Society B, 361, 1857–1875.
Westall, F., de Vries, S. T., Nijman, W., Rouchon, V., Orberger, B.,
Pearson, V., Watson, J., Verchovsky, A., Wright, I., Rouzaud,
J.-N., Marchesini, D., and Severine, A., 2006b. The 3.446 Ga
“Kitty’s Gap Chert”, an early Archean microbial ecosystem.
Geological Society of America, Special Paper, 405, 105–131.
Cross-references
Chert
Isua Supracrustal Belt, West Greenland: Geochronology
Mass Spectrometry
Molecular Clocks
Molecular Dating of Evolutionary Events
Quartz
ELECTRON SPIN RESONANCE (ESR) DATING
OF CORAL
Gerhard Schellmann1 and Ulrich Radtke2
1
Department of Physical Geography & Landscape
Studies, University of Bamberg, Bamberg, Germany
2
University of Duisburg-Essen, Essen, Germany
Synonyms
Electron paramagnetic resonance (EPR) dating
Definition
Electron spin resonance (ESR) dating. Radiation exposure
dating method using radiation-sensitive ESR signals,
which operate like natural dosimeters. An ESR age calcu-
lation derives from the ratio of the accumulated dose (DE)
of a sample to the background natural radioactivity within
a range of 30 cm and cosmic doses.
Introduction
Electron spin resonance (ESR; sometimes called electron
paramagnetic resonance, EPR) dating of corals has
become an efficient tool in earth sciences for paleoenvir-
onmental studies on coral reef terraces in the last 20 years.
The first systematic studies and geochronological applica-
tions of ESR dating of corals started in the 1980s and early
1990s. They were carried out on sites with raised coral reef
 ELECTRON SPIN RESONANCE (ESR) DATING OF CORAL
tracts in southern Japan (e.g., Ikeya and Ohmura, 1983;
Ikeya, 1983), in Haiti (Skinner, 1985), and in Barbados
(Radtke and Grün, 1988; Radtke et al., 1988; Grün et al.,
1992). Since then, several methodological improvements
and the development of more stable and high-resolution
ESR spectrometers have significantly increased the
quality of ESR dating results. The latter is strongly
supported by the comparison of ESR dating results with
radiocarbon (14C) and TIMS 230Th/234U data (see below).
By ESR dating of raised coral reef terraces, it is possible to
distinguish between the major periods of high sea level
during the last 500,000 years but also to differentiate
between submaxima, for example, the submaxima MIS
5a-1 (c. 74 ka), MIS 5a-2 (c. 85 ka), MIS 5c (c. 105 ka),
MIS 5e-1 (c. 118 ka), and MIS 5e-2,3 (ca. 128–132 ka),
during the MIS 5 sea-level highstand (e.g., Schellmann
et al., 2004; Schellmann and Radtke, 2004a). However,
an accurate differentiation between MIS 9, MIS 11, and
older sea-level maxima is often limited by considerable
alterations and recrystallization processes within these
older coral samples. Also, the problem of natural “ESR
rejuvenation,” which is the recombination of electrons that
increases until a natural equilibrium is reached, must be
addressed (e.g., Grün, 1989; Rink, 1997; Jonas, 1997).
Physically, the upper dating limit of the ESR method is
defined by this natural equilibrium limit. Despite this,
weak weathering and slight recrystallization from
a primary aragonite to a calcite coral fabric, which defines
age rejuvenation effects, often cause a progressive under-
estimation of ESR ages before this upper dating limit is
reached.
Methodology
The electron spin resonance (ESR) dating technique is
one of several radiation exposure dating methods based
on radiation dosimetry such as thermoluminescence
(TL), optically stimulated luminescence (OSL), and
radioluminescence (RL). All these methods use the phe-
nomenon of common minerals, like the aragonite fabric
of corals, acting as natural dosimeters. The radiation
(natural radioactivity and cosmic rays) causes charges
(electrons, free radicals) which are trapped at defects in
the crystal lattice of a wide range of minerals. This natu-
rally accumulated dose can be determined by ESR. The
basic principles of the physics and the application of
the ESR dating technique have been provided, for exam-
ple, by Walther et al. (1992), Ikeya (1993), Rink (1997),
Jonas (1997), Sharaf and Hassan (2004), Blackwell
(2006), and Grün (2007). Schellmann et al. (2008) have
reviewed the accuracy of ESR dating of corals, mollusk
shells, and quartz. Blackwell et al. (2007) used coupled
230
Th/234U and ESR coral dating for reconstructing
sea-level changes since the last interglacial maximum
(MIS 5e).
Here, we will focus on ESR dating aragonite coral sam-
ples. The time-dependent accumulations of trapped
charges form paramagnetic centers, which can be detected
235
by the rise of the ESR signal at g ¼ 2.0007 (Figure 1).
Multiple scans (e.g., 5–40 scans) increase the precision
of ESR signal amplitude measurements. The amount of
trapped charge accumulation (paleodose, DE) increases
with time and can be quantified by ESR.
An ESR age is derived from the ratio of the accumu-
lated radiation dose DE (equivalent dose) to the natural
radiation rate D0 (dose rate) within a range of 30 cm, to
which the sample was exposed in the past (Figure 1).
The DE is determined by an additive dose method. For this
method, more than 12 aliquots (Schellmann and Radtke
2001) of the same sample are stepwise irradiated
with increasing artificial g-doses (e.g., using a 60Co
gamma source). The growing amplitude of the ESR signal
at g ¼ 2.0007 of a coral sample is proportional to the num-
ber of trapped charges (DE), which is dependent on the
strength of natural irradiation (dose rate, D0 ) and on the
radiation time (¼age of the sample). A living coral does
not contain the ESR dating signal at g ¼ 2.0007, which
means the ESR intensity is zero. The DE is determined
by extrapolating the resulting dose-response curve to zero
ESR intensity. In the case of dating corals or mollusk
shells, the DE–Dmax plot procedure (D–DP procedure)
should be used for the calculation of DE (Figure 1) as
described by Schellmann and Radtke (2001).
The former natural dose rate (D0 ) of a coral sample is
estimated based on the content of uranium (U) (other
radioactive elements like thorium and potassium are of
no relevance) and the intensity of the cosmic dose rate
(Prescott and Hutton, 1994). Most colonial corals of sub-
modern, Holocene, and Pleistocene corals have mean
U contents of about 2.9–3.2 ppm (Figure 2). This illus-
trates a prevailing U uptake of corals from seawater during
their lifetime or shortly after death. Consequently, ESR
ages of corals should be calculated using a recent or early
U uptake model. However, some extremely high (up to
5 ppm) and low (below 2.5 ppm) U concentrations indi-
cate significant U loss or gain; both limit a reliable age
determination. But U-series dating results of corals are
strongly influenced by U migrations than ESR. Therefore,
the application of both methods can be valuable. For dose
rate calculation, an alpha efficiency (k-factor) of
0.06  0.02 should be used (Grün, 2007). The correct
determination of D’ and DE is essential for any ESR age
estimation. The average error range when dating corals
is between 5 % and 8 %.
Improvements in annual dose rate (D’) estimation and
the newly developed approach of equivalent dose (DE)
determination (DE–Dmax plot procedure) increase the pre-
cision of ESR dating Holocene and Pleistocene corals
(Schellmann et al., 2008). This is strongly supported by
comparing ESR dating results with other numeric dating
methods such as radiocarbon (14C) and TIMS U-series
analysis (TIMS 230Th/234U). Such correlations are
published, e.g., by Radtke et al. (2003), Schellmann
et al. (2004, 2008), and Muhs et al. (2012). They are exem-
plarily illustrated in the following case study from
Barbados.
236 ELECTRON SPIN RESONANCE (ESR) DATING OF CORAL

Electron Spin Resonance (ESR) Dating of Coral, Figure 1 Generalized principle of ESR dating of aragonitic coral.

Electron Spin Resonance (ESR) Dating of Coral, Figure 2 Uranium content (ppm) of modern, Holocene, and Pleistocene corals
(Details are provided in Schellmann et al. (2008)).
 ELECTRON SPIN RESONANCE (ESR) DATING OF CORAL 237

Electron Spin Resonance (ESR) Dating of Coral, Figure 3 ESR and TIMS Th/U ages of uplifted Middle and Late Pleistocene coral reef
terraces in a transect on southern Barbados (Details are provided in Schellmann and Radtke (2004a, b)).

Case study: ESR dating of pleistocene coral reef
terraces
The elevated coral reef terraces on the Caribbean island of
Barbados are a typical region for both the reconstruction
of Late and Middle Pleistocene sea-level changes (classi-
cal “Barbados Model”) and, as one of the most common
applications of ESR dating, the accuracy of ESR dating
results on fossil corals up to c. 600 ka. Since early system-
atic studies by Radtke et al. (1988) and Radtke (1989),
new geological, morphostratigraphic, and chronostra-
tigraphic investigations during the past two decades led
to several methodological improvements of ESR dating
on coral and deliver some impressive examples of the
application of this method (e.g., Schellmann and Radtke,
2004a, Schellmann and Radtke, 2004b).
On southern Barbados, several uplifted Pleistocene
coral reef terraces formed during oscillating MIS 5 and
MIS 7 sea-level maxima are well preserved between the
present coastline and the inland (Figure 3). They rise up
to 67 m above modern sea level (asl) and are accompanied
to the center of southern Barbados by more elevated
older Middle Pleistocene coral reefs formed during MIS
9 to MIS 11 sea-level highstands. Based on geological,
morphostratigraphic, and chronostratigraphic methods,
the uplifted MIS 5 and MIS 7 coral reefs of southern
Barbados can be subdivided into up to seven clearly distin-
guishable main coral reef terraces (T1 to T7) and some less
prominent sublevels (e.g., T-6a and T-6b, T-1a1 and T-1a2).
The youngest reef terraces are located at the lowest eleva-
tions close to the present coast, while older reef tracts have
higher elevations and generally a greater distance from the
coast. Due to the relatively slow uplift rate in southern Bar-
bados (approx. 0.27  0.02 m/1,000 a), the youngest Late
Pleistocene coral reefs (T-1a1) have been formed during
MIS 5a-1 at approx. 74 ka (ESR) or 77 ka (TIMS Th/U)
and the oldest and highest MIS 7 reef terrace (T-7) during
238 ELECTRON SPIN RESONANCE (ESR) DATING OF CORAL
Electron Spin Resonance (ESR) Dating of Coral, Figure 4 ESR and TIMS Th/U ages (d234U >141 and <157%) of MIS 5a-1 coral
reef terrace near Inch Marlowe Point, southern coast of Barbados (Modified after Schellmann and Radtke (2004a), and Schellmann
et al. (2004)).
a MIS 7 sea-level maximum approx. 224 ka (ESR). Both
dating methods support in general a geochronological dif-
ferentiation of the MIS 5a reefs into two sea-level
submaxima (T-1a1, T-1a2) around 74 ka (ESR) or 77 ka
(Th/U), and 85 ka (ESR) or 84 ka (Th/U). The three
T1b, T2, and T3 coral reef terraces were formed during
the MIS 5c around 104–105 ka (ESR) or 102–103 ka
(Th/U). The three coral reef terraces (T4, T5a, T5b) are
of MIS 5e age around c. 118 ka (ESR) or 120 ka (Th/U),
and 132 ka (ESR) or 128 ka (Th/U) for the oldest MIS
5e reef terrace, respectively. However, an age differentia-
tion of the three MIS 7 reef terraces (T6a, T6b, T7), which
were formed around 200–224 ka (ESR), is still not possi-
ble. The correlation between age and height of the reef
terraces allows to calculate neotectonic movements and
to estimate paleo-sea-level changes (e.g., Schellmann
et al., 2004; Radtke and Schellmann, 2005).
The example shows that generally the timing of coral
reef growth determined by both dating methods agrees
well, especially when the error ranges are considered. How-
ever, ESR and TIMS Th/U age estimates from an MIS 5a-1
coral reef near the south point of Barbados (Figure 4) illus-
trate that many ESR ages are up to 5–10 % younger than the
corresponding U-series dates, respectively. The U-series
data shows a slightly overestimation of the expected age.
This data set illustrates very well that the apparently minor
analytical errors (precision) of U-series data (less than 1 %,
ESR between 5 % and 8 %) give no evidence of a better
time resolution (Schellmann et al., 2004). The age estimates
were carried out on paired samples. They show that the
spread of individual ages (not considering errors) for both
methods is about 4,000 year, although all ages were pro-
duced on individuals from the same branch of coral, which
are hence most likely of the same age (grown within not
more than a few hundred years). Therefore, age differentia-
tions between single Late Pleistocene coral reef deposits
that have developed in time periods of less than
4,000–5,000 years are little reliable, independent from the
applied dating technique (ESR or TIMS U-series).
Summary and conclusions
Due to methodological advances, which significantly
increase the accuracy of ESR dating of Pleistocene and
Holocene corals, ESR dating has become an efficient tool
in earth sciences in the past two decades, especially for geo-
chronological studies on coral reef terraces. ESR dating of
Pleistocene corals permits to differentiate between the main
Marine Isotope Stages (MIS) 5, 7, 9, 11, and 13 as well as
between the substages 5e-2,3 and 5e-1, 5c, and 5a-2 and
5a-1. The average error is about 5–8 %. Late Pleistocene
corals can be dated with a similar accuracy as with the
TIMS U-series method in this time range, although the indi-
vidual errors of ages are significantly larger than those of
U-series data. The advantage of ESR, however, is that the
upper dating limit for corals is probably above 500,000
year. If weathering and recrystallization from primary ara-
gonite to calcite of the coral material could be excluded,
dating of coral with an age of several million years should
be feasible from the physical point of view. All in all, future
research in ESR dating corals should be focused on the cre-
ation of a single-aliquot dating technique, which uses orien-
tated aragonitic polycrystals.
Bibliography
Bender, M. L., Taylor, F. W., Matthews, R. K., Goddard, J. G., and
Broecker, W. S., 1979. Uranium-series dating of the Pleistocene
reefs tracts of Barbados, West Indies. Geological Society of
America Bulletin, 90, 577–594.
Blackwell, B. A. B., 2006. Electron Spin Resonance (ESR) dating in
Karst environments. Acta Carsologica, 35(2), 123–153;
Ljubljana.
Blackwell, B. A. B., Teng, S. J. T., Lundberg, J. A., Blickstein, J. I. B.,
and Skinner, A. R., 2007. Coupled 230Th/234U-ESR analyses for
corals: a new method to assess sealevel change. Radiation Mea-
surements, 42, 1250–1255.
Grün, R., 2006. Direct dating of human fossils. Yearbook of Physi-
cal Anthropology, 49, 2–48.
Grün, R., 2007. Electron spin resonance dating. In Elias, S. A. (ed.),
Encyclopedia of Quaternary Science. Amsterdam: Elsevier,
Vol. 2, pp. 1505–1516.
 ELECTRON SPIN RESONANCE (ESR) DATING OF FOSSIL TOOTH ENAMEL
Grün, R., Radtke, U., and Omura, A., 1992. ESR and U-series ana-
lyses on corals from Huon Peninsula, New Guinea. Quaternary
Science Reviews, 11, 197–202.
Ikeya, M., 1993. New Applications of Electron Spin Resonance.
Dating, Dosimetry and Microscopy. Singapore: World
Scientific.
Ikeya, M., and Ohmura, K., 1983. Comparison of ESR ages of
corals from marine terraces with 14C and 230Th/234U ages. Earth
and Planetary Science Letters, 65, 3438.
Jonas, M., 1997. Concepts and methods of ESR dating. Radiation
Measurements, 27, 943–973.
Muhs, D. R., Pandolfi, J. M., Simmons, K. R., and Schumann, R. R.,
2012. Sea-level history of past interglacial periods from ura-
nium-series dating of corals, Curaçao, Leeward Antilles islands.
Quaternary Research, 78, 157–169.
Pirazzoli, P., Radtke, U., Hantoro, W. S., Jouannic, C., Hoang, C. T.,
Causse, C., and Borel-Best, M., 1991. Quaternary raised coral-
reef terraces on Sumba Island, Indonesia. Science, 252,
1834–1836.
Prescott, J. R., and Hutton, J. T., 1994. Cosmic ray contributions to
dose rates for luminescence and ESR dating: large depths and
long-term time variations. Radiation Measurements, 23,
497–500.
Radtke, U., and Grün, R., 1988. ESR dating of corals. Quaternary
Science Reviews, 7, 465–470.
Radtke, U., and Schellmann, G., 2005. Timing and magnitude of sea
level change during MIS 5 derived from Barbados coral reef ter-
races: a critical literature review and new data. Journal of
Coastal Research, SI 42, 52–62.
Radtke, U., Grün, R., and Schwarcz, H. P., 1988. Electron spin res-
onance dating of the Pleistocene coral reef tracts of Barbados.
Quaternary Research, 29, 197–215.
Radtke, U., Schellmann, G., Scheffers, A., Kelletat, D., Kasper,
H. U., and Kromer, B., 2003. Electron spin resonance and radio-
carbon dating of coral deposited by Holocene tsunami events on
Curaçao, Bonaire and Aruba (Netherlands Antilles). Quaternary
Science Reviews, 22, 1309–1315.
Rink, W. J., 1997. Electron Spin Resonance (ESR) dating and ESR
applications in quaternary science and archaeometry. Radiation
Measurements, 27, 975–1025.
Schellmann, G., and Radtke, U., 2001. Progress in ESR dating of
Pleistocene corals – a new approach for DE determination. Qua-
ternary Science Reviews, 20, 1015–1020.
Schellmann, G., and Radtke, U., 2004a. The Marine Quaternary of
Barbados. Köln: University of Cologne. Kölner Geographische
Schriften, Vol. 81, 137 pp.
Schellmann, G., and Radtke, U., 2004b. A revised morpho- and
chronostratigraphy of the Late and Middle Pleistocene coral reef
terraces on Southern Barbados (West Indies). Earth-Science
Reviews, 64, 157–187.
Schellmann, G., Radtke, U., Potter, E.-K., Esat, T. M., and
McCulloch, M. T., 2004. Comparison of ESR and TIMS U/Th
dating of marine isotope stage (MIS) 5e, 5c, and 5a coral from
Barbados – implications for palaeo sea-level changes in the
Caribbean. Quaternary International, 120, 41–50.
Schellmann, G., Beerten, K., and Radtke, U., 2008. Electron spin
resonance (ESR) dating of quaternary materials. Quaternary Sci-
ence Journal (E&G), 57, 150–178.
Sharaf, M. A., and Hassan, G. M., 2004. ESR dosimetric properties
of modern coral reef. Nuclear Instruments and Methods in Phys-
ics Research, B217, 603–610.
Skinner, A. F., 1985. Comparison of ESR and 230Th/234U ages on
fossil aragonitic corals. In Ikeya, M., and Miki, T. (eds.), ESR
Dating and Dosimetry. Tokyo: Ionics, pp. 135–138.
Walther, R., Barabas, M., and Mangini, A., 1992. Basic ESR
studies on recent corals. Quaternary Science Reviews, 11,
191–196.
239
Cross-references
Electron Spin Resonance (ESR) Dating, General Principles
Electron Spin Resonance Spectrometer
Marine Isotope Stratigraphy
Radiation and Radioactivity
Radiation Defect
Radiation Dose Rate
Radiocarbon Dating of Marine Carbonates
U-Series Dating
ELECTRON SPIN RESONANCE (ESR) DATING OF
FOSSIL TOOTH ENAMEL
Mathieu Duval
ESR dating laboratory, Geochronology program, Centro
Nacional de Investigación sobre la Evolución Humana
(CENIEH), Burgos, España
Synonyms
Electron paramagnetic resonance (EPR) dating of fossil
tooth enamel
Definitions
Electron spin resonance (ESR) dating: a chronometric dat-
ing method based on the evaluation of the exposure of
some materials to natural radioactivity, which is quantified
in terms of absorbed radiation dose and corresponds to the
energy deposited in the matter by ionizing radiations.
DE: equivalent dose. The total absorbed radiation dose as
measured in the laboratory.
ka: 1,000 years.
Ma: 1,000,000 years.
Introduction
The first ESR dating applications to fossil bones were
published in the early 1980s (e.g., Ikeya and Miki,
1980). Given the potential of this kind of material, studies
were quickly reoriented toward fossil tooth enamel, which
has more suitable characteristics for dating (Grün and
Schwarcz, 1987). Then, the method progressively gained
in accuracy over the following decades, especially via
a better understanding of the ESR signal of fossil enamel
and its behavior with the absorbed dose. A major advance
in ESR dating of enamel resulted from the introduction of
a method combining ESR and U-series data in order to
model the uranium (U) uptake into dental tissues, thus
constraining the uncertainty of the resulting dose rate
(Grün et al., 1988). This definitely helped to convert the
method into a valuable tool to date fossil remains beyond
the 14C time range.
Tooth enamel: a natural dosimeter
From a mineralogical point of view, enamel is mainly
made of carbonated hydroxyapatite (Ca10(PO4)6(OH)2)
(90 %, vs. ~70 % for dentine and bone), water (~3 %),
240 ELECTRON SPIN RESONANCE (ESR) DATING OF FOSSIL TOOTH ENAMEL

Electron Spin Resonance (ESR) Dating of Fossil Tooth Enamel, Figure 1 ESR spectrum of fossil tooth enamel. In addition to the
main radiation-induced ESR signal generated by CO2
radicals, there are several minor contributions from other species (see Grün,
2000b and references therein): (i) there is a septet centered on the main CO2
signal at g ¼ 2.0032 (labeled “a”; only five lines are
visible on that magnetic field range) formed by a free dimethyl radical; (ii) another isotropic line (marked “b”) at g ¼ 2.0115 might be
attributed to CO3
; and (iii) the isotropic line at g ¼ 2.0056 (marked “c”) is usually attributed to a free radical, probably SO2
. In ESR
dating, ESR intensities are usually extracted by peak-to-peak measurements between T1 and B2.

and organic matter (<1 %) (Elliott, 2002). Enamel is
denser and more compact than dentine or bone, with larger
hydroxyapatite crystals, making it less sensitive to
diagenetic changes in terms of microstructure, chemical
composition and dissolution/recrystallization processes
(e.g., Kohn et al., 1999; Dauphin and Williams, 2004).
The dosimetric properties of tooth enamel have been
known for more than 40 years (e.g., Brady et al., 1968),
and ESR is capable of measuring absorbed dose from only
a few tens of mGy to several thousands of Gy (e.g., Duval
et al., 2009; Fattibene and Callens, 2010). Enamel is now
internationally recognized as a valuable natural ESR
dosimeter, especially used in the field of retrospective
dosimetry for persons accidentally exposed to ionizing
radiations (e.g., IAEA, 2002 and references therein).
anisotropic CO2
radicals, an axial one (g┴ ~ 2.003;
g║ ~ 1.997) and an orthorhombic one (gx ~ 2.003;
gy ~ 1.997 and gz ~ 2.001). These CO2
radicals are usu-
ally divided in two categories: the non-oriented radicals,
which exhibit a powder spectrum, and the oriented CO2

radicals, whose ESR signal may significantly change
according to the angular position of the enamel fragment
in the ESR cavity. Both groups have different characteris-
tics in terms of thermal stability or microwave saturation.
Furthermore, their relative proportions in the ESR signal
may not be the same in the natural and irradiated spectrum.
Irradiation or heating processes might induce some
transformation from one category to another, which would
generate some uncertainties in the evaluation of DE (Brik
et al., 2000; Joannes-Boyau and Grün, 2011).

ESR signal of tooth enamel Age and dose rate equations

The ESR signal of tooth enamel is an asymmetric compos-
ite signal defined by three peaks at g ~ 2.0032 (T1),
g ~ 2.0006 (B1), and g ~ 1.9971 (B2) (Figure 1). Although
many contributors to this signal have been identified
(mainly carbonate-derived radicals and some oxygen rad-
icals: Callens et al., 1998; Fattibene and Callens, 2010),
the major contribution is related to CO2
radicals.
Up to seven types of CO2
radicals have been identi-
fied as contributing to the main radiation-induced ESR
signal (Brik et al., 2000), but the system is nevertheless
usually simplified by considering three main types of
CO2
(Fattibene and Callens, 2010; Joannes-Boyau and
Grün, 2011): one isotropic at g ¼ 2.0006 and two
The basic ESR age equation for tooth dating may be
expressed as follows:
ZT
DE ¼ Dðt Þdt ð1Þ
0
where DE is the equivalent dose, i.e., the total dose (Gy)
absorbed by the enamel during the time elapsed between
the death of the animal (t ¼ 0) and the sampling (t ¼ T),
and D(t) is the mean dose rate (in Gy/ka or mGy/a).
Basically, the total dose accumulated in the enamel over
time is the result of a combination of various contributions
 ELECTRON SPIN RESONANCE (ESR) DATING OF FOSSIL TOOTH ENAMEL 241

Electron Spin Resonance (ESR) Dating of Fossil Tooth Enamel, Figure 2 Tooth geometry for dose rate calculations (Modified from
Rink 1997). Shown here is the sediment-cementum-enamel-dentine geometry. (*) is the removed enamel thicknesses (a few tens
of mm) from both sides of the enamel layer. Top right: examples of occlusal views of horse fossil teeth (a, b, and c) from various Spanish
archaeological sites (scale ¼ 1 cm).

coming from radioactivity in the tooth itself and the sur-
rounding environment. The specificity of tooth dating
relies on the complex system that has to be considered.
Indeed, a tooth is usually made by several tissues of varied
thickness and composition. Therefore, accurate age calcu-
lation must rely on careful consideration of the geometry
of the enamel layer and its surrounding environment
(Figure 2).
Except for the cosmic dose, each component of the
dose rate is directly derived from the contributions of the
a-particles, b-particles and g-rays emitted by the radioac-
tive elements located within the sample and its surrounding
environment (mainly U and Th decay products as well
as 40K). Because each kind of ionizing radiation has
a specific penetration range in matter (~20–40 mm for
a-particles, ~2 mm for b-particles, and ~30 cm for g-rays),
some of their respective contribution has to be attenuated or
even removed depending on the size of the sample dated.
For dental tissues, only the U decay products contribute
significantly to the enamel dose, because tissues are usu-
ally assumed to be free of Th decay products as well as
40
K (e.g., Grün and McDermott, 1994). Given the typical
thickness of the enamel (a few mm), it is usually assumed
that self-absorption of internal g-rays is negligible. Thus,
the internal dose rate from the enamel is only made by
a- and b-components, which are corrected by an alpha
efficiency factor and a self-absorption factor, respectively.
For the other tissues, we need to consider only an attenu-
ated beta component, since the external a-particle contri-
bution is removed by abrading both sides of the enamel
layer over a few tens of mm. The magnitude of the beta
component depends on the initial and removed thick-
nesses of the enamel layer as well as water content in the
tissues. Finally, the sediment typically provides only
a g-ray contribution to dose rate (unless the enamel is in
direct contact with sediment), corrected by water attenua-
tion, and derived from the measured elemental concentra-
tions of U, Th, and K. Consequently, in case of a geometry
sediment-cementum-enamel-dentine (Figure 2), the dose
rate equation may be expressed as follows:
   
Dðt Þ ¼ Da ðt Þ þ Db ðt Þ enamel þ Db ðt Þ dentine þ
   
Db ðt Þ cementum þ Dg ðt Þ sediment þ
  
Dcosmic ðtÞ ¼ C U ðt Þ F U a k þ F U b S U b ena: þ
 
C ðt ÞF U b Ab1 W b dent: þ
 U 
C U ðt ÞF U b Ab2 W b cem: þ
 
C U ðt ÞF U g W g þ C Th ðt ÞF Thg W g þ C K ðt ÞF Kg W g sed: þ
Dcos : ðt Þ
ð2Þ
where CU, Th, K is the concentration in U, Th, and K mea-
sured in each dental tissue and sediment; Wa, b, g is the
242 ELECTRON SPIN RESONANCE (ESR) DATING OF FOSSIL TOOTH ENAMEL
water attenuation for a-particles, b-particles, and g-rays;
FU, Th, K; a, b, g is the dose rate conversion factor depending
on the kind of ionizing radiation and the emitter; k is the
alpha efficiency; SUb is the internal beta self-absorption;
and Ab1,2 is the external beta attenuation for U-series ele-
ments. Equation (2) is valid when equilibrium is assumed
within the U-series or Th-series decay products in the den-
tal tissues and sediment. However, if disequilibrium is
observed, more elements in each decay series have to be
considered in the equation. Specific issues with disequilib-
rium in dental tissues and sediment may be found in Grün
(1989) and Olley et al. (1996), respectively.
In case of enamel in direct contact with the sediment,
Eq. (2) has to be modified by considering an external beta
contribution from the sediment, instead of a tissue.
U-uptake modeling
Dental tissues behave as open systems for U-series
elements. It is therefore crucial not only to measure the
U-content but also to know its past evolution. To address
this point, several parametric U-uptake models were pro-
posed. The early uptake (EU) model assumes that U is
incorporated shortly after burial (Bischoff and
Rosenbauer, 1981), i.e., an approximation of a closed sys-
tem. The linear U-uptake (LU) model describes uptake
with a constant rate since the time of burial (Ikeya,
1982). The recent uptake (RU) model considers that all
measured U-concentration in the sample was taken up in
very recent times (Blackwell et al., 1992). These paramet-
ric models allow some easy age calculations, but are based
on unrealistic assumptions (i.e., the closed system
assumption for the EU uptake). Additionally, the models
oversimplify the complexity of U-uptake into the different
tissues in some teeth (e.g., Grün, 2009a). These models
cause large uncertainties when uranium concentrations in
the dental tissues are high, and the calculated age value
may be then very sensitive to the U-uptake assumption
(e.g., Grün and McDermott, 1994; Duval et al., 2012).
To address the problem of the unknown U-uptake, Grün
et al. (1988) suggested to combine ESR and U-series ana-
lyses and introduced a parameter p to describe the kinetics
of the U-uptake in the dental tissue (US model). In com-
parison with the conventional models, the combined
US-ESR method has the advantage that U-uptake history
is not a priori assumed, but mathematically assessed
according to the ESR and present-day U-series
(U-concentration, 230Th/234U and 234U/238U) data mea-
sured in each tissue. As a consequence, only one com-
bined US-ESR age fits the available dataset (Grün et al.,
1988). Moreover, the US-ESR approach considers the
possibility of different uptake histories among the various
dental tissues of a unique tooth, by calculating a specific
p-parameter for each tissue. This approach also helps to
overcome some limitations of the U-series dating method
in some specific situations, beyond the time range of the
U-series dating method (Grün et al., 2010) or in case of
very recent U-uptake (e.g., Falguères et al., 2006).
However, the combined US-ESR approach has also
several limitations. First, U-leaching from dental tissue
cannot be modeled (Grün et al., 1988). By definition, the
U-uptake history is mainly modeled from the U-series data
measured in dental tissues. Consequently, the reliability of
the calculated age is highly dependent on the accuracy of
these U-series data. However, the latter may be influenced
by many factors, such as the spatial homogeneity and pres-
ervation of the sample, or recent U-mobilization that may
overprint past U-uptake history (Duval et al., 2011). Other
models based on the combination of U-series and ESR
data have been also developed, in particular to specifically
address the issue raised by the occurrence of U-leaching in
dental tissue (e.g., Shao et al., 2012) or the spatial varia-
tion of the U-series data (Pike et al., 2001), but their poten-
tial and limitations are not clearly defined yet. Finally,
another variation of the US model was also proposed by
Grün (2000a), the closed system (CS)-US model. This is
based on a very specific assumption, i.e., the measured
uranium content was incorporated into the tissues at the
time corresponding to the apparent U-series age, provid-
ing thus a maximum possible age for the sample.
Standard analytical procedure: a brief overview
An ESR age estimate is the result of a long and complex
analytical process, made by several steps associating field-
work and laboratory procedures, which may be resumed
as follows:
Sampling. Fossil teeth are collected directly from the
site or chosen from collections. In situ measurements of
the natural radioactivity should be performed at the exact
place, or at least the closest possible, where the sample
was collected.
Sample preparation. In the laboratory, the tooth is pre-
pared by separating mechanically each dental tissue. The
enamel is then cleaned on both sides to remove external
a-particles contribution and powdered, in order to avoid
angular dependence of the ESR signal within the ESR res-
onator and to improve sample homogeneity.
DE reconstruction by ESR spectrometry. The enamel
powder is usually divided in one or several subsample ali-
quots that are irradiated at various increasing dose steps
(additive dose method). Each aliquot is then measured at
room temperature by X-band ESR spectrometry in order
to study the behavior of the ESR signal with the increasing
dose values. The ESR intensity is extracted from each
spectrum, usually by peak-to-peak measurements between
T1 and B2 (Figure 1). A mathematical function, usually
a single saturating exponential or a double saturating
exponential function (Duval et al., 2009), is fitted through
the experimental data points and its extrapolation to the
abscissa axis (Y ¼ 0) provides the DE value,
corresponding the total dose absorbed by the tooth enamel
since the death of the animal (Figure 3).
Evaluation of dose rate. If the gamma dose rate has to
be assessed in situ, the external beta dose rate contribution
coming from the sediment should be assessed in the
 ELECTRON SPIN RESONANCE (ESR) DATING OF FOSSIL TOOTH ENAMEL
Electron Spin Resonance (ESR) Dating of Fossil Tooth Enamel, Figure 3 Example of ESR dose–response curve. A double saturating
exponential (DSE) function is fitted through the experimental points, and DE is obtained by extrapolation of the function to Y ¼ 0.
laboratory, from the sediment sample that was collected
around the tooth. Usually, the dose rate is derived from
the activities or concentrations of the radioactive elements
present in the sample and its surroundings. These concen-
tration data are converted into dose values using known
factors (Guérin et al., 2011) and corrected according to
several parameters, such as the water content (e.g., Grün,
1994), the geometry of the sample, its thickness and den-
sity for b-attenuations (Brennan et al., 2000), and the
alpha efficiency (Grün and Katzenberger-Apel, 1994). In
case of the combined US-ESR approach, U-series data
from each dental tissue (U-concentration, 234U/238U and
230
Th/234U) is required in order to model the uranium
uptake. Finally, the cosmic dose rate contribution is usu-
ally assessed via tables and is corrected according to vari-
ous factors like the ground density, depth of the sample,
altitude, and latitude (Prescott and Hutton, 1994).
ESR age estimate. Basically, two main non-commercial
programs are most widely used for age calculation on
teeth: these are the ROSY software which enables age cal-
culation based on the parametric U-uptake models
(Brennan et al., 1999) and the DATA program, providing
combined US-ESR age estimates (Grün, 2009b). ESR
ages are usually quoted  one standard deviation. Relative
age uncertainties are rarely <10 % (Rink, 1997) and are
more likely between 10 % and 20 %.
Time range applicability
It is usually considered that the optimum time range appli-
cation for ESR dating of tooth enamel lies between ~50
and ~600 ka. Nevertheless, under some specific
243
conditions, the time range limits may be potentially
pushed from <1 ka to ~2–3 Ma.
For modern ESR samples, the lower dating range is in
general mainly driven by (i) the radiation sensitivity of
the material and (ii) the sensitivity of the tools employed
for dose reconstruction, including the ESR spectrometer.
Recent developments in ESR dosimetry of enamel lead
to define a detection limit of several tens of mGy
(Fattibene and Callens, 2010). The detection of back-
ground dose values with ESR spectrometry (e.g., Toyoda
et al., 2011) suggests the possibility of dating almost
present-day (i.e., last millennium) materials, as soon as
radionuclide concentrations in the sample and its sur-
rounding environment can be accurately assessed. How-
ever, a rapid overview of the published works indicates
that ESR dating is rarely applied to fossil teeth younger
than ~20 ka.
The upper limit of the datable age range is mainly
constrained by two parameters: (i) the saturation of the
ESR signal with the absorbed dose and (ii) the thermal sta-
bility of the paramagnetic centers. Depending on the num-
ber of traps available in the crystal lattice, the nature of the
analyzed paramagnetic center and the intensity of the dose
rate, the intensity of the ESR signal may saturate rapidly or
more slowly when the material is absorbing ionizing radi-
ations. In the case of teeth, the saturation level in enamel is
very high and absorbed dose values up to several thou-
sands of Gy have been detected (e.g., Duval et al.,
2009). The second parameter is the thermal stability of
the paramagnetic centers, which describes the mean life
of a trapped electron at the defect site. Schwarcz (1985)
assessed the lifetime of the ESR signal of tooth enamel
244 ELECTRON SPIN RESONANCE (ESR) DATING OF FOSSIL TOOTH ENAMEL
to ~1 Ga at 25  C, suggesting then a possible application
over a few Ma. As a rule of thumb, the ESR signal must
have a mean life of at least one order of magnitude higher
than the age of the sample dated (Grün, 2007) in order to
avoid systematic uncertainties due to decay of the signal
during burial. However, when dealing with old samples
(>800 ka), the main complication lies in the U-uptake
modeling for the dose rate reconstruction. Indeed, U-
leaching may be frequently observed for such time range,
making the application of the combined US-ESR
approach extremely difficult (e.g., Duval et al., 2012).
Only two situations are possible in which ESR dating
could potentially be accurate to ages up to ~2–3 Ma: (i)
no apparent U-leaching is detected in dental tissues, which
may sometimes occur for Early Pleistocene sites (e.g.,
Duval et al., 2012), or (ii) the tooth samples show very
low U-concentration in dental tissues (e.g., Schwarcz
et al., 1994), decreasing the relative contribution of dental
tissues in the total dose rate calculation in comparison with
sediment.
Main strengths and weaknesses of the ESR dating
method applied to fossil teeth
ESR dosimetry of tooth enamel is now an international
reference method (Fattibene and Callens, 2010), demon-
strating the great potential of this material for dose recon-
struction. In addition, ESR is one of the very few dating
methods that may be applied to fossil remains and can pro-
vide a direct measurement of age of hominid or animal
occupation in any sedimentary context. ESR dating is also
one of the very few alternatives for dating fossils, includ-
ing hominid remains, beyond the 14C and U-series dating
time range (Grün et al., 2010).
In contrast, the complexity and the length of the analyt-
ical process are likely the main limitations of this dating
method. Many parameters (>15) have to be considered
and assessed for the age calculation, with a direct conse-
quence on the age precision (final age errors are usually
> 10 %). In addition, at least several months are usually
required to fully process a sample, and the analytical pro-
cedure involves an array of equipment that must be care-
fully and accurately calibrated (e.g., U-series analyses
facilities, ESR spectrometer, gamma irradiation source,
high-resolution gamma spectrometer, portable gamma
spectrometer). Finally, another limitation lies in the sam-
pling itself. In the case of historic excavations, it may be
difficult to precisely locate (geographically and stratigra-
phically) the sample’s provenance, which induce many
uncertainties on the external dose rate associated to sedi-
ment and thus on the final age estimate.
Some examples of recent developments in the
field
ESR measurement of enamel fragments
Working with enamel fragments significantly reduces
the destructive aspect of the method. However, this tech-
nique implies a complex analytical procedure
(e.g., Joannes-Boyau and Grün, 2011). Nevertheless,
ESR analyses of enamel fragments, unlike powder, may
provide useful information about the composition of the
ESR signal: each component contributing to the radia-
tion-induced ESR signal may be potentially isolated and
studied separately, which may be of special interest for
reducing potential source of error in the DE assessment
(Joannes-Boyau and Grün, 2011).
High-resolution mapping of radioelement
distribution in dental tissues
The combination of a laser ablation system with an
ICP-MS (LA-ICP-MS) recently opened huge perspectives
for the U-series analyses of dental tissues (e.g., Eggins
et al., 2003), yielding high-resolution U-series data in
a short acquisition time and with minor sample prepara-
tion. For example, maps of spatial distribution of U-series
elements in dental tissues may be obtained (e.g., Duval
et al., 2011). This is of special interest for studying
U-mobility into dental tissues and may be particularly use-
ful to identify domains in the teeth that are suitable for
ESR dating.
Direct dating of hominid remains
The recent development of ESR measurements on enamel
fragments combined with LA-ICP-MS U-series analysis
now permits analysis of human teeth without causing
any visible damage (e.g., Grün et al., 2006). Such an
approach offers new perspectives for the ESR method,
by giving access to direct dating of inestimable hominid
remains that are beyond the 14C time range.
ESR dating of early pleistocene fossil teeth
“Old” samples result in many challenges that are not
encountered in younger samples and which make the
application of the combined US-ESR approach especially
complex (see Duval et al., 2012 for further details). Never-
theless, recent methodological developments showed that
ESR may be a useful tool to constrain the chronology of
the earliest hominid occupations in Eurasia and Africa
during Early Pleistocene times (e.g., Duval et al., 2012).
Summary
ESR dating is one of the very few dating methods that may
be applied to fossil remains, resulting in a direct measure-
ment of the age of hominid or animal occupations in any
sedimentary context. The challenge in applying ESR
dating to fossil teeth lies in the requirement to model
U-uptake kinetics into dental tissues, which are known
to behave as open systems for U-series elements. So far,
the combined (or coupled) U-series/ESR dating approach
provides the most reliable ESR age estimates. Under some
specific conditions, this application may potentially cover
the whole Quaternary time range.
 ELECTRON SPIN RESONANCE (ESR) DATING OF FOSSIL TOOTH ENAMEL 245

Bibliography Grün, R., 2000b. Methods of dose determination using ESR spectra
of tooth enamel. Radiation Measurements, 32(5–6), 767–772.
Bischoff, J. L., and Rosenbauer, R. J., 1981. Uranium-series dating
Grün, R., 2007. LUMINESCENCE DATING | Electron Spin Reso-
of human skeletal remains from the Del Mar and Sunnyvale
nance Dating. In Scott, A. E. Editor-in-Chief, Encyclopedia of
sites, California. Science, 213, 1003–1005.
Quaternary Science. Oxford: Elsevier, pp. 1505–1516.
Blackwell, B., Porat, N., Schwarcz, H., and Debenath, A., 1992.
Grün, R., 2009a. The relevance of parametric U-uptake models in
ESR dating of tooth enamel – comparison with Th-230/U-234
ESR age calculations. Radiation Measurements, 44(5–6),
speleothem dates at La-Chaise-de-Vouthon (Charente), France.
472–476.
Quaternary Science Reviews, 11, 231–244.
Grün, R., 2009b. The DATA program for the calculation of ESR age
Brennan, B. J., Rink, W. J., Rule, E. M., Schwarcz, H. P., and Prest-
estimates on tooth enamel. Quaternary Geochronology, 4(3),
wich, W. V., 1999. The ROSY ESR dating program. Ancient TL,
231–232.
17(2), 45–53.
Grün, R., and Katzenberger-Apel, O., 1994. An alpha irradiator for
Brennan, B. J., Prestwich, W. V., Rink, W. J., Marsh, R. E., and
ESR dating. Ancient TL, 12(2), 35–38.
Schwarcz, H. P., 2000. Alpha and beta dose gradients in tooth
Grün, R., and McDermott, F., 1994. Open system modelling for
enamel. Radiation Measurements, 32(5–6), 759–765.
U-series and ESR dating of teeth. Quaternary Science Reviews,
Brady, J. M., Aarestad, N. O., Swartz, H. M., 1968. In vivo dosim-
13(2), 121–125.
etry by electron spin resonance spectroscopy. Health Physics,
Grün, R., and Schwarcz, H. P., 1987. Some remarks on ESR dating
15(1), 43–47.
of bones. Ancient TL, 5(2), 1–9.
Brik, A. B., Rosenfeld, L. G., Haskell, E. H., Kenner, G. H., and
Grün, R., Maroto, J., Eggins, S., Stringer, C., Robertson, S., Taylor,
Brik, V. B., 2000. Formation mechanism and localization places
L., Mortimer, G., and McCulloch, M., 2006. ESR and U-series
of CO2
radicals in tooth enamel. Mineral Journal (Ukraine),
analyses of enamel and dentine fragments of the Banyoles
22(5/6) 57–67.
mandible. Journal of Human Evolution, 50(3), 347–358.
Callens, F., Vanhaelewyn, G., Matthys, P., and Boesman, E., 1998.
Grün, R., Aubert, M., Hellstrom, J., and Duval, M., 2010. The
EPR of carbonate derived radicals: applications in dosimetry,
challenge of direct dating old human fossils. Quaternary Inter-
dating and detection of irradiated food. Applied Magnetic
national, 223–224, 87–93.
Resonance, 14(2–3), 235–254.
Guérin, G., Mercier, N., Adamiec, G., 2011. Dose-rate conversion
Dauphin, Y., and Williams, C. T., 2004. Diagenetic trends of dental
factors: update. Ancient TL, 29(1), 5–8.
tissues. Comptes Rendus Palevol, 3, 583–590.
IAEA, 2002. Use of the electron paramagnetic resonance dosimetry
Duval, M., Grün, R., Falguères, C., Bahain, J. J., and Dolo, J. M.,
with tooth enamel for retrospective dose assessment. In IAEA-
2009. ESR dating of Lower Pleistocene fossil teeth: limits of the
TECDOC-1331, International Atomic Energy Agency, Vienna,
single saturating exponential (SSE) function for the equivalent
p. 57.
dose determination. Radiation Measurements, 44(5–6), 477–482.
Ikeya, M., 1982. A model of linear uranium accumulation for ESR
Duval, M., Aubert, M., Hellstrom, J., and Grün, R., 2011. High res-
age of Heidelberg (Mauer) and Tautavel bones. Japanese Jour-
olution LA-ICP-MS mapping of U and Th isotopes in an early
nal of Applied Physics, 21, 690–692.
Pleistocene equid tooth from Fuente Nueva-3 (Orce, Andalusia,
Ikeya, M., and Miki, T., 1980. Electron spin resonance dating of ani-
Spain). Quaternary Geochronology, 6(5), 458–467.
mal and human bones. Science, 207(4434), 977–979.
Duval, M., Falguères, C., and Bahain, J.-J., 2012. Age of the oldest
Joannes-Boyau, R., and Grün, R., 2011. A comprehensive model
hominin settlements in Spain: contribution of the combined U-
for CO2
radicals in fossil tooth enamel: implications for ESR
series/ESR dating method applied to fossil teeth. Quaternary
dating. Quaternary Geochronology, 6(1), 82–97.
Geochronology, 10, 412–417.
Kohn, M. J., Schoeninger, M. J., and Barker, W. W., 1999. Altered
Eggins, S., Grün, R., Pike, A. W. G., Shelley, M., and Taylor, L.,
states: effects of diagenesis on fossil tooth chemistry.
2003. 238U, 232Th profiling and U-series isotope analysis of
Geochimica and Cosmochimica Acta, 63, 2737–2747.
fossil teeth by laser ablation-ICPMS. Quaternary Science
Olley, J. M., Murray, A., and Roberts, R. G., 1996. The effects of
Reviews, 22(10–13), 1373–1382.
disequilibria in the uranium and thorium decay chains on burial
Elliott, J., 2002. Calcium phospahte biominerals. In Kohn, M. J.,
dose rates in fluvial sediments. Quaternary Science Reviews,
Rakovan, J., and Hughes, J. M. (eds.), Phosphates – Geochemi-
15(7), 751–760.
cal, Geobiological and Material Importance. Washington, DC:
Pike, A. W. G., and Hedges, R. E. M., 2001. Sample geometry and
Mineralogical Society of America, p. 427e454.
U uptake in archaeological teeth: implications for U-series and
Falguères, C., Bahain, J.-J., Pérez-Gonzlez, A., Mercier, N., Santonja,
ESR dating. Quaternary Science Reviews, 20, 1021–1025.
M., and Dolo, J.-M., 2006. The Lower Acheulian site of Ambrona,
Pike, A. W. G., and Hedges, R. E. M., 2001. Sample geometry and
Soria (Spain): ages derived from a combined ESR/U-series model.
U uptake in archaeological teeth: implications for U-series and
Journal of Archaeological Science, 33(2), 149–157.
ESR dating. Quaternary Science Reviews, 20(5–9), 1021–1025.
Fattibene, P., and Callens, F., 2010. EPR dosimetry with tooth
Prescott, J. R., and Hutton, J. T., 1994. Cosmic ray contributions to
enamel: a review. Applied Radiation and Isotopes, 68(11),
dose rates for luminescence and ESR dating: large depths and
2033–2116.
long-term time variations. Radiation Measurements, 23(2–3),
Grün, R., Schwarcz, H. P., and Chadam, J., 1988. ESR dating of
497–500.
tooth enamel: coupled correction for U-uptake and U-series
Rink, W. J., 1997. Electron spin resonance (ESR) dating and ESR
desiquilibrium. Nuclear Tracks Radiation Measurements, 14,
applications in quaternary science and archaeometry. Radiation
237–244.
Measurements, 27(5–6), 975–1025.
Grün, R., 1989. Electron spin resonance (ESR) dating. Quaternary
Schwarcz, H. P., 1985. ESR study of tooth enamel. Nuclear Tracks,
International, 1, 65–109.
10, 865–867.
Grün, R., 1994. A cautionary note: use of ‘water content’ and ‘depth
Schwarcz, H. P., 1985. ESR studies of tooth enamel. Nuclear Tracks
for cosmic ray dose rate’ in AGE and DATA programs. Ancient
and Radiation Measurements (1982), 10(4–6), 865–867.
TL, 12(2), 50–51.
Schwarcz, H. P., Grün, R., and Tobias, P. V., 1994. ESR dating stud-
Grün, R., 2000a. An alternative model for open system U-series/
ies of the australopithecine site of Sterkfontein, South Africa.
ESR age calculations: (closed system U-series)-ESR, CSUS-
Journal of Human Evolution, 26(3), 175–181.
ESR. Ancient TL, 18(1), 1–4.
246 ELECTRON SPIN RESONANCE (ESR) DATING, GENERAL PRINCIPLES
Shao, Q., Bahain, J.-J., Falguères, C., Dolo, J.-M., and Garcia, T.,
2012. A new U-uptake model for combined ESR/U-series dating
of tooth enamel. Quaternary Geochronology, 10, 406–411.
Toyoda, S., Kondo, A., Zumadilov, K., Hoshi, M., Miyazawa, C.,
and Ivannikov, A., 2011. ESR measurements of background
doses in teeth of Japanese residents. Radiation Measurements,
46(9), 797–800.
Cross-references
Alpha Spectroscopy
Bones (U-Series)
Electron Spin Resonance Spectrometer
Electron Spin Resonance (ESR) Dating, General Principles
Electron Spin Resonance (ESR) Dating of Coral
Electron Spin Resonance (ESR) Dating of Fossil Tooth Enamel
Hominid Evolution Timescale
Laser Ablation Inductively Coupled Mass Spectrometer
(LA ICP-MS)
Neutron Activation Analysis
Radiation and Radioactivity
Radiation Dose Rate
Radiation Defect
Sediment, ESR
Thermal Ionization Mass Spectrometer (TIMS)
ELECTRON SPIN RESONANCE (ESR) DATING,
GENERAL PRINCIPLES
Anne Skinner
Chemistry Department, Williams College, Williamstown,
MA, USA
Synonyms
Electron paramagnetic resonance (EPR)
Definitions
Electron spin resonance: The identification of paramag-
netic species through absorption of energy in an external
magnetic field.
Electron spin resonance dating: The determination of
burial age through measurement of the concentration of
free radicals created by natural radiation.
Introduction
To understand electron resonance (ESR) dating, one must,
clearly, first understand the principles of electron spin res-
onance spectroscopy. As the name shows, these principles
begin with the nature of atomic structure – a positively
charged nucleus surrounded by negatively charged elec-
trons. The technique of ESR spectroscopy was first devel-
oped by Zavoisky in 1944. Since then applications have
been developed in multiple fields of chemistry and phys-
ics, as well as the potential for dating in geology, paleon-
tology, and archaeology. An excellent summary can be
found in Rink (1997).
Quantum theory tells us that all systems have a definite
energy. Every spectroscopic method involves transitions
between two states due to input of energy to a material.
If the energy provided to the system exactly matches the
difference between two states (resonance), the system
can be moved from one state to a higher one, and this
yields an absorption signal. Transitions from higher to
lower levels also occur, and the emitted energy again pro-
vides a measure of the difference in energy between levels.
Common techniques in chemistry and physics, defined by
the relevant range of the electromagnetic spectrum, are
optical spectroscopy, infrared (IR) spectroscopy, and
ultraviolet (UV) spectroscopy. ESR spectroscopy almost
always involves absorption of energy in the microwave
region.
For atoms and molecules, these states can be defined by
a series of quantum numbers that are related to angular
momentum. ESR spectroscopy, as the name suggests, is
concerned with transitions of electrons between states of
differing electron spin. As a spinning electric charge, elec-
trons possess a magnetic moment. The interaction of this
moment with an applied external magnetic field leads to
Zeeman splitting of possible states as shown in Eq. (1),
where b is the electron Bohr magneton, H is the applied
field, g is the splitting factor for a given system, and Ms
is the electron spin quantum number:
DE ¼ gbHMs ð1Þ
For a “free” electron, the simplest case, g ¼ 2.00232,
and we consider only the two possible spin states
M ¼ 1/2. The associated magnetic moments are equal
and opposite in direction. Figure 1 shows the effect of
applied magnetic field on the separation between the pos-
sible states. The lower state represents a system in which
the magnetic moment of the electron is aligned parallel
to the external field, while in the upper state they are
opposed. Quantum mechanical calculations can determine
the energy of a given state. Fortunately, quantitative
knowledge of absolute energies is not needed in order to
use this technique for dating.
Now we need to remember another chemical principle.
Electrons are normally paired, either in atomic orbitals or
in chemical bonds. By the Pauli principle, the spins of
the paired electrons must be opposite in direction. If, then,
such a material is put in a magnetic field, each state will be
populated by one electron. If we put in energy to raise the
lower electron to the upper state, we have reversed its spin
and we now have a forbidden situation, a level with two
electrons of the same spin. Thus electron spin resonance
can only be observed in systems with unpaired electrons.
Such systems are called “paramagnetic,” and an alternate
name for the method is electron paramagnetic resonance
(EPR). Which term is used is a matter of preference; phys-
icists are more likely to use EPR. The principles are the
same; the only problem that this introduces is the need to
search under two names to obtain all the available refer-
ences in the field.
Those readers with experience in optical, IR, or UV
spectroscopy will be accustomed to scanning by changing
 ELECTRON SPIN RESONANCE (ESR) DATING, GENERAL PRINCIPLES
hv
H
Electron Spin Resonance (ESR) Dating, General Principles,
Figure 1 Effect of Zeeman splitting on electron energy. As the
magnetic field strength (H) increases, the energy difference
between the spin states increases. The unpaired electron (blue
arrow) will move from the lower level to the upper if an amount
of energy, hn, is added to the system.
wavelength. The atomic or molecular orbitals have fixed
energies. ESR spectrometers scan by changing the mag-
netic field. Figure 1 shows that the splitting between levels
is a function of magnetic field. So we change H until the
splitting matches a constant energy, hn. In theory we could
hold H constant instead, but experiment has shown that we
get better precision the other way.
There is a second difference between other methods
and ESR spectra. In other methods, the resonance signal
is indicated by an absorption (or transmission) peak such
as that in Figure 2. Largely because of details of signal
acquisition by the spectrometer, ESR spectra are collected
as the first derivative of the peak. This has one significant
advantage for chemists and physicists. Frequently their
interest is in the dependence of the g-value on electronic
environment. Using the derivative spectrum, one can more
easily see small changes in g-values by measuring the
change in position of the point where the signal crosses
the baseline rather than by attempting to find the maxi-
mum of the peak itself.
Figure 2, of course, shows only the simplest spectrum,
isotropic absorption by a free electron. Few ESR spectra
match this simplicity. Multiple unpaired electrons, such
as those on manganese, result in a spectrum of multiple
lines with equal intensities (fine structure). In many cases,
g-values are affected by interactions with nuclear spins
(hyperfine splitting), and for solid materials, such as those
normally used for ESR dating, by variable orientations of
crystals with the magnetic field (Ikeya, 1993). These, with
symbols such as g⊥ and g k, will be mentioned under spe-
cific applications.
Fortunately most of the dating applications can ignore
this complication. For a given substance, dating specialists
have established the optimum line to measure, whether it
be the complex spectrum of aluminum or the anisotropic
carbonate spectrum. More problematic is the mixture of
elements found in most natural materials. This can yield
a complex spectrum of overlapping lines from which the
247
n = 9388.2 MHz
FWHM
Absorbance
Signal
First Derivative
3346 3348 3350 3352 3354
Magnetic Field Strength (G)
Electron Spin Resonance (ESR) Dating, General Principles,
Figure 2 Relationship of absorbance to first derivative spectrum
as seen in ESR spectroscopy.
desired information must be extracted by suitable adjust-
ment of experimental parameters, including temperature
of measurement. Figure 3, a spectrum of raw flint, is an
example where the spectrum contains contributions from
the E’ silica peak, a peroxy peak, and organics (plus some
others).
Complex spectra can sometimes be resolved by
increasing the resolution, separating overlapping signals.
The more intense the magnetic field, the easier one can
resolve two signals whose g-values are virtually identical.
ESR spectrometers are defined by the frequency of the
microwave source (see Table 1). The most commonly
used type is X-band, but higher bands, especially
K and Q, also have a role in ESR dating. There is, how-
ever, a trade-off. The higher the band, the smaller the
amount of sample used, so that while the precision of the
spectrum, in terms of separating signals with very similar
g-values, is improved, the quantitation of the signal dimin-
ishes. The most useful dating application of Q-band spec-
troscopy is determining whether signals are related to the
same radical species, and therefore, the less-well separated
peaks in the X-band can be treated as a single signal
(Skinner et al., 2001b).
ESR-active materials
So where do we find paramagnetic systems? There are
three major classes. “Point defects” are electrons trapped
during the formation of certain crystals. They are often
responsible for colors. Some atoms, mostly transition
metals, have an uneven number of electrons and therefore
show paramagnetism. Iron is, of course, a well-known
example here. Finally there are molecules which can
develop paramagnetism in two ways. First, the molecular
structure can be paramagnetic, usually because it incorpo-
rates one or more paramagnetic atom. Heme proteins,
248 ELECTRON SPIN RESONANCE (ESR) DATING, GENERAL PRINCIPLES
Electron Spin Resonance (ESR) Dating, General Principles, Figure 3 Spectrum of natural flint with complex peaks.
Electron Spin Resonance (ESR) Dating, General Principles,
Table 1 Comparison of ESR spectrometric bands. n, micro-
wave frequency; Ho, field strength
Band n (GHz) H0 (mT)
L 1.5 53.5
S 3.2 114
X 9.5 339
K 25 892
Q 35 1,250
which contain iron, exhibit an ESR spectrum. Second,
a molecule can be made paramagnetic by removing an
electron from a molecular orbital. These are called “free
radicals.” This class of materials is the one we look to
for dating applications.
The free radicals with which we are concerned are cre-
ated by radiation. There are three forms of nuclear radia-
tion. Alpha (a) radiation is caused by emission of
a helium nucleus from the radioactive atom. As this is
a large particle, it is easily stopped with an effective range
of approximately 20 mm. Beta (b) radiation consists of
nuclear electrons. The energy of these particles can vary
considerably; in materials common in dating, they are
assumed to penetrate about 2 mm. Finally, gamma (g)
radiation is simply energetic photons that, again for this
application, penetrate approximately 30 cm. Additionally,
radicals can be induced by cosmic radiation, high energy
particles with extensive penetration capability (Prescott
and Hutton, 1988).
Nuclear radiation, whether alpha, beta, or gamma, can
break chemical bonds to form radicals; it can also remove
an electron from an atom or molecule. One interesting
result is the production of the point defects mentioned ear-
lier. Irradiation of colorless gems has been used to simu-
late rarer colored versions, often without mentioning the
artificial transformation (e.g., Steinhauser et al., 2013).
Common radicals in current scientific discussion are those
in the atmosphere that result in the “ozone hole.”
Low-level natural radioactivity in the environment is
continually “damaging” materials. Living systems have
mechanisms to repair this damage at these low levels,
but once the system has died, radicals begin to accumu-
late. Not all radicals can be used for dating, however.
Before delving deeper into ESR, one might note that
there are other dating methods based on the creation of
trapped charge by radiation. Usually electrons are trapped
at radiation defects in this process. Collectively they are
called “trapped charge” methods. The oldest of these is
thermoluminescence (TL), where the accumulated dam-
age is detected by heating the sample and measuring the
energy released as the electrons are released from trapped
charge sites and migrate to luminescence centers. Opti-
cally stimulated luminescence (OSL) uses photon excita-
tion to release the energy (Aitken et al., 1993).
Geological, paleontological, and archaeological
dating
General considerations
A good dating method requires three things. Firstly the
dating signal must be zero, or must be determinable, at
the time of the event being dated. Secondly, the signal
must change in a predictable and consistent manner with
time. Thirdly, the signal must be stable over the time
period in question. Paramagnetic systems can be annealed
(revert to diamagnetic, or paired electrons) in several
ways. A paramagnetic molecule may react with another,
taking an electron from the second one. The second one
may then become paramagnetic but the first has lost this
property. Paramagnetic systems tend to be unstable and
anneal spontaneously with time (more on this later). This
process can be accelerated by heat.
As examples of the first factor, consider flint artifacts.
For most of human history, these were made by flaking
natural cobbles. While flint has an ESR signal, the signal
we would see would be that accumulated since the flint
was formed, not since the artifact was created. Hence we
only date flint that was heated in the process of
manufacturing an artifact. On the other hand, biological
materials such as shell, coral, and teeth do not have
a “native” signal that prevents dating. This has been inves-
tigated by studying modern examples.
 ELECTRON SPIN RESONANCE (ESR) DATING, GENERAL PRINCIPLES
The change in signal intensity with time depends on
two factors. The first is the sensitivity of the material to
radiation and is relatively easy to discover. Initially
a linear dependence on dose was assumed, but that was
soon shown to be incorrect (Skinner, 1983). For most of
the last 30 years, the growth curve shape has been
assumed to be a saturating exponential. While this is still
largely the case, other dependencies have also been deter-
mined. For the aluminum center, for example, it appears
that there are two components, and a linear term has been
added to the exponential. In other cases, two exponential
terms have been suggested (Haskell et al., 1997; Duval,
2012). Improvements in instrumental precision are likely
to continue to improve our understanding of dose
dependence.
The other factor is the dose rate, the amount of radiation
delivered per unit time. Dose rates are found by summing
a, b, g, and cosmic contributions. Unfortunately, the dose
rate may change with time. Unlike radiocarbon ages, ESR
ages depend on the environment of the sample, and that
may vary during burial. A cave deposit is perhaps the most
stable, but even then the moisture content within the cave
may not be constant. A sample taken from a riverbank
may not be in its original deposit, so the modern environ-
ment we measure to establish an age may not reflect the
actual burial history. At present, environmental variability
is the greatest source of uncertainty in ESR dating, made
worse by an inability to quantify it with a level of precision
comparable to that we can achieve in the laboratory when
we are measuring ESR spectra. This will be covered in
more detail later in this section.
In comparison, determining signal stability is relatively
straightforward. Recall that unstable systems, such as our
paramagnetic radicals, can be annealed with heat. The
degree of heat required, measured both as time of heating
and temperature, depends on the stability. For back-
ground, consider the kinetics of a reaction. For every reac-
tion there exists a rate law defining the change in
concentration of reactant (or product) with time. For the
annealing of radicals, that law is very simple. In integrated
form,
C
C0 ¼
kt ð2aÞ
where C is the concentration of radicals at time t, C0 is the
radical concentration at t ¼ 0, and k is the rate constant
which depends on temperature. Conventionally, one mea-
sures the signal intensity (proportional to concentration of
radicals) as a function of heating time for a fixed tempera-
ture. From that one can calculate the half-life (mean life-
time) at that temperature, the time needed to reduce the
signal to half its original value, and from that the rate con-
stant, k:
0:5C0
C0 ¼
kt1=2 , or k ¼ 0:5C0 =t1=2 ð2bÞ
In 1889, the Swedish chemist Svante Arrhenius derived
a relationship between temperature and reactivity. Arrhe-
nius’ law can conveniently be written as
249
ln k ¼ ln A
ðE=RTÞ ð3Þ
where E is the energy needed to restore the system to sta-
bility, A and R are constants, k is the rate constant, and T is
temperature in degrees Kelvin. To use this equation, one
measures k at a series of temperatures well above the envi-
ronment, usually around 100–300  C, in order to speed
the reaction. One then plots ln k against (1/T). Extrapolat-
ing this plot to ambient temperature provides the mean
lifetime at this temperature. The extrapolation is usually
quite imprecise, but it provides at least the order of magni-
tude of the mean lifetime in the external environment. The
results also depend on the assumption that A remains con-
stant over the temperature range between measured and
ambient, but this holds well for this system. Other compli-
cations include changes in the ESR spectrum with temper-
ature. Shells and teeth, for example, char at elevated
temperatures and the resulting carbon radical can obscure
the dating signal (Skinner et al., 2000).
ESR spectra
With these general principles in mind, let us turn to the
details involved in determining the age of a material that
has a suitable signal and reasonable signal stability. For
ESR dating the mean life of the signal should be an order
of magnitude greater than the age being determined (Grün,
2007). In theory, what we wish to know is the number of
radicals produced as a function of radiation. For this we
would normally look to the area underneath the absorption
signal. As noted, ESR spectrometers produce the deriva-
tive of the signal. It is possible to integrate the signal,
but instead we use the properties of the original curve.
There is a mathematical relationship between the area
under the peak and the full width at half maximum
(FWHM). And as can be seen from Figure 2, FWHM cor-
responds to the height of the derivative. Thus we can plot
the peak height from the signal against radiation dose and
obtain a proxy for area. This is only really true for well-
shaped lines that are not influenced by other ones. As will
be seen in later sections, this is almost never true, but again
we can find approximations that are sufficiently close to
the actual values that our results are reliable.
As noted earlier, crystalline symmetry affects the
g-value. If the radical rotates freely within the crystal,
the asymmetry averages out. If not, one can define
a series of lines as, for example, g⊥ or gk. The first and still
best-known ESR dating signal is that of a carbonate-
derived radical, and as commented briefly earlier, this
has an anisotropic component. Experiments with 13C-
doped calcite (Callens et al., 1989) suggest that a signal
at g ¼ 1.9973 arises from hindered rotation of an ortho-
rhombic CO2 radical. The larger signal at g ¼ 2.0007
has a contribution from the hindered rotating species, but
also arises from a freely rotating radical. Both signals
grow with radiation, so in principle either can be used.
Experimental parameters can be used to accentuate one
signal over another in a complex spectrum. The intensity
250 ELECTRON SPIN RESONANCE (ESR) DATING, GENERAL PRINCIPLES

of a signal would be expected to grow with microwave

Saturation
power, and it does so up to a point. For every signal with
Intensity
which we will be concerned, there is a saturation power,
above which the signal not only does not grow, but may
actually decrease. The explanation for this is reasonably

ESR Intensity
straightforward. In any spectroscopic system, not just
ESR, there is a distribution of populations among energy
states. For ESR, we need fewer electrons in the higher
energy state than in the lower, since otherwise the electron
in the lower state, even if excited by the resonant energy,
has no place to go. Transitions are constantly occurring
between states. The transition from the higher energy spin
state to the lower is called the “relaxation time.” As long AD
as the relaxation time is short, electrons will return to the
lower state by thermal transfer to the crystal lattice fast −20 0 20 40 60 80 100 120
enough to leave room for new electrons to be promoted to Radiation Dose (Gy) Relative to Natural Sample
the higher state by microwave energy. For any new system, Electron Spin Resonance (ESR) Dating, General Principles,
the signal intensity must be measured as a function of micro- Figure 4 Example of a growth curve in the form of a saturating
wave power. While measurements of a specific line must be exponential.
taken at powers below saturation, sometimes it is convenient
to use saturation to simplify the spectrum. For example, by
using powers >5 mW, the E’ signal in silica can be
can be used, but then one should take measurements at
suppressed (and conversely, if that line is to be measured,
a variety of orientations. Either way, the first measurement
the power must generally be <0.1 mW) (Ikeya, 1993).
is, of course, the spectrum of the native material. Then one
Another experimental parameter is the modulation
takes either aliquots of the powder or the single solid piece
amplitude (also called modulation field width). To sharpen
and applies known artificial radiation doses. Now plotting
the output of the detector, the output current is modulated
the peak height against dose creates a “growth curve.”
by a pair of coils mounted on the cavity (see Section C for
The example in Figure 4 assumes a simple saturating
details). Roughly speaking, this modulation amounts to
exponential growth. As noted earlier, various groups have
sampling the output at a given interval, large if a large
found that other ways are appropriate to express the
modulation amplitude, small if not. The larger the modula-
dependence of signal on dose in different materials,
tion amplitude (i.e., the larger the output sampled), the
including double exponentials (Haskell et al., 1997) and
larger the resulting signal. If, however, the signal is com-
exponentials with a linear component (Duval, 2012).
plex, then a large modulation amplitude may distort the
Extrapolating the growth curve (whatever the function
peak shape, mixing two signals. Some experimenters have
used) to the x-axis gives the accumulated dose, the amount
deliberately overmodulated the signal, assuming that the
of radiation that the material experienced since formation,
mixed signals arise from the same radical and have the
or burial. This quantity is variously known as AD (for
same half-lives (Molodkov, 1998). Generally this is not
accumulated dose, or occasionally archaeological dose)
true, and this technique is regarded with suspicion by most
or ED (equivalent dose or environmental dose) or De
dating specialists.
(as before). Careful measurement and using a statistically
How then does one set the spectroscopic parameters?
significant number of points usually means that the error
Here we need to rely on experiment. If, using a given set
on AD is small.
of parameters, the peak heights obtained yield ages that
Converting this information to a sample age can be
are consistent within experimental error with other methods
complex. In the broad sense, the AD is the sum of dose
over a wide time range, then those parameters are the ones
rates times the time that has passed. Mathematically this
to use. This is the test that matters. Whatever “error” is
can be expressed as
introduced by assuming a Gaussian or Lorentzian line Z r
shape for a signal that is clearly neither must be much lower
than the error introduced by other factors such as the dose De ¼ Dðt Þdt ð4Þ
a
rate discussed in the following section.
As a practical matter, this requires knowing the sample
environment. Recall that the contributors to environmen-
Dating procedure tal dose include a-, b-, and g-radiation from radioisotopes
So to begin a dating experiment, select your sample. Gen- inside the sample and in the sediment surrounding it, plus
erally the sample is powdered in order to ensure that the cosmic radiation. As a first approximation, we can express
signal is independent of orientation within the cavity, this as
and the powder is divided into multiple aliquots. If
avoiding this sample destruction is essential, a solid piece AD ¼ ADext þ ADint ¼ Dext t þ Dint t ð5Þ
 ELECTRON SPIN RESONANCE (ESR) DATING, GENERAL PRINCIPLES
Consider first Dext, the external dose rate. g-radiation
can penetrate 30 cm, so essentially one needs to know
the complete environment in a sphere of 30 cm radius
around the sample. In most cases one cannot determine
this absolutely precisely. For one thing, the “top” half of
the sphere has probably been removed in the course of
excavation. More detrimental, the sediment at most sites
is inhomogeneous, the so-called lumpy site.
Several methods have been used to obtain dose rate
information. None of them are free of flaws. In situ gamma
spectroscopy at the position of the sample can provide
both the immediate gamma dose, including the cosmic
component, and concentrations of the major radioisotopes
from which b doses can be calculated. One difficulty is
that this provides only an estimate on the remnant sedi-
ment, which may or may not be identical to that removed.
Further, calibration of these spectrometers has been
a problem (Lyons, 1989; for a more recent view see Duval
and Arnold, 2013).
In a second method, dosimeters are inserted into the
site, which measure the cumulative external dose, again
the g plus cosmic contributions. These are based on one
of the other trapped charge methods, either TL or OSL.
Both types have their proponents and detractors
(McKeever and Moscovitch, 2003; Richter et al., 2010).
Either type of dosimeter must be left in place. Depending
on the site, they could be retrieved after as short a time as
a week or must remain for 6 months to a year. Thus placing
a dosimeter at the precise sample location, needed for
accuracy, in most sites would imply suspending excava-
tion in that area, hardly desirable. As a result, usually these
are put in the walls, or other areas where they do not inter-
fere with continued excavation. They therefore may not
measure the local dose rate.
Another alternative samples a number of sedimentary
units in the site and does a volumetric average of the dose
rate from each. This may be needed for sites where
a gamma spectrometer is not available, or the cost or com-
plexity of retrieving dosimeters is a problem. However,
both collection of a statistically significant number of sam-
ples and calculation of the resulting average are complex
procedures. In principle, in calculating the volumetric
average, one should consider, for example, exactly where
every rock or bone is in relation to the sample.
If sedimentary analysis is chosen, there are again differ-
ent ways of obtaining the analytical information. Gamma
spectroscopy can be used for direct measurement of sedi-
mentary radioisotope content in the lab. Another method
commonly used is neutron activation analysis (NAA) of
sediments. In NAA, samples taken from the site are
bombarded with neutrons to stimulate nuclear reactions.
Those associated with such elements as uranium, potas-
sium, and thorium, the three most common contributors
to environmental dose, give off energy of specific wave-
length, with intensity proportional to concentration. The
weaknesses in laboratory measurements are first that there
may be other isotopes that are not captured, second that
the actual intensity of radiation can be attenuated by water
251
and thus the sediment moisture content has to be mea-
sured, and third that converting the concentration of,
say, uranium to a dose requires assumptions about the
chain of daughter products (Murray et al., 1992; Grün,
1992). In particular, one has to make an assumption about
whether radon has been captured in the sediment and
therefore contributed to the dose or whether, as a gas, it
has escaped. In addition, one has to estimate the contribu-
tion of cosmic radiation.
The approximations do not invalidate the results. When
two methods such as TL dosimetry and sedimentology
have been used on the same site, they yield similar ages
for archaeological materials (Dibble et al., 2012), despite
being based on different assumptions.
In comparison one would think that finding the internal
dose rate, Dint, would be straightforward. For some mate-
rials it is. Unfortunately, most of the important materials
for ESR dating are more or less porous. We can determine,
again, the current level of radioisotopes in, say, a tooth, but
from experience we know that this level has accumulated
over time, perhaps quickly, immediately after burial, or per-
haps slowly and continuously. Determining this uptake is
a significant complication, most notably for teeth but also
for shells. Historically one modeled three possible types
of uptake: early uptake (EU), in which uptake of uranium
was essentially instantaneous; linear uptake (LU), in which
the rate of uptake is constant over time; and recent uptake
(RU), in which virtually all the uranium is absorbed by
the sample in a short period before the sample is recovered.
Figure 5 represents these options, using a parameter, p, to
define them. Which option is most plausible depends on
the sample environment. A constant environment is, natu-
rally, likely to produce an LU age. A sample that is initially
buried in a lake that then dries up may well fit the EU
model. RU models can occur in two ways. The environ-
ment may change suddenly to one with more available ura-
nium, perhaps due to tectonic activity. Or the sample itself
may change. There is evidence that by 1 Ma, tooth enamel
may develop microcracks, allowing more rapid absorption
of uranium (Skinner et al., 2001b). And, of course, interme-
diate values of p are always possible (Blackwell
et al., 2002). Recently a number of researchers have devel-
oped a coupled ESR/U Series method for teeth that can
determine the time-averaged value of p and thus eliminate
much of the uncertainty in uptake (Grün and McDermott,
1994).
Determining the true uncertainty in the ESR age is dif-
ficult. Of course we know the uncertainty in measure-
ments of, e.g., ED and the modern concentrations of
radioisotopes, but we cannot know with the same certainty
the history of the site. No matter what method is used to
determine Dext, it does not capture the overall variation
in moisture, nor possible changes in overburden with
accompanying change in cosmic dose. Even radon escape
may vary with time. And we cannot quantify what we do
not know. Two approaches can increase confidence in
the results. One is the “isochron” method for tooth enamel
that relies on multiple subsamples from a single tooth
252 ELECTRON SPIN RESONANCE (ESR) DATING, GENERAL PRINCIPLES

Now The 110

Tooth EU LU
Now! 100
Early
Uptake 90
(EU) 80
p = −1

Age (ka)
U)
(L
70

e
ak
U Concentration

pt
60
= s) U
u
uo

0 50
in
p
nt
Co

40
(

5
ar

0.
ne

30
=

1
Li

100 200 300 400 500 600 700

=

2
p
p=

10 External Dose Rate (mGy/y)

5

The
0
Electron Spin Resonance (ESR) Dating, General Principles,
p=

p=2
p=

Tooth Recent Figure 6 Dependence of calculated age on external dose rate.

at
Death
0 about 250 to 600 mGy/year.
Time
0
Death
Electron Spin Resonance (ESR) Dating, General Principles,
Figure 5 Different possible uranium uptake models are defined
by the parameter “p” (Figure courtesy of Dr. Bonnie Blackwell).
(Blackwell et al., 2001). More generally, we can
recalculate the age as we vary factors such as average
water content or cosmic dose rate (Deely et al., 2011).
A simplified example is shown in Figure 5 where dose
rates from approximately 250–600 mGy/year still yield
ages within the errors on the factors we can quantify. This
exceeds variability found in virtually all sites (Figure 6).
Materials used in ESR dating
Carbonate
The most successful examples of ESR dating have
involved carbonate-containing compounds. The major
sample type, mollusks, deserves its own section and will
not be discussed further. It is worth commenting, however,
on speleothems, which were the first samples dated
(Ikeya, 1978). “Speleothem” is a general term for a variety
of cave deposits resulting from dissolution and
reprecipitation in limestone rocks. The general features
of speleothem spectra resemble those of some mollusks.
The primary difficulty in using them is contamination by
inorganic and organic detritus which can lead to interfer-
ing peaks in the spectrum and misleading internal dose
rates. An interlaboratory cross-calibration (Hennig
et al., 1982) showed considerably inconsistencies, includ-
ing disagreement as to the appropriate signal. Lyons
(1989) undertook a systematic study of possible causes.
There continue to be a number of studies (Bahain
et al., 2011), but also suggestions that ESR is of limited
utility for dating speleothems (Döppes et al., 2008).
Uptake As measured, the external dose rate for this sample was
(RU) 400 mGy/year. Error bars show 1 s; within 2 s, ages are
consistent for a given model over a range of dose rates from
l1
Now
A number of early studies of continental uplift used
ESR dating of coral (Skinner, 1985; Schellmann and
Radtke, 2004). Large coral reefs can serve as both the sam-
ple material and the surrounding “sediment.” Although the
signal stability in coral should restrict its use to <500 ka,
good agreement between ESR and other methods on coral
has been found up to about 650 ka without correction for
fading (Radtke et al., 1988). This suggests a large uncer-
tainty in mean lifetimes determined by isothermal
annealing. U-uptake may also be an issue here.
Other carbonate-containing materials have been con-
sidered for dating. Caliche and travertine have contamina-
tion problems. Foraminifera from deep-sea cores have
been studied, with some success, but more work is needed
(Schellmann et al., 2008).
Phosphates
The most useful substance for ESR dating has been tooth
enamel. Hydroxyapatite (HAP), the structural mineral in
teeth, is a phosphate. However, the ESR signal arises from
carbonate inclusions in the HAP crystal, and hence the
parameters for measuring are the same as for carbonates.
The stability of the signal in enamel has been shown to
exceed 1015y (Schwarcz, 1985; Skinner et al., 2000),
allowing dating from as young as 30 ka to 3–4 Ma.
Both ends of this range have been cross-checked
with radiometric methods, 14C for the younger samples
(e. g., Rink et al., 1996) and Ar/Ar at Olduvai Gorge for
the older (Skinner et al., 2001a).
Bone contains the same hydroxyapatite (HAP) as teeth
and presumably the same carbonate impurities. Therefore,
it would seem that bones could be used for dating. Two
factors mitigate against this. The success of dating enamel
is in part due to the signal stability attributable to large
 ELECTRON SPIN RESONANCE (ESR) DATING, GENERAL PRINCIPLES
HAP crystals. Crystal size in bone is considerably smaller.
More importantly, bone is quite porous. Uranium can be
absorbed more easily, but also leached from bone during
burial. This largely prevents precise determination of an
internal dose. Early studies (Mascarenhas et al., 1982;
Kai et al., 1988) seemed to show that ages could be mea-
sured within an order of magnitude and were more reliable
for young bones, presumably because there was less time
for absorption/leaching episodes. Currently, however,
ESR dates on bone should not be considered dependable
(Döppes et al., 2008).
Silicates
Quartz grains in sediment were initially studied using the
E’ signal measureable at room temperature (e.g., Xu and
Zhou, 2009). A recent development has been the measure-
ment in liquid nitrogen (LN2) of optically bleachable sig-
nals in quartz sediment building on pioneering
experiments by Toyoda et al. (2000) and others. There
are three of them, one deriving from aluminum centers,
one from a Ti-H center, and one from Ti-Li. If ages deter-
mined for the three centers agree, then one has confidence
in the results. These applications will be covered in more
detail in a later section.
Quartz, in microcrystalline form, also exists as flint or
chert. The presence of flint in many archaeological sites
encouraged the dating community to look to all possible
dating methods. Unfortunately, since trapped charge
methods record all radiation since formation, ESR dating
of unheated flint would at best date the creation of the
raw material, not its conversion into a tool. The “skinflint”
method hoped to get around this by comparing the damage
to the outermost layer with that of the core, on the hypoth-
esis that b-radiation from the sediment, present only after
knapping, would penetrate the first 2 mm of the sample
and provide a contrast to the overall dose (Schwarcz and
Rink, 2001). Despite some initial success, the lack of addi-
tional studies suggests this method is no longer considered
valid. The dose history of heated flint, however, begins
with the time of heating. Dating heated flint by thermolu-
minescence (TL) is well established. Efforts to use ESR
have been less successful. The quartz-based E’ signal used
for dating must be extracted from a complex spectrum
derived from other minerals and organic inclusions. Its
behavior depends somewhat on the heating temperature
(Toyoda et al., 1993). If the heating temperature was over
about 650 oC, a phase transition prevents the regrowth of
the signal. Perhaps most problematic, there is a base
“unbleachable” signal (to use the TL terminology) which
varies from sample to sample (Skinner and Rudolph,
1997). Thus neither the precision nor the accuracy of dat-
ing by ESR can match that of TL.
In a related study, heated quartz grains were dated to
establish the age of a Lower Paleolithic site in Brittany
(Monnier et al., 1994).
Last, but far from least, is the use of the E’ signal to
study the movement of earthquake faults. Along with the
253
study of speleothems, this was one of the earliest applica-
tions. The principle is that the movement of an earthquake
fault anneals radiation damage in quartz grains primarily
by heat and pressure. Thus the observed modern dose
reflects the age since last faulting event. Results may be
affected by incomplete resetting – large grains may retain
some of the original signal (Buhay et al., 1988). As well as
the E’ center, signals attributed to germanium, aluminum,
titanium and the “OHC” center have been studied. The Ge
signal has a short lifetime (in terms of this type of study),
estimated to be 100 ka. As noted for the optical bleaching,
the Ti center is less stable than that for Al. While good
concordance of ages from different centers has been
observed, the ESR ages can exceed the geological esti-
mates, presumably due to incomplete resetting (Ikeya,
1993). Other studies have agreed better with geology
(Fukuchi et al., 1986).
Sulfate minerals
Gypsum (calcium sulfate) is found in a number of envi-
ronments as the result of precipitation or evaporation. Sul-
fite radicals have well-characterized g-values around g ¼
2.004. This is, unfortunately, the same region as the car-
bonate spectra, so natural gypsum may show both.
A second signal at g ¼ 2.008 can also be used for dating
(Kasuya et al., 1991). Although the mean lifetime appears
to exceed 1012 years, applications to date have concen-
trated on the time period <500 ka. They include additional
efforts to study tectonics (Mathew et al., 2004). However,
a comparison of ESR and TL ages showed consistent
underestimation by ESR relative to TL, tentatively
ascribed to uncertainty in the relative efficiencies of a-,
b-, and g-radiation in inducing defects (Nambi, 1982).
In an interesting extension of ESR dating to other
regions of the earth, barite (barium sulfate) dating has elu-
cidated some aspects of hydrothermal vent formation in
the depths of the ocean. Although the calculations are
extremely complex, due to the shape of these vents, the
proof of concept is encouraging (Sato et al., 2011).
Space science
Ikeya (1993) championed the concept of using ESR dating
to study formations on the outer planets, using the OH.
radical. While too unstable to be measured at ambient
Earth temperatures, he believed that the low temperatures
in space would allow one to measure an accumulated
dose. He did not think that this technique would be ready
for use in less than a century, however.
Non-radiolytic dating
So far we have discussed only free radicals created by
radioactivity. In principle, any increase in free radical con-
centration with time could be used for dating. Oxidation of
natural products also creates radicals. Ikeya and
coworkers experimented with some preliminary work on
organic degradation due to light and/or oxygen. Among
the substances tested were oils on potato chips, and blood.
254 ELECTRON SPIN RESONANCE (ESR) DATING, GENERAL PRINCIPLES

Electron Spin Resonance (ESR) Dating, General Principles, Bibliography

Table 2 Materials commonly used in ESR dating. The first
Aitken, M. J., Stringer, C. B., and Mellars, P. A. (eds.), 1993. The
column notes whether the material has a signal that is either
Origins of Modern Humans and the Impact of Chronometric
zero at t ¼ 0 or determinable. The second notes whether this
Dating. Princeton, NJ: Princeton University Press.
signal grows in a predictable manner so that an accumulated
Bahain, J. J., Burrafato, G., Falguères, C., Gueli, A. M., Stella, G.,
dose can be established. Finally the third column notes
Troja, S. O., and Zuccarello, A. R., 2011. ESR and U/Th dating
whether the signal stability is adequate for practical uses. The
methodologies applied to carbonates from Southern Italy. In
reliability of ages has been cross-checked by other methods for
Turbanti-Memmi, I. (ed.), Proceedings of the 37th International
these materials as well. The greater the number of emoticons
Symposium on Archaeometry. Heidelberg: Springer.
(positive or negative), the more confidence in the opinion
Blackwell, B. A. B., Skinner, A. R., and Blickstein, J. I. B., 2001.
Signal AD Age ESR isochron exercises: how accurately do modern dose rate
measurements reflect paleodose rates? Quaternary Geochronol-
Carbonates ogy, 20, 1031–1039.
Coral ☺☺☺ ☺☺☺ ☺☺☺ Blackwell, B. A. B., Skinner, A. R., Brassard, P., and Blickstein,
Mollusk ☺☺ ☺☺ ☺☺ J. I. B., 2002. Uptake in tooth enamel: Lessons from isochron
Speleothem ☺☺ ☺☺ ☺☺ analyses and laboratory simulation experiments. Advances in
Travertine ☺ ☺☺ ESR Applications, 18, 65–86.
Foraminifera ☺ ☺ ☺☺ Buhay, W. M., Schwarcz, H. P., and Grün, R., 1988. ESR dating of
Caliche ☺ ☺ ☺☺ fault gouge: the effect of grain size. Quaternary Science
Hydroxyapatite Reviews, 7, 515–522.
Tooth ☺☺☺ ☺☺☺ ☺☺☺ Callens, F. J., Verbeeck, R. M., Naessens, D. E., Matthys, P. F., and
Bone ☺ ☺ ☺☺ Boesman, E. R., 1989. Effect of carbonate content on the ESR
Silicates spectrum near g ¼ 2 of carbonated calcium apatites synthesized
Sediment from aqueous media. Calcified Tissue International, 44(2),
E’ ☺☺ ☺☺ ☺ 114–124.
Al, Ti, ☺☺ ☺☺ ☺ Deely, A. E., Blackwell, B. A. B., Mylroie, J. E., Carew, J. L.,
Faults, E’ ☺☺ ☺ ☺ Blickstein, J. I. B., and Skinner, A. R., 2011. Testing cosmic dose
Chert, E’, Al ☺☺ ☺☺ ☺☺☺ rate models for ESR: dating corals and molluscs on San Salva-
dor, Bahamas. Radiation Measurements, 46, 853–859.
Dibble, H., Aldeias, V., Alvarez-Fernandez, E., Blackwell, B. A. B.,
Hallett-Desguez, E., Jacobs, A., Goldberg, P., Lin, A., Morala,
A., Olszewski, D., Reed, K., Reed, D., Rezek, A., Richter, D.,
Roberts, R. G., Sandgathe, D., Schurmans, U., Skinner, A. R.,
Steele, T., and El-Hajraoui, M., 2012. New excavations at the
In general, however, the signal lifetimes and saturation site of contrebandiers Cave, Morocco. PaleoAnthropology,
behavior severely limit the dating potential. A comprehen- 2012, 145–201.
sive review can be found in Ikeya (1993). There has been Döppes, D., Kempe, S., and Rosendahl, W., 2008. Dated paleonto-
essentially no further work in this type of application logical cave sites of Central Europe from Late Middle Pleisto-
since then. cene to early Upper Pleistocene (OI5 to OIS8). Quaternary
International, 187, 97–104.
Duval, M., 2012. Dose response curve of the ESR signal of the
Aluminum center in quartz grains extracted from sediment.
Summary Ancient TL, 30(2), 41–50.
This outline by no means exhausts the efforts that Duval, M., and Arnold, L. J., 2013. Field gamma dose-rate assess-
have been made to use ESR for dating. The following ment in natural sedimentary contexts using LaBr3(Ce) and NaI
table can be used for an overview of the current state (Tl) probes: a comparison between the “threshold” and “win-
of ESR dating. It is, naturally, oversimplified, but dows” techniques. Applied Radiation and Isotopes, 74, 36–45.
Duval, M., Falguères, C., and Bahain, J.-J., 2012. Age of the oldest
generally indicates what materials have been good hominin settlements in Spain: contribution of the combined
sources for dating sites and which have serious flaws U-series/ESR dating method applied to fossil teeth. Quaternary
(Table 2). Geochronology, 10, 412–417.
A literature search will soon show that aside from the Fukuchi, T., Imai, N., and Shimokawa, K., 1986. ESR dating of fault
examples quoted above, many other minerals have movement using various defect centres in quartz; the case in
been tested for their ESR response to radiation. Dating the western South Fossa Magnam, Japan. Earth & Planetary
Science Letters, 78, 121–128.
applications are the result of low dose rates delivered Grün, R., 1992. Suggestions for minimum requirements for
over a long period of time. Another major use of ESR reporting ESR age estimates. Ancient TL, 10, 37–41.
for radiation sensitivity is determining high doses deliv- Grün, R., 2007. Electron spin resonance dating. In Scott A. Elias
ered over a short period of time, as in accident dosimetry. (ed) Encyclopedia of Quaternary Science. Oxford: Elsevier,
Since in this latter case signal stability is not an issue, pp. 1505–1516.
useable materials abound. The future will undoubtedly Grün, R., and McDermott, F., 1994. Open system modeling for
U-Series and ESR dating of teeth. Quaternary Geochronology
demonstrate that some of these may be transferrable to (Quaternary Science Reviews), 13, 121–125.
the dating category. Certainly there is plenty of scope Haskell, E., Kenner, G., Hayes, R., Sholom, S., and Chumak, V.,
for experimentation; no scientific field should ever be 1997. An EPR intercomparison using teeth irradiated prior to
considered closed. crushing. Radiation Measurements, 27(2), 419–424.
 ELECTRON SPIN RESONANCE SPECTROMETER
Hennig, G. J., Geyh, M. A., and Grün, R., 1982. The first inter-
laboratory ESR comparison project phase II: evaluation of
equivalent doses (ED) of calcites. Nuclear Tracks and Radiation
Measurements, 10(4–6), 945–952.
Ikeya, M., 1978. Electron spin resonance as a method of dating.
Archaeometry, 20, 147–158.
Ikeya, M., 1993. New Applications of Electron Spin Resonance.
Singapore: World Scientific.
Kai, A., Miki, T., and Ikeya, M., 1988. ESR dating of teeth, bones
and eggshells excavated at a Paleolithic site of Doura Cave,
Syria. Quaternary Science Reviews, 7, 503–507.
Kasuya, M., Brumby, S., and Chappell, J., 1991. ESR signals
from natural gypsum single crystals: implications for ESR
dating. Nuclear Tracks and Radiation Measurements, 18,
329–333.
Lyons, R., 1989. Contributions to Electron Spin Resonance Dating.
PhD thesis, University of Auckland.
Mascarenhas, S., Baffa Filho, O., and Ikeya, M., 1982. Electron spin
resonance dating of human bones from Brazilian shell-mounds
(Sambaquis). American Journal of Physical Anthropology, 59,
413–417.
Mathew, G., Rao, T.K.G., Shoni, P.S., Karanth, P.V., 2004. ESR
dating of interfault gypsum from Katrol hill range, Kachchh,
Gujarat: Implications for neotectonism. Current Science, 89,
1269–1274.
McKeever, S. W. S., and Moscovitch, M., 2003. On the advantages
and disadvantages of optically stimulated luminescence dosime-
try and thermoluminescence dosimetry. Radiation Protection
Dosimetry, 104, 263–270.
Molodkov, A., 1998. ESR dating of quaternary shells: recent
advances. Quaternary Science Reviews, 7, 477–484.
Monnier, J. L., Hallégouët, B., Hinguant, S., Laurent, M.,
Auguste, P., Bahain, J.-J., Falguères, C., Gebhardt, A.,
Marguerie, D., Molines, N., Morzadec, H., and Yokoyama, Y.,
1994. A new regional group of the Lower Palaeolithic in
Brittany (France), recently dated by Electron Spin Resonance.
Comptes Rendus de l'Académie des sciences de Paris, 319,
155–160.
Murray, A. S., Olley, J. M., and Wallbrink, P. J., 1992. Natural radio-
nuclide behavior in the fluvial environment. Radiation Protec-
tion Dosimetry, 45, 285–288.
Nambi, K. S. V., 1982. ESR and TL dating studies on some marine
gypsum crystals. PACT Journal, 6, 314–321.
Prescott, J. R., and Hutton, J. T., 1988. Cosmic ray and gamma ray
dosimetry for TL and ESR. Nuclear Tracks and Radiation Mea-
surements, 14, 223–227.
Radtke, U., Grün, R., and Schwarcz, H. P., 1988. Electron spin res-
onance dating of the pleistocene coral reef tracts of barbados.
Quaternary Research, 29, 197–215.
Richter, D., Dombrowski, H., Neumaier, S., Guibert, P., and Zink,
A. C., 2010. Environmental gamma dosimetry with OSL of
a-Al2O3:c for in situ sediment measurements. Radiation Protec-
tion Dosimetry, 141, 27–35.
Rink, W. J., 1997. Electron spin resonance (ESR) dating and ESR
applications in quaternary science and archaeometry. Radiation
Measurements, 27, 975–1025.
Rink, W. J., Schwarcz, H. P., Lee, H. K., Valdés, V. C., de Quirós,
F. B., and Hoyos, M., 1996. ESR dating of tooth enamel: com-
parison with AMS 14C at El Castillo Cave, Spain. Journal of
Archaeological Science, 23, 945–951.
Sato, F., Toyoda, S., Banerjee, D., and Ishibashi, J., 2011. Thermal
stability of ESR signals in hydrothermal barite. Radiation
Measurements, 46, 866–870.
Schellmann, G., and Radtke, U., 2004. A revised morpho- and
chronostratigraphy of the Late and Middle Pleistocene coral reef
terraces on Southern Barbados (West Indies). Earth-Science
Reviews, 64(3–4), 157–187.
255
Schellmann, G., Beerten, K., and Radtke, U., 2008. Electron spin
resonance (ESR) dating of Quaternary materials. Eiszeitalter
und Gegenwart (Quaternary Science Journal), 57, 150–178.
Schwarcz, H. P., 1985. ESR studies of tooth enamel. Nuclear
Tracks, 10, 865–867.
Schwarcz, H. P., and Rink, W. J., 2001. Skinflint dating. Quaternary
Science Reviews, 20, 1047–1050.
Skinner, A. R., 1983. Overestimate of stalagmitic calcite ESR dates
due to laboratory heating. Nature, 304, 152.
Skinner, A. F., 1985. Comparison of ESR and 230Th/234U ages in
fossil aragonitic corals. In Proceedings of the First International
Symposium on ESR Dating, Ionics Ube, Japan, pp. 135–138.
Skinner, A. R., and Rudolph, M. N., 1997. The use of the Eℜ signal
in flint for ESR dating. Applied Radiation and Isotopes, 47,
1399–1404.
Skinner, A. R., Blackwell, B. A. B., Chasteen, N. D., Shao, J., and
Min, S. S., 2000. Improvements in dating tooth enamel by
ESR. Applied Radiation and Isotopes, 52, 1337–1344.
Skinner, A. R., Blackwell, B. A. B, Lothian, V.S., 2001a. Calibrat-
ing ESR dating using 2 Myr-old teeth Paleoanthropology Soci-
ety Abstracts, P.16
Skinner, A. R., Chasteen, N. D., Shao, J., and Blackwell, B. A. B.,
2001b. Q-band studies of the ESR signal in tooth enamel. Qua-
ternary Science Reviews (Quaternary Geochronology), 20,
1027–1030.
Steinhauser, G., Sterba, J. H., and Hammer, V. M., 2013. Nuclear
forensics of a colored gemstone: evidence of proton bombard-
ment of a blue topaz. Applied Radiation and Isotopes, 75, 18–21.
Toyoda, S., Ikeya, M., Dunnell, R. C., and McCutcheon, P. T., 1993.
The use of electron spin resonance (ESR) for the determination
of prehistoric lithic heat treatment. Applied Radiation and
Isotopes, 44, 227–231.
Toyoda, S., Voinchet, P., Falguères, C., Dolo, J. M., and Laurent,
M., 2000. Bleaching of ESR signals by the sunlight:
a laboratory experiment for establishing the ESR dating of
sediments. Applied Radiation and Isotopes, 52, 1357–1362.
Walther, R., and Zilles, D., 1994. ESR studies on bleached sedimen-
tary quartz. Quaternary Science Reviews, 13, 611–614.
Xu, L., and Zhou, S., 2009. Quaternary glaciations recorded by gla-
cial and fluvial landforms in the Shaluli Mountains, Southeastern
Tibetan Plateau. Geomorphology, 103(2), 268–275.
Zavoisky, E., 1945. Spin-magnetic resonance in paramagnetics.
Journal of Physics, 9, 245.
Cross-references
Electron Spin Resonance (ESR) Dating of Coral
Electron Spin Resonance (ESR) Dating of Fossil Tooth Enamel
Electron Spin Resonance Spectrometer
Radiation and Radioactivity
Sediment, ESR
ELECTRON SPIN RESONANCE SPECTROMETER
Jeroen W. Thompson
Department of Medical Physics and Applied Radiation
Sciences, McMaster University, Hamilton, ON, Canada
Synonyms
Electron paramagnetic resonance spectrometer; EPR
spectrometer; ESR spectrometer
256 ENVIRONMENTAL RELEASES
Definition
Instrument for the detection of electron paramagnetic tran-
sitions, consisting of a tunable source of electromagnetic
radiation, a resonant cavity, an electromagnet, and
a detector.
The electron spin resonance (ESR) spectrometer is
employed to detect electron paramagnetic resonance tran-
sitions. Although a wide variety of designs exist, the typ-
ical commercial continuous-wave spectrometer consists
of (1) a microwave source tunable within a limited range
of frequencies near 9.4 GHz; (2) a sample cavity, often
cylindrical, capable of maintaining an electromagnetic
standing wave; (3) an electromagnet capable of generating
and ramping a uniform magnetic field, along with
superimposed rapid oscillation up to 100 kHz in fre-
quency; (4) a detection system which amplifies the
reflected microwave signal; and (5) an internal magnetic
field standard such as MnO.
Transitions between electronic energy levels may be
induced by polarized electromagnetic radiation such that
the energy of the incident radiation is equal to the differ-
ence in energy between the energy levels in question. In
other words, the microwave source is used to generate
a fixed energy (E) of incident radiation with frequency n
(where E ¼ hn) which is reflected from the sample in the
resonant cavity, and the energy reflected is recorded by
the detector. The energy levels will vary with the strength
of an external magnetic field. Therefore, “sweeping” the
external magnetic field strength from low to high will
allow for resonant transitions to occur at the appropriate
combination of (fixed) microwave energy (frequency)
and external magnetic field. Resonance involves a net
absorption of incident energy, which is detected as
a relative loss in reflected energy.
The faithful reproduction of any given resonant absorp-
tion signal requires careful attention to instrumental
parameters, including (1) incident microwave power,
(2) center field (magnetic field strength at the midpoint
of the sweep), (3) scan width (range of the swept magnetic
field), (4) modulation amplitude (amplitude of the
superimposed 100 kHz oscillating magnetic field),
(5) sweep time, (6) time constant (roughly, the characteris-
tic time over which instrumental data is averaged),
(7) number of sweeps, and (8) detector gain.
Bibliography
Weil, J. A., Bolton, J. R., and Wertz, J. E., 1994. Electron Paramag-
netic Resonance: Elementary Theory and Practical Applica-
tions. New York: Wiley.
Cross-references
Electron Spin Resonance (ESR) Dating, General Principles
ENVIRONMENTAL RELEASES
Robert Morrison
Hawi, HI, USA
Definition
Environmental Forensics. The systematic and scientific
evaluation of physical, chemical, and historical informa-
tion for the purpose of developing defensible scientific
and legal conclusions regarding the source or age of
a contaminant release into the environment (Morrison
and Murphy, 2013).
Introduction
The study of contaminant releases into the environment is
termed environmental forensics. Forensic techniques used
to date and identify the origin of a contaminant release
include dendroecology, aerial photogrammetry,
compound-specific isotopic analysis (CSIA), congener
analysis, contaminant transport modeling (liquid and col-
loidal), molar ratio analysis, biological, isotopic, and
chemical pattern recognition, corrosion modeling, lipid
and associated biological techniques, scanning electron
microscopy (SEM), and advanced multicomponent analy-
sis of chemical data. The following text describes selected
techniques used to identify the source and age of a con-
taminant release into the environment.
Source identification
Forensic techniques for identifying a contaminant source
are often similar to environmental investigations designed
to define the areal extent of a contaminant release
for remediation purposes. Environmental forensics tran-
scends traditional investigative techniques, however, by
employing scientific techniques in combination to identify
and discriminate between sources of a contaminant
release. Three approaches used to identify contaminant
sources include (1) association of a diagnostic chemical
with a particular use and/or time frame; (2) chemical,
biological, and isotopic techniques; and (3) congener
analysis.
Diagnostic chemicals used for source identification
Examples of diagnostic chemicals indicative of a discrete
production time frame for source identification and release
dating in the United States (USA) include (USEPA, 1978;
Kaplan et al., 1997; ATSDR, 2002a; ATSDR, 2002b;
Morrison and Murphy, 2006):
• 1,4-dioxane with methyl chloroform indicative of a
post-1951 release
• Monsanto Aroclor 1016 (produced between 1971 and
1974)
• Aroclor 1260 as the predominant Aroclor used in
electrical transformers prior to 1971
 ENVIRONMENTAL RELEASES
• Cessation of DDT production in 1972, aldrin/dieldrin in
1974, and mirex in 1962 to 1978
• The gasoline oxygenate methyl tertiary butyl ether
(MTBE) as indicative of a post early-1980s release
Diagnostic chemicals in a formulation specific to
a particular application offer the potential to link
a chemical with a particular temporal activity; an example
is the presence of the stabilizer thymol with trichloroethy-
lene (TCE) as indicative of pharmaceutical grade TCE
used prior to 1984 (Totonidis, 2005). The presence of
octene, nonene, and n-pyrrole with tetrachloroethylene
(PCE) in an environmental sample is indicative of its use
as a dielectric fluid in electrical transformers, likely in
the 1980s (Morrison and Murphy, 2013). A substantial
body of peer-reviewed literature is available that describes
using diagnostic chemicals for source identification and/or
dating of contaminant releases (Morrison, 1999; Murphy
and Morrison, 2002; Morrison and Murphy, 2006; Mur-
phy and Morrison, 2007; Peters et al., 2007; Morrison
and O’Sullivan, 2010; Morrison and O’Sullivan, 2012;
Morrison, 2013; Murphy and Morrison, 2014; Morrison
and O’Sullivan, 2014).
Chemical, biological, and isotopic techniques
Chemical techniques for source identification include
molar ratio analysis and degradation relationships, includ-
ing the chlorinated solvent PCE degradation series (PCE
! TCE ! 1,2-DCE ! vinyl chloride) for which common
plots include [TCE/PCE], [DCE/TCE], [DCE]/[PCE +
TCE], [cis + trans 1,2 DCE + VC]/[PCE + TCE],
[VC]/[PCE + TCE], [cis + trans- 1,2DCE]/[PCE +
TCE], and [VC]/[cis + trans-1,2 DCE] as a means to iden-
tify and discriminate between multiple sources of PCE.
Techniques used to distinguish sources for crude petroleum
hydrocarbons include diagnostic biomarker ratios, includ-
ing acyclic isoprenoids (e.g., (pristine)/(phytane), (pris-
tine)/(n-C17) and (phytane)/(n-C18), terpanes (e.g., (C21)/
(C23 tricyclic terpane) and (C28bisnorhopane)/(C30ab
hopane)) and steranes (e.g., (C27abb/C29abb steranes at
m/z 218) and (C28abb)/(C27abb + C28abb + C29abb) at
m/z 218) (Morrison and Murphy, 2006). The normalized
distribution, ratios and cross-plots of polyaromatic hydro-
carbon (PAHs) sources are similarly available for source
identification (Figure 1), some of which are used to distin-
guish between petrogenic and anthropogenic sources
(e.g., (fluoranthene + pyrene)/(C2+C3+C4 phenanthrenes)
and (phenanthrene/anthracene) versus (fluoranthrene/
pyrene)) (Fig. 2) (Morrison and Murphy, 2006; Wang and
Stout, 2007; Wenning and Martello, 2014).
For distinguishing a release from a sanitary sewer from
other sources, detergent metabolites including
4-n-octylphenol, 4-tert-octylphenol, diethoxynonylphenol,
NPEO-2, OPEO-2, ethoxynonylphenol NPOE-1,
ethoxyoctylphenol OPEO-1, and 4-nonylphenol are avail-
able. Other indicators include pharmaceutical indicators,
257
0.7
Fluoranthene / (Fluoranthene + Pyrene)
0.6 Petroleum or Combustion Fugitive Combustion
0.5
0.4 Petroleum
Petroleum or Combustion
0.3
Petroleum
0.2
Petroleum or Combustion
0.1
0.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
(Phenanthene + Anthracene)
(Phenanthrene + Anthracene + C1 Phenanthrene)
Environmental Releases, Figure 1 Normalized cross-plots of
polyaromatic hydrocarbons used to distinguish between
petrogenic and anthropogenic sources of petroleum
hydrocarbon releases.
trihalomethanes, and human fecal bacteria such as
Bacteroidales and viruses (i.e., hepatitis A, adenovirus,
and enterovirus). Other biological markers include sterols
(i.e., cholesterol, b-sitosterol, 3b-coprostanol, and
stigmastanol, b- and a-cholestanone, campesterol, stigmas-
terol, and epicoprostanol) as indicators of fecal contamina-
tion. The fecal sterol 3b-coprostanol is produced in the gut
of some mammals, including humans, pigs and cats, by the
microbially mediated reduction of cholesterol under anaer-
obic conditions. Sterols and stanols can be combined to dis-
tinguish between the release of feces containing swine and
cow waste. The specificity of steroids as a means to differ-
entiate cow feces from pig slurries is available by consider-
ing fecal stanol diagnostic compounds, including
copropstanol, epicopropstanol, campenstanol, sitostanol,
24-ethylocoprostanol and 24-ethylepicoprostanol and/or
their ratios. Another forensic opportunity is the detection
of nanoparticles (1–100 nm), such as nanosilver (n-Ag)
and nano-titanium oxide (n-TiO2) that are associated with
fabric and exterior paints, respectively, in environmental
samples as surrogates for the releases of contaminants from
these potential sources (Benn and Westerhoff, 2008).
Bulk and compound-specific isotopic analysis (CSIA)
can be used to identify and discriminate between different
sources of contamination. Examples of the use of CSIA
and bulk isotope analysis include whether methane gas is
biogenic or anthropogenic (1H/2H); the origin of sulfur
from naturally occurring hydrothermal sulfate or from the
oxidation of reduced sulfur in pyrite-rich ore materials
using four stable isotopes (32S, 33S, 34S, 36S); discrimina-
tion between sources of crude oil releases (13C/12C); char-
acterization of the biochemical evolution of dissolved
organic carbon; examination of biodegradation of RDX
258 ENVIRONMENTAL RELEASES

1.0
Bunker C

Relative Abundance
0.8

0.6

0.4

0.2

0.0
N0
N1
N2
N3
N4
Acl
Ace
F0
F1
F2
F3
AN
P0
P1
P2
P3
P4
D0
D1
D2
D3
D4
FL
PY
FP1
FP2
FP3
BaA
C0
C1
C2
C3
C4
BbF
BkF
BeF
BaP
ID
DA
BgP
1.0
Relative Abundance

0.8 Creosote

0.6

0.4

0.2

0.0
N0
N1
N2
N3
N4
Acl
Ace
F0
F1
F2
F3
AN
P0
P1
P2
P3
P4
D0
D1
D2
D3
D4
FL
PY
FP1
FP2
FP3
BaA
C0
C1
C2
C3
C4
BbF
BkF
BeF
BaP
ID
DA
BgP
Environmental Releases, Figure 2 Normalized histogram using the relative abundance of various polyaromatic hydrocarbons
(PAHs) listed along the x-axis to distinguish between creosote and Bunker C fuel.

(hexahydro-1,3,5-trinitro-1,3,5-triazine) using d15N
18
and d O enrichment; MTBE (methyl tertiary butyl
ether) source identification using (13C/12C); the
origin (naturally occurring or anthropogenic) of
perchlorate using 18O/16O, 35Cl/37Cl, and 17O/16O
ratios; the kinetics of natural attenuation/degradation of
contaminants; the origin of PCBs, pesticides, and lead; the
identification of sulfur sources using stable sulfur isotope
signatures; and the origin of PCE and TCE using differ-
ences in carbon, chlorine, and/or hydrogen isotope ratios.
In hydrology investigations, isotopes are used to estimate
ages of groundwater based on 18O/16O and assumptions
about past temperatures and recharge compositions. In
environmental forensic investigations, the primary isotopes
and ratios include hydrogen (2H/1H), carbon (13C/12C),
nitrogen (15N/14 N), oxygen (18O/16O), sulfur (34S/32S),
and for chlorinated solvents, chlorine (37Cl/35Cl).
CSIA is also used to distinguish between naturally occur-
ring and anthropogenic sources of chloroform in soil gas
and groundwater. In one study, d13C values of chloroform
(
22.8 to
26.2 %) in soil gas collected in a forested area
were within the same range as values from the soil organic
matter (
22.6 to
28.2 %) but were more enriched in
13
C compared to industrial chloroform (d13C values of

43.2 to
63.6 % (Hunkeler et al., 2012)).
Congener analysis
The term congener refers to individual compounds,
regardless of their homologue class. PCBs, dioxins, and
furans are conducive to congener analysis for source dis-
crimination. Figure 3 is a normalized histogram for virgin
PCE solvent and sludge residue from a distillation still at
a facility in Japan (Fukai and Nakata, 2001). Many of
the chlorinated dibenzodioxins (CDDs) and chlorinated
dibenzofurans (CDFs) in the new PCE solvent in Figure 3
are OCDD and HpCDD enriched in the distillation sludge,
thereby providing a potential means to (a) develop
a working hypothesis regarding the source of the PCE
and CDD/CDF compounds found with a particular
PCE/dioxin/furan profile and (b) whether the PCE origi-
nated from new or spent PCE.
Dating a contaminant release
Forensic techniques used for source identification
can often bracket the time when a release occurred.
Another approach is contaminant transport modeling
of atmospheric, surface, and subsurface releases of con-
taminants. Forensic contaminant transport models
include simulations to identify the age of contaminant
releases in the atmosphere, liquid releases on
a paved/unpaved surface, liquid transport through the
unsaturated zone, vapor transport, and intrusion models
in soil and groundwater models. Groundwater models
used to date a contaminant release include (a) measuring
the leading edge of a contaminant plume, (b) hydrolysis
modeling of methyl chloroform, and (c) inverse
modeling.
 ENVIRONMENTAL RELEASES 259

100%

90% Octachlorodibenzofuran

80% Heptachlorodibenzofuran

70% Hexachlorodibenzofuran

Pentaachlorodibenzofuran
60%
Tetrachlorodibenzofuran
50%
Octachlorodibenzo-p-dioxin
40%
Heptachlorodibenzo-p-dioxin
30% Hexachlorodibenzo-p-dioxin

20% Pentaachlorodibenzo-p-dioxin

10% Tetrachlorodibenzo-p-dioxin

0%
New PCE Solvent PCE Solvent Distillation
Sludge

Environmental Releases, Figure 3 Normalized histogram of TCDD and PCDD congeners in new PCE solvent and in distillation
sludge from a dry cleaning facility using PCE.

Leading edge of a contaminant plume in plume. The groundwater seepage velocity and retardation
groundwater is expressed as:
The term L is the length of the contaminant plume. Some KI
portion of L, DL, is due to longitudinal dispersion. If the u¼ ð5Þ
groundwater seepage velocity is u and the retardation fac- yϵ
tor for the contaminant in question is R, so that the contam- and
inant transport velocity is u/R, then the release date, t, is
estimated as rb K d
R¼1þ ð6Þ
yϵ
R
t ¼ ðL
DLÞ : ð1Þ where K and I is the hydraulic conductivity and gradient
u
over the plume length, respectively, ye is the effective
The longitudinal dispersion can be estimated as porosity or interconnected pore space of the aquifer, rb
rffiffiffiffiffi rffiffiffiffiffiffiffiffiffi is the aquifer material dry bulk density, and Kd is the
Dt aL ut pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi adsorption distribution coefficient. Kd is often expressed
DL ¼ 2 ¼2 ¼ 2 aL ðL
DLÞ as
R R
pffiffiffiffiffi K d ¼ K oc f oc ð7Þ
ffi 2 aL L ð2Þ
where D is the coefficient of hydrodynamic dispersion, aL where Koc is the distribution coefficient for partitioning of
is the longitudinal dispersivity, and molecular dispersion the solvent between water and organic carbon and foc is the
is neglected. The factor of two is included because this fraction organic carbon in the soil. With substitutions, the
appears as part of the characteristic length scale in two- expression for the elapsed time since a release to ground-
and three-dimensional analytical solutions. The approxi- water becomes
  
mate solution for DL, for large L, is yϵ r K oc f oc pffiffiffiffiffiffiffiffi

pffiffiffiffiffi pffiffiffiffiffiffiffiffi t¼ 1þ b L
2 aL : ð8Þ

DL ffi 2 aL L
2a ffi 2 aL L ð3Þ I KI yϵ

and The contaminant plume length is related to the source

term, groundwater velocity, and contaminant degradation.
 
R pffiffiffiffiffi
t¼ ðL
2 aL LÞ ð4Þ Hydrolysis of methyl chloroform
u
Methyl chloroform (1,1,1-TCA) is unique in that its degra-
which illustrates the importance of correcting for longitu- dation is hydrolysis driven which can be used to date its
dinal dispersion as it decreases with the length of the release. The method assumes that the rate of chemical
260 ENVIRONMENTAL RELEASES
hydrolysis of methyl chloroform to acetic acid is known,
as is the rate of the elimination reaction to form
1,1-dichloroethylene (1,1-DCE). Other assumptions
include that the hydrolysis rate and groundwater tempera-
ture are constant, the biodegradation of TCA and 1,1-DCE
and the effects of chromatographic separation are insignif-
icant, and that different volatilization rates on the
DCE/TCA ratio are negligible. Given these assumptions,
the reactions describing the abiotic degradation of methyl
chloroform begin as
CH3 CCl3 ! CH3 C þ Cl2 þ Cl
, ð9Þ
which is followed by a substitution reaction yielding
hydrochloric acid and acetic acid (CH3COOH) as
CH3 C þ Cl2 þ Cl
þ 2H2 O ! CH3 COOH þ 3HCl:
ð10Þ
Alternatively, the hydrogen can be eliminated to yield
DCE (CCl2) as
CH3 C þ Cl2 þ Cl
! CH2 ¼ CCl2 þ HCl
:
ð11Þ
These reactions are considered to be first order and tem-
perature dependent. As a consequence, the reaction can
be described by the general form of the Arrhenius equa-
tion to describe the temperature dependence of the rate
of methyl chloroform degradation where the first-order
rate constant for TCA (k) is
k ¼ Ae
Ea=RT ð12Þ
where A is the Arrhenius constant; Ea is the activation
energy, estimated from laboratory data as 8.7  1013 s
and 122.8 kJ/mol- K respectively; R is the universal gas
constant (8.3145  10
3 kJ/mol-K); and T is temperature
( K). The rate constant for the abiotic degradation of TCA
is equal to the sum of the individual rate constants in the
substitution and elimination reactions as
k ¼ ks þ ke ð13Þ
and
 
½TCA ¼ TCAðt ¼ 0Þ e
kt ð14Þ
and
ke    degree of uncertainty (Michalak, 2001).
½DCE ¼ TCAðt¼0Þ 1
e
kt ð15Þ
k
where t is time and the molar values of TCA and DCE are
bracketed. Equations (14) and (15) solve for t as
 
1 k ½DCE
t ¼ ln 1 þ : ð16Þ
k k e ½TCA
In Eq. (16), t is the estimated time that methyl chloroform
entered the groundwater (i.e., without accounting for the
time required to migrate through the vadose zone) and is
obtained by substituting a single measured value for the
molar ratio of [DCE]/[TCA] and groundwater temperature
or by substituting time series field measurements. The
field estimate of the abiotic rate constant for k is estimated
by plotting time series measurements.
When TCE is present and its biodegradation is
significant, the cis-1,2 dichloroethene (cis-1,2 DCE) con-
centration from this process is approximately two orders
of magnitude greater than the 1,1-DCE concentration.
Since a small amount of TCE biodegrades to 1,1-DCE
relative to the concentration of cis-1,2 DCE, this reaction
usually affects the 1,1-DCE/TCA ratio only slightly.
Inverse modeling
Inverse modeling is used to determine the historical distri-
bution of observed contamination, identify a contaminant
source, and/or to reconstruct the release history from
a known source (Morrison, 1999; Michalak and Kitanidis,
2003, 2004; Clement, 2011). Inverse models include tech-
niques used to determine (1) the location and magnitude of
a steady-state point source (Sciortino et al., 2000), (2) the
addition of temporal variables to describe a source term(s),
(3) the location or time of release of instantaneous point
sources, and (4) the use of a function estimate to character-
ize the historical contaminant distribution, source loca-
tion, or release history using either a deterministic or
stochastic approach (Woodbury et al., 1998; Alapati and
Kabala, 2000; Aral et al., 2001; Neupauer and Wilson,
2002; Butera and Tanda, 2003; Morrison, 2014). In
stochastic approaches to the solution of inverse problems,
the unknown parameters are described through probability
distributions where the estimated uncertainty is recog-
nized and its importance is often determined, in contrast
to deterministic approaches, where a single value is
acquired for the unknown parameters (Michalak and
Kitanidis, 2003).
The development of stochastic approaches where
parameters are viewed as a jointly distributed random field
is one means of accounting for the uncertainty in contam-
inant concentration estimates, release history, and release
history in a statistical manner, especially if the uncertainty
associated with the estimate is quantified (McLaughlin
and Townley, 1996). The application of inverse models
to identify the source of a contaminant release or source
loading is nonunique (equifinality) with a potentially large
Conclusion
Numerous forensic techniques are available to identify the
source of a contaminant release and/or bracket when it
occurred. Wherever possible, the use of multiple corrobo-
rative forensic evidence should be considered in order to
establish the most accurate estimate of the origin and date
of a contaminant release into the environment.
 EXHUMATION (THERMOCHRONOLOGY) 261

Bibliography Morrison, R. D., and Murphy, B. L., 2013. Chlorinated Solvents

A Forensic Evaluation. Cambridge, UK: Royal Society of
Agency for Toxic Substance and Disease Registry (ATSDR), 2002a.
Chemistry Publishing, p. 572.
Toxicological Profile for DDT, DDE, and DDD. Atlanta, GA:
Morrison, R.D., and O’Sullivan, G., 2010. Environmental Foren-
U.S. Department of Health and Human Services, Public Health
sics Proceedings of the 2009 INEF Annual Conference. Calgary,
Service, p. 403. http://www.atsdr.cdc.gov/toxprofiles/tp35.pdf.
CA: Royal Society of Chemistry Publishing, p. 313.
Accessed April 13, 2014.
Morrison, R. D., and G. O’Sullivan, 2012. Environmental Forensics
Agency for Toxic Substance and Disease Registry (ATSDR),
Proceedings of the 2011 INEF Conference. St. John’s College,
2002b. Toxicological Profile for aldrin/dieldrin. Atlanta, GA:
Cambridge, UK: Royal Society of Chemistry Publishing, p. 345.
U.S. Department of Health and Human Services, Public Health
Morrison, R. D., and O’Sullivan, G., 2014. Environmental Foren-
Service, p. 303. http://www.atsdr.cdc.gov/toxprofiles/tp1.pdf.
sics Proceedings of the 2013 INEF Conference. Pennsylvania
Accessed April 13, 2014.
State University, University Park, PA. Royal Society of Chemis-
Alapati, S., and Kabala, Z. J., 2000. Recovering the release history
try Publishing, p. 214
of a groundwater contaminant using a non-linear least-squares
Murphy, B. L., and Morrison, R. D., 2002. Introduction to Environ-
method. Hydrology Processes, 14(6), 1003–1016.
mental Forensics. Oxford, UK: Elsevier Academic Press, p. 560.
Aral, M. M., Guan, J. B., and Malia, M. L., 2001. Identification of
Murphy, B. L., and Morrison, R. D., 2007. Introduction to Environ-
contaminant source location and release history in aquifers.
mental Forensics, 2nd edn. Oxford, UK: Elsevier Academic
Journal Hydrology Engineering, 6(3), 225–234.
Press, p. 746.
Benn, B. T., and Westerhoff, P., 2008. Nanoparticle silver released
Murphy, B. L., and Morrison, R. D., 2014. Introduction to Environ-
into water from commercial available sock fabrics. Environmen-
mental Forensics, 3rd edn. Oxford, UK: Elsevier Academic
tal Science & Technology, 42(11), 4133–4139.
Press, p. 704.
Butera, I., and Tanda, M. G., 2003. A geostatistical approach to
Neupauer, R. M., and Wilson, J. L., 2002. Backward probabilistic
recover the release history of groundwater pollutants. Water
model of groundwater contamination in non-uniform and tran-
Resources Research, 39(12), 1372, doi:10.1029/2003WR002314.
sient flow. Advances in Water Resources, 25(7), 733–746.
Clement, P., 2011. Complexities in hindcasting models – when
Peters, K. E., Walters, C. C., and Moldowan, J. M., 2007. The Bio-
should we say enough is enough? Groundwater, 49(5), 620–629.
marker Guide: volume 1, Biomarkers and Isotopes in the Envi-
Fukai, T., and Nakata, Y., 2001. A Study of Dry Cleaning Solvents
ronment and Human History, 2nd edn. Cambridge, UK:
and Clothing: Source of Dioxin Exposure? Nagano: Clean Water
Cambridge University Press, p. 451.
World Initiative Institute of Sosei, p. 24. http://www.bredl.org/
Sciortino, A., Harmon, T. C., and Yeh, W., 2000. Inverse modeling for
pdf/DryCleaning_DioxinStudy.pdf. Accessed April 13, 2014.
locating dense nonaqueous pools in groundwater under steady
Hunkeler, D., Laier, T., Breider, G., and Jacobson, O. S., 2012.
flow conditions. Water Resources Research, 36(7), 1723–1735.
Demonstrating a natural origin of chloroform in groundwater
Totonidis, S., 2005. A role for trichloroethylene in developing
using stable carbon isotopes. Environmental Science & Technol-
nation anaesthesia. Kathmandu University Medical Journal,
ogy, 46(11), 6096–6101.
3(2), 181–190.
Kaplan, I. R., Galperin, Y., Lu, S.-T., and Lee, R. P., 1997. Forensic
USEPA, 1978. Review of the Environmental Effects of Pollutants:
environmental geochemistry: differentiation of fuel types, their
1. Mirex and kepone. EPA-600/1-78-013. Health Effects
sources and release time. Organic Geochemistry, 27(5),
Research Laboratory, Cincinnati, OH, p. 235.
289–317.
Wang, Z., and Stout, S., 2007. Oil Spill Environmental Forensics.
McLaughlin, D., and Townley, L. R., 1996. A reassessment of the
Oxford, UK: Academic Press, p. 554.
groundwater inverse problem. Water Resources Research,
Wenning, R. L., and Martello, L., 2014. POPs in marine and fresh-
32(5), 1131–1161.
water environments. Chapter 8. In O’Sullivan, G., and Sandau,
Michalak, A., 2001. Feasibility of contaminant source identification
C., (eds.), Environmental Forensics for Persistent Organic Pol-
for property rights enforcement. In Anderson, T. L., and Hall,
lutants. Oxford, UK: Academic Elsevier Press, pp. 357–390.
P. J. (eds.), The Technology of Property Rights. Lanham, MD:
Woodbury, A. D., Sudicky, E., Ulrych, T. J., and Ludwig, R., 1998.
Rowman and Littlefield, pp. 125–146.
Three dimensional plume source reconstruction using minimum
Michalak, A. M., and Kitanidis, P. K., 2003. A method for enforcing
relative entropy inversion. Journal of Contaminant Hydrology,
parameter nonnegativity in Bayesian inverse problems with an
32(1–2), 131–158.
application to contaminant source identification. Water
Resources Research, 39(2), 1033, doi:10.1029/2002WR001480.
Michalak, A. M., and Kitanidis, P. K., 2004. Estimation of historical
groundwater contaminant distribution using the adjoint state Cross-references
method applied to geostatistical inverse modeling. Water
Dendrochronology, Progress
Resources Research, 40, W08302, doi:10.1029/
2004WR003214.
Morrison, R. D., 1999. Environmental Forensics. Boca Raton, FL:
CRC Press, p. 351.
Morrison, R. D., 2013. Overview, analysis and interpretation of EXHUMATION (THERMOCHRONOLOGY)
environmental forensic evidence. In Siegel, J. A., and Saukko,
P. J. (eds.), Encyclopedia of Forensic Sciences, 2nd edn. Oxford,
UK: Academic Press, pp. 45–50. Pieter van der Beek
Morrison, R.D., 2014. Forensic applications of subsurface contam- Institute des Sciences de la Terre (ISTerre), CNRS,
inant transport models. Chapter 16. In Murphy, B. L., and Mor- Université Joseph Fourier, Grenoble, France
rison, R.D. (eds). Introduction to Environmental Forensics, 3rd
edn. Oxford, UK: Academic Press, pp. 520–551.
Morrison, R. D., and Murphy, B. L., 2006. Environmental Forensics
Contaminant Specific Guide. Oxford, UK: Academic Press, Synonyms
p. 541. Denudation
262 EXHUMATION (THERMOCHRONOLOGY)

Definition exhumation is reserved for rocks or surfaces that have

Exhumation is the vertical upward movement of a rock
particle toward the Earth’s surface.
Introduction
Exhumation is the vertical upward movement of a rock
particle toward the Earth’s surface or, more generally,
the vertical upward component of a rock-particle path with
respect to the Earth’s surface. The opposite vertical move-
ment (downward with respect to the Earth’s surface) is
called burial. Exhumation rate is the vertical upward com-
ponent of the velocity at which rocks move with respect to
the Earth’s surface.
Exhumation, denudation, and erosion
Usage of the term exhumation with respect to the similar
terms erosion and denudation is not rigorously defined.
“Erosion” is often used to refer to the physical processes
that remove material from the Earth’s surface, whereas
denudation and exhumation are commonly used in refer-
ence to the movement of rocks with respect to the Earth’s
surface and are therefore geometrical quantities. How-
ever, the term erosion is also used in the latter context
when referring to relatively short (historical to millen-
nial) timescales. In contrast, exhumation or denudation
refers to longer (geological) timescales and/or larger
(regional) spatial scales. The distinction between exhu-
mation and denudation is even more subtle: in one usage
(Ring et al., 1999), exhumation refers to the upward
movement of rocks with respect to a reference frame
fixed to the Earth’s surface, whereas denudation refers
to the downward movement of the Earth’s surface with
respect to a reference frame fixed to the rock. In another
usage (Summerfield and Brown, 1998), the term
previously been buried, whereas denudation is the more
general term (i.e., a granite can only be denuded, not
exhumed, unless it was denuded in a previous cycle and
subsequently reburied; a sedimentary rock or erosion
surface can be buried and subsequently exhumed).
Most workers, however, consider exhumation and
denudation to be synonyms and use the two terms
interchangeably.
Mechanisms of exhumation
Exhumation of rocks can be caused by various mecha-
nisms (Figure 1): (1) erosion, by removing rocks from
the Earth’s surface, diminishes the vertical distance
between the underlying rocks and the surface (erosional
exhumation); (2) rocks in the footwalls of normal faults
will be exhumed by extensional faulting; (3) a less gener-
ally inferred mechanism is ductile stretching of the crust,
which will lead to pure-shear extension and thinning,
thereby diminishing the vertical distance between the sur-
face and a rock sample; (4) continental rocks that are
brought to great depths by subduction can be subsequently
buoyantly exhumed in a subduction channel, by replacing
denser surrounding mantle rocks. The latter three mecha-
nisms collectively contribute to tectonic exhumation.
The most generally inferred exhumation mechanism is
erosional exhumation. Tectonic exhumation by normal
faulting is an efficient mechanism in metamorphic core
complexes, which are exhumed along extensional detach-
ments that can accumulate several tens of kilometers of
extension. Buoyant exhumation of deep rocks is limited
to subduction settings, while exhumation by ductile
stretching of the crust is rarely invoked. The opposite
mechanisms will lead to burial of rocks: sedimentary
burial beneath an accumulating pile of sediments, tectonic

Exhumation in
Erosion Normal Faulting Ductile thinning subduction channel

Erosional Exhumation
Tectonic Exhumation
Exhumation (Thermochronology), Figure 1 Schematic depiction illustrating exhumation of rocks by different mechanisms: erosion
(erosional exhumation), normal faulting, ductile thinning, exhumation in subduction channel (collectively known as tectonic
exhumation). Top panels show initial situation, bottom panels show situation after exhumation of rock particle (black dot). Double
arrows show depth of rock particle below surface before and after exhumation. Shaded regions represent the crust; other arrows
indicate tectonic movements.
 EXHUMATION (THERMOCHRONOLOGY) 263

burial in the footwall of a thrust fault, burial by ductile Initial rock column after uplift/erosion
pure-shear shortening and thickening, and deep burial in
subduction zones. e
ur
us
ur
Measuring exhumation e
There is no direct way of measuring the amounts or rates
of erosional exhumation. The amount of tectonic exhuma-
tion by normal faulting or ductile extension and thinning
can, in theory, be quantified using structural or tectonic
markers such as extensional fault offsets or strain markers.
In practice, however, these are rarely sufficiently pre-
served to allow such reconstruction. If rocks have been
buried sufficiently deeply to be metamorphosed, meta-
morphic pressure-temperature estimates allow quantifying
the amount of exhumation; such measures are commonly
used to quantify buoyant exhumation of rocks in subduc-
tion settings, for instance. The most generally employed
techniques for quantifying the amount and rates of
exhumation on geological timescales rely on
Thermochronology data. Different thermochronometers
(including (U-Th)/He in apatite, titanite, and zircon; fis-
sion tracks in apatite, titanite, and zircon; K-Ar or
40
Ar-39Ar in K-feldspar, mica, and hornblende) are char-
acterized by different closure temperatures (varying from
~70  C for the apatite (U-Th)/He system to ~500  C for
the hornblende 40Ar-39Ar system). Determining the ages Exhumation (Thermochronology), Figure 2 Schematic
of any subset of these different systems in an exhumed depiction showing relationships between rock uplift (ur), surface
rock sample will provide information on whether that uplift (us), and erosion/exhumation (e). Shaded rectangles
sample was buried deeply enough to have experienced represent rock columns; open rectangle represents eroded
sufficiently high temperatures to reset a specific material; black dot indicates exhumed rock particle. Note that
thermochronological or thermobarometric methods record the
thermochronometric system (allowing a minimum amount of exhumation, not (either rock or surface) uplift.
amount of exhumation to be established) and when the
sample last cooled through the closure temperature while
being exhumed toward the surface (allowing an exhuma-
tion rate to be estimated). All of these methods provide
a measure of cooling (rate) in a thermal reference frame,
Uplift versus exhumation
however; converting this to an exhumation (rate) in a spa- Confusion remains widespread between the terms and
tial reference frame requires estimating (or guessing) concepts of uplift versus exhumation, even though the dif-
the geothermal gradient at the time of cooling through ference between these has been made clear more than
the closure temperature. Additional constraints on exhu- 20 years ago (England and Molnar, 1990). While exhuma-
mation rates and pathways can be obtained by combining tion refers to the vertical upward movement of rocks with
samples with a known spatial relationship to each other, respect to the Earth’s surface, uplift refers to vertical
for instance, by using their age-elevation relationship. upward movement of rocks (rock uplift) or the Earth’s sur-
Note that exhumation refers to the vertical upward face (surface uplift) with respect to an external reference
component of motion of a rock particle with respect to frame such as sea level (Figure 2). The above-cited tech-
the surface. Because both temperature and pressure niques are not able to determine either surface or rock
gradients are (close to) vertical in the Earth’s crust, meta- uplift, which require a measure of paleo-elevation with
morphic pressure-temperature estimates combined with respect to a distinct datum, such as sea level. The three
thermochronology yield estimates of exhumation. How- measures are not totally independent, however: by defini-
ever, in many orogens, particle paths have large horizontal tion, rock uplift is the sum of surface uplift and
components as well. Therefore, cumulative amounts of exhumation.
surface erosion may be very much larger than the total
amount of exhumation (Batt and Brandon, 2002). Summary
On much shorter (millennial) timescales and smaller Exhumation is the vertical upward movement of a rock
(metric) spatial scales, in situ produced terrestrial cosmo- particle toward the Earth’s surface. Although subtle differ-
genic nuclide dating also allows quantification of ences exist in meaning, most workers consider the terms
erosion/denudation rates. exhumation and denudation as synonyms; both are
264 EXTRATERRESTRIAL MATERIALS (K–Ar/Ar–Ar)
geometrical quantities representing a diminution of over-
burden above a rock particle. Different mechanisms can
lead to exhumation of rocks; these are classified as ero-
sional exhumation or tectonic exhumation. The most com-
mon method to quantify and date exhumation is by
thermochronology; if the exhumed rocks were previously
buried sufficiently deeply to be metamorphosed, petro-
logic pressure-temperature estimates can also be used to
quantify the amount of exhumation. Exhumation and
uplift are different quantities and no direct measure of
(either surface or rock) uplift currently exists.
Bibliography
Batt, G. E., and Brandon, M. T., 2002. Lateral thinking: 2-D inter-
pretation of thermochronology in convergent orogenic settings.
Tectonophysics, 349, 185–201.
England, P., and Molnar, P., 1990. Surface uplift, uplift of rocks, and
exhumation of rocks. Geology, 18, 1173–1177.
Ring, U., Brandon, M. T., Willett, S. D., and Lister, G. S., 1999.
Exhumation processes. In Ring, U., Brandon, M. T., Willett,
S. D., and Lister, G. S. (eds.), Exhumation processes: normal
faulting, ductile flow and erosion. London: Geological Society.
Special publications, Vol. 154, pp. 1–27.
Summerfield, M. A., and Brown, R. W., 1998. Geomorphic factors
in the interpretation of fission-track data. In Van den Haute, P.,
and De Corte, F. (eds.), Advances in fission-track geochronol-
ogy. Dordrecht: Kluwer, pp. 269–284.
Cross-references
Alpine Terranes (K–Ar/Ar–Ar)
Apatite
Ar–Ar and K–Ar Dating
Fault Zone (Thermochronology)
Fission Track Dating and Thermochronology
Metamorphic Terranes (K–Ar/40Ar/39Ar)
Minerals (40Ar–39Ar)
Luminescence, Earthquake and Tectonic Activity
Terrestrial Cosmogenic Nuclide Dating
Thermochronology, Detrital Zircon
Thermochronology, Landform Evolution
U–Th/He Dating
EXTRATERRESTRIAL MATERIALS (K–Ar/Ar–Ar)
Timothy Swindle
Lunar and Planetary Laboratory, University of Arizona,
Tucson, AZ, USA
Definitions
Potassium-argon (K–Ar) dating is a technique of finding
the ages of rocks based on the decay of 40K to 40Ar.
Argon-argon (40Ar–39Ar, or just Ar–Ar) dating is a type
of potassium-argon dating in which a sample is irradiated
with neutrons, converting a fraction of the atoms of 39K,
the most abundant isotope of potassium, into 39Ar,
enabling both the potassium abundance and the 40Ar
abundance to be measured with a single noble gas mass
spectrometer (McDougall and Harrison, 1999). The extra-
terrestrial materials to which the potassium-argon and
argon-argon techniques have been applied include only
meteorites and samples returned from the surface of the
Moon by either the American Apollo astronauts or the
Russian Luna robotic sample-return missions – age deter-
minations based on the decay of radioactive isotopes have
never been attempted in-situ on any planetary surface, but
several concepts based on the K–Ar technique have been
proposed since 2000.
Introduction
The potassium-argon system, which is at the heart of both
K–Ar and Ar–Ar dating, is notable among the major geo-
chronology systems with respect to easy diffusion of the
daughter product, 40Ar. As a result, it can be applied in
cases where a rock is heated, but not to a high enough tem-
perature for a long enough period of time for other systems
to be reset. It is this quality that makes it very useful for
impact age determinations.
The rarest of the three naturally-occurring potassium
isotopes, 40K, decays to either 40Ca or 40Ar, with a total
halflife of 1.25 Ga (Renne et al., 2011). Although the
majority of the decays are to Ca, it is the branch involving
Ar that is far more widely used as a geochronometer,
because Ar is far rarer than Ca.
The possibility of using the K–Ar system for age deter-
minations was discussed in the 1940s, and heavily
implemented as K–Ar dating in the 1950s. Although the
possibility of something like Ar–Ar dating was discussed
late in that decade, the first real Ar–Ar experiment was
performed on the Bruderheim meteorite in the mid-
1960s (Merrihue and Turner, 1966). It was quickly applied
to other meteorites, and when the Apollo lunar samples
were returned just 3 years later, the technique was already
mature enough that it was used heavily.
Because Ar does not engage in chemical bonds,
a silicate melt will tend to lose its Ar, so most rocks con-
tain little or no Ar when they form. However, Ar diffuses
only sluggishly through solids, so 40Ar will then build
up, and in an extremely simple system, a K–Ar age
records the length of time since the rock was formed.
However, Ar diffusion is temperature-dependent – if
a rock is heated up, Ar will begin to escape long before
the rock is melted. How much Ar is lost depends on the
diffusion characteristics of the host mineral as well as the
time and temperature history of the rock, so in a rock that
has been heated at some time after it cooled, some or all of
the radiogenic (decay-produced) 40Ar will be have been
lost, and the K–Ar age will be between the formation
age and the length of time since the reheating event. Or,
if the time-temperature history is sufficient, the rock can
lose all of its radiogenic 40Ar in a thermal event and the
K–Ar age will record the time since that event. Actually,
it will record the time since the rock was last at
a temperature high enough to lose Ar by diffusion, a tem-
perature that depends on the diffusion characteristics of
 EXTRATERRESTRIAL MATERIALS (K–Ar/Ar–Ar) 265

the mineral and on the cooling rate, but is typically a few An initial study of ordinary chondrites did not find such
hundred degrees Celsius. Diffusion characteristics have a relationship overall (Turner et al., 1978). A later study,
been determined for many minerals, and this type of calcu- focused on a single type of ordinary chondrites, the
lation is at the heart of “thermochronology” techniques H (high-iron) chondrites, and concentrated on meteorites
(McDougall and Harrison, 1999). with little evidence for the complicating effects of impact
In the most common version of Ar–Ar dating, the irra- heating, and found the predicted correlation between age
diated sample is heated to progressively higher tempera- and degree of metamorphism (Trieloff et al., 2003), both
tures, and the gas released at each step is analyzed. This for Ar–Ar ages and Pb–Pb ages, with Ar–Ar ages
sort of step-wise heating approach has several advantages extending over a ~0.1 Ga period. This suggests that the
over classical K–Ar dating. First, most of the release is H chondrite parent body, at least, may have had such an
through diffusion, so mineral sites from which diffusion onion-shell structure. However, experiments on the other
occurs more readily, because of the grain size, the diffu- two groups of ordinary chondrites have not shown such
sion characteristics of the mineral, or the proximity of a relationship (Turner et al., 1978; Dixon et al., 2004).
the site to the edge of the rock or to a fast-diffusion path-
way, are degassed more readily. This makes it possible to Determining the ages of lunar impact craters
distinguish Ar from different mineral sites. In addition,
For most of its history, the dominant geological process on
the diffusion characteristics can be determined in the
the surface of the Moon has been impact cratering. There
experiment itself. Finally, in a situation where the rock
are enough craters that it is possible to build up a detailed
has been only partially degassed, or where some mineral
relative chronology based on the density of impact craters
sites have been degassed, but not others, different steps
on a surface. Such relative chronologies can be constructed
will give different ages, so it is possible to determine that
for the surfaces of many planets and moons, but only for the
it is a complex system, and possibly even to sort out some
Moon do we have samples on which we can determine ages
of the complexity. The most detailed review of Ar–Ar
based on isotopic systems, so the Moon and its cratering
ages of meteorites is given by Bogard (2011).
rate serve as a standard for the determination on other
The heating caused by impact cratering often leads to
surfaces (see “Planetary Surfaces (Cratering Rate)”). The
a complex system. For one thing, the heating can be to
primary way by which ages of craters are determined is
a very high temperature but for a very brief period of time,
Ar–Ar dating, although other techniques can sometimes
often leading to only incomplete Ar degassing. Further-
be applied (Stöffler and Ryder 2001).
more, the time-temperature history can lead to some coun-
Within 4 years of the first samples having been returned
ter-intuitive results. For example, a glassy vein of impact
from the Moon, it was clear that there is a concentration in
melt material may retain more Ar than unmelted material
the Ar–Ar ages of Apollo samples at about 3.9 Ga (Turner
nearby (McConville et al., 1988), or a mineral in which
et al., 1973). Almost simultaneously, Tera et al. (1974)
diffusion is slower at low temperatures but faster at high
determined that many of the Apollo samples, from various
temperatures may suffer more Ar loss than one that
sites, appear to have uranium-thorium-lead ages that were
degasses more readily at laboratory temperatures (Cassata
disturbed at about the same time. This led Tera et al.
et al., 2011).
(1974) to suggest that there had been a “lunar cataclysm,”
a large number of cratering events on the Moon in
Ordinary chondrite metamorphism a relatively short interval of time. Subsequent research
has shown that there were events both earlier (Fernandes
Ordinary chondrites, the most common types of meteor-
et al., 2013) and later (Cohen et al., 2005) than the
ites, can be divided into three groups, presumably
~3.9 Ga spike in ages seen in the early Apollo data. How-
representing different parent asteroids, based on their
ever, evidence from asteroidal meteorites also suggests
chemical composition. However, within each group, the
that the impact cratering rate in the Solar System was
ordinary chondrites can be divided into metamorphic
higher between ~4.1 and 3.4 Ga than either earlier or later
types, reflecting the amount of high-temperature
(see below).
processing (such as chemical equilibration and mineral
In parallel with the chronology studies, investigations
recrystallization) they have experienced. A very simple
into early Solar System orbital dynamics revealed
model to explain the metamorphic types envisions all the
a plausible mechanism for a cataclysm, involving the
meteorites of a single group on a single parent asteroid that
migration of the orbits of Jupiter and Saturn (Gomes
heats up and then slowly cools by radiating its heat away
et al., 2005). As more details of the complexity of the
from the surface. In this case, the rocks near the surface
impact cratering history have been revealed, the orbital
would experience the lowest heating, and would cool past
dynamics models have become more complex, as well
any particular temperature first, while those in the center
(Morbidelli et al., 2012).
would be heated the most and take the longest to cool
down. In this case, the Ar–Ar ages, which reflect the time
at which the rock cooled past a particular temperature, Impact ages of asteroidal meteorites
should be younger for samples that are more highly Impact ages have been determined for more than 100 mete-
metamorphosed, reflecting this “onion-shell” structure. orites from asteroids (Bogard, 1995; Swindle et al., 2013),
266 EXTRATERRESTRIAL MATERIALS (K–Ar/Ar–Ar)
including all three groups of ordinary chondrites, the HED
meteorites that are presumed to come from the asteroid 4
Vesta, the mesosiderites (a kind of stony iron meteorite)
and the IIE iron meteorites. The HED meteorites predomi-
nantly give impact ages of 3.4–4.1 Ga, although the impact
crater distribution on Vesta observed by the Dawn space-
craft has been interpreted as demonstrating a massive
impact event at ~1 Ga (Schenk et al., 2012), an unresolved
puzzle. Most other meteorite groups are also roughly con-
sistent with the lunar record, with more impact ages of
3.4–4.1 Ga than either before or after, except for the most
recent 1 Ga. In the recent past, the best recorded impact is
a large event that affected the L (Low-iron) chondrite parent
body at 0.47 Ga, an event that is also recorded in fossil
meteorites found in terrestrial quarries of that age (Schmitz
et al., 2001). It is not clear whether the higher number of
recent ages reflects an increase in cratering rate or a short
average survival lifetime.
Ages of martian meteorites
Argon-argon dating has also been applied to meteorites
from Mars, although the most important result was not
an age. In an attempt to determine the age of Elephant
Moraine A79001, one of a group of igneous meteorites
with remarkably young crystallization ages based on other
chronometers, D. D. Bogard and P. Johnson found that the
apparent age of some samples was greater than the age of
the Solar System, unless the data was corrected for
a “trapped” component with a very large ratio of 40Ar to
36
Ar. Further study revealed that the samples with the high
40
Ar contents were also enriched in the other noble gases
(neon, krypton, and xenon), and had relative abundances
of the noble gases, as well as key ratios of isotopes of Ar
and Xe, that matched measurements of the Martian atmo-
sphere by the Viking spacecraft in the 1970s (Bogard and
Johnson, 1983). This was taken as confirmation of an ear-
lier suggestion that this group of meteorites indeed comes
from Mars. High amounts of argon, apparently trapped
either from the atmosphere or from magmas, are found
in many of the Martian meteorites, making it difficult to
determine accurate argon-argon ages (Bogard et al.,
2009). However, on those meteorites where ages can be
determined, it appears that the rocks are relatively young,
mostly 1.3 Ga or younger (Nyquist et al., 2001), and were
not subjected to high temperatures on Mars (Weiss et al.,
2002).
In-situ K–Ar dating
Several concepts for performing K–Ar dating in situ on
planetary surface, have been suggested (Swindle et al.,
2003; Talboys et al., 2009; Cohen et al., 2013; Farley
et al., 2013), but none has yet flown on a spacecraft. Mars
is a particularly appealing target, because the relative ages
of Martian surfaces can be determined by geological
superposition relationships or by the relative abundance
of impact craters, but the absolute ages are poorly known.
Ar–Ar dating cannot be applied because of the high
neutron fluences required, but metamorphism is not
expected to occur, so K–Ar dating might be expected to
date formation ages.
Summary and conclusions
Potassium-argon dating, specifically the Ar–Ar variant,
has been widely applied to extraterrestrial samples. The
relative ease of diffusion of Ar, compared to the elements
involved in other isotopic dating techniques, has made
Ar–Ar the most widely used system to determine the
timing of heating events, particularly those from impact
cratering.
Bibliography
Bogard, D. D., 1995. Impact ages of meteorites: a synthesis. Mete-
oritics, 30, 244–268.
Bogard, D. D., 2011. K-Ar ages of meteorites: clues to parent-body
thermal histories. Chemie der Erde, 71, 207–226.
Bogard, D. D., and Johnson, P., 1983. Martian gases in an Antarctic
meteorite? Science, 221, 651–654.
Bogard, D., Park, J., and Garrison, D., 2009. 39Ar-40Ar “ages” and
origin of excess 40Ar in Martian shergottites. Meteoritics &
Planetary Science, 44, 905–923.
Cassata, W. S., Renne, P. R., and Shuster, D. L., 2011. Argon diffu-
sion in pyroxenes: implications for thermochronometry and
mantle degassing. Earth & Planetary Science Letters, 304,
407–416.
Cohen, B. A., Swindle, T. D., and Kring, D. A., 2005. Geochemistry
and 40Ar-39Ar geochronology of impact-melt clasts in feld-
spathic lunar meteorites: implications for lunar bombardment
history. Meteoritics & Planetary Science, 40, 755–777.
Cohen B. A., Li Z.–H., Miller J. S., Brinckerhoff W. B., Clegg
S. M., Mahaffy P. R., Swindle T. D., and Wiens R. C., 2013.
Update on development of the potassium-argon laser experiment
(KArLE) instrument for in situ geochronology. Lunar & Plane-
tary Science, 44, Abstract #2363.
Dixon, E. T., Bogard, D. D., Garrison, D. H., and Rubin, A. E.,
2004. 39Ar-40Ar evidence for early impact events on the LL par-
ent body. Geochimica et Cosmochimica Acta, 68, 3779–3790.
Farley, K. A., Hurowitz, J. A., Asimow, P. D., Jacobson, N. S., and
Cartwright, J. A., 2013. A double-spike method for K-Ar mea-
surement: a technique for high precision in situ dating on Mars
and other planetary surfaces. Geochimica et Cosmochimica
Acta, 110, 1–12.
Fernandes, V. A., Fritz, J., Weiss, B. P., Garrick-Bethell, I., and
Shuster, D. L., 2013. The bombardment history of the Moon as
recorded by 40Ar-39Ar chronology. Meteoritics & Planetary
Science, 48, 1–29.
Gomes, R., Levison, H. F., Tsiganis, K., and Morbidelli, A., 2005.
Origin of the cataclysmic Late Heavy Bombardment period of
the terrestrial planets. Nature, 435, 466–469.
McConville, P. S., Kelley, S., and Turner, G., 1988. Laser probe
40
Ar-39Ar studies of the Peace River shocked L6 chondrite.
Geochimica et Cosmochimica Acta, 52, 2487–2499.
McDougall, I., and Harrison, T. M., 1999. Geochronology and
Thermochronology by the 40Ar/39Ar Method. New York: Oxford
University.
Merrihue, C., and Turner, G., 1966. Potassium-argon dating by acti-
vation with fast neutrons. Journal of Geophysical Research, 71,
2852–2857.
Morbidelli, A., Marchi, S., Bottke, W. F., and Kring, D. A., 2012.
A sawtooth-like timeline for the first billion years of lunar bom-
bardment. Earth & Planetary Science Letters, 355–356,
144–151.
 EXTRATERRESTRIAL MATERIALS (K–Ar/Ar–Ar)
Nyquist, L. E., Bogard, D. D., Shih, C.-Y., Greshake, A., Stöffler,
D., and Eugster, O., 2001. Ages and geologic histories of Mar-
tian meteorites. Space Science Reviews, 96, 105–164.
Renne, P. R., Balco, G., Ludwig, K. R., Mundil, R., and Min, K.,
2011. Response to the comment by W. H. Schwarz et al. on
“Joint determination of 40 K decay constants and 40Ar*/40 K for
the Fish Canyon sanidine standar, and improved accuracy for
40
Ar/39Ar geochronology” by P. R. Renne et al. (2010).
Geochimica et Cosmochimica Acta, 75, 5097–5100.
Schenk, P., O’Brien, D. P., Marchi, S., Gaskell, R., Preusker, F.,
Roatsch, T., Jaumann, R., Buczkowski, D., McCord, T.,
McSween, H. Y., Williams, D., Yingst, A., Raymond, C., and
Russell, C., 2012. The geologically recent giant impact basins
at Vesta’s South Pole. Science, 336, 694–697.
Schmitz, B., Tassinari, M., and Peucker-Ehrenbrink, B., 2001.
A rain of ordinary chondritic meteorites in the early Ordovician.
Earth & Planetary Science Letters, 194, 1–15.
Stöffler, D., and Ryder, G., 2001. Stratigraphy and isotope ages of
lunar geologic units: chronological standard for the inner solar
system. Space Science Reviews, 96, 9–54.
Swindle T. D., Bode R., Boynton W. V., Kring D. A., Williams M.,
Chutjian A., Darrach M. R., Cremers D. A., Wiens R. C., and
Baldwin S. L., 2003. AGE (Argon Geochronology Experiment):
an instrument for in situ geochronology on the surface of Mars.
Lunar & Planetary Science, 34, Abstract #1488.
Swindle T. D., Kring D. A., and Weirich J. R. 2013. 40Ar-39Ar ages
of impacts involving ordinary chondrite meteorites. In Jourdan,
F., Mark, D., and Verati, C. (eds.), 40Ar/39Ar Dating: From Geo-
chronology to Thermochronology, From Archaeology to Plane-
tary Sciences. London: Geological Society (in press).
267
Talboys, D. L., Barber, S., Bridges, J. C., Kelley, S. P., Pullan, D.,
Verchovsky, A. B., Butcher, G., Fazel, A., Fraser, G. W.,
Pillinger, C. T., Sims, M. R., and Wright, I. P., 2009. In situ radio-
metric dating on Mars: investigation of the feasibility of K-Ar
dating using flight-type mass and X-ray spectrometers. Plane-
tary and Space Science, 57, 1237–1245.
Tera, F., Papanastassiou, D. A., and Wasserburg, G. J., 1974. Isoto-
pic evidence for a terminal lunar cataclysm. Earth & Planetary
Science Letters, 22, 1–21.
Trieloff, M., Jessberger, E. K., Herrwerth, I., Hopp, J., Fiéni, C.,
Ghelis, M., Bourot-Denise, M., and Pellas, P., 2003. Structure
and thermal history of the H-chondrite parent asteroid revealed
by thermochronometry. Nature, 422, 502–506.
Turner G., Cadogan P. H., and Yonge C. J., 1973. Argon
selenochronology. In Proceedings of the Lunar Science Confer-
ence, 4th, pp. 1889–1914.
Turner G., Enright M. C., and Cadogan P. H., 1978. The early history
of chondrite parent bodies inferred from 40Ar-39Ar ages. In Pro-
ceedings of the Lunar Science Conference, 9th, pp. 989–1025.
Weiss, B. P., Shuster, D. L., and Stewart, S. T., 2002. Temperatures
on Mars from 40Ar/39Ar thermochronology of ALH84001. Earth
& Planetary Science Letters, 201, 465–472.
Cross-references
Meteorites (U–Pb)
Noble Gas Mass Spectrometer
Planetary Surfaces (Cratering Rate)
Thermochronology, Meteorites
Uranium–Lead, Extraterrestrial, Planetary Materials
 F
FAULT ZONE (THERMOCHRONOLOGY)
Ben van der Pluijm and Chris Hall
Department of Earth and Environmental Sciences,
University of Michigan, Ann Arbor, MI, USA
Synonyms
40
Ar-39Ar dating; Ar-Ar dating; Brittle fault dating; Clay
gouge; Illite
Introduction and definition of brittle fault dating
Dating of deformation revolutionized our understanding
of crustal evolution by inserting the element of time and
rate into strain and regional geology. Middle and deep
crustal faults (>10–15 km depth) deform in a temperature
regime that promotes growth of various mineral phases
containing radiogenic elements, which allow direct and
indirect dating of deep fault zones (e.g., van der Pluijm
et al., 1994). On the other hand, shallow crustal faulting
occurs in the brittle regime, where growth of suitable
phases for dating is much more limited. The exception is
the K-bearing mineral illite, which is widespread and can
be used for dating with K-Ar and Ar-Ar methods in
a range of settings, from diagenesis to deformation
(Kelley, 2002). K-Ar dating has been successful in many
areas (e.g., Clauer et al., 1997; Zwingmann et al., 2010;
Clauer et al., 2012), while others focus on Ar-Ar dating
(e.g., Haines and van der Pluijm, 2008; Yun et al., 2010).
This entry describes the use of Ar-Ar chronology to date
brittle faults, including the mineralogic characterization
of samples, Ar-Ar systematics and clay dating, and an
example application. Application of illite age analysis
(IAA) to fault dating has the following benefits: very small
sample volumes can be analyzed, potassium and argon are
measured simultaneously, and it has high precision.
Regardless of the dating approach, separation of grain-size
J.W. Rink, J.W. Thompson (eds.), Encyclopedia of Scientific Dating Methods, DOI 10.1007/978-94-007-6304-3,
© Springer Science+Business Media Dordrecht 2015
fractions and characterization and quantification of illite
are key to reliable ages, requiring the integration of quan-
titative X-ray analysis.
Illite characterization and quantification
Illite-bearing samples of fault rock, herein called clay
gouge (Vrolijk and van der Pluijm, 1999), are separated
into multiple size fractions. Crushed rock is placed in an
ultrasonic bath for deflocculation and separated into size
fractions using centrifuge settling following Stoke’s law.
A sample of fault rock is separated into three or four size
fractions, in the range of <2.0 to <0.02 mm; samples
containing feldspar may require <0.5 mm as the upper size
limit to avoid contamination. To characterize bulk miner-
alogy, oriented clay slurry mounts are analyzed with
X-ray diffractometer (XRD) by scanning from 2
to 50

2Y (Cu Ka) of both air-dried and glycolated samples
(Moore and Reynolds, 1997). If carbonate minerals
obscure peaks required for quantification of illite
polytypes, size fractions of gouges are treated with
a weak (~1 N) acetic acid solution after separating material
for use in 40Ar-39Ar dating. A separate aliquot is used to
limit any effects of acid treatment on Ar and K retention.
To determine the crystallite size and diagenetic grade of
gouge, oriented slurry samples of each size fraction are
scanned from 2
to 20
2Y using an X-ray diffractometer.
If appropriate, the percent illite in illite-smectite (%I in
I-S) is determined using the method of Środoń (1980)
and diffraction patterns are modeled using the program
NEWMOD© (Reynolds and Reynolds, 1996). Illite crys-
tallinity (IC), which reflects the thickness distribution of
illite crystallites, is determined by measuring the full
width at half maximum (FWHM) of the 001 illite peak
in
2Y (Kübler, 1968), calibrated by Warr and Rice’s
(1994) method and standards. Crystallite thickness is cal-
culated using the Scherrer equation (Drits et al., 1997).
270 FAULT ZONE (THERMOCHRONOLOGY)

Fault Zone (Thermochronology), Figure 1 XRD patterns (black curve) for six mixtures of Owl Creek muscovite (2M1) and IMt-1 illite
(1Md) and the best-matched models (gray lines). Note that for 2M1-rich mixtures, the best matches are obtained from 2M1-specific
peaks, while for 1Md-rich mixtures, the best matches fit the entire pattern between 21
and 36
2y (From Haines and van der Pluijm,
2008).

Illite age analysis requires quantifying the percentages
of the two main polytypes of illite: the higher temperature
(detrital) 2M1 polytype and the lower temperature
(authigenic) 1M/1Md polytype. Standard-oriented X-ray
mounts tend to obscure peaks that have non-(00l) indices,
so randomized powder mounts are made using an
end-packer device (after Moore and Reynolds, 1997).
Illite polytypes are quantified by matching natural XRD
patterns with synthetic XRD patterns created using
WILDFIRE© software (Reynolds, 1993). The program
allows different properties for 2M1 and 1M/1Md illite
and varying parameters such as randomness, crystallite
size, hydration state, percentage cis- and trans-vacant
interlayers, and smectite interlayers. We use
a spreadsheet program to match measured XRD patterns
to the synthetic XRD patterns, as illustrated in Figure 1.
Typical precision for illite polytype quantification is
2–3 %.
 FAULT ZONE (THERMOCHRONOLOGY)
40
Ar-39Ar chronology of clays
The K-Ar dating technique is based upon the assumption
that radiogenic 40Ar (i.e., 40Ar*) produced by the beta-
decay of 40K is quantitatively retained within a mineral
structure. Given this assumption, the 40Ar*-40K ratio is
effectively a measure of the time since a mineral became
closed to Ar loss. The 40Ar-39Ar technique extends the
K-Ar method by using the fast neutron reaction
39
K(n,p)39Ar to produce 39Ar, which is a proxy for 40K,
and therefore, the 40Ar*-39Ar ratio is used as a measure
of the age of mineral. However, unlike the decay reaction
that produces 40Ar*, the reaction that forms 39Ar is highly
energetic and it causes the resulting nucleus to recoil an
average distance of 162 nm within silicate minerals
(Turner and Cadogan, 1974; Onstott et al., 1995).
For most minerals that are dated using the 40Ar-39Ar
method, this recoil distance is considered to be insignifi-
cant relative to the average grain size, but in the case of
fine-grained clay particles, the average thickness of
a grain might be much less than the average recoil dis-
tance. This means that 39Ar recoil effects can dominate
argon systematics for clay-sized particles, and an
approach for collecting any 39Ar lost due to recoil is
needed in order to use the 40Ar-39Ar technique to date
clays.
Seidemann (1976) first described a vacuum encapsula-
tion technique, where a sample is placed within an evacu-
ated fused silica capsule. After neutron irradiation, the
capsule was cracked open within a vacuum system and
the released argon isotopes were analyzed using an
attached noble gas mass spectrometer. Subsequently, the
technique has been used in several laboratories (e.g., Hess
and Lippolt, 1986; Foland et al., 1992; Smith et al., 1993),
but the most extensive use of the method has been at the
University of Michigan.
Smith et al. (1993) noted that the quantity of 39Ar
released during neutron irradiation, called the recoil gas
fraction, was less than anticipated purely from geometrical
considerations due to the sample grain size. They attrib-
uted the difference to efficient reimplantation of 39Ar
between neighboring grains and postulated that the domi-
nant control on the retention of 39Ar during neutron irradi-
ation was the depth of penetration of the recoiled nucleus.
In effect, 39Ar retention depends on where the recoiled
nucleus stops and does not depend strongly upon grain
size or the recoiled distance. This point was further
supported by the observation that the retained fraction of
37
Ar was virtually identical to that for 39Ar despite 37Ar
having over twice the expected recoil distance compared
to 39Ar (Onstott et al., 1995). Dong et al. (1995) confirmed
these observations and observed that the fraction of 39Ar
released due to recoil from illite was correlated with the
crystallinity index as derived from XRD analysis, with
well-crystallized illite having extremely low loss of 39Ar,
regardless of grain size. This suggested that the amount
of Ar loss due to recoil depends mostly on the nm-scale
structural state of illite.
271
Total gas ages and retention ages of illite
Dong et al. (1995) postulated that the retention of 39Ar is
controlled by the density of dislocations, defects, and
exposed mica surfaces within illite crystal packets. It has
long been suggested that Ar can only diffuse along the
interlayer regions between the structurally strong 2:1 unit
sheets that comprise micaceous minerals (e.g., Hames and
Bowring, 1994; Kramar et al., 2003). Therefore, it is likely
that an Ar atom that arrives within a bounded interlayer will
be retained. Given an average illite packet thickness of n 2:1
structural units, there are a total of 2n surfaces on these
n sheets. Two adjacent 2:1 structural units comprise an
interlayer region, and there should be 2n – 2 internal surfaces
and 2 exposed surfaces. Dong et al. (1995) suggested that in
this case, the fraction of non-retentive sites is 2/2n ¼ 1/n.
The ages calculated from both the 39Ar ejected due to
recoil and the Ar retained within the sample in the Dong
et al. (1995) study (called total gas ages) were often too
young when compared with other geological constraints.
Specifically, well-crystallized epizonal-grade illite was
found to give much older ages than lower-grade diagenetic
illite within the same sedimentary basin, indicating that
the lower-grade, poorly crystallized illite had already lost
a significant fraction of its radiogenic 40Ar. Dong
et al. (1995, 1997) found that ages calculated without the
recoil gas fraction, meaning ages calculated only from
Ar that was retained within the sample, gave geologically
reasonable values; these were called retention ages. This
was further confirmed for hydrothermally produced clay
minerals by Hall et al. (1997, 2000). These retention ages
assume that there is K within both retained interlayer sites
as well as in exposed sites in defects, dislocations, and
exposed surfaces. K residing in these non-retentive sites
will contribute to the total 39Ar inventory; however, any
40
Ar produced at these sites will not have been retained
over geological time. In effect, the Ar retention ages cor-
rect for the loss of radiogenic 40Ar by ignoring any Ar
from non-retentive sites.
Subsequently, Dong et al. (2000) found that total gas
ages gave stratigraphically correct ages for smectite-rich
illite-smectite (I/S), and van der Pluijm et al. (2001,
2006) found that the total gas age model yielded reliable
ages for authigenic I/S formed during shallow fault
motion. This indicates that in weakly crystallized I/S,
defect sites are almost entirely composed of K-poor smec-
titic layers. Thus, total gas ages and retention ages repre-
sent end-member models that are linked to the crystal
structure of illite, with total gas ages assuming there is
no K in non-retentive sites and with retention ages assum-
ing that non-retentive sites have K concentrations equal to
that in internal, retentive interlayer sites.
Illite age analysis: example
The Sierra Mazatán fault is part of a Cordilleran metamor-
phic core complex in Sonora, Mexico, that separates
mid-crustal Paleocene granites and minor gneisses in
272 FAULT ZONE (THERMOCHRONOLOGY)

Fault Zone (Thermochronology), Figure 2 Ar degassing spectra for four size fractions of the Mazatán fault, showing total gas ages
and retention ages for each, as well as Ca/K and Cl/K ratios (Modified from Haines and van der Pluijm, 2008).

the footwall from Cretaceous metavolcanics and
unmetamorphosed Tertiary sedimentary rocks in the hang-
ing wall (Wong and Gans, 2003). XRD pattern matching
of size fractions from the fault shows that gouge composi-
tions range from 84 % to 18 % 2M1 illite and that compo-
sitions are positively correlated with grain-size fraction
(Haines and van der Pluijm, 2008). The corresponding
Ar degassing spectra and total gas and retention ages for
four samples of coarse to very fine material are shown in
Figure 2.
The retention ages vary nonsystematically, from
19.6 to 20.2 Ma, while the total gas ages are younger
 FAULT ZONE (THERMOCHRONOLOGY) 273

Fault Zone (Thermochronology), Figure 3 Illite age analysis plot for gouge from the Sierra Mazatán detachment. Numbers in bold
are the extrapolated ages for the authigenic (14.8 Ma) and host illite population (18.7 Ma) mixed in the gouge. Percentage host rock
illite (2M1) is plotted against the function (elt, 1) because it is proportional to 40Ar*-40K, which in turn is a linear function of the
proportion of authigenic and detrital grains. l is the total decay constant for 40K and t is the apparent age (Reprocessed data from
Haines and van der Pluijm, 2008).

and vary systematically with decreasing grain size
(18.0–15.5 Ma). Illite crystallinity measurements, record-
ing a small average illite crystallite size, indicate that total
gas ages should be used for the dating of these samples,
which result in systematically younger ages as the percent-
age of detrital 2M1 illite decreases and authigenic 1M/1Md
illite increases in finer-grained samples. Plotting age ver-
sus % 2M1 illite for each size fraction and using York
regression (Mahon, 1996) produces a lower intercept,
representing 100 % authigenic illite, of 14.8  0.2 Ma
and an upper intercept, representing 100 % mica in the
granitic footwall, of 18.7  0.7 Ma (Figure 3). The extrap-
olated authigenic illite age supports well-known field con-
straints on faulting and the upper intercept age matches
feldspar cooling ages from these samples (Wong and
Gans, 2003). The use of four size fractions significantly
improves the statistics of the age determination over using
the minimum of three size fractions, with an age error that
is mostly defined by the error in clay quantification and
not Ar ages of each size fraction.
Summary and progress
Illite age analysis combines quantitative X-ray analysis
of clay separates with encapsulated sample 40Ar-39Ar
dating. Both illite in illite-smectite (I in I/S) and illite
polytype methods offer robust, internally consistent
approaches to dating of upper crustal deformation struc-
tures that involve the (trans)formation of illitic phases,
which is common in natural rocks. It has been reliably
applied to shallow faults in multiple regional settings that
include reverse, normal, and strike-slip systems and was
recently extended to include the dating of fold systems
(Fitz-Diaz and van der Pluijm, 2013).
Bibliography
Clauer, N., Srodon, J., Francu, J., and Sucha, V., 1997. K-Ar dating
of illite fundamental particles separated from illite-smectite.
Clay Minerals, 32, 181–196.
Clauer, N., Zwingmann, H., Liewig, N., and Wendling, R., 2012.
Comparative 40Ar/39Ar and K–Ar dating of illite-type clay min-
erals: a tentative explanation for age identities and differences.
Earth-Science Reviews, 115, 76–96.
Dong, H., Hall, C. M., Peacor, D. R., and Halliday, A. N., 1995.
Mechanisms of argon retention in clays revealed by laser
40
Ar-39Ar dating. Science, 267, 355–359.
Dong, H., Hall, C. M., Halliday, A. N., and Peacor, D. R., 1997.
40
Ar-39Ar dating of Late-Caledonian (Acadian) metamorphism
and cooling of K-bentonites and slates from the Welsh Basin,
U.K. Earth and Planetary Science Letters, 150, 337–351.
Dong, H., Hall, C. M., Peacor, D. R., Halliday, A. N., and Pevear,
D. R., 2000. Thermal 40Ar/39Ar separation of diagenetic from
detrital illitic clays in Gulf Coast shales. Earth Planetary Science
Letters, 175, 309–325.
Drits, V. A., Środoń, J., and Eberl, D. D., 1997. XRD measurement
of mean crystallite thickness of illite and illite/smectite;
reappraisal of the Kubler index and the Scherrer equation. Clays
and Clay Minerals, 45, 461–475.
Fitz-Diaz, E., and van der Pluijm, B., 2013. Fold dating:
a new approach to constrain the timing of deformation in shallow
crustal rocks. Journal of Structural Geology, 54, 174–179.
Foland, K. A., Hubacher, F. A., and Arehart, G. B., 1992. Ar-40/Ar-39
dating of very fine-grained samples – an encapsulated-vial
procedure to overcome the problem of Ar-39 recoil loss. Chemical
Geology, 102, 269–276.
Haines, S. H., and van der Pluijm, B. A., 2008. Clay quantification
and Ar-Ar dating of synthetic and natural gouge – application to
274 FAULTS (U-SERIES)
the Miocene Sierra Mazatán detachment fault, Sonora, Mexico.
Journal of Structural Geology, 30, 525–538.
Hall, C. M., 2013. Direct measurement of recoil effects on
40
Ar/39Ar standards. In Jourdan, F., Mark, D. F., and Verati, C.
(eds.), Advances in 40Ar/39Ar Dating: From Archaeology to
Planetary Sciences. London: Geological Society. Geological
Society, Special Publications, Vol. 378.
Hall, C. M., Higueras, P., Kesler, S. E., Lunar, R., Dong, H., and
Halliday, A. N., 1997. Dating of alteration episodes related to
mercury mineralization in the Almadén district, Spain. Earth
and Planetary Science Letters, 148, 287–298.
Hall, C. M., Kesler, S. E., Simon, G., and Fortuna, J., 2000.
Overlapping cretaceous and Eocene alteration, twin creeks
Carlin-type deposit, Nevada. Economic Geology, 95,
1739–1752.
Hames, W. E., and Bowring, S. A., 1994. An empirical evaluation of
the argon diffusion geometry in muscovite. Earth and Planetary
Science Letters, 124, 161–167.
Hess, J. C., and Lippolt, H. J., 1986. Kinetics of Ar isotopes during
neutron irradiation; 39Ar loss from minerals as a source of error
in 40Ar/39Ar dating. Chemical Geology: Isotope Geoscience Sec-
tion, 59, 223–236.
Kelley, S., 2002. K-Ar and Ar-Ar dating. Reviews of Mineralogy
and Geochemistry, 47, 785–818.
Kramar, N., Cosca, M. A., Buffat, P.-A., and Baumgartner, L. P.,
2003. Stacking fault-enhanced argon diffusion in naturally
deformed muscovite. In Vance, D., Müller, W., and Villa, I. M.
(eds.), Geochronology: Linking the Isotopic Record with Petrol-
ogy and Textures. London: Geological Society. Geological Soci-
ety, Special Publications, 220, 249–260.
Kübler, B., 1968. Évaluation quantitative du métamorphisme para
la cristallinité de l'illite; état des progrès realisés ces dernières
années. Bulletin du Centre de Recherches de Pau, 2, 385–397.
Mahon, K., 1996. The new “York” regression; application of an
improved statistical method to geochemistry. International
Geology Review, 38, 293–303.
Moore, D. M., and Reynolds, R. C., Jr., 1997. X-ray Diffraction and
the Identification and Analysis of Clay Minerals. Oxford, UK:
Oxford University Press, p. 378 p.
Onstott, T. C., Miller, M. L., Ewing, R. C., Arnold, G. W., and
Walsh, D. S., 1995. Recoil refinements: implications for the
40
Ar/39Ar dating technique. Geochimica et Cosmochimica Acta,
59, 1821–1834.
Reynolds Jr., R. 1993. WILDFIRE- A Computer Program for the
Calculation of Three-Dimensional Powder X-ray Diffraction
Patterns for Mica Polytypes and Their Disordered Variations.
Hanover, NH.
Reynolds, Jr., R., and Reynolds, III., R. 1996. NEWMOD-for-
Windows. The Calculation of One-Dimensional X-ray Diffrac-
tion Patterns for Mixed-Layered Clay Minerals. Hanover, NH.
Seidemann, D. E. 1976. K-Ar and 40Ar/39Ar dating of deep-sea
rocks. PhD thesis, Yale University, 196p.
Smith, P. E., Evensen, N. M., and York, D., 1993. First successful
40
Ar-39Ar dating of glauconies: argon recoil in single grains of
cryptocrystalline material. Geology, 21, 41–44.
Środoń, J., 1980. Precise identification of illite/smectite interstratifi-
cations by X-ray powder diffraction. Clays and Clay Minerals,
28, 401–411.
Turner, G., and Cadogan, P. H. 1974. Possible effects of 39Ar recoil
in 40Ar-39Ar dating. In Proceedings of the Fifth Lunar
Science Conference. Chemical and Isotope Analyses, Organic
Chemistry, Vol. 2, Houston, TX, March 18–22, 1974,
pp. 1601–1615.
van der Pluijm, B. A., Mezger, K., Cosca, M. A., and Essene, E. J.,
1994. Determining the significance of high-grade shear zones by
using temperature-time paths, with examples from the Grenville
Orogen. Geology, 22, 743–746.
van der Pluijm, B. A., Hall, C. M., Vrolijk, P. J., Pevear, D. R., and
Covey, M. C., 2001. The dating of shallow faults in the Earth’s
crust. Nature, 412, 172–175.
van der Pluijm, B. A., Vrolijk, P. J., Pevear, D. R., Hall, C. M., and
Solum, J., 2006. Fault dating in the Canadian rocky mountains:
evidence for late cretaceous and early Eocene orogenic pulses.
Geology, 34, 837–840.
Vrolijk, P., and van der Pluijm, B. A., 1999. Clay gouge. Journal of
Structural Geology, 21, 1039–1048.
Warr, L. N., and Rice, A. H. N., 1994. Interlaboratory standardiza-
tion and calibration of clay mineral crystallinity and crystallite-
size data. Journal of Metamorphic Geology, 12, 141–152.
Wong, M., and Gans, P., 2003. Tectonic implications of early Mio-
cene extensional unroofing of the Sierra Mazatán metamorphic
core complex, Sonora, Mexico. Geology, 31, 953–956.
Yun Jianbing, Shi Hesheng, Zhu Junzhang, Zhao Linghao and Qiu
Huaning, 2010. Dating petroleum emplacement by illite
40
Ar/39 Ar laser stepwise heating AAPG Bulletin, 94, 759–771.
Zwingmann, H., Mancktelow, N., Antognini, M., and Lucchini, R.,
2010. Dating of shallow faults: new constraints from the
AlpTransit tunnel site (Switzerland). Geology, 38, 487–490.
Cross-references
Ar–Ar and K–Ar Dating
Clays and Glauconites (K–Ar/Ar–Ar)
Mass Spectrometry
Metamorphic Terranes (K–Ar/40Ar/39Ar)
Minerals (40Ar–39Ar)
Noble Gas Mass Spectrometer
Potassium–Argon (Argon–Argon), Structural Fabrics
Uranium–Lead, Igneous Rocks
FAULTS (U-SERIES)
Perach Nuriel
Department of Geological and Environmental Sciences,
Stanford University, Stanford, CA, USA
Synonyms
Dating coseismic calcite; Dating fault-related precipitate;
Paleoseismicity; U–Th geochronology of syn-tectonic calcite
Definition
The use of U-series geochronology applied to fault-related
calcite for dating brittle fault activity.
Introduction
Absolute time constraints on fault activity are an essential
part of many plate tectonic and paleoseismic studies, and
their potential interpretation in terms of fault seismicity
and seismic hazards is well recognized (e.g., Hee-Kwon
and Schwarcz, 1995; Grant et al., 1999). The introduction
of paleoseismological exploratory trenching studies
offered the opportunity to determine the number and
timing of large paleo-earthquakes and their estimated
magnitude and recurrence intervals (e.g., Clark
et al., 1972; Sieh, 1984). Other studies provided important
paleoseismic records recovered in laminated sediments
 FAULTS (U-SERIES) 275

Faults (U-Series), Figure 1 Fault-related calcite precipitates. (a) Tension gash filled with calcite from the Dead Sea fault zone (DSFZ),
Southern Israel (Nuriel et al., 2011); (b) calcite-cemented breccia; (c) calcite-filled dilation vein from the DSFZ, Northern Israel (Nuriel
et al., 2012a); (d) calcite coating striation marks from the DSFZ, Northern Israel (Nuriel et al., 2012b); and (e) calcite injection vein from
the Pamukkale geothermal fields (Uysal et al., 2009). All fault-related calcite are secondary phases within limestone host rock.

(Marco et al., 1996; Ken-Tor et al., 2001) and damaged
speleothems (Kagan et al., 2005; Plan et al., 2010). How-
ever, most paleoseismological studies use indirect
methods for dating fault movement based on the geochro-
nology of material collected from the displaced or dam-
aged stratigraphic section and not from the fault zone
itself. In addition, most paleoseismological dating studies
are based on 14C and optically stimulated luminescence
(OSL) dating methods, which are both limited to relatively
young events [50 ka for 14C (Ramsey et al., 2012) and
100 ka for OSL (Huntley et al., 1985)]. U-series dating
of fault-related calcite precipitates is therefore an impor-
tant source of information of both processes occurring in
the fault zone during seismic events and of paleoseismic
events occurring further back in the geological history.
The use of U-series method for dating fault activity
implies temporal association among faulting, fluid circu-
lation, and precipitation processes in the fault zone. For
example, calcite dissolution and reprecipitation are com-
mon processes in fault zones within host-rock carbonates
due to enhanced fluid infiltration processes in the frac-
tured damage zone. In this setting, the formation of sec-
ondary calcite may occur during (syn-tectonic), or
following (post-tectonic), a faulting event. While
U-series ages of syn-tectonic calcite provide constraint
on the timing of seismically active period, ages of post-
tectonic calcite can be used to constrain the timing of
inter-seismic period and/or provide a minimal age for the
timing of the deformation event. Identifying syn-tectonic
from post-tectonic precipitates therefore represents a
276 FAULTS (U-SERIES)

Faults (U-Series), Figure 2 Microstructures of fault-related calcite. (a) Polished slab of a hand specimen taken perpendicular to fault
plane and grooves showing the occurrences of syn-tectonic calcite (within striation and breccia) and post-tectonic calcite (coating
layer), (b) close-up of calcite striation (location is shown in a) showing characteristic morphology of syn-tectonic calcite (calcite twins
with euhedral crystals and radial growth texture) and post-tectonic calcite (crack-seal texture). Sample is from the DSFZ, Northern
Israel (Nuriel et al., 2012b).

fundamental step toward accurate interpretation of the
U-series age results. Making these interpretations requires
a combination of structural, isotopic-geochemical, and
geochronological approaches.
Fault-related calcite precipitates in brittle fault
zones
Fault-related structures are formed during brittle deforma-
tion events and are commonly used as kinematic indica-
tors for strain orientation (Boullier et al., 2004; De Paola
et al., 2008) and sense of displacement (Petit, 1987). Ten-
sion gashes, fault breccias, dilation veins, striation marks,
and injection veins are examples for fault-related struc-
tures that are potential sites for calcite precipitation
(Figure 1). Calcite precipitated in these fault-related fea-
tures is a good candidate for the implementation of
U-series geochronology for dating fault activity; however,
further microstructural and isotopic-geochemical charac-
terization is needed in order to determine the temporal
association between calcite precipitation and faulting
(e.g., syn- or post-tectonic).
Microstructures of fault-related calcite
precipitates
Microstructural studies of fault-related precipitates pro-
vide important information on the progressive develop-
ment of a structure, on temporal changes in precipitation
conditions and on growth mechanisms. In particular, the
size of displacement along fault asperities will result in
the occurrence of either syn-tectonic or post-tectonic pre-
cipitates (Gratier and Gamond, 1990). Large displacement
is associated with a cataclastic-seismic event and the
formation of large cavities (several mm in size). These
open cavities can be filled with syn-tectonic calcite, char-
acterized with morphologies including twining, euhedral
crystals, and radial growth texture (Figure 2a, b). In con-
trast, small displacement is associated with aseismic slip,
involving dissolution and sealing of successive
microcracks (mm in size) by crack-seal mechanism
(Ramsay, 1980). Calcite that precipitated by crack-seal
mechanism should be considered as post-tectonic
(Gratier and Gamond, 1990) and is characterized with
fibrous crystals (Figure 2b) and surface-parallel arrays of
detached overgrowths containing fluid inclusions (Cox
and Etheridge, 1983; Uysal et al., 2011).
Geochemistry of fault-related calcite precipitates
The geochemistry of fault-related calcite precipitates can
provide important information on the fluids that are respon-
sible for their formation (e.g., Pili et al., 2002). The type and
source of fluids and their transport mechanism and resi-
dence time in the system can also help determine the time
relation between faulting, precipitation, and fluid circula-
tion. Common analytes used to determine fluid source and
transportation mechanisms include stable isotopes (d18O
vs. d13C; Figure 3a), strontium isotopes (87Sr/86Sr;
Figure 3b), and rare earth element (REE) concentrations
(Figure 3c). Based on these and other parameters, various
fluid sources and transportation mechanisms have been
proposed for the formation of fault-related calcite precipi-
tates in different tectonic settings, including: (1) earthquake-
mobilized fluids (Uysal et al., 2007; 2011), (2) pressurized
solutions (Gratier et al., 2002; Nuriel et al., 2012b), (3) mete-
oric water interacting with host-rock carbonates to various
 FAULTS (U-SERIES) 277

Faults (U-Series), Figure 3 Geochemistry of fault-related calcite precipitates and host-rock samples from the Dead Sea fault zone
(DSFZ), Northern Israel (Modified after Nuriel et al., 2012b). (a) d13C versus d18O with values given in per mil (%) relative to Vienna Pee
Dee Belemnite (V-PDB) carbonate standard and Vienna Standard Mean Ocean Water (V-SMOW) standard, (b) 87Sr/86Sr versus 1/Sr
concentration (ppm) plot, and (c) chondrite-normalized REE patterns of the different types of calcite precipitates relative to their host
rock. Calcite types are marked with different symbols (Chondrite values from Haskin et al. (1971)).

degrees (Verhaert et al., 2003, 2004; Labaume et al., 2004;
Nuriel et al., 2011; 2012a), and (4) a mixing of several
sources (Boles and Grivetti, 2000; Verhaert et al., 2009).
In addition, the geochemical-isotopic signature of
syn-tectonic and post-tectonic calcite precipitates from the
same site (or same sample) can be very different and can
reflect differences in the formation mechanisms (see exam-
ple in Figure 3).
U-series dating of fault-related calcite precipitates
The U-series dating technique is based on the fact that the
238
U decay chain has two intermediate radiogenic and
radioactive daughters (234U and 230Th). Both have geolog-
ically short half-lives (~105 years), meaning that the
method is appropriate only for dating processes that
occurred in the last ~500,000 years (Edwards et al., 1987).
Basic assumptions that apply to the U-series dating
method include two main points (e.g., Ivanovich and
Harmon, 1992; van Calsteren and Thomas, 2006):
(1) daughter nuclides (i.e., 230Th) were not present in the
sample at the time of precipitation, and (2) the sample
must be a closed system to post-formation migration or
addition of any of the radionuclides (i.e., 234U, 230Th).
Thus, changes in U/Th isotopic abundance in the system
occur only by radioactive decay and growth processes.
The first assumption can be validated by the presence of
232
Th (Kaufman and Broecker, 1965; Ku and Liang,
1983), the more common of the two Th isotopes that
behaves like a stable isotope due to its long half-life.
According to the Kaufman-Broecker concept, if measured
230
Th/232Th activity ratios are low (<20), it is likely that
non-authigenic 230Th is also present (i.e., initial 230Th pre-
sent at the time the calcite was formed rather than forming
subsequently by in situ decay of 234U or 238U), and a cor-
rection procedure is needed. However, if 230Th/232Th
activity ratios are high (>20), the correction for initial
230
Th is negligible. The second assumption is, however,
much harder to validate, particularly in syn-tectonic cal-
cite. Unlike dating of corals and speleothems, which com-
monly exhibit measured activity ratios (234U/238U and
230
Th/238U) predicted by U-series isotope evolution in
material formed by progressive deposition (continuous
growth, see Ku, 1965), syn-tectonic precipitates form
278 FAULTS (U-SERIES)

Faults (U-Series), Figure 4 U-series ages of fault-related calcite from the Dead Sea fault zone, Northern Israel (Modified after Nuriel
et al., 2012b). U-series ages, in ka, are shown for syn-tectonic (squares) and post-tectonic (triangles) calcite from different samples
(sample names are denoted at the left-side axes). Solid vertical lines indicate well-defined (high-robustness) deformation events
characterized by coeval development of syn-tectonic precipitates. Post-tectonic calcite is interpreted to form following deformation
events during inter-seismic periods. Dashed vertical lines indicate moderately defined (low-robustness) deformation events based on
single-sample age. Symbols are larger than 2s age errors.

episodically during different precipitation events and may
result in complex patterns of physical growth. As such, it
can be much more difficult to evaluate whether U-series
isotopes remained closed to secondary mobility. One
way to help evaluate resulting ages is to assume closed
system behavior and compare ages calculated for
syn-tectonic and associated post-tectonic precipitates,
which formed continuously during the inter-seismic
period from the same site and preferably from the same
sample. Dating post-tectonic and syn-tectonic precipitates
from the same site may also provide a more complete his-
tory for fault activity that includes both the active
(seismic) and inactive (inter-seismic) periods. An example
for such a record from the Dead Sea fault zone in Northern
Israel is shown in Figure 4.
Conclusion
Microstructural, geochemical, and isotopic signatures of
syn-tectonic and post-tectonic calcite precipitated in brit-
tle fault zones reflect different formation mechanisms
and temporal relations to faulting events. Therefore,
assessment of these features is important for making accu-
rate interpretations of U-series ages of fault-related calcite.
Resulting ages have been used to directly determine the
timing of faulting events in many recent studies (Flotté
et al., 2001; Verhaert et al., 2003, 2004; Boles
et al., 2004; Uysal et al., 2007; Nuriel et al., 2009, 2011,
2012a; Nuriel et al., 2012b). In addition, U-series ages of
both syn-tectonic and post-tectonic calcite precipitates
from the same site can shed light on seismic and inter-
seismic fault activity.
Bibliography
Boles, J. R., and Grivetti, M., 2000. Calcite cementation along the
Refugio/Carneros fault, Coastal California: a link between defor-
mation, fluid movement and fluid-rock interaction at a basin
margin. Journal of Geochemical Exploration, 69–70, 313–316.
Boles, J. R., Eichhubl, P., Garven, G., and Chen, J., 2004. Evolution
of a hydrocarbon migration pathway along basin-bounding
faults; evidence from fault cement. American Association of
Petroleum Geologists Bulletin, 88, 947–970.
Boullier, A.-M., Fujimoto, K., Ohtani, T., Roman-Ross, G., Lewin,
E., Ito, H., Pezard, P., and Ildefonse, B., 2004. Textural evidence
for recent co-seismic circulation of fluids in the Nojima fault
zone, Awaji island, Japan. Tectonophysics, 378, 165–181.
Clark, M. M., Grantazn, A., and Rubin, M., 1972. Holocene activity
of the Coyote Creek fault as recorded in sediments of Lake
Cahuilla. U.S. Geological Survey Professional Paper, 787,
112–130.
Cox, S. F., and Etheridge, M. A., 1983. Crack-seal fibre growth
mechanisms and their significance in the development of ori-
ented layer silicate microstructures. Tectonophysics, 92,
147–170.
De Paola, N., Holdsworth, R. E., and Collettini, C., 2008. The inter-
nal structure of dilational stepovers in regional transtension
zones. International Geology Review, 50, 291–304.
Edwards, R. L., Chen, J. H., and Wasserburg, G. J., 1987.
238
U-234U-230Th-232Th systematics and the precise measurement
of time over the past 500,000 years. Earth and Planetary Science
Letters, 81, 175–192.
Flotté, N., Plagnes, V., Sorel, D., and Benedicto, A., 2001. Attempt
to date Pleistocene normal faults of the Corinth-Patras Rift
(Greece) by U/Th method, and tectonic implications. Geophysi-
cal Research Letters, 28, 3769–3772.
Grant, L. B., Mueller, K. J., Gath, E. M., Cheng, H., Edwards, R. L.,
Munro, R., and Kennedy, G. L., 1999. Late Quaternary uplift and
earthquake potential of the San Joaquin Hills, southern Los
Angeles basin, California. Geology, 27, 1031–1034.
Gratier, J. P., and Gamond, J. F., 1990. Transition between seismic
and aseismic deformation in the upper crust. In Knipe, R. J.,
and Rutter, E. H. (eds.), Deformation Mechanisms, Rheology
and Tectonics. London: The Geological Society of London. Geo-
logical Society special publication, 54, pp. 461–473.
Gratier, J. P., Favreau, P., Renard, F., and Pili, E., 2002. Fluid pres-
sure evolution during the earthquake cycle controlled by fluid
flow and pressure solution crack sealing. Earth, Planets and
Space, 54, 1139–1146.
Haskin, L. A., Helmke, P. A., Paster, T. P., and Allen, R. O., 1971.
Rare earths in meteoritic, terrestrial, and lunar matter.
In Brunfelt, A., and Steinnes, E. (eds.), Activation Analysis in
 FELDSPAR, INFRARED-STIMULATED LUMINESCENCE
Geochemistry and Cosmochemistry. Oslo: Universitetsforlaget,
pp. 201–218.
Hee-Kwon, L., and Schwarcz, H. P., 1995. Fractal clustering of fault
activity in California. Geology, 23, 377–380.
Huntley, D. J., Godfrey-Smith, D. I., and Thewalt, M. L. W., 1985.
Optical dating of sediments. Nature, 313, 105–107.
Ivanovich, M., and Harmon, R. S., 1992. Uranium-series Disequi-
librium: Applications to Earth, Marine, and Environmental Sci-
ences, 2nd edn. Oxford: Clarendon, p. 910.
Kagan, E. J., Agnon, A., Bar-Matthews, M., and Ayalon, A., 2005.
Dating large infrequent earthquakes by damaged cave deposits.
Geology, 33, 261–264.
Kaufman, A., and Broecker, W., 1965. Comparison of Th230 and
C14 ages for carbonate materials from lakes Lahontan and
Bonneville. Journal of Geophysical Research, 70, 4039–4054.
Ken-Tor, R., Agnon, A., Enzel, Y., Stein, M., Marco, S., and
Negendank, J. F. W., 2001. High-resolution geological record
of historic earthquakes in the Dead Sea basin. Journal of Geo-
physical Research, B: Solid Earth, 106, 2221–2234.
Ku, T.-L., 1965. An evaluation of the U234/U238 method as a tool for
dating pelagic sediments. Journal of Geophysical Research, 70,
3457–3474.
Ku, T.-L., and Liang, Z.-C., 1983. Dating of impure carbonates with
decay-series isotopes. Nuclear Instruments and Methods in
Physics Research, 223, 563–571.
Labaume, P., Carrio-Schaffhauser, E., Gamond, J. F., and Renard,
F., 2004. Deformation mechanisms and fluid-driven mass trans-
fers in the recent fault zones of the Corinth Rift (Greece).
Comptes Rendus Geoscience, 336, 375–383.
Marco, S., Stein, M., Agnon, A., and Ron, H., 1996. Long-term
earthquake clustering: a 50,000-year paleoseismic record in the
Dead Sea Graben. Journal of Geophysical Research, B: Solid
Earth, 101, 6179–6191.
Nuriel, P., Weinberger, R., Rosenbaum, G., Golding, S. D., Zhao, J.-
X., Bar-Matthews, M., Uysal, I. T., and Gross, M. R, 2009.
Timing and mechanism of calcite-filled vein formation in a con-
tractional strike-skip setting, the Dead Sea fault (abstract). In
Barrell, D., and Tulloch, A. (eds.), Joint Geological and Geo-
physical Societies Conference, 2009 Programme & Abstracts.
Geological Society of New Zealand Annual Conference,
Oamaru, November 23–27, 2009, p. 160.
Nuriel, P., Rosenbaum, G., Uysal, I. T., Zhao, J.-X., Golding, S. D.,
Weinberger, R., Karabacak, V., and Avni, Y., 2011. Formation of
fault-related calcite precipitates and their implications for dating
fault activity in the East Anatolian and Dead Sea fault zones. In
Fagereng, A., Toy, V. G., and Rowland, J. V. (eds.), Geology of
the Earthquake Source – A Volume in Honor of Rick Sibson.
London: Geological Society of London. Geological Society spe-
cial publication, 359, pp. 229–248.
Nuriel, P., Rosenbaum, G., Zhao, J.-X., Feng, Y., Golding, S. D.,
Villemant, B., and Weinberger, R., 2012a. U-Th dating of stri-
ated fault planes. Geology, 40, 647–650.
Nuriel, P., Weinberger, R., Rosenbaum, G., Golding, S. D., Zhao,
J.-X., Uysal, I. T., Bar-Matthews, M., and Gross, M. R., 2012b.
Timing and mechanism of late-Pleistocene calcite vein forma-
tion across the Dead Sea Fault Zone, northern Israel. Journal
of Structural Geology, 36, 43–54.
Petit, J. P., 1987. Criteria for the sense of movement on fault
surfaces in brittle rocks. Journal of Structural Geology, 9,
597–608.
Pili, É., Poitrasson, F., and Gratier, J.-P., 2002. Carbon-oxygen iso-
tope and trace element constraints on how fluids percolate
faulted limestones from the San Andreas Fault system:
partitioning of fluid sources and pathways. Chemical Geology,
190, 231–250.
Plan, L., Grasemann, B., Spötl, C., Decker, K., Boch, R., and
Kramers, J., 2010. Neotectonic extrusion of the Eastern Alps:
279
constraints from U/Th dating of tectonically damaged
speleothems. Geology, 38, 483–486.
Ramsay, J. G., 1980. The crack-seal mechanism of rock deforma-
tion. Nature, 284, 135–139.
Ramsey, C. B., Staff, R. A., Bryant, C. L., Brock, F., Kitagawa, H.,
van der Plicht, J., Schlolaut, G., Marshall, M. H., Brauer, A.,
Lamb, H. F., Payne, R. L., Tarasov, P. E., Haraguchi, T.,
Gotanda, K., Yonenobu, H., Yokoyama, Y., Tada, R., and
Nakagawa, T., 2012. A complete terrestrial radiocarbon record
for 11.2 to 52.8 kyr B.P. Science, 338, 370–374.
Sieh, K. E., 1984. Lateral offsets and revised dates of large earth-
quakes at Pallett Creek, California. Journal of Geophysical
Research, 89, 7641–7670.
Uysal, I. T., Feng, Y., Zhao, J.-X., Altunel, E., Weatherley, D.,
Karabacak, V., Cengiz, O., Golding, S. D., Lawrence, M. G.,
and Collerson, K. D., 2007. U-series dating and geochemical
tracing of late Quaternary travertine in co-seismic fissures. Earth
and Planetary Science Letters, 257, 450–462.
Uysal, I. T., Feng, Y.-X., Zhao, J.-X., Isik, V., Nuriel, P., and Gol-
ding, S. D., 2009. Hydrothermal CO2 degassing in seismically
active zones during the late Quaternary. Chemical Geology,
265, 442–454.
Uysal, I. T., Feng, Y.-X., Zhao, J.-X., Bolhar, R., Isik, V., Baublys,
K. A., Yago, A., and Golding, S. D., 2011. Seismic cycles
recorded in late Quaternary calcite veins: geochronological, geo-
chemical and microstructural evidence. Earth and Planetary Sci-
ence Letters, 303, 84–96.
van Calsteren, P., and Thomas, L., 2006. Uranium-series dating
applications in natural environmental science. Earth-Science
Reviews, 75, 155–175.
Verhaert, G., Muchez, P., Sintubin, M., Similox-Tohon, D., Vandycke,
S., and Waelkens, M., 2003. Reconstruction of neotectonic activity
using carbonate precipitates: a case study from the northwestern
extremity of the Isparta Angle (SW Turkey). Journal of Geochem-
ical Exploration, 78–79, 197–201.
Verhaert, G., Muchez, P., Sintubin, M., Similox-Tohon, D.,
Vandycke, S., Keppens, E., Hodge, E. J., and Richards, D. A.,
2004. Origin of palaeofluids in a normal fault setting in the
Aegean region. Geofluids, 4, 300–314.
Verhaert, G., Muchez, P., Keppens, E., and Sintubin, M., 2009.
Fluid impact and spatial and temporal evolution of normal
faulting in limestones. A case study in the Burdur-Isparta region
(SW Turkey). Geologica Belgica, 12, 59–73.
Cross-references
Carbonates, Lacustrine (U-Series)
Carbonates, Pedogenic (U-Series)
Fault Zone (Thermochronology)
U-Series Dating
FELDSPAR, INFRARED-STIMULATED
LUMINESCENCE
Mayank Jain
Center for Nuclear Technologies, Technical University of
Denmark, DTU Risø Campus, Roskilde, Denmark
Definition
This entry primarily concerns the characteristics and the
origins of infrared-stimulated luminescence in feldspars.
280 FELDSPAR, INFRARED-STIMULATED LUMINESCENCE
Introduction
Feldspar is a dominant (~60 %) rock-forming mineral
group in the Earth’s crust. Plagioclase feldspars
(NaAlSi3O8–CaAl2Si2O8) are also commonly observed in
meteorite samples and are a major mineral constituent on
the surface of other extraterrestrial bodies such as the Moon
and Mars. In luminescence dating, K-feldspar (KAlSi3O8)
is the most commonly used dosimeter after quartz.
Feldspars have a direct band gap (with parabolic con-
duction/valence band) which varies with mineralogical
composition: 7.86, 7.70, and 7.62 eV in NaAlSi3O8,
KAlSi3O8, and CaAl2Si2O8, respectively (Malins et al.,
2004; Poolton et al., 2006). Additionally, feldspars have
low-mobility band-tail states which may extend up to >1
eV in individual samples (Poolton et al., 2006, 2009).
Both structural and atomic defects states occupy the band
gap, which give feldspars their dosimetric
(geochronometric) characteristics. Exposure to ionizing
radiation creates free charge (electrons and holes) that
can be trapped at these defects and be subsequently
excited either thermally or optically in the UV–vis-NIR
(ultraviolet–visible–near infrared) range. This entry con-
cerns the near-infrared (NIR, henceforth IR) resonance
feature observed between 750 and 1,000 nm during opti-
cal stimulation of these trapped charges and its dosimetric
characteristics.
The main interest in the IR resonance in feldspars fol-
lows from the fact that IR-stimulated luminescence
(IRSL) signal grows to typically more than 1 kGy with
ionizing dose, thus giving the potential for increasing
the upper end of the optically stimulated luminescence
(OSL) dating range by a factor of about 5–10. Moreover,
feldspar IRSL is typically the only preferable dating sig-
nal either in samples that lack quartz (e.g., basic volcanic
provinces, Moon, Mars, meteorites, etc.) or in samples
that have insensitive quartz. Being mineral specific, there
is also a significant methodological value in using IR res-
onance for targeting, identifying, or cleaning signals
from feldspar phases in untreated polymineral samples
(e.g., fine silt or clay-sized sediments, or field dating
samples).
Historical development
To the author’s knowledge, the first published study
demonstrating the phenomenon of OSL was based on
IR stimulation of prior excited strontium sulfide and
strontium selenide (rare earth element doped) phosphors
(Ellickson, 1946). Ellickson demonstrated that there is
a relationship between the IRSL signal and the thermolu-
minescence (TL) traps in these materials and that the
shape of the IRSL signal decay is dependent on the IR
intensity and it followed second-order kinetics. However,
it took several decades to realize the potential of OSL or
IRSL in dosimetry and dating. IRSL was independently
discovered by Hutt et al. (1988) while measuring
excitation spectra of previously irradiated K-feldspar
samples. These authors observed strong multiple,
overlapping peaks in the region of 1.2–1.6 eV
(~750–1,000 nm) and proposed that the IR resonance in
feldspars arises by a two-step process: IR photon absorp-
tion by a trapped electron (photoionization energy of
~2.5 eV) causing transition to its excited state, followed
by a thermally assisted transition to the conduction band.
This study kick-started widespread use of feldspar IRSL
for luminescence dating because of cheap availability
IR light-emitting diodes (Poolton and Bailiff, 1989) that
could be easily implemented in a TL/OSL reader
(Bøtter-Jensen et al., 1991). Subsequent studies
established that IRSL was thermally stable and bleached
easily with daylight, although somewhat less rapidly
compared to the OSL of quartz (see Murray et al.,
2012), making it a suitable signal for dating. However,
unfortunately, IRSL signal suffered from a special type
of instability popularly known as “anomalous fading”
(loss of thermally stable charge at room temperature;
Wintle, 1973). In the last two decades or so, the single
biggest issue in IRSL dating has been tackling the anom-
alous fading problem. For example, there has been focus
on defining the mechanisms of fading (Spooner, 1994;
Visocekas et al., 1994; Huntley, 2006; Larsen et al.,
2009), fading correction methods (Huntley and Lamothe,
2001; Kars et al., 2008), and a hunt for more stable IRSL
signals (Fattahi and Stokes, 2004; Tsukamoto et al.,
2006; Thomsen et al., 2008; Buylaert et al., 2009;
Jain and Ankjærgaard, 2011; Buylaert et al., 2012). There
has also been a significant thrust, both experimentally
and theoretically, to understand the basic physical mech-
anisms involved in IRSL production with the ultimate
objective of obtaining a better insight into the fading
problem (e.g., Poolton et al., 1994, 1995a; Poolton
et al., 1995b; Poolton et al., 2002a; Poolton et al.,
2002b, 2006; Jain and Ankjærgaard, 2011; Andersen
et al., 2012; Jain et al., 2012; Kitis and Pagonis, 2013).
See Duller (1997) for an extensive review on the subject.
Characteristics of the IRSL
The IRSL emission occurs in several different bands cen-
tered at ~1.7, 2.2, 3.1, 3.7, and 4.1 eV (see review by
Krbetschek et al. (1997) and the later extensive survey
by Baril and Huntley (2003a)). Baril and Huntley
observed for preheated samples that the 2.2 eV band
was dominant in plagioclase feldspars, while the 3.1 eV
band was dominant in alkali feldspars. They also
observed a band at 1.7 eV that was primarily present in
the samples that did not receive any preheating after irra-
diation. The 3.1 eV band is probably due to the native lat-
tice defects, while 2.2 and 1.7 eV emissions are
considered to arise from Mn2+ and Fe3+ transition metal
ion impurities, respectively (see Poolton et al., 2001;
Krbetschek et al., 1997). The 3.1 eV band is commonly
 FELDSPAR, INFRARED-STIMULATED LUMINESCENCE
used in IRSL dating, and it suffers from anomalous fad-
ing, which is now widely accepted to be caused by
tunneling of electrons from the ground state of the dosi-
metric trap to the holes trapped at luminescence centers
(Spooner, 1994). It has been proposed that the orange-
red emission in IRSL is more stable in terms of anoma-
lous fading (Fattahi and Stokes, 2004).
The UV band shows a strong thermal dependence at
high temperatures (0.15 eV activation energy) and
a small or no thermal dependence at low temperatures
(0.05 eV–0.0005 eV activation energy) in both K- and
Na-feldspar samples (Rieser et al., 1997; Poolton et al.,
2009). It has also been also shown that IRSL involves
higher thermal activation energy at the end of the IRSL
decay curve than at the beginning (Li, 2010). The peak
energy of the IR resonance is quite stable with tempera-
ture (Andersen et al., 2012); the main effect of tempera-
ture is an apparent broadening of the resonance
structure. The IRSL decay rate generally increases with
temperature (see McKeever et al., 1997); however, the
extent of this change can vary significantly from sample
to sample. An intriguing thermal characteristic of feld-
spar IRSL is that the net signal intensity as well as the
net depletion of the trapped charge increases with stimu-
lation temperature. Jain and Singhvi (2001) found that
the dosimetric trap giving rise to OSL and IRSL in K-
feldspar can be gradually emptied by high-temperature
IR stimulations, suggesting a strong thermal dependence
on trap emptying in the IRSL process. They attributed
this effect to existence of a light sensitive trap deeper
than the IRSL trap and speculated that this trap may not
show anomalous fading because of expected greater
localization. This thermal characteristic, general to all
feldspars, implies that the IRSL signal can be regenerated
thermally after IR bleaching at a lower temperature. For
example, it is possible to do an IRSL measurement at
room temperature followed by another measurement at
a higher temperature (Thomsen et al., 2008) or by
a sequence of systematically increasing temperatures
(Li and Li, 2011a); these signals are often known as ele-
vated temperature “post-IR-IRSL” (pIR-IRSL) signals.
The general characteristic of these high-temperature sig-
nals is that they show less anomalous fading compared to
the first IRSL signal measured at room temperature
(Thomsen et al., 2008; Li and Li, 2011a; Buylaert et al.,
2012).
Mechanism: IR resonance and luminescence
generation
The resonance feature observed in the excitation spectra
of prior irradiated feldspar rock mineral or sediment
samples has been widely studied by various workers
(e.g., Bøtter-Jensen et al., 1994; Poolton et al., 1995a,
1997; Barnett and Bailiff, 1997; Baril and Huntley,
2003b; Poolton et al., 2009; Andersen et al., 2012). The
281
broad resonance spectrum consists of 1–3 overlapping
peaks superimposed on a monotonously rising continuum.
Spectral analysis suggests that there are two resonance
peaks in the region ~1.42–1.56 eV, which are generally
characteristic of most feldspar samples, and an occasional
peak at about 1.25 or 1.34 eV.
The dominant resonance in the region of ~1.44 eV is
widely accepted to arise due to ground-to-excited state
transition in a deep dosimetric electron trap upon direct
absorption of an IR photon and subsequent recombina-
tion. The exact peak energy of the resonance has been
found to change with the feldspar composition. Poolton
et al. (1995a) modeled this peak shift assuming
a hydrogenic donor center (1s-2p transition) and an effec-
tive electron mass of 0.757 me to obtain peak energies of
1.44, 1.42, and 1.27 eV in K, Na, and Ca end-members,
respectively. The effective electron mass was later con-
firmed to be 0.79 me using optically detected cyclotron
resonance in a sample of Na-feldspar (Poolton et al.,
2001).
Despite a general agreement on the origin of the reso-
nant energy, there has been a considerable debate on the
luminescence generating mechanism subsequent to reso-
nant excitation. The Hutt model (Hütt et al., 1988)
assumes recombination via the conduction band after
a thermally activated transition from the excited state;
however, the thermal barrier of about 0.8 eV for such
a process is rather high and is not likely to be overcome
on the measurement time scales (few tens of seconds)
at or below room temperature. Studies using exo-
electron emission have confirmed that the conduction
band does not participate in the IRSL process at room
temperature (Ankjærgaard et al., 2006). The most widely
accepted model today is the “band-tail state” model pro-
posed by Poolton et al. (1994). These authors investi-
gated the relationship between crystal structure and
the OSL properties of feldspar and suggested existence
of low-mobility band-tail states arising due to high den-
sity of atomic or structural perturbations (impurity
atoms, variations in bonding angles, lattice distortions,
etc.). The band-tail state model is based on a donor–
acceptor recombination, where an excited state electron
undergoes tunneling or phonon-assisted band-tail
hopping in order to recombine with a trapped hole.
This model also explains the broad continuum underly-
ing the resonance peak (Poolton et al., 2002a; Poolton
et al., 2002b).
The proof for the existence of the band-tail states and
their role in charge transport by thermal hopping or tunnel-
ing comes in more recent studies (Poolton et al., 2009; Jain
and Ankjærgaard, 2011). Jain and Ankjærgaard (2011)
developed a comprehensive phenomenological model
based on a single dosimetric trap. They proposed that
two dominant recombination routes in the band-tail states
(0.05 eV phonon-assisted diffusion and quantum mechan-
ical tunneling) together with energy-dependent, localized
282 FELDSPAR, INFRARED-STIMULATED LUMINESCENCE
or delocalized recombinations give rise to a range of
observable luminescence phenomena (both thermal and
optical) in feldspars. This model is able to qualitatively
relate all observable luminescence behavior of feldspar
to a single dosimetric trap and the known available energy
levels such as the excited state (~1.44 eV) of the trap and
the band-tail states.
The IRSL trap
There has also been a significant debate in literature on
whether the IRSL signal originates from a single or mul-
tiple electron traps. The multiple trap hypothesis is gen-
erally based on examination of thermal characteristics
of the IRSL signals; these include a systematic depletion
in the TL signal with the duration of IR exposure (e.g.,
Duller, 1995) and an increasing depletion in the IR inten-
sity with preheat (anneal) temperature (the so-called
“pulse-anneal” curves; Duller, 1994). A gradual system-
atic decrease in such data, either as a function of expo-
sure time or preheat temperature, suggests that IRSL
arises from multiple discrete or a continuum of traps in
feldspars. However, based on a systematic study of sam-
ple ages with preheat temperatures, Murray et al. (2009)
concluded that decrease in the intensities of the low-
temperature TL peaks due to IR exposure reflects reduc-
tion in the hole population rather than the trapped
electron population. They identified the source of IRSL
to be the trap responsible for the TL peak at ~400
C.
There has also been a significant discussion on the origin
of the high-temperature or PIR-IRSL signals. Jain and
Singhvi (2001) attributed that high-temperature IR stim-
ulation accesses a deeper (type B in their terminology)
trap compared to the IR at room temperature. They also
speculated that the deeper trap may be more stable in
terms of anomalous fading. The same hypothesis has
been advocated also in later works to explain the pulse-
anneal behavior of the PIR-IRSL signals (Li and Li,
2011b). However, a consideration of localized recombina-
tion through the excited state in feldspars suggests that
pulse-anneal curves can be explained by both single and
multiple trap hypothesis and such data are, therefore, incon-
clusive for determining the types of trapping states
(Thomsen et al., 2011). Jain and Ankjærgaard (2011) dem-
onstrate that it is phenomenogically possible to explain all
thermal and optical behavior of feldspars in terms of
a single dosimetric electron trap with different recombina-
tion routes/mechanisms.
A more direct examination of the single vs. multiple
trap issue is acquired through excitation spectroscopy.
For a wide range of excitation energies (visible to IR),
the luminescence output has been observed to be consis-
tent with a single trap model (Baril and Huntley, 2003b;
Andersen et al., 2012; Kars et al., 2013). Similarly, the res-
onance peak does not vary with stimulation temperature
up to 290
C apart from the expected thermal broadening
effect suggesting that the same trap is sampled even
at very high temperatures (Andersen et al., 2012).
There has been observed a minor blue shift in the excita-
tion spectra measured after a prior IR bleach (i.e., pIR-
IRSL resonance) of about 10–20 meV; this shift is
compatible with the measured phonon assistance required
for electron transport in the band-tail states (Jain and
Ankjærgaard, 2011) confirming that IRSL and the
PIR-IRSL signals arise from the same dosimetric trap
(Andersen et al., 2012).
There have been several attempts to estimate thermal
trap depth using the first-order analysis of the pulse-anneal
data; these give a range of values: 1.7 eV (Murray et al.,
2009), 1.66 eV (Li and Tso, 1997), and 1.8 eV
(Li and Li, 2011b). The values for PIR-IRSL range from
2.2 to 2.4 eV or 1.9 to 2.1 eV depending on whether the
continuum of states in the band tails is considered or not
(Li and Li, 2011b, 2013). It is, however, questionable
whether the first-order analysis is valid for feldspars in
view of the complex energy structure and different recom-
bination routes (Jain and Ankjærgaard, 2011). Jain et al.
(2012) have developed a new numerical model for analy-
sis of such systems, which seems to be able to explain the
shape of the TL and IRSL decay curve using a single dosi-
metric trap (Kitis and Pagonis, 2013). Further tests of the
model to predict luminescence behavior of feldspars is
necessary.
The presence of the band-tail states makes it difficult to
measure the trap depth directly by optical methods. Some
estimates of optical trap depth in literature are 2.5 eV
(Clark and Sanderson, 1994), >2.5 eV(Baril and Huntley
2003b), and >2.4 eV (Short 2005; Jain and Ankjærgaard
2011). In a recent extensive study, Kars et al. (2013) have
given tentative estimates of optical trap depth to be ~2.1
for Na-feldspar and ~2.5 eV for K-feldspar.
Dating methods using IRSL
Use of IRSL in dating requires detection and, if neces-
sary, correction for anomalous fading. Today the most
commonly used dose measurement method is SAR (sin-
gle-aliquot regenerative dose), which is based on sensi-
tivity monitoring using a test dose (Murray and Wintle,
2000; Wallinga et al., 2000). Estimation of precise fading
rate for age correction has been an issue. Auclair et al.
(2003) adapted the SAR method to measurement of fad-
ing rates, which are subsequently used to correct ages
using a simple integration of charge depletion (t1 depen-
dence) and trapping over the geological time scales
(Huntley and Lamothe, 2001) or a more complex model
using distance distributions (Kars et al., 2008). Age cor-
rections are typically 20–40 % and the correction models
are highly extrapolative. In the past 5 years, several
new dating methods, based on very low or nonfading
IRSL signals, have been proposed. Of these, the method
based on the pIR-IRSL signals is becoming increasingly
routine in dating applications; this method relies on
preferential selection of electrons that are distant
 FELDSPAR, INFRARED-STIMULATED LUMINESCENCE
enough from any holes so as not to undergo ground-state
tunneling in nature (Thomsen et al., 2008; Buylaert
et al., 2012).
Bibliography
Andersen, M. T., Jain, M., and Tidemand-Lichtenberg, P., 2012.
Red-IR stimulated luminescence in K-feldspar: single or multi-
ple trap origin? Journal of Applied Physics, 112, 043507.
Ankjærgaard, C., Murray, A. S., Denby, P. M., and Bøtter-Jensen,
L., 2006. Measurement of optically and thermally stimulated
electron emission from natural minerals. Radiation Measure-
ments, 41, 780–786.
Auclair, M., Lamothe, M., and Huot, S., 2003. Measurement of
anomalous fading for feldspar IRSL using SAR. Radiation Mea-
surements, 37, 487–492.
Bailiff, I. K., and Poolton, N. R. J., 1991. Studies of charge transfer
mechanisms in feldspars. Nuclear Tracks and Radiation Mea-
surements, 18, 111–118.
Baril, M. R., and Huntley, D. J., 2003a. Infrared stimulated lumines-
cence and phosphorescence spectra of irradiated feldspars. Jour-
nal of Physics. Condensed Matter, 15(46), 8029–8048.
Baril, M., and Huntley, D. J., 2003b. Optical excitation spectra of
trapped electrons in irradiated feldspars. Journal of Physics:
Condensed Matter, 15, 8011–8027.
Barnett, S. M., and Bailiff, I. K., 1997. Infrared stimulation spectra
of sediments containing feldspars. Radiation Measurements, 27,
237–242.
Bøtter-Jensen, L., Ditlefsen, C., and Mejdahl, V., 1991. Combined
OSL (infrared) and TL studies of feldspars. Radiation Measure-
ments, 18, 257–263.
Bøtter-Jensen, L., Duller, G. A. T., and Poolton, N. R. J., 1994.
Excitation and emission spectrometry of stimulated lumines-
cence of quartz and feldspar. Radiation Measurements, 23,
613–616.
Buylaert, J. P., Murray, A. S., Thomsen, K. J., and Jain, M., 2009.
Testing the potential of an elevated temperature IRSL signal
from K-feldspar. Radiation Measurements, 44, 560–565.
Buylaert, J.-P., Jain, M., Murray, A. S., Thomsen, K. J., Thiel, C.,
and Sohbati, R., 2012. A robust feldspar luminescence dating
method for Middle and Late Pleistocene sediments. Boreas, 41,
435–451.
Clark, R. J., and Sanderson, D. C. W., 1994. Photostimulated lumi-
nescence excitation spectroscopy of feldspars and micas. Radia-
tion Measurements, 23, 641–646.
Duller, G. A. T., 1994. A new method for the analysis of infrared
stimulated luminescence data from potassium feldspars. Radia-
tion Measurements, 23(2–3), 281–285.
Duller, G. A. T., 1995. Infrared bleaching of the thermolumines-
cence of four feldspars. Journal of Physics D: Applied Physics,
28, 1244–1258.
Duller, G. A. T., 1997. Behavioural studies of stimulated lumines-
cence from feldspars. Radiation Measurements, 27, 663–694.
Ellickson, R. T., 1946. Light sum of phosphors under thermal and
infra-red stimulation. Journal of Optical Society of America,
36, 264–269.
Fattahi, M., and Stokes, S., 2004. Absorbed dose evaluation in feld-
spar using a single-aliquot regenerative-dose (SAR) infrared-
stimulated red luminescence protocol. Radiation Measurements,
38, 127–134.
Huntley, D. J., 2006. An explanation of the power-law decay of
luminescence. Journal of Physics. Condensed Matter, 18,
1359–1365.
Huntley, D. J., and Lamothe, M., 2001. Ubiquity of anomalous
fading in K-feldspars and the measurement and correction for it
283
in optical dating. Canadian Journal of Earth Sciences, 38,
1093–1106.
Hütt, G., Jaek, I., and Tchonka, J., 1988. Optical dating: K-feldspars
optical response stimulation spectra. Quaternary Science
Reviews, 7, 381–385.
Jain, M., and Ankjærgaard, C., 2011. Towards a non-fading signal
in feldspar: insight into charge transport and tunnelling from
time-resolved optically stimulated luminescence. Radiation
Measurements, 46, 292–309.
Jain, M., and Singhvi, A. K., 2001. Limits to depletion of blue–
green light stimulated luminescence in feldspars: implications
for quartz dating. Radiation Measurements, 33, 883–892.
Jain, M., Guralnik, B., and Andersen, M. T., 2012. Stimulated lumi-
nescence emission from localized recombination in randomly
distributed defects. Journal of Physics: Condensed Matter, 24,
385402.
Kars, R. H., Wallinga, J., and Cohen, K. M., 2008. A new approach
towards anomalous fading correction for feldspar IRSL dating-
tests on samples in field saturation. Radiation Measurements,
43, 786–790.
Kars, R. H., Poolton, N., Jain, M., Ankjærgaard, C., Dorenbos, P.,
Wallinga, J., (2013). On the trap depth of the IR-sensitive
trap in Na- and K-feldspar. Radiation Measurements, 59,
103–113.
Kitis, G., and Pagonis, V., 2013. Analytical solutions for stimulated
luminescence emission from tunnelling recombination in ran-
dom distributions of defects. Journal of Luminescence, 137,
109–115.
Krbetschek, M. R., Goetze, J., Dietrich, A., and Trautmann, T.,
1997. Spectral information from minerals relevant for lumines-
cence dating. Radiation Measurements, 27, 695–748.
Larsen, A., Greilich, S., Jain, M., and Murray, A. S., 2009. Develop-
ing a numerical simulation for fading in feldspar. Radiation Mea-
surements, 44, 467–471.
Li, B., 2010. The relationship between thermal activation energy,
infrared stimulated luminescence and anomalous fading of
K-feldspars. Radiation Measurements, 45, 757–763.
Li, B., and Li, S.–. H., 2011a. Luminescence dating of K-feldspar
from sediments: a protocol without anomalous fading correction.
Quaternary Geochronology, 6, 468–479.
Li, B., and Li, S.–. H., 2011b. Thermal stability of infrared stimu-
lated luminescence of sedimentary K-feldspar. Radiation Mea-
surements, 46, 29–36.
Li, B., and Li, S.–. H., 2013. The effect of band-tail states on the
thermal stability of the infrared stimulated luminescence from
K-feldspar. Journal of Luminescence, 136, 5–10.
Li, S.-H., and Tso, M.-Y. W., 1997. Lifetime determination of osl
signals from potassium feldspar. Radiation Measurements, 27,
119–121.
Malins, E. R., Poolton, N. R. J., Quinn, F. M., Johnsen, O., and
Denby, P. M., 2004. Luminescence excitation characteristics of
Ca, Na and K-aluminosilicates (feldspars) in the stimulation
range 5–40 eV: determination of the band-gap energies. Journal
of Physics D: Applied Physics, 37, 1439–1450.
McKeever, S. W. S., Bøtter-Jensen, L., Agersnap, L. N., and Duller,
G. A. T., 1997. Temperature dependence of OSL decay curves:
experimental and theoretical aspects. Radiation Measurements,
27, 161–170.
Murray, A. S., and Wintle, A. G., 2000. Luminescence dating of
quartz using an improved single- aliquot regenerative-dose pro-
tocol. Radiation Measurements, 32, 57–73.
Murray, A. S., Buylaert, J. P., Thomsen, K. J., and Jain, M., 2009.
The effect of preheating on the IRSL signal from feldspar. Radi-
ation Measurements, 44, 554–559.
Murray, A. S., Thomsen, K. J., Masuda, N., Buylaert, J. P., and Jain,
M., 2012. Identifying well-bleached quartz using the different
284 FELDSPARS
bleaching rates of quartz and feldspar luminescence signals.
Radiation Measurements, 47, 688–695.
Poolton, N. R. J., and Bailiff, I. K., 1989. The use of LEDs as an
excitation source for photoluminescence dating of sediments.
Ancient TL, 7, 18–20.
Poolton, N. R. J., Bøtter-Jensen, L., Ypma, P. J. M., and Johnsen, O.,
1994. Influence of crystal structure on the optically stimulated
luminescence properties of feldspars. Radiation Measurements,
23, 551–554.
Poolton, N. R. J., Bøtter-Jensen, L., and Johnsen, O., 1995a. Influ-
ence on donor electron energies of the chemical composition of
K, Na and Ca aluminosilicates. Journal of Physics. Condensed
Matter, 7, 4571–4762.
Poolton, N. R. J., Bøtter-Jensen, L., and Johnsen, O., 1995b.
Thermo-optical properties of optically stimulated luminescence
in feldspars. Radiation Measurements, 24, 531–534.
Poolton, N. R. J., Bøtter-jensen, L., and Johnsen, O., 1997. A search
for IRSL-active dosimeters with enhanced sensitivity: a spectro-
scopic survey. Radiation Measurements, 27, 279–290.
Poolton, N. R. J., Nicholls, J. E., and Bøtter-jensen, L., 2001. Obser-
vation of free electron cyclotron resonance in NaAlSi 3 O 8 Feld-
spar: direct determination of the effective electron mass. Physica
Status Solidi, 225, 467–475.
Poolton, N. R. J., Wallinga, J., Murray, A. S., Bulur, E., and Bøtter-
Jensen, L., 2002a. Electrons in feldspar I: on the wavefunction of
electrons trapped at simple lattice defects. Physics and Chemis-
try of Minerals, 29, 210–216.
Poolton, N. R. J., Ozanyan, K. B., Wallinga, J., Murray, A. S., and
Bøtter-Jensen, L., 2002b. Electrons in feldspar II:
a consideration of the influence of conduction band-tail states
on luminescence processes. Physics and Chemistry of Minerals,
29, 217–225.
Poolton, N. R. J., Mauz, B., Lang, A., Jain, M., and Malins, A. E. R.,
2006. Optical excitation processes in the near band-edge region
of and feldspar. Radiation Measurements, 41, 542–548.
Poolton, N. R. J., Kars, R. H., Wallinga, J., and Bos, A. J. J.,
2009. Direct evidence for the participation of band-tails
and excited-state tunnelling in the luminescence of irradi-
ated feldspars. Journal of Physics: Condensed Matter, 21,
485–505.
Rieser, U., Hutt, G., Krbetschek, M. R., and Stolz, W., 1997. Feld-
spar IRSL emission spectra at high and low temperatures. Radi-
ation Measurements, 27, 273–278.
Short, M. A., 2005. The simultaneous measurement of lumines-
cence and photocurrent of an irradiated K-feldspar excited with
1.45 or 2.41 eV photons. Radiation Measurements, 39, 197–201.
Spooner, N. A., 1994. Infrared-stimulated luminescence. Radiation
Measurements, 23, 625–632.
Thomsen, K. J., Murray, A. S., Jain, M., and Bøtter-Jensen, L.,
2008. Laboratory fading rates of various luminescence signals
from feldspar-rich sediment extracts. Radiation Measurements,
43, 1474–1486.
Thomsen, K. J., Murray, A. S., and Jain, M., 2011. Stability of IRSL
signals from sedimentary feldspars. Geochronometria, 38(1),
1–13.
Tsukamoto, S., Denby, P. M., Murray, A. S., and Bøtter-Jensen, L.,
2006. Time-resolved luminescence from feldspars: new insight
into fading. Radiation Measurements, 41, 790–795.
Visocekas, R., Spooner, N. A., Zink, A., and Blanc, P., 1994. Tunnel
afterglow, fading and infrared emission in thermoluminescence
of feldspars. Radiation Measurements, 23, 377–385.
Wallinga, J., Murray, A. S., and Wintle, A. G., 2000. The single-
aliquot regenerative-dose (SAR) protocol applied to coarse-
grain feldspar. Radiation Measurements, 32, 529–532.
Wintle, A. G., 1973. Anomalous fading of thermoluminescence in
mineral samples. Nature, 245, 143–144.
Cross-references
Feldspars
Luminescence Dating
FELDSPARS
Michel Lamothe
Université du Québec à Montréal, Montréal,
Québec, Canada
Definition
Feldspars are the most ubiquitous minerals in the Earth’s
crust. They are unique dating targets, in the fields of lumi-
nescence and K-Ar/Ar-Ar geochronology, as mineral
grains in sediments and rocks. Both as grains of sediment
and as grains in rocks.
Feldspars are compositionally symbolized as MT4O8
where M are K, Na, and Ca and T are Si and Al.
End-members composition for alkali feldspars are
(K, Al) Si3O8 (microcline, sanidine) and (Na, Al) Si3O8
(albite). Plagioclases represent a continuous solid solution
with end-members composition between albite and anor-
thite (CaAl2Si2O8).
Feldspars are known as framework silicates, in which
corner-sharing (Al, Si) O4 tetrahedra form crankshaft-like
chains, assembled along their a axis. Cross-linking gener-
ates large coordination sites occupied by the charge-
compensating cations (Figure 1).
A most interesting aspect in feldspar mineralogy is in
the distribution of Al and Si in tetrahedral sites. For the
K-feldspars, complete disorder is observed for the high-
temperature monoclinic sanidine, with an equal probabil-
ity of finding Al at any T sites. At the other extreme, in
ordered triclinic microcline, Al is found fitting preferen-
tially in T1 sites.
Due to the inherent complexity of natural crystals, the
solid-state physics of feldspar luminescence is poorly
known (Krbetschek et al., 1997; Baril, 2002). Charge
defects have been mostly detected along Al-O1-Al brid-
ges, a likely candidate at the origin of the blue lumines-
cence emission commonly selected in dating K-feldspar.
Yellow-green (Mn2+) and red (Fe3+) emissions are clearly
linked to specific recombination centers (Krbetschek
et al., 1997; Baril, 2002). Identification of electron traps
has proven challenging. A good candidate at this stage is
substitution of Ti+4 for Al3+ (T site) (Krbetschek
et al., 1997; Baril, 2002). The phenomenology of feldspar
luminescence comprises a distinct instability of latent
luminescence over geological time known as anomalous
fading. Proposed culprits include Al/Si disorder,
tunneling-prone electron traps, and instability of recombi-
nation centers (Feldspar, Infrared-Stimulated
Luminescence).
 FISSION TRACK DATING AND THERMOCHRONOLOGY 285

Feldspars, Figure 1 Idealized feldspar structure, modified from Nesse (2012). Large circles are M sites, which are locations of K, Na, or
Ca; T sites are locations for either Al or Si; Small circles are O oxygen locations.

A minute fraction of the interstitial K undergoes radio-

active decay to 40Ar and 40Ca. For thermochronometry FISSION TRACK DATING AND
applications, the argon diffusion processes are parameter- THERMOCHRONOLOGY
ized through laboratory step-heating experiments.
Thermal diffusion of Ar in feldspar is controlled Andrew J. W. Gleadow and Christian Seiler
by M-O and O-O lattice configurations, vacancies, School of Earth Sciences, University of Melbourne,
and linear and planar defects (Cassata and Renne, 2013). Melbourne, VIC, Australia

Bibliography
Baril, M. R., 2002. Spectral investigations of luminescence in feld-
spars. PhD thesis, Burnaby, BC, Department of Physics, Simon
Fraser University, 296p.
Cassata, W. S., and Renne, P. R., 2013. Systematic variations of
argon diffusion in feldspars and implications for
thermochronometry. Geochimica et Cosmochimica Acta, 112,
251–287.
Krbetschek, M. R., Götze, J., Dietrich, A., and Trautmann, T., 1997.
Spectral information from minerals relevant for luminescence
dating. Radiation Measurements, 27, 695–748.
Nesse, W. D., 2012. Introduction to Mineralogy. New York: Oxford
University Press, 480p.
Cross-references
Feldspar, Infrared-Stimulated Luminescence
Luminescence Dating
Definitions
Fission Track: A linear trail of radiation damage in an
insulating solid resulting from the passage of fission frag-
ments formed by spontaneous or induced fission of
a heavy nucleus, especially of isotopes of uranium.
Fission Track Dating: A technique for geological or
archaeological dating based on the accumulation of fission
tracks from the spontaneous nuclear fission of 238U in nat-
ural minerals and glasses.
Fission Track Density: The number of fission tracks
counted per unit area on a given surface.
Fission Track Length: The length of a fission track mea-
sured under a microscope after chemical enlargement.
Fission Track Annealing: The property of fission
tracks that results in the gradual repair of the radiation
damage in the host material upon exposure to elevated
temperatures.
286 FISSION TRACK DATING AND THERMOCHRONOLOGY
Fission Track Analysis: The measurement of different fis-
sion track parameters including track density and track
length that are used to monitor the degree of fission track
annealing for the purpose of reconstructing rock thermal
histories.
Fission Track Thermochronology: The reconstruction of
rock thermal histories based on the fission track annealing
properties of one or more constituent minerals.
Introduction
The use of accumulated radiation damage tracks from
spontaneous nuclear fission in natural minerals was first
suggested more than 50 years ago as the basis for
a geological dating technique by Price and Walker
(1963). Since that time fission track analysis has evolved
to serve a range of geological, and occasionally archaeo-
logical, dating applications, with particularly widespread
utility in the field of thermochronology. In this approach,
the unique characteristic of fission tracks to fade, or
anneal, at elevated temperatures is used to obtain both
time and temperature information. The development of
powerful modelling techniques now allows for recon-
struction of the thermal histories of rocks by analysis of
fission tracks in their constituent minerals. Fission track
analysis has therefore emerged as one of the most impor-
tant methods for thermochronology, especially in the
low-temperature (<150
C) environment characteristic
of approximately the upper 5 km of the Earth’s crust.
Fission tracks were first observed by transmission elec-
tron microscopy as linear trails of radiation damage in
mica by Silk and Barnes (1959) and Price and Walker
(1962). These tracks appeared as cylindrical regions of
highly disordered or amorphous material averaging
around 10 nm across within an undisturbed crystalline
matrix. Price and Walker (1963) went on to show that
these tracks could be enlarged by a simple chemical etch-
ing treatment so that they could then be observed much
more conveniently under an optical microscope. They also
showed that many natural micas had a background of
tracks from the spontaneous fission of 238U and suggested
that this was the basis for a new geological dating method.
The simplicity of revealing fission tracks by etching led to
a rapid expansion of activity by many researchers that
explored the application of fission tracks, and other
nuclear particle tracks, in many different branches of sci-
ence as summarized by Fleischer et al. (1975). Of these,
fission track dating has proved to be by far the most
important.
Early work on fission track dating showed that fission
tracks could be observed in a variety of natural minerals
and glasses (for a review, see Wagner and Van den haute,
1992), but most of these contain insufficient uranium or
are too rare to be useful in routine dating studies. The
pioneering studies by Naeser (1967) and Wagner (1968)
focused attention on commonly occurring, uranium-
bearing accessory minerals such as apatite, titanite, and
zircon. These accessory minerals occur as trace
constituents in a wide variety of common igneous, meta-
morphic, and sedimentary rocks and mostly contain suit-
able amounts of uranium within their structures
(typically in the ppm to hundreds of ppm range) to be use-
ful for fission track dating. Of these minerals, just one,
apatite, has come to account for more than 80 % of all fis-
sion track analyses because of its common occurrence,
generally suitable uranium concentration, consistent etch-
ing properties, and well-understood thermal annealing
behavior.
The use of a particular mineral for fission track dating
requires estimates of three quantities: (1) the number of
fission tracks in the material, (2) the amount of fissionable
parent isotope present, and (3) the rate of spontaneous fis-
sion decay. In addition, it must either be assumed that no
fission tracks have been lost since the system was formed
or that the degree of loss can be estimated. It was soon rec-
ognized, however, that fission tracks could undergo
annealing if exposed to elevated temperatures, resulting
in fission track ages that were “too young” compared to
independent formation age estimates. Considerable effort
was therefore devoted to understanding and quantifying
this fission track annealing process, initially with the
objective of “correcting” the observed ages in order to
recover the true age of the material. Such attempts resulted
in some success, particularly in dating natural glasses,
such as tektites (Storzer and Wagner, 1969).
It was soon realized, however, that this concept could
be turned around so that fission track annealing could be
used to recover information about the conditions of
annealing rather than being used to correct ages. The three
American physicists who originated the fission track
method, Fleischer et al. (1965a, p. 389), observed: “As
in all age measurements, one must ask what time interval
is being measured. It is our opinion that the “low” ages
which are sometimes obtained by the fission-track method
represent definite physical events, such as heating epi-
sodes.” This was one of the key insights that initiated the
study of thermochronology – the use of temperature-
sensitive dating methods to reconstruct rock thermal histo-
ries – a field that has come to dominate the use of fission
track analysis over the last 30 years.
The basic principles and applications of fission track
dating and thermochronology have been extensively
reviewed by Wagner and Van den haute (1992), Gallagher
et al. (1998), Gleadow et al. (2002), Donelick et al. (2005),
and Tagami and O’Sullivan (2005).
Spontaneous fission and nuclear track formation
Spontaneous nuclear fission is one of the rarest forms of
radioactive decay and occurs naturally only in the isotopes
of uranium and thorium. In practice, however, only 238U is
responsible for the fission tracks observed in natural mate-
rials, for which the spontaneous fission half-life is approx-
imately 8  1015 years. Other isotopes of these heavy
elements, 235U and 232Th, can effectively be ignored as
a source of fission tracks because they are either much less
 FISSION TRACK DATING AND THERMOCHRONOLOGY
abundant or have significantly longer half-lives. Even in
238
U, the rate of spontaneous fission decay is vastly slower
than the normal alpha decay mode, so that only about one
in every two million decay events in this isotope is the
result of nuclear fission. The much more common
alpha decay is the basis for three other dating methods
(see “Uranium–Lead Dating”; “U-Series Dating”;
“U–Th/He Dating” ).
Spontaneous fission results in a great variety of middle-
weight isotopes as daughter products, most of which are
extremely difficult to measure due to the rarity of these
decay events. However, fission track dating is extremely
sensitive as it measures the effect of the decay in the form
of radiation damage, rather than the decay products them-
selves. Each fission track represents a single 238U nucleus
that has decayed in this way, so that counting fission tracks
is equivalent to counting individual radioactive decay
events.
In spontaneous fission the parent 238U nucleus splits
apart into two sub-equal and massively charged particles
that carry away most of the enormous energy released by
nuclear fission as kinetic energy. These massively ionizing
particles travel rapidly apart at 180
to each other, to pro-
duce a single linear trail of radiation damage in the host
material, the fission track (see “Radiation Defect”). The
result is a narrow track core of amorphous material about
5–10 nm across but continuing for lengths of up to
20 mm in some materials. In apatite, recent evidence sug-
gests that the track core may even be a void
(Li et al., 2012).
Revealing and observing fission tracks
Fission track etching
Fission tracks are revealed for counting and measurement
under a microscope by a chemical etching procedure.
The etchant preferentially attacks the damaged core
of the track, which is chemically more reactive than
the surrounding undamaged matrix. This contrast in reac-
tivity is to be expected in crystalline minerals but also
exists to a lesser degree in natural glasses, where the track
core is more disordered than the surrounding glass
matrix.
Etching is typically carried out on an internal surface of
the material to be dated. Internal surfaces are exposed
either by grinding and polishing or by exploiting the natu-
ral cleavage of minerals such as mica. The surface is
immersed in a suitable chemical etchant for the particular
mineral to enlarge latent fission tracks to optical dimen-
sions of ~1–2 mm across.
Initially, etching of the track core is controlled by the
radiation damage itself, but further enlargement of the
track is controlled by the properties of the material. In
the case of etching tracks in glasses, the rate of etching
along the track is only slightly faster than the bulk etching
rate of the surrounding material, so that the track is simul-
taneously widening while the etchant is still penetrating
toward the end of the track. The outcome is a broad,
287
a c-axis b
c-axis
Dper
1 µm
Dpar
Fission Track Dating and Thermochronology, Figure 1 The
shape of etched fission tracks on different crystal surfaces in
apatite (a). The scanning electron microscope image (b) shows
the elongated appearance of etched fission track openings on
a surface parallel to the c-axis. The long and short axes of etch
pits are known as Dpar and Dper (Donelick et al., 2005).
conical-shaped track. In minerals, the etchant penetrates
almost instantaneously along the whole length, so that
the track subsequently widens as a nearly parallel-sided
hole. As the etching rates are not uniform in different crys-
tallographic directions, fully etched tracks are essentially
faceted “negative crystals” forming shapes like knife
blades, daggers, prisms, and pyramids depending on their
orientation (Figure 1). Figure 2 shows the appearance of
well-etched tracks on a c-axis-parallel surface in an apatite
crystal.
Prior to etching, minerals are typically separated from
their host rock by crushing and then using standard heavy
liquid and magnetic separation techniques, before mount-
ing the concentrated mineral grains in a suitable medium
such as epoxy resin or, for use with highly reactive etch-
ants, transparent FEP teflon. Some typical etching treat-
ments for different materials are listed in Table 1.
Fission track microscopy
Observation and measurement of fission tracks requires
a high-quality research microscope typically configured
for illumination by both transmitted and incident light,
a rotating nosepiece with various objectives up to 100
with an appropriate substage condenser and a motorized
x-y scanning stage. Track lengths are conventionally mea-
sured using a drawing tube device that superimposes an
image of a digitizing tablet on the visible image, although
modern equipment generally uses digital cameras
attached to the microscope for this purpose (see section
“Automation of Fission Track Measurements”).
The choice of objectives is important for fission track
analysis and all objectives should be corrected for slides
288 FISSION TRACK DATING AND THERMOCHRONOLOGY

Fission Track Dating and Thermochronology, Figure 2 Etched spontaneous fission tracks in an apatite grain from the Grassy
Granodiorite, King Island, in southeastern Australia. The tracks appear as dark, randomly oriented lines of up to about 15 mm length.
Faint lines running continuously across the surface are etched polishing scratches.

Fission Track Dating and Thermochronology, of interest and marking of tracks on screen using
Table 1 Commonly used etching conditions for revealing a computer mouse. Digital image analysis may also
fission tracks be used as discussed below (see section “Automation of
Mineral Etchant Temperature Time
Fission Track Measurements”).

Apatite 5 N HNO3 20
C 20 s Fission track measurements

Zircon KOH:NaOH eutectic 220
C 4–100 h Measurements of fission tracks fall into two categories –
Titanite 50 N NaOH 120
C 30 min–5 h counting the number of etched fission tracks intersecting
1HF:2HCl:3HNO3:6H2O 20
C 1–25 min
Muscovite HF (40 %) 20
C 20 min a prepared surface and measuring various dimensions of
mica the etched tracks. These two types of measurements
Natural HF (48–15 %) 20
C 15 s–5 min respectively contain the time and the temperature compo-
glasses nents that are required for thermal history reconstruction.

with no cover glass. Most researchers prefer to use dry
objectives, rather than oil immersion lenses, especially
for observation of fission tracks in apatite and mica whose
refractive indices are almost the same as the immersion oil
making the tracks difficult to see. Oil immersion lenses
are, however, sometimes used for observing tracks in
higher refractive index minerals, such as zircon and
titanite.
The number of fission tracks present on a particular sur-
face is most commonly counted at a magnification of
1,000 or more by scanning across a calibrated grid of
known area in the microscope eyepiece and recording
the number of tracks with a hand tally device. In modern
systems, digital cameras attached to the microscope gener-
ate a calibrated live image that can be analyzed directly on
a computer, allowing counting of more complex regions
Track counting
The primary parameter measured in fission track analysis
is the track density, which provides an estimate for the
number of fission events accumulated over time in
a geological material. The fission track density is defined
as the number of track intersections per unit area of
a given surface, usually expressed as tracks per cm2. As
such, the track density is a two-dimensional measurement,
which is related to the three-dimensional distribution of
fission events via the average length of the fission tracks
in that material.
Due to the anisotropic etching properties of minerals
(see section “Fission Track Etching”), track densities are
usually measured on specific crystal planes. In apatite
and in zircon, the optimum surfaces for counting fission
tracks are those that are cut parallel to the crystallographic
c-axis. Such surfaces reveal tracks with a well-developed
form and a high detection efficiency, which may be
defined as the number of tracks actually counted as
 FISSION TRACK DATING AND THERMOCHRONOLOGY
a function of the total number of tracks crossing that sur-
face. In mica the optimum surface is the basal cleavage
plane.
Track densities are of two kinds, spontaneous and
induced. Spontaneous, or fossil, track densities record
the number of tracks originating from spontaneous fission
of 238U contained within the target material. These are the
fission tracks that accumulate over the geological lifetime
of the material, their number being proportional to the
time over which that accumulation has been taking place,
and the uranium concentration. Induced track densities, on
the other hand, record the number of tracks arising from
neutron-induced fission of 235U and are used during fis-
sion track dating as a proxy for the amount of 238U present
in a sample (see section “Fission Track Methods”).
In determining the track density, it is important to reli-
ably discriminate true fission tracks from non-track fea-
tures such as dislocations, etching artifacts, inclusions,
surface dust, and scratches. In most cases, the tracks are
obvious and discrimination is straightforward, but occa-
sionally very difficult grains with complex dislocation
swarms are encountered. Such grains are best avoided
altogether. Three commonly used criteria to discriminate
tracks from non-track features are as follows: Fission
tracks are straight-line defects, are randomly oriented,
and are of limited length, determined by the range of fis-
sion fragments in that particular material (Fleischer and
Price, 1964). Dislocations and other spurious etch pits,
on the other hand, are often preferentially oriented in par-
allel arrays, may be much longer than fission tracks, and
often show complex curvature and branching paths.
Interference from some of the other artifacts can be min-
imized by clean handling and careful observation.
Track length measurements
A second group of measurements relates to the overall
dimensions of fission tracks, which can give valuable
information on the degree of thermal annealing they have
experienced. Different dimension parameters that have
been used include the diameter of track etch pits in glasses,
the projected lengths of tracks in minerals (the apparent
length of dipping tracks projected onto the horizontal sur-
face), and the lengths of confined fission tracks.
Most fission track studies in minerals since the
mid-1980s have measured the lengths of horizontal con-
fined tracks, especially in apatite (Gleadow et al., 1986).
Confined fission tracks are tracks with their entire etchable
length contained within the body of the surrounding min-
eral. In this case, the etchant penetrates into the confined
track, not directly from the surface but from a surface-
intersecting track or fracture. Depending on the way they
are etched, confined tracks can be either track-in-track
events (TINTs) or track-in-cleavage events (TINCLEs,
Lal et al., 1969) as illustrated in Figure 3. TINTS are
now favored over TINCLEs, because the latter introduce
additional observational biases and may be anomalously
stabilized (e.g., Barbarand et al., 2003), making their
289
lengths unrepresentative of the thermal history. The orien-
tations of these tracks relative to the crystallographic
c-axis are also usually measured to account for differences
in both etching and annealing characteristics of tracks in
different orientations. Confined tracks show the closest
approximation to the true lengths of unetched fission
tracks, although their measurement is still subject to vari-
ous unavoidable, but well-understood, sampling biases
(Galbraith, 2005).
Due to the limited accuracy of vertical movement in
earlier-generation microscopes, measurements are usually
made only for horizontal confined track lengths, thereby
avoiding the need to correct for track inclination (Laslett
et al., 1982). Modern microscopes have overcome this
problem, so that it is now possible to accurately measure
confined tracks of any orientation, which obviously
increases the sampling size. Typically the lengths of around
100 confined track lengths are measured, although this may
not be possible in samples with low track densities.
The number of TINTs in a particular sample can be
enhanced by implanting artificial fission tracks from
a 252Cf source (Donelick and Miller, 1991). This increases
the number of etch channels to maximize the likelihood of
TINT intersections below the surface and is especially
useful where too few confined tracks can be found
naturally.
Track diameter measurements
Measurements of track diameters were first used in natural
tektite glasses as a means for correcting the apparent fis-
sion track ages for partial loss, or fading, of the spontane-
ous tracks (e.g., Storzer and Wagner, 1969; see “Impact
Glass (Fission Tracks)”). This followed the observation
that etched diameters of spontaneous tracks were smaller
than induced tracks, and experiments revealed
a relationship between the reduction in track diameter
and the track density. Although his approach worked well
for glasses, a similar approach for minerals was much less
successful as corrected ages rarely relate to the formation
age of the host sample.
Fission track fading or annealing in apatites is
influenced by their chemical composition (see section
“Experimental Fission Track Annealing Studies”), primar-
ily variations in chlorine concentration (Green
et al., 1985). The rate of etching is also observed to vary
with composition in apatite, such that the tracks appear
more strongly etched in some cases than others after con-
stant etching. Measuring track diameter parameters such
as Dpar and Dper – the average diameters of etch pits paral-
lel and perpendicular to the crystallographic c-axis,
respectively (Figure 1) – can thus be used to indirectly
monitor the annealing sensitivity of apatite (Donelick
et al., 2005). Dpar values typically vary between about
1.2 and 3.5 mm across the range of natural apatite samples.
Although there is a coarse correlation between Dpar and
chlorine content of different apatites (Donelick
et al., 2005), this relationship becomes less clear for the
290 FISSION TRACK DATING AND THERMOCHRONOLOGY

Fission Track Dating and Thermochronology, Figure 3 Photomicrographs of confined fission tracks in apatite in reflected (left) and
transmitted light (right). Two track-in-track (TINT) features are highlighted by arrows in (a) and (b), and two track-in-cleavage (TINCLE)
features in (c) and (d). The dashed arrows show the crystallographic c-axis direction defined by the elongation of the surface track
etch pits in reflected light.

most common fluorine-rich varieties of apatite (e.g.,
Green et al., 2005) leading to some debate about the use-
fulness of Dpar alone to monitor track annealing.
Automation of fission track measurements
The latest generation of fully motorized digital research
microscopes allows for comprehensive computer control
of all aspects of microscopy and the acquisition of high-
quality digital images. This new technology has con-
verged with rapid increases in computing power and cheap
availability of abundant digital image storage to bring new
opportunities for automation to the laborious process of
fission track data collection.
Gleadow et al. (2009) have described a procedure that
makes use of these new technologies to implement autono-
mous digital imaging and automatic counting of fission
tracks using image analysis. The emphasis has been on
counting tracks in apatite and mica to support the great bulk
of fission track analysis conducted today, but the system is
readily adaptable to other minerals, glasses, and plastic
track detectors. The central procedure of coincidence map-
ping relies on the digital comparison of pairs of reflected
and transmitted light images for each microscopic field of
view, as illustrated schematically in Figure 4. A detailed
Z-stack of transmitted light images is also captured at the
same location to provide a fully focusable image of each
grain for later inspection. The system also includes
a comprehensive set of partially automated tools for locat-
ing suitably oriented grains in the mount as well as measur-
ing various geometric parameters, including c-axis
orientation, track length, and orientation, and Dpar.
Methods of dating
Fission track methods
As in any radiometric dating method, determining
a fission track age involves obtaining a measure of both
 FISSION TRACK DATING AND THERMOCHRONOLOGY 291

Fission Track Dating and Thermochronology, Figure 4 Automated counting of fission tracks by coincidence mapping. The
procedure digitally combines a slightly defocused transmitted light image (a) and a focused reflected light image (b) of the same
tracks to obtain the overlapping features shown as black spots in (c), which are then analyzed automatically to obtain the number of
tracks, discriminate track overlaps, and determine dimensions such as Dpar.

the parent isotope and the daughter products of the decay
scheme. In this case, the number of decay events
(daughter products) is measured using the spontaneous
track density as described in section “Track Counting.”
The parent isotope, 238U, is normally estimated indirectly
by using 235U as a proxy, the concentration of which
is obtained by counting the number of induced fission
tracks produced during thermal neutron irradiation
(see section “Track Counting”). Induced tracks are related
to the 238U concentration via the known 235U/238U ratio,
which can be taken as constant (0.73 %) in natural geolog-
ical materials, and the total neutron fluence received.
Most fission track dating techniques, therefore,
require determination of the ratio of spontaneous to
induced track densities in a sample and measurement of
the neutron fluence received. There are a number of ways
in which these variables can be determined (Gleadow,
1981; Hurford and Green, 1982; Wagner and Van den
haute, 1992; Gleadow et al., 2002). In grain-population
methods, both spontaneous and induced track densities
are determined as averages over a large number of min-
eral grains from different aliquots, while grain-by-grain
methods allow individual ages to be determined for each
analyzed mineral grain. Since the 1980s, just one of the
grain-by-grain methods, the external detector method
(EDM, see section “The External Detector Method”),
has become standard practice for the great majority
of mineral dating studies due to its analytical conve-
nience and the individual grain data that are acquired.
An important recent development is the emergence of
laser ablation inductively coupled plasma mass spectrome-
try (LA-ICP-MS) as routine tool for the precise analysis of
trace elements in minerals. This method can measure the
concentration of the parent isotope 238U directly, rather than
via 235U, thereby eliminating the need for neutron irradiation
while maintaining many of the benefits of the EDM method
(see section “The Laser Ablation ICP-MS Method”).
The fission track age equation and calibration
The fission track age equation is a specialized form of the
general radiometric age equation that expresses the fission
track age (t in years) in terms of the ratio of the spontane-
ous fission track density (rs cm2) to the induced track
density (ri cm2) as follows (Fleischer and Price, 1964;
Naeser, 1967):
 
1 lD fsI rs
t ¼ ln 1 þ ð1Þ
lD lf ri
where
lD ¼ total decay constant for 238U (1.55125 
10
10 year1)
lf ¼ spontaneous fission decay constant of 238U (8.46
 1017 year1)
292 FISSION TRACK DATING AND THERMOCHRONOLOGY

I ¼ isotopic ratio 235U/238U (7.2527  103) As for the zeta calibration approach described by
s ¼ thermal neutron cross section for 235U (580.2  Eq. (4) above, the aggregate constant x (Xi) in Eq. (6)
24
10 cm2) can also be calibrated empirically against a set of age stan-
f ¼ thermal neutron fluence (n cm2) dards. However, all of the constituent constants are
A calibration factor g must be introduced if the sponta- known, at least in principle, or can be determined explic-
neous and induced tracks are measured on surfaces with itly, so that an absolute determination of the age is also
different track registration geometries (spherical or half- possible.
spherical; see section “The External Detector Method”)
and detection efficiencies, as is the case in the external The external detector method
detector method.
In the external detector method (EDM), spontaneous
For methods that require neutron dosimetry, the neutron
tracks are measured on an internal surface in the etched
fluence is calculated from the following:
mineral grains and induced tracks in an external track
f ¼ Brd ð2Þ detector held firmly in contact with the mineral surface
during irradiation. The external detector is almost always
where rd is the induced track density measured in a mica a cleaved surface of low-uranium muscovite. During neu-
external detector that was in contact with a uranium-bear- tron irradiation, some of the induced fission fragments
ing standard glass in the same irradiation package. The from 235U in the mineral grains leave the surface and enter
constant B is calibrated by comparison with independent the external detector. It is important to note that an external
thermal neutron fluence dosimetry. detector only receives tracks from one side (2p or hemi-
Substituting Eq. (2) into Eq. (1) gives the general work- spherical track registration geometry), compared to the
ing form of the fission track age equation expressed in case of spontaneous tracks etched on an internal surface,
terms of the three track densities: which receives tracks from both sides (4p or spherical reg-
  istration geometry).
1 rs
t ¼ ln 1 þ lD zg rd ð3Þ After irradiation, the mica external detectors are etched
lD ri to reveal patches of induced tracks corresponding to each
grain in the mineral mount. As the grain mount is not
where
etched again after irradiation, only spontaneous tracks
sIB are visible there, while only induced tracks are visible in
z¼ ð4Þ the external detector. The etched surfaces have a mirror-
lf
image relationship to each other and are affixed side by
The aggregate constant z (zeta) is calibrated empirically side to a microscope slide for counting of exactly
against a set of age standards and was introduced to cir- matching areas of spontaneous and induced tracks.
cumvent uncertainties in the spontaneous fission decay Matching between grains and mica is usually automated
constant and the neutron dosimetry calibration (Hurford using a computer-controlled stage.
and Green, 1983). Suitable age standards are minerals An age determination by the EDM proceeds by
from rocks that have cooled rapidly, such as undisturbed selecting suitable grains in the mount, counting the spon-
volcanic rocks, and have been precisely dated by other taneous tracks present, and then counting induced tracks
methods, most commonly by Ar/Ar dating. over the corresponding area on the external detector.
In the case where the 238U concentration of a mineral Poorly orientated grains, those with nonuniform uranium
grain is determined explicitly using the LA-ICP-MS distributions, or those with interference from dislocations
method (section “The Laser Ablation ICP-MS Method”), or other defects, are simply avoided.
Eq. (3) above is modified to An important advantage of the EDM over alternative
  fission track methods is that data are collected over indi-
1 rs vidual grains. Normally, about 20 grains are counted so
t ¼ ln 1 þ lD x ð5Þ
lD CU that the resultant single-grain age distribution can be
tested statistically to see whether they represent
where a uniform age population or not. Since the 1980s, the
A EDM has become the standard method for fission track
x¼ ð6Þ analyses, being applied almost universally of studies of
lf N 0 Ra apatite, zircon, and titanite, as well as occasional analysis
of other uranium-bearing minerals, such as garnet and
and
epidote.
CU ¼ concentration of 238U
A ¼ atomic weight of 238U
N0 ¼ Avogadro’s number Neutron dosimetry
D ¼ density of the mineral For the external detector, and all other methods of fission
R ¼ average etchable range of one fission fragment track analysis that utilize neutron irradiation, it is vital to
a ¼ detection efficiency for the etched surface monitor the total neutron fluence received by each grain
 FISSION TRACK DATING AND THERMOCHRONOLOGY
mount during irradiation in order to know how much of
the 235U has undergone fission during irradiation. This is
accomplished by including neutron flux monitors in the
same irradiation package as the prepared mounts. Flux
monitors are positioned at the top and bottom of each
package to monitor the total neutron fluence received
and correct for any flux gradients. Once the neutron
fluence is known for each sample in the irradiation pack-
age, uranium concentrations can be calculated for each
analyzed mineral grain based on the corresponding num-
ber of induced tracks recorded in the external detector.
Typically these flux monitors are simply muscovite
external detectors attached to pieces of reference glass
with a well-known and homogeneous uranium concentra-
tion, allowing standard glass track densities to be deter-
mined in the etched external detector. Although flux
monitors may be calibrated against the absolute thermal
neutron fluence (Wagner and Van den haute, 1992), the
fluence calibration is usually included in the aggregate
zeta constant (see section “The Fission Track Age
Equation and Calibration”) so that the standard glass track
density is applied directly in the age calculation (Eq. 3).
Neutron dosimetry-based methods have the advantage
that they are all based on a ratio of two track density mea-
surements, which have similar properties and are therefore
likely to be subject to similar errors that ideally should
cancel out. Significant disadvantages include very long
sample turn-around times while the samples are away at
the reactor and the necessity of handling radioactive mate-
rials after they are returned from irradiation.
The laser ablation ICP-MS method
Over the last decade, the emergence of LA-ICP-MS to
directly measure the 238U concentration has the potential
to rival, and even replace, the EDM (Hasebe
et al., 2004). Sample preparation and etching proceeds in
much the same way as for the EDM, and, as before,
a series of well-oriented mineral grains are selected on
an internal surface. Once the spontaneous tracks are
counted, the sample is placed into a laser ablation cham-
ber, and an area of each counted grain is ablated and ana-
lyzed in the mass spectrometer.
As grains are typically small (~100 mm), a single laser
spot with a diameter of 20–40 mm is usually used to ablate
part of the mineral. An inert gas transports the aerosol
from the ablation chamber into the plasma torch of an
ICP-MS, where the ablated molecules are dissociated
and ionized before being bled into a quadrupole mass
spectrometer for analysis. The measured 238U signal is
then normalized using a stoichiometric isotope of known
(or assumed) concentration such as Ca or Si to correct
for variations in the ablation volume and the absolute con-
centration determined relative to a standard of known
composition (see “Laser Ablation Inductively Coupled
Mass Spectrometer (LA ICP-MS)”).
The LA-ICP-MS method can be applied to a range of
different minerals and produces single-grain analytical
293
data of comparable precision to the EDM. This new
approach has the additional advantages of simplified sam-
ple preparation procedures, rapid and largely automated
analysis, and needs only one (spontaneous) instead of
three (spontaneous, induced, and standard glass) track
densities. Most importantly, however, it does not require
neutron irradiation and therefore eliminates the need to
handle irradiated materials, resulting in dramatically
reduced sample turn-around times. Although significant
work still needs to be done on calibration, standardization,
and testing, this technique is likely to be adopted much
more widely in the near future.
One disadvantage of the LA-ICP-MS method is that it
is a destructive technique on a microscopic scale,
destroying part of each grain during analysis so that later
inspection of the counting results is not possible. How-
ever, this shortcoming can easily be overcome if image
sets are captured before ablation (see section “Automation
of Fission Track Measurements”), thus creating
a permanent digital record of each grain.
Fission track annealing and modelling
It was discovered early in the development of fission track
dating that fission tracks are stable over geological time
only at relatively low, near-surface temperatures. As the
temperature is raised, the radiation damage making up
the core of each track is progressively repaired until the
track no longer forms a continuous defect that can be
revealed by etching. This process of fission track fading
upon exposure to elevated temperatures is called thermal
annealing, or often simply annealing.
Thermal annealing is a kinetic process in which temper-
ature and time can to some extent be interchanged with
each other, longer exposure to lower temperatures produc-
ing the same effect as shorter exposure to higher tempera-
tures. Each mineral shows a different characteristic
temperature range over which fission track annealing
occurs, often referred to as the partial annealing zone.
For apatite this annealing range is approximately
60–120
C over geological heating times, with signifi-
cantly higher characteristic annealing temperatures of
around 150–200
C for zircon and 200–250
C for
titanite, although these are not as well constrained as they
are for apatite. A quantitative understanding of the track
annealing process is critical to the use of fission track anal-
ysis for thermochronology.
Environmental parameters other than time and temper-
ature (e.g., pressure, ionizing radiation) have no signifi-
cant effect on fission track annealing (Fleischer
et al., 1965b), at least at the upper crustal pressures under
which thermal annealing of fission track occurs in com-
monly used minerals such as apatite (Schmidt et al., 2014).
Experimental fission track annealing studies
Thermal annealing of fission tracks in a particular mineral
can be studied in the laboratory by observing the effects of
different time and temperature treatments on freshly
294 FISSION TRACK DATING AND THERMOCHRONOLOGY

Borehole Laboratory
1.0

Track Length Reduction (l/l0)

0.5

107 years

106 years

96 hours
30 days

9 hours

20 min
1 hour
0.0
0 100 200 300 400

Fission Track Dating and Thermochronology, Figure 5 Comparison of the effect of thermal annealing on fission tracks on
laboratory (right) and geological time scales (left). Laboratory annealing data (From Green et al., 1986) show the reduction in mean
confined track length (l) relative to the length of unannealed induced tracks in the same apatite (lo) after heating at various
temperatures and times. Similar length reduction is shown by spontaneous tracks in apatites from deep boreholes in the Otway Basin
of SE Australia (From Gleadow and Duddy, 1981).

induced 235U tracks on the assumption that these will
behave identically to 238U tracks. The degree of annealing
is then monitored by measuring the length distribution of
confined fission tracks.
The results of numerous such studies have led to quan-
titative models of fission track annealing in different min-
erals that may be extrapolated from laboratory to
geological time scales (e.g., Laslett et al., 1987; Carlson
et al., 1999; Barbarand et al., 2003 for apatite; Tagami
et al., 1998 for zircon). Figure 5 shows the effects of labo-
ratory annealing on induced track lengths in Durango apa-
tite over different time scales (Green et al., 1986). The
longer the exposure time, the lower the temperature
required to produce a given degree of track shortening
due to annealing. Extrapolations of the annealing curve
for this apatite are also shown for heating times of
106–107 years using the annealing model of Laslett
et al. (1987), along with observational data from deep
borehole samples from the Otway Basin (Gleadow and
Duddy, 1981). The Otway apatites show a range of com-
positional variation, which broadens their annealing
behavior compared to the single Durango apatite composi-
tion. However, the general consistency and similar form of
the annealing curves between the borehole and laboratory
results strongly suggests that annealing is being caused by
the same mechanism in both cases.
In apatite, another important control of fission track
annealing is the mineral composition. Apatites from most
geological environments are dominated by fluorapatites,
but also grade in composition toward chlorapatite, and,
less commonly, hydroxyapatite. Fission tracks in more
chlorine-rich apatites are more resistant to thermal
annealing than those in fluorapatite as illustrated in
Figure 6. As described earlier (see section “Track Diame-
ter Measurements”), etched track dimensions such as Dpar
also correlate to some degree with composition and may
thus be used as a kinetic parameter to assess the annealing
sensitivity of a particular apatite.
Thermal history modelling
Track shortening during thermal annealing has two impor-
tant consequences for fission track analysis. One is that
each individual spontaneous track in a natural sample
experiences a different fraction of the host rock’s total
thermal history and will shorten to a degree determined
by the maximum temperature experienced since track for-
mation. Because all of the tracks are of different ages,
being added progressively by radioactive decay,
a spectrum of track lengths results that contains
a detailed record of the entire annealing history within
the temperature range of the partial annealing zone. The
second effect is that fewer tracks will intersect a given sur-
face as the average track length gets shorter during
annealing, so that the apparent fission track age is reduced
proportionately during annealing. Thus, partial thermal
annealing will have the effect both of reducing the fission
track age and changing the track length distribution. Uti-
lizing fission track analysis for thermochronology
involves modelling the information in both the apparent
 FISSION TRACK DATING AND THERMOCHRONOLOGY
15
Mean Track Length ( m)
6% Cl
10
5 0.8% Cl
0.4% Cl
0% Cl
0 from common crystalline rocks. In this later phase, fission
0 5 10 15 20
Annealing Time (hrs)
Fission Track Dating and Thermochronology, Figure 6 The
effect of chemical composition on experimental fission track
annealing in apatites. The diagram shows the results of an
isothermal annealing experiment where four apatites with
different chlorine concentrations were annealed at the same
temperature, 330
C, for different times from 1 to 24 h. The
higher the chlorine concentration of the apatite, the more
resistant are the fission tracks to thermal annealing (Data from
PF Green).
ages and track lengths to reconstruct the thermal history,
based on a quantitative understanding of the annealing
kinetics from laboratory experiments (see section “Exper-
imental Fission Track Annealing Studies”).
Based on these annealing models, it is possible to for-
ward model the distribution of fission track lengths and
ages, whereby the two parameters are predicted for any
given temperature-time path, and manually compared to
the measured data. While this approach is useful to guide
further interpretation, it is also relatively inefficient. In
inverse modelling, a realistic thermal history is
reconstructed from a given set of observed fission track
measurements of age and length. Such inverse modelling
is achieved by conducting numerous, random thermal his-
tory trials and testing the simulated forward models for
their closeness of fit to the observed fission track parame-
ters. Eventually after a very large number of trials,
a number of good-fitting thermal histories emerge that
match the observational data within defined statistical
limits. However, it is important to note that such model
solutions may not be unique, as different thermal histories
can result in similar age and track length distributions.
Thermal histories can be further constrained by adding
independent geological information (e.g., geochronologi-
cal data, depositional age, etc.) or by modelling fission
track data in conjunction with other thermochronometers
such as (U/Th)–He or vitrinite reflectance. Efforts are also
being undertaken to allow more sophisticated modelling
of samples within their geological context, for example,
by modelling a series of neighboring samples simulta-
neously or by integrating thermal history modelling with
structural or geological models. The strategies for inverse
295
modelling of fission track data for reconstructing thermal
histories are discussed in detail by Gallagher (1995,
2012) and Ketcham (2005).
Summary and conclusions
Fission track dating has been applied to a wide diversity of
geological and occasionally archaeological problems,
with an early emphasis on dating such exotic materials
as meteorite impact glasses and volcanic glasses associ-
ated with important hominin fossil sites. These early stud-
ies have largely been replaced by a much larger body of
work on accessory minerals, such as apatite and zircon,
25 track analysis has developed as one of the core methodol-
ogies of thermochronology, and applications to strati-
graphic dating have largely given way to other techniques.
The progressive development of a quantitative
understanding of fission track annealing under
elevated temperatures has led to the major application of
fission track dating today in thermochronology, particu-
larly centered on apatite, which is by far the best
understood and most commonly used mineral system.
Fission track thermochronology has been widely
applied to studies of tectonic and exhumation processes
in both convergent and extensional settings (see “Lumi-
nescence, Earthquake and Tectonic Activity”; “Exhuma-
tion (Thermochronology)”), fault movements (see “Fault
Zone (Thermochronology)”), sedimentary basin analysis,
provenance studies, and landscape evolution (see
“Thermochronology, Landform Evolution”). Various
applications of fission track dating have been reviewed
in Wagner and Van den haute (1992), Gallagher
et al. (1998), Gleadow et al. (2002), and Reiners and
Ehlers (2005).
Rapid advancements are still taking place in fission
track techniques, and in particular, the emergence of new
methods involving laser ablation ICP mass spectrometry
and automated image analysis are likely to lead to
a further transformation of fission track thermochronology
in the coming years.
Bibliography
Barbarand, J., Carter, A., and Hurford, A. J., 2003. Compositional
and structural control of fission-track annealing in apatite. Chem-
ical Geology, 198, 107–137.
Carlson, W. D., Donelick, R. A., and Ketcham, R. A., 1999. Vari-
ability of apatite fission track annealing kinetics:
I. Experimental results. American Mineralogist, 84, 1212–1223.
Donelick, R. A., and Miller, D. S., 1991. Enhanced TINT fission track
densities in low spontaneous track density apatites using 252Cf-
derived fission fragment tracks: a model and experimental observa-
tions. Nuclear Tracks and Radiation Measurements, 18, 301–307.
Donelick, R. A., O’Sullivan, P. B., and Ketcham, R. A., 2005. Apa-
tite fission-track analysis. Reviews in Mineralogy and Geochem-
istry, 58, 49–94.
Fleischer, R. L., and Price, P. B., 1964. Techniques for geological
dating of minerals by chemical etching of fission fragment
tracks. Geochimica et Cosmochimica Acta, 28, 1705–1712.
296 FISSION TRACK DATING AND THERMOCHRONOLOGY
Fleischer, R. L., Price, P. B., and Walker, R. M., 1965a. Tracks of
charged particles in solids. Science, 149, 383–393.
Fleischer, R. L., Price, P. B., and Walker, R. M., 1965b. Effects of
temperature, pressure, and ionization on the formation and sta-
bility of fission tracks in minerals and glasses. Journal of Geo-
physical Research, 70, 1497–1502.
Fleischer, R. L., Price, P. B., and Walker, R. M., 1975. Nuclear
Tracks in Solids. Berkeley: University of California Press.
Galbraith, R. F., 2005. Statistics for Fission Track Analysis. Boca
Raton: Chapman & Hall.
Gallagher, K., 1995. Evolving thermal histories from apatite fission-
track data. Earth and Planetary Science Letters, 136, 421–435.
Gallagher, K., 2012. Transdimensional inverse thermal history
modeling for quantitative thermochronology. Journal of Geo-
physical Research, 117, B02408.
Gallagher, K., Brown, R. W., and Johnson, C., 1998. Fission track
analysis and its applications to geological problems. Annual
Reviews of Earth and Planetary Sciences, 26, 519–572.
Gleadow, A. J. W., 1981. Fission track dating methods: what are the
real alternatives? Nuclear Tracks, 5, 3–14.
Gleadow, A. J. W., and Duddy, I. R., 1981. A natural long-term track
annealing experiment for apatite. Nuclear Tracks, 5, 169–174.
Gleadow, A. J. W., Duddy, I. R., Green, P. F., and Lovering, J. F.,
1986. Confined fission track lengths in apatite – a diagnostic tool
for thermal history analysis. Contributions to Mineralogy and
Petrology, 94, 405–415.
Gleadow, A. J. W., Belton, D. X., Kohn, B. P., and Brown, R. W.,
2002. Fission track dating of phosphate minerals and the
thermochronology of apatite. Reviews in Mineralogy and Geo-
chemistry, 48, 579–630.
Gleadow, A. J. W., Gleadow, S. J., Belton, D. X., Kohn, B. P., and
Krochmal, M. S., 2009. Coincidence mapping a key strategy
for automated counting in fission track dating. In Ventura, B.,
Lisker, F., and Glasmacher, U. A. (eds.), Thermochronological
Methods: From Palaeotemperature Constraints to Landscape
Evolution Models. Geological Society of London Special Publi-
cation, Vol. 324, pp. 25–36.
Green, P. F., Duddy, I. R., Gleadow, A. J. W., Tingate, P. R., and
Laslett, G. M., 1985. Fission track annealing in apatite: track
length measurements and the form of the Arrhenius plot.
Nuclear Tracks, 10, 323–328.
Green, P. F., Duddy, I. R., Gleadow, A. J. W., Tingate, P. R.,
and Laslett, G. M., 1986. Thermal annealing of fission tracks in
apatite 1. A qualitative description. Chemical Geology, 59,
237–253.
Green, P. F., Duddy, I. R., and Hegarty, K. A., 2005. Comment on
“Compositional and structural control of fission track annealing
in apatite”. Chemical Geology, 214, 351–358.
Hasebe, N., Barberand, J., Jarvis, K., Carter, A., and Hurford, A. J.,
2004. Apatite fission-track chronometry using laser ablation
ICP-MS. Chemical Geology, 207, 135–145.
Hurford, A. J., and Green, P. F., 1982. A user’s guide to fission track dat-
ing calibration. Earth and Planetary Science Letters, 59, 343–354.
Hurford, A. J., and Green, P. F., 1983. The zeta age calibration of fis-
sion track dating. Chemical Geology, 1, 285–317.
Ketcham, R. A., 2005. Forward and inverse modeling of
low-temperature thermochronometry data. Reviews in Mineral-
ogy and Geochemistry, 58, 275–314.
Lal, D., Rajan, R. S., and Tamhane, A. S., 1969. Chemical compo-
sition of nuclei of Z > 22 in cosmic rays using meteoritic min-
erals as detectors. Nature, 221, 33–37.
Laslett, G. M., Kendall, W. S., Gleadow, A. J. W., and Duddy, I. R.,
1982. Bias in measurement of fission track length distributions.
Nuclear Tracks, 6, 79–85.
Laslett, G. M., Green, P. F., Duddy, I. R., and Gleadow, A. J. W.,
1987. Thermal annealing of fission tracks in apatite: 2 –
a quantitative analysis. Chemical Geology, 65, 1–13.
Li, W., Lang, M., Gleadow, A. J. W., Zdorovets, M. V., and Ewing,
R. C., 2012. Thermal annealing of unetched fission tracks in apa-
tite. Earth and Planetary Science Letters, 321–322, 121–127.
Naeser, C. W., 1967. The use of apatite and sphene for fission track
age determinations. Geological Society of America Bulletin, 78,
1523–1526.
Price, P. B., and Walker, R. M., 1962. Observations of charged parti-
cle tracks in solids. Journal of Applied Physics, 33, 3400–3406.
Price, P. B., and Walker, R. M., 1963. Fossil tracks of charged par-
ticles in mica and the age of minerals. Journal of Geophysical
Research, 68, 4847–4862.
Reiners, P. W., and Ehlers, T. A. (eds.), 2005. Low-temperature
thermochronology. Reviews in Mineralogy and Geochemistry, 58.
Schmidt, J. S., LeLarge, M. L. M. V., Conceircao, R. V., and
Balzaretti, N. M., 2014. Experimental evidence regarding the
pressure dependence of fission track annealing in apatite. Earth
and Planetary Science Letters, 390, 1–7.
Silk, E. C. H., and Barnes, R. S., 1959. Examination of fission frag-
ment tracks with a transmission electron microscope. Philosoph-
ical Magazine, 4, 970–972.
Storzer, D., and Wagner, G. A., 1969. Correction of thermally
lowered fission track ages of tektites. Earth and Planetary Sci-
ence Letters, 5, 463–468.
Tagami, T., and O’Sullivan, P. B., 2005. Fundamentals of fission
track thermochronology. Reviews in Mineralogy and Geochem-
istry, 58, 19–47.
Tagami, T., Galbraith, R. F., Yamada, R., and Laslett, G. M., 1998.
Revised annealing kinetics of fission tracks in zircon and geolog-
ical applications. In Van den Haute, P., and De Corte, F. (eds.),
Advances in Fission-Track Thermochronology. Dordrecht:
Kluwer Academic, pp. 99–112.
Wagner, G. A., 1968. Fission track dating of apatites. Earth and
Planetary Science Letters, 4, 411–415.
Wagner, G., and Van den haute, P., 1992. Fission Track Dating.
Dordrecht: Kluwer Academic.
Cross-references
Apatite
Ar–Ar and K–Ar Dating
Exhumation (Thermochronology)
Fault Zone (Thermochronology)
Hominid Evolution Timescale
Impact Glass (Fission Tracks)
Laser Ablation Inductively Coupled Mass Spectrometer (LA ICP-
MS)
Luminescence, Earthquake and Tectonic Activity
Radiation Defect
Thermochronology, Detrital Zircon
Thermochronology, Landform Evolution
U-Series Dating
Uranium–Lead Dating
U–Th/He Dating
Volcanic Glass (Fission Track)
Zircon
 G
GENE SEQUENCING
Jessica A. Thomas
Department of Biology, University of York, York, UK
Definition
Gene sequencing is the process of determining the order of
nucleotides (adenine, guanine, cytosine, and thymine) in
a strand of DNA. The resulting DNA sequences can be
analyzed using phylogenetic methods, molecular clocks,
and other techniques.
In living cells, DNA is copied by the enzyme DNA
polymerase. The two strands of the DNA double helix,
which are complementary to each other, are separated
and used as templates. New DNA strands are synthesized
by the sequential addition of nucleotides complementary
to each strand.
The sequence of nucleotides in DNA can provide use-
ful information for various biological analyses. Over the
past few decades, the traditional method of determining
DNA sequences in the laboratory has been the Sanger
chain-termination method. This was developed by Freder-
ick Sanger and colleagues in the late 1970s (Sanger et al.,
1977). The method uses modified versions of the four
nucleotides that terminate the extension of the new DNA
strand and that are labelled with a nucleotide-specific fluo-
rescent marker. Using a mixture of modified and regular
nucleotides, DNA synthesis proceeds until a modified
base is incorporated into the new DNA strand by chance.
Each strand will terminate at a different nucleotide along
the gene sequence, depending on the progress of DNA
synthesis. At the end of the reaction, fragments are
separated according to length and a laser determines the
fluorescence of each fragment. This produces a series of
fluorescence peaks, allowing the nucleotide sequence to
be determined. Sequences of up to 900 nucleotides can
J.W. Rink, J.W. Thompson (eds.), Encyclopedia of Scientific Dating Methods, DOI 10.1007/978-94-007-6304-3,
© Springer Science+Business Media Dordrecht 2015
be produced in a single Sanger-sequencing reaction.
To determine longer sequences of nucleotides, multiple
reactions are required.
DNA sequencing has been transformed by the develop-
ment of next-generation sequencing technologies, which
parallelize the sequencing process on a massive scale.
These methods allow entire genomes to be sequenced in
a time- and cost-effective manner. In “amplified single
molecule” sequencing – also termed “second-generation”
sequencing – DNA is fragmented into short molecules,
which are amplified and then sequenced in massively
paralleled reactions. Several biotechnology companies
have developed second-generation sequencing platforms,
each using technologies for DNA amplification and
sequencing (Margulies et al., 2005; Shendure et al., 2005;
Bentley et al., 2008) (see also Mardis, 2008; Quail et al.,
2012). An ideal platform would maximize the number of
sequencing reactions per run, be able to sequence long
stretches of DNA, and have a very low error rate.
More recently, “true single molecule” (third-
generation) sequencing methodologies have become via-
ble (Clarke et al., 2009; Eid et al., 2009; Thompson and
Steinmann, 2010). These methods can detect fluorescence
at the level of individual sequencing reactions, enabling
DNA molecules to be sequenced directly without the need
for amplification. These reactions are able to produce
DNA sequences of several thousand nucleotides, in
comparison to the much shorter fragments produced by
second-generation sequencing.
Bibliography
Bentley, D. R., Balasubramanian, S., Swerdlow, H. P., Smith, G. P.,
Milton, J., et al., 2008. Accurate whole human genome sequenc-
ing using reversible terminator chemistry. Nature, 456(7218),
53–59. http://www.ncbi.nlm.nih.gov/pubmed/18987734.
Clarke, J., Wu, H.-C., Jayasinghe, L., Patel, A., Reid, S., and
Bayley, H., 2009. Continuous base identification for
298 GEOMAGNETISM
single-molecule nanopore DNA sequencing. Nature Nanotech-
nology, 4, 265–270, doi: 10.1038/nnano.2009.12.
Eid, J., Fehr, A., Gray, J., Luong, K., Lyle, J., et al., 2009. Real-time
DNA sequencing from single polymerase molecules. Science,
323(5910), 133–138, doi: 10.1126/science.1162986.
Mardis, E. R., 2008. The impact of next-generation sequencing
technology on genetics. Trends in Genetics, 24(3), 133–141,
doi: 10.1016/j.tig.2007.12.007.
Margulies, M., Egholm, M., Altman, W. E., Attiya, S., Bader, J. S.,
et al., 2005. Genome sequencing in microfabricated high-density
picolitre reactors. Nature, 437, 376–380, doi: 10.1038/
nature03959.
Quail, M. A., Smith, M., Coupland, P., Otto, T. D., Harris, S. R.,
Connor, T. R., Bertoni, A., Swerdlow, H. P., and Gu, Y., 2012.
A tale of three next generation sequencing platforms: compari-
son of Ion Torrent, Pacific Biosciences and Illumina MiSeq
sequencers. BMC Genomics, 13, 341. http://www.
biomedcentral.com/1471-2164/13/341.
Sanger, F., Nicklen, S., and Coulson, A. R., 1977. DNA sequencing
with chain-terminating inhibitors. Proceedings of the National
Academy of Sciences of the United States of America, 74,
5463–5467.
Shendure, J., Porreca, G. J., Reppas, N. B., Lin, X., McCutcheon,
J. P., et al., 2005. Accurate multiplex polony sequencing of an
evolved bacterial genome. Science, 309, 1728–1732, doi:
10.1126/science.1117389.
Thompson, J. F., Steinmann, K. E., 2010. Single molecule sequenc-
ing with a HeliScope genetic analysis system. Current Protocols
in Molecular Biology, Chapter 7, Unit 7.10.
GEOMAGNETISM
Vadim A. Kravchinsky
University of Alberta, Edmonton, AB, Canada
Synonyms
Earth’s magnetic field
Definition
Geomagnetism is a domain of geophysics which studies
the origin and nature of the Earth’s magnetic field. Geo-
magnetism involves integrated approaches to decipher
ancient and contemporary magnetic fields of the Earth
and other planets of the solar system.
Structure and properties of the earth’s magnetic
field
The Earth’s magnetic field or geomagnetic field surrounds
the planet where 98 % of the field is generated by the
geodynamo, creating the magnetosphere (Merrill et al.,
1996). Geodynamo is the compositional and thermal con-
vection in the outer liquid core of the conductive material
(mainly iron). Convective electrical current produces the
geomagnetic field. The main part of the field is produced
in the Earth’s core and is described near the Earth’s surface
as a magnetic field of a dipole (Jones, 1977; McFadden and
Merrill, 1984). The dipole has its magnetic South Pole ori-
ented toward the geographical North Pole (see Figure 1).
The field structure is deformed by the solar wind above
the Earth’s atmosphere (1–2 % of a total measured field
at the Earth surface) (Courtillot and Le Mouel, 1988). In
reality, there is a deviation between the best-fit dipolar mag-
netic field on the Earth and the actual field. In the mantle,
the shape of the magnetic field is changed to a quadrupole
or multipole system. Additional disruption of the field
(1–2 %) comes from the magnetized crustal rocks
(Merrill, 2010).
Presently, the horizontal component of the field is
directed roughly into the geographic north–south direc-
tion. The local deviation from this alignment is called dec-
lination (see Figure 2). In the middle and high latitudes,
the magnetic flux lines change from the pointing horizon-
tally vector to a much stronger more vertical vector
(see Figure 1). The inclination angle of the field lines can
be measured with a compass needle, whose axis of rotation
is mounted horizontally (see Figure 2). At the magnetic
North and South Poles, the inclination is 90
, and at the
magnetic equator, it is 0
. Currently, the axis of the geomag-
netic dipole field deviates at about 11.5
relative to the
Earth’s axis of rotation. The present-day dipole moment of
the geomagnetic field is M ¼ 7.746  1022 Am2 and can
be calculated from the most recent International Geomag-
netic Reference Field of 2010 (IGRF-11) (International
Association of Geomagnetism and Aeronomy, Working
Group V-MOD, 2010). It changes constantly with the
current annual decrease of 0.006  1022 Am2/year.
Paleomagnetism
Paleomagnetism is the study of the past geomagnetic field
recorded in rocks and sediments. Since magnetic minerals
are abundant in the Earth’s crust, paleomagnetic studies
can be applied to a wide range of rocks and sediments.
When a rock, containing ferromagnetic minerals, is heated
to temperatures above the Curie point and then cooled in
the ambient geomagnetic field, it is magnetized in the
direction of the existing geomagnetic field (McElhinny
and McFadden, 1999; Tauxe, 2010). This applies to volca-
nic rocks, bricks, pottery, and terracotta. During sediment
formation, the magnetic grains align themselves with the
local magnetic field until they become locked into place.
The acquired magnetization is called a remnant magneti-
zation and can be described by its inclination, declination,
and intensity or by the related values of paleolatitude and
paleolongitude which define a virtual geomagnetic pole
(VGP). When a sequence of VGPs (paleomagnetic poles)
has reliable ages, they can be used to produce an apparent
polar wander path (APWP). This path is constructed using
different poles of known age for a stable continental block
and represents the APWP through time for that continent
(Tauxe, 2010).
Geomagnetic reversals, excursions, and
secular variations
The direction of the ancient magnetic field has been con-
tinuously registered in the mid-ocean ridges when the
 GEOMAGNETISM 299

Geomagnetism, Figure 1 The geomagnetic dipole position in the center of the Earth with its axis tilted relative to the Earth’s
axis. Magnetic field lines are shown with horizontal magnetic compass needle pointing along the lines. The needle indicates
magnetic inclination, the angle measured from the horizontal surface at the point of observation. Positive values of inclination
indicate that the magnetic field of the Earth is pointing downward, into the Earth, and negative values indicate that it is
pointing upward.

erupted magma cools down. It enables one to study the
geomagnetic polarity reversals, i.e., when the Earth dipo-
lar field reverses 180
relatively to the present-day orien-
tation. Such reversals are common throughout Earth’s
entire geological history. The most recent polarity reversal
occurred at 780,000 years ago (Brunhes-Matuyama rever-
sal) (Langereis et al., 2010). Recent study suggests that
an actual pole shift takes around 1,000 years (Valet
et al., 2012). Reversals are very well documented until
about 160 million years ago from the marine magnetic
anomalies (Cande and Kent, 1995; Gee and Kent, 2007)
and are less reliably known for older geological times.
A geomagnetic excursion is a significant change in the
prevailing geometry of the geomagnetic field, which lasts
on the order of the first thousands of years. The geomag-
netic field, however, does not flip to a stable position at
180
during excursion. The most common contemporary
conception states that the excursion occurs due to local-
ized disturbances in the geomagnetic field caused by the
convection of electrical currents in the outer core and its
interaction with the mantle (Olson et al., 2013).
Geomagnetic secular variations are the changes in
the field over periods of a year or more; they reflect
shorter-term changes in the Earth’s outer core. The variations
that predate historical observations and recorded in archaeo-
logical and geological materials are called paleomagnetic
secular variations or paleosecular variations (PSV).
Paleomagnetic dating
For paleomagnetic dating, the APWP is used to determine
the age of a paleomagnetic pole obtained from rocks or
sediments of undefined age by projecting the pole on
a sphere to the nearest point on the APWP (Symons and
Sangster, 1991; Blanco et al., 2013) (see Figure 3).
A similar idea arises with PSV. The secular variation
reference curves are created to display inclination, decli-
nation, and the intensity of the magnetic field for a given
region through time. The PSV curves are typically defined
for only the last few thousand years, which is useful when
dating young material such as archaeological artifacts
(Hangstrum and Blinman, 2010; Pavon-Carrasco et al.,
2011; Tanguy et al., 2003). Recently there have been great
strides made in creating well-defined techniques and use-
ful computer software to date archaeological artifacts
using inclination, declinations, and intensities (Le Goff
et al., 2002; Pavon-Carrasco et al., 2011).
300
north
eti
c
gn n 30 N
ma ia
rid
me
X H
D
I east
Y
B
Z
vertical
Geomagnetism, Figure 2 Illustration of the coordinate systems
used for representing the Earth’s magnetic field. The
geomagnetic field can be described by north (X), east (Y), and
vertically downward (Z) Cartesian components and by the
angles of declination (D) and inclination (I) together with the
total field intensity (B).
Magnetostratigraphy
Polarity change is a global phenomenon which occurs
simultaneously all over the world. Magnetostratigraphy
studies polarity change registered in the geological sec-
tions, drilling cores, and magnetic anomalies in the ocean.
The polarity change pattern can be used for stratigraphic
correlation and dating of the geological section separated
from meters to thousands of kilometers. Creating
a universal reference scale of polarity changes using bio-
stratigraphic and radiometric ages produces the geomag-
netic polarity time scale (GPTS). This scale provides
valuable stratigraphic markers for short- and long-range
correlations of geological sections (Langereis et al.,
2010). Oceanic and marine magnetic anomalies are very
important for determining the spreading velocities
between two plates on either side of a mid-ocean ridge.
Paleointensity
The value of ancient intensity of the geomagnetic field is
called geomagnetic paleointensity. Independent records
of paleointensity from sediment cores in different areas
of the world are stacked together and represent the refer-
ence curve of the variations of the geomagnetic dipole
moment (Valet et al., 2005; Channell et al., 2009). This
provides information regarding the processes governing
the geodynamo. So far, this procedure has been limited
to the last two million years. The reference curve from
the sedimentary records is calibrated with the absolute
dipole moments derived from volcanic lavas. Intervals of
low intensity often correspond to the geomagnetic
GEOMAGNETISM
150°E
°
90°E 120 E°
intersection1
APWP
paleomagnetic
pole minimum
distance
mid point
0°
intersection2
30°S
Geomagnetism, Figure 3 Illustration of the paleomagnetic age
determination (Modified from Blanco et al. (2013) with Elsevier
permission). The pole and the circle of confidence are projected
over the minimum distance point on the APWP. The intersection
points 1 and 2 will constrain the error in age.
excursions and reversals. Comparison of the continued
paleointensity records for sediments with unknown age
to the reference paleointensity curve results in a possibility
to date sediments when other means of dating are not reli-
able (e.g., Lake Baikal sedimentary cores, Kravchinsky
et al., 2007).
Archeomagnetism
Archeomagnetism is based on the variation of the intensity
and direction of the geomagnetic field, and on the property
of baked clay to record the situation during firing, as it
remains fixed in the arrangement of the particles of iron
oxides in it, immobilized by heating temperatures of about
650–700
.
The direction of the Earth’s magnetic field does not
vary in a consistent manner over time or space, and the
changes have been recorded by scientists from only the
sixteenth century. It is therefore necessary for each region
to determine what were the geomagnetic changes through
the examination of samples which remained in the same
position from the time of heating. In the field of archaeol-
ogy, once the reference curves are established, this can be
used for dating other artifacts (Tarling, 1975). The accu-
racy of dating can vary depending on the rapidity of
changes in the Earth’s magnetic field at different times,
and in general, it is possible to obtain ages with accuracy
of decades or quarter century. While other forms of dating
 GLACIAL LANDSCAPE (COSMOGENIC NUCLIDE)
are based at the decay of radioactive elements, magnetic
dating allows archaeologists to figure out the age of more
varied and older artifacts.
Summary
The main geomagnetic field is formed in the outer core of
the Earth and can be measured on the surface or in space.
In the first approximation, the geomagnetic field of the
Earth can be represented by the field of a dipolar magnet
placed in the center of the planet. Geomagnetic poles
migrate at the Earth’s surface continuously, and reversals
of the geomagnetic field direction occurred many times
in the Earth’s geological history. Short-term and long-term
changes of the geomagnetic field are constantly monitored
and studied by geophysicists because they carry important
information about the current state of the geomagnetic
field and past magnetic events.
The magnetic field has an effect on a variety of phe-
nomena on Earth, in the atmosphere, and closest space.
Magnetic measurements are commonly used for chronol-
ogy and are important in geophysical exploration, for
navigation, aerospace, and geodesy.
Bibliography
Blanco, D., Kravchinsky, V. A., Konstantinov, K. M., and Kabin,
K., 2013. Paleomagnetic dating of Phanerozoic kimberlites in
Siberia. Journal of Applied Geophysics, 88, 139–153.
Cande, S. C., and Kent, D. V., 1995. Revised calibration of the geo-
magnetic polarity timescale for the Late Cretaceous and
Cenozoic. Journal of Geophysical Research: Solid Earth,
100, 6093–6095.
Channell, J. E. T., Xuan, C., and Hodell, D. A., 2009. Stacking
paleointensity and oxygen isotope data for the last 1.5 Myr
(PISO-1500). Earth and Planetary Science Letters, 283(1),
14–23.
Courtillot, V., and Le Mouel, J. L., 1988. Time variations of the
Earth’s magnetic field: from daily to secular. Annual Review of
Earth and Planetary Sciences, 16, 389–476.
Gee, J. S., and Kent, D. V., 2007. Source of oceanic magnetic anom-
alies and the geomagnetic polarity time scale. In Kone, M. (ed.),
Treatise on Geophysics. Amsterdam: Elsevier. Geomagnetism,
Vol. 5, pp. 455–507.
Hagstrum, J. T., and Blinman, E., 2010. Archeomagnetic dating in
western North America: an updated reference curve based on
paleomagnetic and archeomagnetic data sets. Geochemistry
Geophysics Geosystems, 11: doi: 10.1029/2009GC002979.
International Association of Geomagnetism and Aeronomy, Work-
ing Group V-MOD, 2010. International geomagnetic reference
field: the eleventh generation. Geophysical Journal Interna-
tional, 183, 1216–1230.
Jones, G. M., 1977. Thermal interaction of the core and the mantle
and long-term behavior of the geomagnetic field. Journal of
Geophysical Research, 82, 1703–1709.
Kravchinsky, V. A., Evans, M. E., Peck, J. A., Sakai, H., Krainov,
M. A., King, J. W., and Kuzmin, M. I., 2007. A 640 kyr geomag-
netic and paleoclimatic record from Lake Baikal sediments.
Geophysical Journal International, 101, 101–116.
Langereis, C. G., Krijgsman, W., Muttoni, G., and Menning, M.,
2010. Magnetostratigraphy concepts, definitions, and applica-
tions. Newsletters on Stratigraphy, 43(3), 207–233.
Le Goff, M., Gallet, Y., Genevey, A., and Warme, N., 2002. On
archeomagnetic secular variation curves and archeomagnetic
301
dating. Physics of the Earth and Planetary Interiors,
134, 203–211.
McElhinny, M. W., and McFadden, P. L., 1999. Paleomagnetism:
Continents and Oceans. San Diego: Academic Press.
McFadden, P. L., and Merrill, R. T., 1984. Lower mantle convection
and geomagnetism. Journal of Geophysical Research,
89, 3354–3362.
Merrill, R. T., McElhinny, M. W., and McFadden, P. L., 1996. The
magnetic field of the earth: paleomagnetism, the core and the
deep mantle, international geophysics series. Academic Press,
San Diego, Vol. 63.
Merrill, R. T., 2010. Our Magnetic Earth: The Science of Geomag-
netism. Chicago: The University of Chicago Press.
Olson, P., Deguen, R., Hinnov, L. A., and Zhong, S., 2013. Controls
on geomagnetic reversals and core evolution by mantle convec-
tion in the Phanerozoic. Physics of the Earth and Planetary
Interiors, 214, 87–103.
Pavon-Carrasco, F. J., Rodriguez-Gonzalez, J., Osete, M. J., and
Torta, J. M., 2011. A Matlab tool for archaeomagnetic dating.
Journal of Archaeological Science, 38, 408–419.
Symons, D. T. A., and Sangster, D. F., 1991. Late Devonian paleo-
magnetic age for the Polaris Mississippi Valley-type Zn-Pb
deposit, Canadian Arctic Archipelago. Canadian Journal of
Earth Sciences, 29, 15–25.
Tanguy, J. C., Le Goff, M., Principe, C., Arrighi, S., Chillemi, V.,
LaDelfa, S., and Patane, G., 2003. Archeomagnetic dating of
Mediterranean volcanics of the last 2100 years: validity and
limits. Earth and Planetary Science Letters, 211, 111–124.
Tarling, D. H., 1975. Archaeomagnetism: the dating of archaeolog-
ical materials by their magnetic properties. World Archaeology,
7(2), 185–197.
Tauxe, L., 2010. Essentials of Paleomagnetism. Berkeley: Univer-
sity of California Press.
Valet, J. P., Meynadier, L., and Guyodo, Y., 2005. Geomagnetic
dipole strength and reversal rate over the past two million years.
Nature, 435(7043), 802-805.
Cross-references
Archaeomagnetic Dating
Magnetic Anomalies
Magnetic Chronology
Magnetostratigraphic Dating
Sediments, Terrestrial (Paleomagnetism)
Volcanic Rocks (Ar/Ar)
GLACIAL LANDSCAPE (COSMOGENIC NUCLIDE)
Joerg M. Schaefer
Lamont-Doherty Earth Observatory, Columbia
University, Palisades, NY, USA
Introduction
Cosmogenic nuclides have become powerful tools to
explore glacial landscapes toward reconstructing
cryospheric change (Balco, 2011) on a wide spectrum of
time scales and are currently revolutionizing the fields of
glacial geology, terrestrial climate, cryosphere science,
and earth surface process geomorphology. Whenever gla-
ciers and ice sheets fluctuate, the mighty ice alters the earth
surface, carves previously unexposed surfaces, and deposits
302 GLACIAL LANDSCAPE (COSMOGENIC NUCLIDE)
glacial sediments, including moraine ridges. Exposure of
such newly formed glacial surfaces to the cosmic radiation
at the moment of ice retreat starts the cosmogenic nuclide
clock (referred to as “Terrestrial Cosmogenic Nuclide Dat-
ing”; See also “Glacial Landscape (Cosmogenic Nuclide)”
overview entry in the same issue and the following seminal
review articles: Cerling and Craig (1994), Gosse and Phil-
lips (2001), Granger et al. (2013), Lal (1991)). About
20 years ago, pioneering studies first applied cosmogenic
nuclide techniques to date moraines, glacially formed
ridges of rock and soil that mark former glacier positions
(e.g., Brook et al., 1993; Gosse et al., 1995; Ivy-Ochs
et al., 1999). Since then, originally paced by progress by
John Stone (University of Washington, Seattle), the sensi-
tivity of cosmogenic nuclide techniques has been transfor-
mationally refined (e.g., Stone et al., 2003). Also, new
cosmogenic nuclide tools, further widening the application
spectrum, have been introduced, most recently the “in situ
14
C” technique, developed and implemented by Nat Lifton
(Lifton et al., 2001). Time scales to cosmogenically clock
land ice change now span from many million years
(Bruno et al., 1997; Ivy-Ochs et al., 1997; Schäfer
et al., 1999; Margerison et al., 2005) to 100 years ago and
less (Kelly et al., 2008; Licciardi et al., 2009; Schaefer
et al., 2009; Schimmelpfennig et al., 2014; Stroup
et al., 2014). Novel cosmogenic nuclide applications
include dating of the first maximum of the Laurentide Ice
Sheet (Balco et al., 2005) using buried tills in the central
USA and constraining past ice sheet fluctuations in Antarc-
tica by cosmogenic nuclide dating of “nunatak dipsticks,”
ice-free rock formations towering above the ice sheet
(Balco et al., 2013; Johnson et al., 2014). Other recent stud-
ies pioneer the application of cosmogenic nuclide tech-
niques to explore pro- and subglacial bedrock as
environmental archive (Goehring et al., 2013, 2011). All
these successes are the product of an integrated science
approach, crosscutting the disciplines of Quaternary Geol-
ogy (geomorphic mapping), Glaciology (glacier dynam-
ics), Geochemistry (cosmogenic nuclide analysis), and
Climate Science (synthesis). This short entry attempts to
illustrate the breathtaking dynamic of progress in cosmo-
genic nuclide science by presenting application examples
toward forefront questions in climate and glacier science.
Here, we primarily focus on the cosmogenic nuclide 10Be.
Other cosmogenic nuclide techniques, in particular 3He
(Kurz, 1986; Goehring et al., 2010; Blard et al., 2013b),
36
Cl (Schimmelpfennig, 2009; Schimmelpfennig et al.,
2011), and in situ 14C (Lifton et al., 2001; Hippe et al.,
2013, 2014; Goehring et al., 2014), are rapidly progressing,
but covering all cosmogenic nuclides would be beyond
scope and space of this entry.
Methods
Cosmogenic nuclide production rates
Here, we review recent progress in our quantitative
understanding of cosmogenic nuclide production rates.
For a detailed methodological overview, see the accom-
panying entry “Glacial Landscape (Cosmogenic
Nuclide)” (REF) and references therein. Cosmogenic
nuclide surface exposure dating is based on the measure-
ment of minute amounts of cosmogenic nuclides, here
10
Be, produced in near-surface rocks as a function of
time, after glacier retreat has exposed the “new” surface
to cosmogenic radiation (time ¼ zero of the cosmogenic
clock). By careful sample selection in the field, we min-
imize secondary influences, compromising this
approach, such as inheritance of cosmogenic nuclides
in the sampled surface from prior periods of exposure,
temporary shielding of the sampled surface by snow or
sediment cover, or considerably erosion of the surface.
These potentially compromising influences can also be
evaluated by measuring many samples of the same gla-
cial feature and thus the same expected cosmogenic
nuclide inventory (see below).
To calculate a “surface exposure age” from the
measured 10Be concentration in the surface sample, we
need to know the 10Be production rate at the sample loca-
tion integrated over the full period of exposure. The
uncertainty of this production rate value linearly propa-
gates into the overall age error. Because incoming cos-
mic radiation is shielded by the Earth’s atmosphere and
the geomagnetic field, production rates of cosmogenic
nuclides at the earth’s surface are a function of altitude,
latitude, and time (see accompanying entry “Glacial
Landscape (Cosmogenic Nuclide)”). As an example,
the altitudinal scaling factor from sea level to 2,000
is ~4.8.
The uncertainty in the knowledge of the time-integrated
production rate at the location of the study has remained
the major source of error in cosmogenic nuclide studies.
However, there is good news! Over the last few years, sev-
eral studies have shown the potential to locally calibrate
10
Be production rates (and 3He production rates) with a
precision of better than 3 % (1s) (e.g., Putnam et al.,
2010; Fenton et al., 2011; Kaplan et al., 2011; Young
et al., 2013). In Figure 1, we plot a compilation of recent
local 10Be production rate calibrations that we believe to
be robust, scaled for reference to sea level and high lati-
tude using the same scaling scheme (Lal/Stone time-
dependent “Lm” after Balco et al. (2008)). The overall pic-
ture is one of striking internal consistency, and these
experiments afford for high-precision cosmogenic dating
studies near the production rate calibration points. Simul-
taneously, new modeling approaches (Argento et al.,
2012; Lifton et al., 2014) to scale between the locations
of production rate calibrations improve and converge as
well toward the original scaling factors published by Lal
in 1991 (Lal, 1991). With the current rate of progress of
production rate calibrations and model interpolations in
between, it appears realistic to project a robust, near-
global network of production rate values allowing for pre-
cise and accurate cosmogenic dating studies anywhere on
earth in the near future.
 GLACIAL LANDSCAPE (COSMOGENIC NUCLIDE) 303

Probability

0
3 4 5
Be- 10 Production Rate [atoms/(g yr)]
Glacial Landscape (Cosmogenic Nuclide), Figure 1 Consistency of recent 10Be production rate calibrations. Production rate
calibration experiments from Norway (Oldedalen Landslide (Goehring et al., 2012a) and Grotlandsura Landslide (Fenton et al., 2011)),
Greenland (Young et al., 2013), New England (Balco et al., 2009), tropical Peru (Kelly et al., in press), tropical Bolivia (Blard et al., 2013a),
New Zealand (Putnam et al., 2010), and Patagonia (Kaplan et al., 2011) shown by thin line, the bold curve gives the integrative
probability function. The blue vertical line is the arithmetic mean, and the black vertical line gives the 1s uncertainties, red line the 2s,
and green line the 3s range. Following the convention, all values are scaled to sea level and high latitudes using the time-dependent
“Lm” scaling after (Balco et al., 2008) (see also text). This tight cluster of 10Be production rate calibrations has been produced over the
last 3 years and reflects the considerable progress relative to the earlier “global 10Be production rate” value of 4.4  0.4 atoms/(g year)
(Balco et al., 2008).

Novel applications of cosmogenic nuclides to
glacial landscapes
The first glacial maximum in North America
When did the first ice age peak? When did Earth’s climate
commence to swing between full glacial and interglacial
modes? These long-standing geological questions have
been elegantly addressed recently by a series of studies
(Balco et al., 2005; Balco and Rovey, 2008; Balco and
Rovey, 2010), dating the lowest, and thus oldest, glacial
tills in the central USA by means of cosmogenic 26Al/10Be
burial dating (Granger and Muzikar, 2001; Balco and
Shuster, 2009). The cosmogenic burial dating method
makes use of the difference in half-life of two (or more) cos-
mogenic nuclides in sediments that had have been buried
after a period of exposure and thus cosmogenic nuclide
production. During the period of burial of sediments,
the nuclide with the shorter half-life, here 26Al
(t1/2 ¼ 717,000 years), decays faster than the longer-lived
nuclide, here 10Be (t1/2 ¼ 1,400,000 years). Hence, the
measured 26Al/10Be ratio in buried sediments quantifies
the burial time and, in this case, dates the deposition of
the oldest glacial tills and the first glacial maximum of the
Laurentide Ice Sheet to about 2.42  0.14 million years.
This result represents the most robust date for the first max-
imum of the predominant northern continental ice sheet and
postdates the onset of the Quaternary.
The structure of the last glacial maximum in
New Zealand’s Southern Alps
A recent example of a comprehensive, high-sensitivity
10
Be chronology from boulders protruding from moraines
at Lake Ohau, New Zealand’s Southern Alps, yielding
new information about the structure of the Last Glacial
Maximum (LGM) in southern mid-latitudes (Putnam
et al., 2013b), is shown in Figure 2. It is the internal con-
sistency of 10Be boulder ages within these multi-sample
data sets that best highlights that glacier moraine records
are uniquely suitable for the cosmogenic SED approach
and that secondary, potentially compromising processes
such as snow and sediment cover of the sampled surface,
turning of the sample after deposition by the glacier on
top of the moraine, and finally inheritance of cosmogenic
nuclides in the sample from prior periods of exposure are
not significant in well-selected settings.
This chronology shows an extended mountain glacier
LGM in New Zealand, with culminations at about
32, 26, 22, and 18 ka ago, and illustrates the potential to
reconstruct glacier fluctuations during the last ice age with
millennial resolution. Beyond, the 10Be ages on the green
moraines in Figure 2 indicate that in some cases, the SED
method can provide age information about the glacier cul-
mination at the end of the penultimate glacial cycle.
From the paleoclimatic perspective, the LGM moraine
records from New Zealand add new and robust information
about the onset, fine structure, and termination of the Last
Glacial Maximum in southern mid-latitude. Because New
Zealand’s glaciers respond most sensitively to summer tem-
peratures (Anderson and Mackintosh, 2006; Anderson and
Mackintosh, 2012; Golledge et al., 2012), the glacier fluctu-
ations can be transformed into a temperature record, which
implies that LGM summers in New Zealand were about
6–7
C cooler than today. Beyond, the long LGM in New
Zealand was punctuated by multiple culminations that show
304 GLACIAL LANDSCAPE (COSMOGENIC NUCLIDE)

Glacial Landscape (Cosmogenic Nuclide), Figure 2 10Be boulder chronology for Last Glacial Maximum glacier culminations in New
Zealand’s Southern Alps (Modified after Putnam et al. (2013b). Red ¼ Last Glacial Maximum (LGM) moraines; green ¼ pre-LGM
moraines. Boxes give 10Be ages of boulders on moraines, outliers plotted in italic. Note the high internal consistency of 10Be ages from
boulders from one moraine sequence, together with the detailed stratigraphic order of the moraine ages (inboard moraines are
younger). The few old ages on the green moraine represent the glacier position at the end of the penultimate ice age, indicating that
the SED method can be used over orbital time scales in some cases).

striking similarity to Antarctic ice core temperature records
(Fudge et al., 2013) and sea surface temperature records in
the southern ocean. This multiple culmination structure of
the LGM rules out southern summer insolation as driver
of the LGM in New Zealand, because 44
S December inso-
lation was rising steadily through the New Zealand LGM
period, from its minimum 32,000 years ago until the insola-
tion peak some 22,000 years ago. Finally, the abrupt termi-
nation of the LGM in the moraine record commencing at
about 18,000 years ago is a near match to the atmospheric
CO2 curve. Together with the timing of the culmination dur-
ing the long LGM period, a northern hemisphere driver
connected to atmospheric greenhouse gas concentrations
appears to be the most promising scenario to explain the
southern mid-latitude LGM pattern (Putnam et al., 2013a).
Proglacial bedrock as climate archive: holocene
fluctuations of Rhone Glacier
Recent studies by Goehring et al. (2011, 2012b) introduce
high-sensitivity cosmogenic nuclide techniques as tools to
explore the bedrock exposed only a few years ago by Rhone
Glacier, Swiss Alps, toward reconstructions of Holocene
glacier dynamics. Here, two cosmogenic nuclides produce
in quartz, in situ 14C and 10Be, were used in concert to eval-
uate the periods during which the Rhone Glacier was
smaller or bigger, respectively, than today over the Holo-
cene period (the last 11,500 years). The cosmogenic nuclide
concentrations through any glacier forefield decrease from
margin to central trough, which is consistent with the ero-
sion profile showing higher erosion rates where the glacier
flows faster (center), compared to the slower-flowing
 GLACIAL LANDSCAPE (COSMOGENIC NUCLIDE)
margins. We can assume that this bedrock was free of cos-
mogenic nuclides at the end of the last ice age some
11,500 years ago, after tens of thousands of years of big gla-
ciers and subglacial erosion. In turn, the 10Be ages of the
bedrock samples at the margin of the Rhone Glacier trough
provide minimum estimates of exposure, so the integrated
period of time during which Rhone Glacier was smaller
than today over the last 11,500 years. The dual-nuclide
approach in this case suggests that Rhone Glacier was
smaller than today for about 55% of the Holocene and that
the most likely scenario describes a small Rhone Glacier to
the early and mid-Holocene, followed by a larger Rhone
glacier over the last 4000 years. This cosmogenic nuclide-
based scenario, updated toward higher precision in a second
isochron-based study (Goehring et al., 2013), agrees well
with climate proxy records from the Alps and elsewhere
in the northern hemisphere, making the case for warmer-
than-today climate during the early and mid-Holocene,
followed by climate deterioration and regrowing glaciers
around 4,000 years ago, peaking during the Little Ice Age
(1300–1850 CE).
Rock dipsticks to measure ice sheet fluctuations:
rapid thinning of the Pine Island Glacier, a major
outlet glacier of the West Antarctic Ice Sheet
The West Antarctic Ice Sheet (WAIS) is potentially insta-
ble in our warming world and could contribute up to 3.5 m
of sudden sea level rise, which would be a disastrous envi-
ronmental change. Applying for the first time 10Be tech-
niques to Holocene glacial deposits along an altitude
transect (dipstick approach), Stone et al. (2003) showed
that the WAIS has been steadily thinning during the Holo-
cene. On top of the long-term thinning trend, modern
observation from the Pine Island Glacier (PIG), one of
the major outlet glaciers of the WAIS, shows very rapid
thinning at a rate of about 1 m per year over the recent
years (Park et al., 2013). The questions whether the cur-
rent rapid thinning is unique, how sustainable it might
be, and what actually drives the PIG thinning are funda-
mental to WAIS stability and global sea level predictions.
A new study has applied cosmogenic nuclide tech-
niques to a rock dipstick (Figure 3) above an unnamed out-
let glacier directly linked to the PIG. The rock formation
towering over the ice sheet and the PIG tributary has been
steadily exposed during the ice sheet thinning, and this
signal is traceable by high-sensitivity 10Be techniques.
The data shown in Figure 3 illustrates that a rapid thinning
of the PIG occurred about 8,000 years ago and lowered the
PIG surface over 100 m over about a century. Thus, the
thinning rate of the PIG 8,000 year ago was similar to
the modern thinning rate but was sustained over 100 years.
The most likely explanation for the rapid and sustained
PIG thinning 8,000 years ago is a breakup of the Pine
Island Bay ice shelves (Figure 3, upper panel). Such ice
shelves buttress the outlet glaciers while intact, and vice
versa, if the buttress disappears, the outlet glacier flow
305
accelerates dramatically, thinning the outlet glacier and
draining the ice sheet. This happened 8,000 years ago,
during a climate period that was not even close to the cur-
rent greenhouse gas-induced climate forcing. While the
comparison between the PIG and WAIS dynamics 8,000
years ago and today is complex, this experiment illustrates
the high sensitivity of the PIG and its ice shelves, and thus
the WAIS, to climate forcing. This is a worrisome result,
given that the warming of our planet accelerates and that
transport of warm ocean waters to the Amundsen Sea
Embayment (Figure 3) is one of the observed conse-
quences (Martinson, 2012). By this argument, we must
expect that current thinning of the PIG continues and
might accelerate, which will translate into a rapidly
increasing melt water signal coming from the WAIS.
Summary
Cosmogenic nuclide techniques have revolutionized our
capabilities to exploit glacial landscapes toward reconstruc-
tion of past land ice change and climate variability. Yet, this
might be just the tip of the iceberg. High-sensitivity cosmo-
genic nuclide techniques applied to the moraine record now
afford for precise reconstruction of past glacier fluctuations
with centennial resolution and on a time scale ranging from
decades to beyond 100,000 years. It has been a long-
standing dream of geologists and climate scientists alike
to precisely map land ice change during the last ice age
through the deglaciation period and into the current inter-
glacial, referred to as the Holocene. This dream appears
now in reach. Of course, there is a tremendous amount of
scientific work ahead to realize this dream, but there is no
daunting scientific challenge ahead; we simply need to
intensify the multidisciplinary efforts toward precise and
accurate cosmogenic nuclide studies of glacial landscapes,
which depends critically on more cutting-edge cosmogenic
production rate calibrations at different latitudes and alti-
tudes and also on funding resources! Beyond the moraine
program, it is a central goal to further improve the sensitiv-
ity of other cosmogenic nuclide systems, in particular 3He,
36
Cl, and in situ 14C, and to develop the multi-nuclide
approaches further. Finally, from the proglacial bedrock
studies described above, it appears a natural next step to
use subglacial bedrock as climate archive. Subglacial bed-
rock holds the isotopic answer to fundamental questions
such as: When did the WAIS collapse the last time? When
was Greenland ice-free? The answer to these questions
form the foundation to evaluate whether the recent and
ongoing polar ice sheet response to global warming might
be the dawn of ice sheet collapse and potential abrupt sea
level rise, which would represent a disaster of epic dimen-
sion for our society at a time when most of earth’s popula-
tion lives near coastal margins. From that perspective,
cosmogenic nuclide techniques applied to glacial land-
scapes can contribute fundamental information toward sus-
tainable development of our ever-growing society on a
rapidly warming planet.
306 GLACIAL LANDSCAPE (COSMOGENIC NUCLIDE)

Glacial Landscape (Cosmogenic Nuclide), Figure 3 Rock outcrops as dipsticks of ice sheet thinning (Modified after Johnson et al.,
in press). Upper left panel: Map of the Amundsen Sea Embayment, showing location of the study area. The grounded ice sheet is
shown in dark gray and ice shelves in light gray. The inset shows the location within Antarctica, and the box shows the study area.
Upper right panel: Map of Pine Island Bay, showing flow velocities of Pine Island Glacier and the unnamed outlet glacier flowing
through the Hudson Mountains. Contours are in meters. The grounding line is represented by the solid black line, and the crest of the
sub-ice shelf ridge as a dashed white line. Lower panel: Dipstick rock outcrop towering above the unnamed outlet glacier connected to
Pine Island Glacier, West Antarctic Ice Sheet, with sample locations along an altitude transect and 10Be ages shown.

Bibliography Argento, D. C., et al., 2012. Modeling the earth’s cosmic radiation.
Nuclear Instruments and Methods in Physics Research B,
Anderson, B., and Mackintosh, A., 2006. Temperature change is the
294, 464–469.
major driver of late-glacial and Holocene glacier fluctuations in
Balco, G., 2011. Contributions and unrealized potential contribu-
New Zealand. Geology, 34(2), 121–124.
tions of cosmogenic-nuclide exposure dating to glacier chronol-
Anderson, B., and Mackintosh, A., 2012. Controls on mass balance
ogy 1990–2010. Quaternary Science Reviews, 30(1–2), 3–27.
sensitivity of maritime glaciers in the Southern Alps, New
Balco, G., and Rovey, C. W., 2008. An isochron method for
Zealand: the role of debris cover. Journal of Geophysical
cosmogenic-nuclide dating of buried soils and sediments.
Research-Earth Surface, 117, doi:F01003 10.1029/
American Journal of Science, 308(10), 1083–1114.
2011jf002064.
 GLACIAL LANDSCAPE (COSMOGENIC NUCLIDE)
Balco, G., and Rovey, C. W., 2010. Absolute chronology for major
Pleistocene advances of the Laurentide Ice Sheet. Geology,
38(9), 795–798.
Balco, G., and Shuster, D. L., 2009. Al-26-Be-10-Ne-21 burial dat-
ing. Earth and Planetary Science Letters, 286(3–4), 570–575.
Balco, G., et al., 2005. The first glacial maximum in North America.
Science, 307(5707), 222.
Balco, G., et al., 2008. A complete and easily accessible means of
calculating surface exposure ages or erosion rates from Be-10
and Al-26 measurements. Quaternary Geochronology, 3(3),
174–195.
Balco, G., et al., 2009. Regional beryllium-10 production rate cali-
bration for northeastern North America. Quaternary Geochro-
nology, 4, 93–107.
Balco, G., et al., 2013. Exposure-age record of Holocene ice sheet
and ice shelf change in the northeast Antarctic Peninsula. Qua-
ternary Science Reviews, 59, 101–111.
Blard, P.-H., et al., 2013a. Cosmogenic 10Be production rate
calibrated against 3He in the high Tropical Andes
(3800–4900 m, 20–22
S). Earth and Planetary Science Letters,
182, 140–149.
Blard, P. H., et al., 2013b. Cosmogenic He-3 production rate in the
high tropical Andes (3800 m, 20 degrees S): implications for the
local last glacial maximum. Earth and Planetary Science Letters,
377, 260–275.
Brook, E. J., Kurz, M. D., Ackert, R. P., Jr., Denton, G. H., Brown,
E. T., Raisbeck, G. M., and Yiou, F., 1993. Chronology of Taylor
Glacier advances in Arena Valley, Antarctica, using in situ cos-
mogenic 3He and 10Be. Quaternary Research, 39, 11–23.
Bruno, L. A., et al., 1997. Dating of Sirius Group tillites in the Ant-
arctic dry valleys with cosmogenic 3He and 21Ne. Earth and
Planetary Science Letters, 147, 37–54.
Cerling, T. E., and Craig, H., 1994. Geomorphology and in-situ cos-
mogenic isotopes. Annual Review of Earth and Planetary Sci-
ences, 22, 273–317.
Fenton, C. R., et al., 2011. Regional (10)Be production rate calibra-
tion for the past 12 ka deduced from the radiocarbon-dated
Grotlandsura and Russenes rock avalanches at 69 degrees N,
Norway. Quaternary Geochronology, 6(5), 437–452.
Fudge, T. J., et al., 2013. Onset of deglacial warming in West
Antarctica driven by local orbital forcing. Nature (London),
500(7463), 440–444.
Goehring, B. M., et al., 2010. A reevaluation of in situ cosmogenic
He-3 production rates. Quaternary Geochronology, 5(4),
410–418.
Goehring, B. M., et al., 2011. The Rhone Glacier was smaller than
today for most of the Holocene. Geology, 39(7), 679–682.
Goehring, B. M., et al., 2012a. Late glacial and holocene 10Be
production rates for western Norway. Journal of Quaternary
Science, 27(1), 89–96.
Goehring, B. M., et al., 2012b. Holocene dynamics of the Rhone
Glacier, Switzerland, deduced from ice flow models and cosmo-
genic nuclides. Earth and Planetary Science Letters,
351, 27–35.
Goehring, B. M., et al., 2013. An in situ 14C–10Be Bayesian iso-
chron approach for interpreting complex glacial histories. Qua-
ternary Geochronology, 15, 61–66.
Goehring, B. M., et al., 2014. Capabilities of the Lamont–Doherty
Earth Observatory in situ 14C extraction laboratory updated.
Quaternary Geochronology, 19, 194–197.
Golledge, N. R., et al., 2012. Last Glacial Maximum climate in
New Zealand inferred from a modelled Southern Alps icefield.
Quaternary Science Reviews, 46, 30–45.
Gosse, J. C., and Phillips, F. M., 2001. Terrestrial in situ cosmogenic
nuclides: theory and application. Quaternary Science Reviews,
20(14), 1475–1560.
307
Gosse, J. C., et al., 1995. Precise cosmogenic 10Be measurements in
western North America: support for a global Younger Dryas
cooling event. Geology, 23(10), 877–880.
Granger, D. E., and Muzikar, P. F., 2001. Dating sediment
burial with in situ-produced cosmogenic nuclides: theory, tech-
niques, and limitations. Earth and Planetary Science Letters,
188(1–2), 269–281.
Granger, D. E., et al., 2013. A cosmic trip: 25 years of cosmogenic
nuclides in geology. Geological Society of America Bulletin,
125(9–10), 1379–1402.
Hippe, K., et al., 2013. An update on in situ cosmogenic C-14 anal-
ysis at ETH Zurich. Nuclear Instruments & Methods in Physics
Research Section B-Beam Interactions with Materials and
Atoms, 294, 81–86.
Hippe, K., et al., 2014. Chronology of Lateglacial ice flow reorgani-
zation and deglaciation in the Gotthard Pass area, Central Swiss
Alps, based on cosmogenic Be-10 and in situ C-14. Quaternary
Geochronology, 19, 14–26.
Ivy-Ochs, S. et al., 1997. 10Be and 26Al exposure ages for the Sirius
Group at Mount Fleming, Mount Feather and Table Mountain,
and for the plateau surface at Table Mountain. In The Antarctic
Region: Geological Evolution and Processes, pp. 1153–1158.
Ivy-Ochs, S., et al., 1999. Moraine exposure dates imply synchro-
nous Younger Dryas glacier advances in the European Alps
and in the southern Alps of New Zealand. Geografiska Annaler,
81 A(2), 313–323.
Johnson, J. S., et al., 2014. Rapid Thinning of Pine Island Glacier in
the Early Holocene. Science 343, 999–1001.
Kaplan, M. R., et al., 2011. In-situ cosmogenic (10)Be production
rate at Lago Argentino, Patagonia: implications for late-glacial
climate chronology. Earth and Planetary Science Letters,
309(1–2), 21–32.
Kelly, M. A., et al., 2008. A 10Be chronology of late-glacial and
Holocene mountain glaciation in the Scoresby Sund region, east
Greenland: Implications for seasonality during late-glacial time.
Quaternary Science Reviews, 27(25–26), 2273–2282.
Kelly, M. A. et al., (in press). A locally calibrated, late glacial 10Be
production rate from a low-latitude, high-altitude site in the
Peruvian Andes. Quaternary Geochronology.
Kurz, M. D., 1986. Cosmogenic helium in a terrestrial igneous rock.
Nature, 320, 435–439.
Lal, D., 1991. Cosmic ray labeling of erosion surfaces: in situ
nuclide production rates and erosion models. Earth and Plane-
tary Science Letters, 104, 424–439.
Licciardi, J. M., et al., 2009. Holocene glacier fluctuations in the
Peruvian Andes indicate Northern climate linkages. Science,
325, 1677–1679.
Lifton, N. A., et al., 2001. A new extraction technique and produc-
tion rate estimate for in situ cosmogenic 14C in quartz.
Geochimica et Cosmochimica Acta, 65(12), 1953–1969.
Lifton, N. A. et al., 2014. Scaling in situ cosmogenic nuclide
production rates using analytical approximations to atmospheric
cosmic-ray fluxes. Earth and Planetary Science Letters
386, 149–160.
Margerison, H. R., et al., 2005. Cosmogenic 3He concentrations in
ancient flood deposits from the Coombs Hills, northern Dry Val-
leys, East Antarctica: interpreting exposure ages and erosion
rates. Earth and Planetary Science Letters, 230, 163–175.
Martinson, D. G., 2012. Antarctic circumpolar current’s role in the
Antarctic ice system: an overview. Palaeogeography Palaeocli-
matology Palaeoecology, 335–336, 71–74.
Park, J. W., et al., 2013. Sustained retreat of the Pine Island Glacier.
Geophysical Research Letters, 40, 2137–2142.
Putnam, A. E., et al., 2010. In situ cosmogenic 10Be production-rate
calibration from the Southern Alps, New Zealand. Quaternary
Geochronology, 5(4), 392–409.
308 GROUNDWATER DATING WITH ATMOSPHERIC HALOGENATED COMPOUNDS
Putnam, A. E., et al., 2013a. Warming and glacier recession in the
Rakaia valley, Southern Alps of New Zealand, during Heinrich
Stadial 1. Earth and Planetary Science Letters, 382, 98–110.
Putnam, A. E., et al., 2013b. The Last Glacial Maximum at 44
S
documented by a 10Be moraine chronology at Lake Ohau, South-
ern Alps of New Zealand. Quaternary Science Reviews,
62, 114–141.
Schaefer, J. M., et al., 2009. High frequency Holocene glacier fluc-
tuations in New Zealand differ from the northern signature.
Science, 324, 622.
Schäfer, J. M., et al., 1999. Cosmogenic noble gas studies in
the oldest landscape on earth: surface exposure ages of the Dry
Valleys, Antarctica. Earth and Planetary Science Letters,
167, 215–226.
Schimmelpfennig, I., 2009. Cosmogenic 36Cl in Ca and K rich min-
erals: analytical developments, production rate calibrations and
cross calibration with 3He and 21Ne. PhD thesis; Université
Aix-Marseille III.
Schimmelpfennig, I., et al., 2011. Calibration of cosmogenic 36Cl
production rates from Ca and K spallation in lava flows from
Mt. Etna (38
N, Italy) and Payun-Matru (36
S, Argentina).
Geochimica et Cosmochimica Acta, 75, 2611–2632.
Schimmelpfennig, I. L., et al., 2014. A chronology of Holocene and
Little Ice Age glacier culminations of the Steingletscher,
CentralAlps, Switzerland, based on high-sensitivity beryllium-10
moraine dating. Earth and Planetary Science Letters 393, 220–230.
Stone, J. O., et al., 2003. Holocene deglaciation of Marie Byrd
Land, West Antarctica. Science, 299(5603), 99–102.
Stroup, J. S., et al., 2014. Late Holocene fluctuations of Qori Kalis
outlet glacier, Quelccaya Ice Cap, Peruvian Andes. Geology,
42, 347–350.
Young, N. E., et al., 2013. A Be-10 production-rate calibration for
the Arctic. Journal of Quaternary Science, 28(5), 515–526.
GROUNDWATER DATING WITH ATMOSPHERIC
HALOGENATED COMPOUNDS
Karl B. Haase and Eurybiades Busenberg
CFC Laboratory, U.S. Geological Survey, Reston,
VA, USA
Synonyms
CFC dating; Chlorofluorocarbon dating; SF6 dating
Definition
Atmospheric environmental releases refer to the emission
of stable, long-lived compounds of solely anthropogenic
origin into the atmosphere and the use of the compounds
to estimate dates of their incorporation into groundwater.
Introduction
The majority of environmental anthropogenic tracers that
can be used to date groundwater were released to the
atmosphere after 1940 (Figure 1). Groundwater dating
relates the measured concentration of these trace atmo-
spheric gases to the reconstructed history of the trace gas
concentration in the atmosphere. Three general types
of tracers can be used to date groundwater: (1) halogenated
tracers, including the chlorofluorocarbons and
hydrochlorofluorocarbons (CFCs and HCFCs), the
perfluorocarbons and hydrofluorocarbons (PFCs and
HFCs), sulfur hexafluoride (SF6), trifluoromethyl sulfur
pentafluoride (SF5CF3), and nitrogen trifluoride (NF3)
(IAEA, 2006); (2) anthropogenic radiogenic tracers intro-
duced by atmospheric nuclear testing and reactor fuel
reprocessing (3H, 14C, 36Cl, and 85Kr) (Cook and Herczeg,
1999; Plummer, 2005); and (3) event markers which
include alkylbenzene sulfonate (ABS) and linear
alkylbenzene sulfonate (LAS) surfactants, herbicides,
pharmaceuticals, pesticides, and their degradation prod-
ucts (Plummer et al., 1993). A number of excellent
reviews discussing groundwater ages and dating methods
using environmental tracers have been published
(Plummer et al., 1993; Cook et al., 1995; Goode, 1996;
Plummer and Friedman, 1999; Höhener et al., 2003;
Bethke and Johnson, 2008; Darling et al., 2010; Newman
et al., 2010). This section discusses the application of
halogenated atmospheric tracers to groundwater dating.
CFCs were first synthesized in the late 1920s, and by the
1940s they became the replacements for ammonia, sulfur
dioxide, and methyl chloride as refrigerants, which had
been implicated in a series of accidents and deaths
(Giunta, 2006). CFC production, applications, and releases
accelerated after World War II, with chlorofluorocarbons
eventually comprising approximately half of the tropo-
spheric chlorine burden (Schauffler et al., 1993). In devel-
oped countries, production of CFCs stopped, and releases
slowed following ratification of the Montreal Protocol on
Substances that Deplete the Ozone Layer treaty. Concentra-
tions in the atmosphere are now in decline (WMO, 2010).
The CFCs have been replaced by hydrochlorofluor-
ocarbons (HCFCs), which will also be phased out by
2030, and by hydrofluorocarbons (HFCs). These com-
pounds have much shorter atmospheric lifetimes and are
less damaging to the ozone layer than the CFCs (WMO,
2010). The concentration of perfluorocarbons (PFCs)
(CF4, C2F6), SF6, and SF5CF3 are lower than CFCs, have
atmospheric lifetimes of hundreds to thousands of years,
and are potent greenhouse gases (IPCC, 2007). PFC analy-
sis in water is difficult and these compounds have not been
routinely used in groundwater dating (Deeds et al., 2008;
Deeds, 2008). In contrast, analyses of CFCs, SF6, and
SF5CF3 are relatively easy by gas chromatography with
an electron capture detector (GC-ECD) (Lovelock, 1958,
1974; Thompson and Hayes, 1979; Bullister and Weiss,
1988; Busenberg and Plummer, 1992; Hofer and Imboden,
1998; Busenberg and Plummer, 2000, 2008).
Dating young groundwater is important for evaluating
groundwater resources. CFC dating was used to obtain
the historical record of nitrate contamination in aquifers
(Böhlke and Denver, 1995; Böhlke et al., 1997; Johnston
et al., 1998; Katz et al., 2001; Lindsey et al., 2003) and
groundwater pollution histories and evaluation suscepti-
bility of aquifers to contamination (Nelms et al., 2003;
Shapiro et al., 2004; Plummer et al., 2008), to evaluate
aquifer resources (Dunkle et al., 1993; Szabo et al.,
1996; Plummer et al., 2000; Bauer et al., 2001;
 GROUNDWATER DATING WITH ATMOSPHERIC HALOGENATED COMPOUNDS 309

Groundwater Dating with Atmospheric Halogenated Compounds, Figure 1 Historic northern hemisphere background air
concentrations of CFC-11, CFC-12, CFC-113, and SF6 (http://water.usgs.gov/lab); global average 85Kr activity per liter of air (Ahlswede
et al., 2009, 2013); northern hemisphere 36Cl deposition rate (Bentley et al., 1982, 1986; Althaus et al., 2009); 3H activity in Washington,
DC, precipitation (IAEA/WMO, 2006); and global average percent modern 14C in atmospheric CO2 (Nydal and Lövseth, 1983;
Winger et al., 2005; Turnbull et al., 2007).

Burton et al., 2002; Lindsey et al., 2003; Cook et al., 2005;
Gooddy et al., 2006; von Rohden et al., 2010b; McMahon
et al., 2011; Manning et al., 2012; Liu et al., 2013), and to
calibrate groundwater models (Reilly et al., 1994;
Weissmann et al., 2002).
Groundwater dating theory
The concentrations of environmental tracers in groundwa-
ter are affected by transport processes. Dating based on
concentrations of atmospheric gases in groundwater
depends on the model used and model-dependent assump-
tions of mixing and transport in aquifers. For preliminary
age interpretations, the piston-flow model is used
(Fontes, 1982; Zuber, 1986). The piston-flow dating
method uses the measured tracer concentrations in water
and their calculated partial pressures and subsequently
compares them to known historical concentrations in
the atmosphere. Several assumptions are implied by the
piston-flow model: the tracers are in equilibrium with the
gas at the water table during recharge (when the water
reaches the water table), the tracer partial pressures at the
water table are the same as the atmosphere (i.e., the unsat-
urated zone is less than 10 m thick), the recharge temper-
ature and elevation are known, and the tracer
concentrations in the aquifer were not significantly modi-
fied by adsorption, contamination, matrix diffusion, or
biodegradation. The piston-flow model assumes that
water moves as closed packets with no mixing or hydrody-
namic dispersion (Fontes, 1982). The concentration of
several environmental tracers in groundwater can also be
described by lumped-parameter mathematical models.
These models use simplified aquifer geometries, flow,
hydrodynamic dispersion, and mixing to describe tracer
concentrations in the aquifer and borehole (Zuber, 1986;
Cook and Böhlke, 1999; Gooddy et al., 2006; Zuber and
Rosanski, 2007; Eberts et al., 2012; Jurgens et al., 2012).
However, groundwater samples are mixtures of packets
of molecules each with a unique age and tracer concentra-
tion. This is an important distinction, as individual compo-
nents have different aquifer residence times. Thus,
a successful reactive transport flow model will calculate
all the observed tracer concentrations in the groundwater.
Tracer concentrations are best used to calibrate transport
and flow models which can be used to determine the aver-
age residence time of the sample in the aquifer (Etcheverry
and Perrochet, 2000; Bethke and Johnson, 2008;
Newman, et al., 2010).
Uncertainty in age determination (disparities between
tracers ranging from several to many years) can derive
from many sources, including vadose zone thickness,
recharge temperature, infiltration rate, dispersion, mixing,
and bacterial activity (IAEA, 2006). Corrections need to
be applied to tracer concentration measurements if
recharge is through thick unsaturated zones since the
partial pressures at the water table can be significantly dif-
ferent from atmospheric concentrations due to diffusion
through the vadose zone (Cook et al., 1995; Cook and
Solomon, 1997, Busenberg and Plummer, 2000;
Engesgaard et al., 2004; Cook et al., 2006; Busenberg
and Plummer, 2008). The recharge temperature may be
different from the water temperature at the time of
310 GROUNDWATER DATING WITH ATMOSPHERIC HALOGENATED COMPOUNDS
Groundwater Dating with Atmospheric Halogenated Compounds, Table 1 Solubility coefficients to the Clarke-Glew-Weiss
(Eq. 4) used for determining solubilities in water (Warner and Weiss, 1985; Bu and Warner, 1995; Bullister et al., 2002; Busenberg
and Plummer, 2008). Units are in mol L1 atm1
Compound A B C D
CFC-11 (CCl3F) 134.1536 203.2156 56.2320
CFC-12 (CCl2F2) 122.3246 182.5306 50.5898
CFC-113 (C2Cl3F3) 134.243 203.898 54.9583
SF6 80.0343 117.232 29.5817
SF5CF3 12.8108 23.9976 3.9715
sampling, since recharge temperature is often within
1
C of the mean annual air temperature of the recharge
area (Collins, 1925; Stute and Schlosser, 1993; Zuber
et al., 1995). Tracer concentrations will exceed the equi-
librium concentration when there is a rapid rise in the
water table, which causes trapped gas bubbles (excess
air) to dissolve due to the increase in hydrostatic pressure
and which requires correction of the dissolved tracer
concentrations (Stute et al., 1995; Aeschbach-Hertig
et al., 1999, 2000; Sun et al., 2010). The recharge temper-
ature, excess air, and recharge elevation can be estimated
from dissolved noble gas concentrations (Ne, Ar, Kr, and
Xe) measured by mass spectrometry (Cook et al., 2006;
Sun et al., 2010). In aerobic waters, the recharge tempera-
ture and excess air can be estimated by gas chromatogra-
phy (GC) from the concentrations of N2 and Ar alone
(Heaton and Vogel, 1981).
Relationships between age tracers and solubility
The equilibrium concentration of gases in water is
described by Henry’s Law (Eq. 1), which equates the
aqueous concentration to the partial pressure (mole frac-
tion) of gases above the water:
C w ¼ K h ðxi Þ
where Cw is the molar concentration of the gas in water, Kh
is the Henry’s Law constant, and xi is the partial pressure
of the gas (Weiss, 1970; Warner and Weiss, 1985). The
equilibrium concentration is dependent on the atmo-
spheric pressure and the solution temperature and salinity.
The mole fraction of the gas above the solution is given by
the equation:
mi
xi ¼
 ð2Þ
K h P  pH 2 O
where xi is the dry mole fraction of the ith tracer, mi is the
molar concentration, P is the atmospheric pressure, and
pH 2 O is the vapor pressure of water, respectively. pH 2 O is
given by the empirical equation:
pH 2 O ¼ eð24:454367:4509ð T Þ4:8489lnð100Þ0:000544S Þ
100 T
E F Ref.
0.144449 0.092952 0.015997 (Bullister and Weiss, 1988)
0.145633 0.092509 0.0156627 (Bullister and Weiss, 1988)
0.02632 0.005874 – (Bu and Warner, 1995)
0.0335183 0.0373942 0.00774862 (Bullister et al., 2002)
– – – (Busenberg and Plummer, 2008)
where S is the salinity of the water in per mill and T is the
temperature in degrees Kelvin (Weiss and Price, 1980).
The Clarke-Glew-Weiss equation (Eq. 4) is commonly
used to parameterize the temperature- and salinity-
dependent gas solubilities (Table 1) (Warner and Weiss,
1985; Bu and Warner, 1995; Bullister et al., 2002;
Busenberg and Plummer, 2008):
   
100 T
lnK h ¼ A þ B  þ C  ln
T 100
   2 ! ð4Þ
T T
xS DþE þF
100 100
where A, B, C, D, E, and F are fit coefficients and T is the
equilibrium (recharge) temperature in degrees Kelvin. The
solubilities for atmospheric age tracers are shown in
Figure 2.
Historical atmospheric concentrations
Ascertaining the historical atmospheric concentrations of
tracers is of critical importance to dating with environmen-
tal tracers. The historic concentrations of halocarbons
have been measured and extensively studied because of
ð1Þ their role in atmospheric chemistry and potent radiative
forcing properties (IPCC, 2007; WMO, 2010; Montzka
et al., 2011; Garcia et al., 2012). Many halocarbons that
are stable in groundwater systems are routinely measured
in a network of ground stations around the world by the
National Aeronautics and Space Administration
(NASA)-funded Advanced Global Atmospheric Gases
Experiment (AGAGE) and the National Oceanic and
Atmospheric Administration Earth System Research
Laboratory’s (NOAA ESRL) Halocarbons and other
Atmospheric Trace Species Group (HATS) (AGAGE,
2013; HATS, 2013). These projects have been collecting
archive samples and reporting in situ and flask measure-
ments since 1978. The pre-1976 atmospheric concentra-
tions have been reconstructed from production and
release estimates (McCarthy et al., 1977; AFEAS, 2013).
Measurements are typically reported on calibration scales
prepared by NOAA (HATS, 2013) or Scripps Oceano-
graphic Institute (AGAGE, 2013). The majority of atmo-
ð3Þ spheric tracers are produced and released in the northern
 GROUNDWATER DATING WITH ATMOSPHERIC HALOGENATED COMPOUNDS
Groundwater Dating with Atmospheric Halogenated Compounds, Figure 2 The temperature-dependent solubility of commonly
used environmental tracers in pure water.
hemisphere; consequently, there is an approximate 1-year
lag in transport between hemispheres (Seinfeld and
Pandis, 1998). Local atmospheric sources can increase
concentrations above background levels in the vicinity
of the source, requiring knowledge of the local history in
these instances in order to reliably interpret groundwater
ages (Oster et al., 1996; Cook and Solomon, 1997;
Ho et al., 1998; Cook et al., 2006).
Analytical methods
Purge-and-trap-based GC-ECD systems are usually used
to quantify CFCs, SF6, and SF5CF3 concentrations
(Lovelock, 1958, 1974; Thompson and Hayes, 1979;
Bullister and Weiss, 1988; Busenberg and Plummer,
1992; Hofer and Imboden, 1998; Busenberg and
Plummer, 2000, 2008). To quantify the HCFCs, HFCs,
and many PFCs, methods employing a GC with mass
spectrometer detectors (GC-MS) are used, as these com-
pounds have low response in ECDs (Sousa and
Bialkowski, 1997; Miller et al., 2008; Deeds, 2008; Deeds
et al., 2008).
Environmental releases of ultra-trace halocarbons
may soon become useful in dating groundwater because
of improved sampling procedures in the field. Several
techniques are being developed that will allow the extrac-
tion of gases from very large volumes of water and enable
the use of tracers that have atmospheric concentrations of
only a few parts per quadrillion (1  1015) in dating stud-
ies (Johnson, 1999; Probst et al., 2007; Loose et al., 2009;
311
Ohta et al., 2009; Busenberg and Plummer, 2008; Gardner
and Solomon, 2009; AGAGE, 2013).
CFC-11, CFC-12, CFC-13, and CFC-113
CFC-11, CFC-12, and CFC-113 are commonly used in
groundwater dating studies (Thompson and Hayes,
1979; Busenberg and Plummer, 1992; Ekwurzel
et al., 1994; Solomon et al., 1996; Cook and Solomon,
1997), as the other CFCs (CFC-114(a), CFC-115, and
CFC-13), have low atmospheric concentrations and
(or) are more difficult to detect (Busenberg and Plummer,
2008). CFC-11 (CCl3F) and CFC-12 (CCl2F2) were exten-
sively used as safe refrigerants, nonflammable aerosol
propellants, and blowing agents for foam plastics. Signif-
icant production of CFC-113 (C2Cl3F3), a liquid degreaser
and cleaning agent, started in the 1950s. CFC-13 (CClF3)
was primarily used as a very low-temperature refrigerant
and has shown promise as a dating tool. CFCs are rela-
tively inert with long atmospheric lifetimes (102, 45, and
85 years for CFC-12, CFC-11, and CFC-113, respec-
tively). CFC-11 and, to a lesser degree, CFC-12 and
CFC-113 are susceptible to degradation by methanogenic
bacteria in aquifers (Lovley and Woodward, 1992; Bauer
et al., 2001; Balsiger et al., 2005; Sebol et al., 2007;
Montzka and Reimann, 2010). CFC atmospheric concen-
trations in the northern hemisphere peaked between 1990
and 2000 (CFC-12 peaked at 546 pptv in 2001, CFC-11 at
272 pptv in 1994, and CFC-113 at 85 pptv in 1996).
Atmospheric concentrations are now slowly decreasing
312 GROUNDWATER DATING WITH ATMOSPHERIC HALOGENATED COMPOUNDS
Groundwater Dating with Atmospheric Halogenated Compounds, Figure 3 Useful dating ranges of CFCs, SF6, and their ratios.
with the 2010 northern hemisphere concentrations of
529, 237, and 74 pptv, for CFC-12, CFC-11, and
CFC-113, respectively (Montzka and Reimann, 2010).
This decrease in atmospheric concentrations can compli-
cate piston-flow age interpretation of modern waters
since there is not a unique age solution. This difficulty
may be overcome by using CFC/CFC or CFC/SF6 ratios
or SF6 concentration to qualify the apparent piston-flow
age (Figure 3) (Busenberg and Plummer, 2000; IAEA,
2006).
SF6 and SF5CF3
Sulfur hexafluoride (SF6) is an inert gas that has the
highest known dielectric constant. The production of SF6
began in the 1940s as part of the war effort; the high elec-
trical insulating properties were crucial in the develop-
ment of million volt x-ray tubes, radar, and the first
atomic bombs (Preston, 2003). Large-scale production
began in 1953 with the introduction of SF6-filled electrical
switches and transformers (Ko et al., 1993; Maiss and
Brenninkmeijer, 1998; Busenberg and Plummer, 2000).
SF6 has also been used extensively in oceanic tracer stud-
ies and as an inert blanketing gas in the melting and pro-
duction of magnesium metal (Maiss et al., 1996; Wilson
and Mackay, 1996; Caplow et al., 2003; Hall
et al., 2011; Smith et al., 2011). A natural atmospheric
steady-state preindustrial background source of <0.05
pptv has been recognized and is thought to originate from
silicate-rich igneous rocks, the mineral fluorite (CaF2),
some natural gases, and ancient groundwater (Maiss
et al., 1996; Harnisch et al., 2000; Busenberg and
Plummer, 2000). These geologic sources can cause ele-
vated levels of SF6 in some aquifers above atmospheric
equilibrium and complicate dating (Busenberg and
Plummer, 2000; Deeds et al., 2008; von Rohden
et al., 2010a). SF6 is an important dating tracer because,
unlike CFCs, it is not known to be susceptible to microbial
degradation and it has an estimated atmospheric lifetime
between 800 and 3,200 years; thus, atmospheric concen-
trations are not expected to decrease in the foreseeable
future (Ko et al., 1993; Levin et al., 2010).
SF5CF3 was first reported in the atmosphere by
Sturges et al. (2000) and is entirely of anthropogenic ori-
gin. SF5CF3 does not have commercial applications and
is believed to be a by-product of production of some
fluorochemicals (Sturges et al., 2000; Santoro, 2000;
Erboy and Smethie, 2012; Sturges et al., 2012). SF5CF3
has been used as a dating tool in groundwater studies
(Busenberg and Plummer, 2008) and as a tracer in hydro-
logic studies (Sturges et al., 2000; Ho et al., 2008;
Busenberg and Plummer, 2010). Samples of air from
polar snowpack (“fern”) and archived air show that the
atmospheric SF5CF3 concentration increased from
<0.005 pptv in the 1960s to 0.158 pptv in 2008 after
which the SF5CF3 concentration plateaued due to
changes in fluorochemical production methods (Sturges
et al., 2000; Busenberg and Plummer, 2008; Sturges
et al., 2012).
 GROUNDWATER DATING WITH ATMOSPHERIC HALOGENATED COMPOUNDS
Perfluoroalkanes (CnF2n+2) and cyclic
perfluorocarbons
The most abundant PFC in the atmosphere is CF4. Like
SF6, it has both natural and anthropogenic sources, with
a preindustrial steady-state atmospheric concentration of
34  1 pptv. After 1960, the concentration increased to
the present concentration of ~80 pptv (Simmonds
et al., 2002; Mühle et al., 2010; AGAGE, 2013). The next
most abundant PFC, C2F6, had a preindustrial background
concentration of <0.3 and the concentration has increased
to ~3.5 pptv as of 2010 (Worton et al., 2007; AGAGE,
2013). CF4 occurs naturally in fluorite minerals and gran-
ite (Harnisch et al., 1996, 2000) and is also present in sig-
nificant concentrations in natural gas and in deep
groundwater (Deeds, 2008). CF4 and C2F6 are produced
at the graphite anodes during Al production (Roberts and
Ramsey, 2013). C3 to C8 perfluorocarbons have been
detected in the atmosphere and their concentrations are
increasing (Mühle et al., 2010; Ivy et al., 2012).
Perfluorocycloalkanes are very stable compounds and
are increasing in the atmosphere (Straume et al., 1998;
Watson et al., 2007) and have been successfully used to
trace of air masses and may be used in future groundwater
studies (Watson et al., 2007). They cannot be easily ana-
lyzed by GC-ECD, and research into detection approaches
using specialized GC mass spectrometry (GC-MS) and
negative chemical ionization mass spectroscopy
(GC-NCI-MS) is ongoing (Galdiga and Greibrokk,
1997; Simmonds et al., 2002; Miller et al., 2008).
Nitrogen trifluoride
Prather and Hsu (2008) recognized the potent greenhouse
gas NF3 in the atmosphere, which has a lifetime of
740 years (Prather and Hsu, 2008). NF3 is extensively
used in the electronics industry for equipment cleaning,
for etching of microcircuits, and in the manufacture of
flat-panel displays and thin-film photovoltaic cells
(Weiss et al., 2008). The NF3 concentrations have
increased from about 0.02 pptv in 1978 at a rate of 11 %
per year to 0.45 in 2008 (Weiss et al., 2008) and may have
uses in future hydrologic investigations.
Summary
Anthropogenic atmospheric tracers have been extensively
used in hydrologic tracer studies and in groundwater dat-
ing studies. CFCs and SF6 were the most extensively used
tracers; however, many other environmental tracers
(SF5CF3, CFC-13, PFCs, and HFCs) show potential as
dating tools pending development of improved analytical
methods. Dating of groundwater using the piston-flow
model, lumped-parameter mathematical models, or reac-
tive transport flow models requires the knowledge of
recharge conditions (recharge temperature and pressure
and vadose zone thickness) and aquifer properties that
affect the dissolved tracer concentrations in the sample
(adsorption, contamination, degradation).
313
Bibliography
Aeschbach-Hertig, W., Peeters, F., Beyerle, U., and Kipfer, R.,
1999. Interpretation of dissolved atmospheric noble gases in nat-
ural waters. Water Resources Research, 35(9), 2779–2792,
doi:10.1029/1999wr900130.
Aeschbach-Hertig, W., Peeters, F., Beyerle, U., and Kipfer, R.,
2000. Palaeotemperature reconstruction from noble gases in
groundwater taking into account equilibration with entrapped
air. Nature, 405(6790), 1040–1044.
AFEAS, 2013. Production, Sales, and Atmospheric Release of
Fluorocarbons Through 2007: Alternative Fluorocarbons Envi-
ronmental Acceptability Study (AFEAS) Administrative Organi-
zation. Accessed at April 21, 2014 at www.afeas.org.
AGAGE. 2013. Advanced Global Atmospheric Gases Experiment.
Accessed on April 12, 2014 at http://agage.eas.gatech.edu/.
Ahlswede, J., Hebel, S., Kalinowski, M. B., and Ross, O., 2009.
Update of the Global Krypton-85 Emission Inventory. Zentrum
Für Naturwissenschaft und Friedensforschung der Universitat
Hamburg Occasional Paper No. 9, 24 p., Accessed on April
12, 2014 at http://www.znf.uni-hamburg.de/OcPaper_No9.pdf.
Ahlswede, J., Hebel, S., Ross, J. O., Schoetter, R., and Kalinowski,
M. B., 2013. Update and improvement of the global krypton-85
emission inventory. Journal of Environmental Radioactivity,
115, 34–42.
Althaus, R., Klump, S., Onnis, A., Kipfer, R., Purtschert, R.,
Stauffer, F., and Kinzelbach, W., 2009. Noble gas tracers for
characterisation of flow dynamics and origin of groundwater:
a case study in Switzerland. Journal of Hydrology, 370(1–4),
64–72, doi:10.1016/j.jhydrol.2009.02.053.
Balsiger, C., Holliger, C., and Höhener, P., 2005. Reductive dechlo-
rination of chlorofluorocarbons and hydrochlorofluorocarbons
in sewage sludge and aquifer sediment microcosms.
Chemosphere, 61(3), 361–373, doi:10.1016/j.chemosphere.
2005.02.087.
Bauer, S., Fulda, C., and Schäfer, W., 2001. A multi-tracer study in
a shallow aquifer using age dating tracers 3H, 85Kr, CFC-113 and
SF6 – indication for retarded transport of CFC-113. Journal of
Hydrology, 248(1–4), 14–34, doi:10.1016/S0022-1694(01)
00381-X.
Bentley, H. W., Phillips, F. M., Davis, S. N., Gifford, S., Elmore, D.,
Tubbs, L. E., and Gove, H. E., 1982. Thermonuclear 36Cl pulse
in natural water. Nature, 300(5894), 737–740.
Bentley, H. W., Phillips, F. M., Davis, S. N., Habermehl, M. A.,
Airey, P. L., Calf, G. E., Elmore, D., Gove, H. E., and Torgersen,
T., 1986. Chlorine 36 dating of very old groundwater: 1. The
Great Artesian Basin, Australia. Water Resources Research,
22(13), 1991–2001, doi:10.1029/WR022i013p01991.
Bethke, C. M., and Johnson, T. M., 2008. Groundwater age
and groundwater age dating. Annual Review of Earth and
Planetary Sciences, 36(1), 121–152, doi:10.1146/annurev.
earth.36.031207.124210.
Böhlke, J. K., and Denver, J. M., 1995. Combined use of groundwa-
ter dating, chemical, and isotopic analyses to resolve the history
and fate of nitrate contamination in two agricultural watersheds,
Atlantic coastal plain, Maryland. Water Resources Research,
31(9), 2319–2339, doi:10.1029/95wr01584.
Böhlke, J. K., Révész, K., Busenberg, E., Deák, J., Deseö, É., and
Stute, M., 1997. Groundwater record of Halocarbon transport
by the Danube river. Environmental Science and Technology,
31(11), 3293–3299, doi:10.1021/es970336h.
Bu, X., and Warner, M. J., 1995. Solubility of chlorofluorocarbon
113 in water and seawater. Deep Sea Research Part I: Oceano-
graphic Research Papers, 42(7), 1151–1161, doi:10.1016/
0967-0637(95)00052-8.
Bullister, J. L., and Weiss, R. F., 1988. Determination of CCl3F and
CCl2F2 in seawater and air. Deep Sea Research Part I:
314 GROUNDWATER DATING WITH ATMOSPHERIC HALOGENATED COMPOUNDS
Oceanographic Research Papers, 35(5), 839–853, doi:10.1016/
0198-0149(88)90033-7.
Bullister, J. L., Wisegarver, D. P., and Menzia, F. A., 2002. The sol-
ubility of sulfur hexafluoride in water and seawater. Deep Sea
Research Part I: Oceanographic Research Papers, 49(1),
175–187, doi:10.1016/S0967-0637(01)00051-6.
Burton, W. C., Plummer, L. N., Busenberg, E., Lindsey, B. D., and
Gburek, W. J., 2002. Influence of fracture anisotropy on ground-
water ages and chemistry, Valley and Ridge province, Pennsyl-
vania. Ground Water, 40(3), 242–257.
Busenberg, E., and Plummer, L. N., 1992. Use of chlorofluorocar-
bons (CCl3F and CCl2F2) as hydrologic tracers and age-dating
tools: The alluvium and terrace system of central Oklahoma.
Water Resources Research, 28(9), 2257–2283, doi:10.1029/
92wr01263.
Busenberg, E., and Plummer, L. N., 2000. Dating young groundwa-
ter with sulfur hexafluoride: natural and anthropogenic sources
of sulfur hexafluoride. Water Resources Research, 36(10),
3011–3030, doi:10.1029/2000wr900151.
Busenberg, E., and Plummer, L. N., 2008. Dating groundwater
with trifluoromethyl sulfurpentafluoride (SF5CF3), sulfur
hexafluoride (SF6), CF3Cl (CFC-13), and CF2Cl2 (CFC-12).
Water Resources Research, 44(2), 18 p, doi:10.1029/
2007wr006150.
Busenberg, E., and Plummer, L. N., 2010. A rapid method for the
measurement of sulfur hexafluoride (SF6), trifluoromethyl sulfur
pentafluoride (SF5CF3), and Halon 1211 (CF2ClBr) in hydro-
logic tracer studies. Geochemistry, Geophysics, Geosystems,
11(11), Q11001, 10 p, doi:10.1029/2010gc003312.
Caplow, T., Schlosser, P., Ho, D. T., and Santella, N., 2003. Trans-
port dynamics in a sheltered estuary and connecting tidal straits:
sf6 tracer study in New York Harbor. Environmental Science and
Technology, 37(22), 5116–5126, doi:10.1021/es034198+.
Collins, W. D., 1925. Temperature of water available for industrial
use in the United States. U.S. Geological Survey Water-Supply
Paper 530-F, 97–104.
Cook, P. G., and Herczeg, A. L., 1999. Environmental Tracers in
Subsurface Hydrology. Norwell: Kluwer Academic, 529 p.,
ISBN: 0-7923-7707-9.
Cook, P. G., and Böhlke, J. K., 1999. Determining timescales for
groundwater flow and solute transport. In Cook, P. G., and
Herczeg, A. L. (eds.), Environmental Tracers in Subsurface
Hydrology. Norwell: Kluwer Academic, 529 p., ISBN: 0-7923-
7707-9.
Cook, P. G., and Böhlke, J. K., 2000. Determining timescales for
groundwater flow and solute transport. In Cook, P. G., and
Herczeg, A. L. (eds.), Environmental Tracers in Subsurface
Hydrology. Norwell: Kluwer, pp. 1–30.
Cook, P. G., and Herczeg, A. L., 2000. Environmental Tracers in
Subsurface Hydrology. Norwell: Kluwer, 529 p.
Cook, P. G., and Solomon, D. K., 1997. Recent advances in dating
young groundwater: chlorofluorocarbons, 3H/3He and 85Kr.
Journal of Hydrology, 191(1–4), 245–265, doi:10.1016/s0022-
1694(96)03051-x.
Cook, P. G., Solomon, D. K., Plummer, L. N., Busenberg, E., and
Schiff, S. L., 1995. Chlorofluorocarbons as tracers of groundwa-
ter transport processes in a shallow, silty sand aquifer. Water
Resources Research, 31(3), 425–434, doi:10.1029/94wr02528.
Cook, P. G., Love, A. J., Robinson, N. I., and Simmons, C. T., 2005.
Groundwater ages in fractured rock aquifers. Journal of Hydrol-
ogy, 308(1–4), 284–301, doi:10.1016/j.jhydrol.2004.11.005.
Cook, P. G., Plummer, L. N., Solomon, D. K., Busenberg, E., and
Han, L. E., 2006. Chapter 4 – Effects and processes that can
modify apparent CFC age. In Gröning, M., Han, L. F., and
Aggarwal, P. (eds.), Use of Chlorofluorcarbons in Hydrology:
A Guidebook. Vienna: International Atomic Energy Agency,
pp. 31–58.
Darling, W. G., Gooddy, D. C., Morris, B. L., and MacDonald, A. M.,
2010. Using CFCs and SF6 for groundwater dating: a SWOT anal-
ysis. In Birkle, P., and Torres-Alvarado, I. (eds.), Water-Rock
Interaction XIII. Boca Raton: CRC Press, pp. 15–22.
Deeds, D. A., 2008. The Natural Geochemistry of Tetrafluor-
omethane and Sulfur Hexafluoride: Studies of Ancient Mojave
Desert Groundwaters, North Pacific Seawaters and the Summit
Emissions of Kilauea Volcano. Dissertation, San Diego, Scripps
Institution of Oceanography, 158 p.
Deeds, D. A., Vollmer, M. K., Kulongoski, J. T., Miller, B. R.,
Mühle, J., Harth, C. M., Izbicki, J. A., Hilton, D. R., and Weiss,
R. F., 2008. Evidence for crustal degassing of CF4 and SF6 in
Mojave Desert groundwaters. Geochimica et Cosmochimica
Acta, 72(4), 999–1013, doi:10.1016/j.gca.2007.11.027.
Dunkle, S. A., Plummer, L. N., Busenberg, E., Phillips, P. J., Denver,
J. M., Hamilton, P. A., Michel, R. L., and Coplen, T. B., 1993.
Chlorofluorocarbons (CCl3F and CCl2F2) as dating tools and
hydrologic tracers in shallow groundwater of the Delmarva Penin-
sula, Atlantic Coastal Plain, United States. Water Resources
Research, 29(12), 3837–3860, doi:10.1029/93wr02073.
Eberts, S., Böhlke, J., Kauffman, L., and Jurgens, B., 2012. Com-
parison of particle-tracking and lumped-parameter
age-distribution models for evaluating vulnerability of produc-
tion wells to contamination. Hydrogeology Journal, 20(2),
263–282, doi:10.1007/s10040-011-0810-6.
Ekwurzel, B., Schlosser, P., Smethie, W. M., Jr., Plummer, L. N.,
Busenberg, E., Michel, R. L., Weppernig, R., and Stute, M.,
1994. Dating of shallow groundwater: comparison of the transient
tracers 3H/3He, chlorofluorocarbons, and 85Kr. Water Resources
Research, 30(6), 1693–1708, doi:10.1029/94wr00156.
Engesgaard, P., Højberg, A. L., Hinsby, K., Jensen, K. H., Laier, T.,
Larsen, F., Busenberg, E., and Plummer, L. N., 2004. Transport
and time lag of chlorofluorocarbon gases in the unsaturated
zone, Rabis Creek, Denmark. Vadose Zone Journal, 3(4),
1249–1261, doi:10.2113/3.4.1249.
Erboy, Y., and Smethie, W. M., Jr., 2012. Trifluoromethyl
sulfur pentafluoride and its relationship to sulfur hexafluoride
and chlorofluorocarbon-12 in the atmosphere near the New York
City metropolitan area. Atmospheric Environment, 55, 135–138,
doi:10.1016/j.atmosenv.2012.02.053.
Etcheverry, D., and Perrochet, P., 2000. Direct simulation of
groundwater transit-time distributions using the reservoir
theory. Hydrogeology Journal, 8(2), 200–208, doi:10.1007/
s100400050006.
Fontes, J. C., 1982. Dating groundwater. In International Atomic
Energy Agency (ed.), Guidebook on Nuclear Techniques in
Hydrology. Vienna: International Atomic Energy Agency.
Galdiga, C. U., and Greibrokk, T., 1997. Simultaneous determination
of trace amounts of sulphur hexafluoride and cyclic perfluor-
ocarbons in reservoir samples by gas chromatography.
Chromatographia, 46(7–8), 440–443, doi:10.1007/bf02490884.
Garcia, R. R., Kinnison, D. E., and Marsh, D. R., 2012. “World
avoided” simulations with the whole atmosphere community cli-
mate model. Journal of Geophysical Research, [Atmospheres],
117(D23303), 16 p, doi:10.1029/2012jd018430.
Gardner, P., and Solomon, D. K., 2009. An advanced passive diffu-
sion sampler for the determination of dissolved gas concentra-
tions. Water Resources Research, 45(6), W06423, 12 p,
doi:10.1029/2008wr007399.
Giunta, C. J., 2006. Thomas Midgley, Jr., and the invention of Chlo-
rofluorocarbon refrigerants: it ain’t necessarily so. Bulletin for
the History of Chemistry, 31(2), 66–73.
Gooddy, D. C., Darling, W. G., Abesser, C., and Lapworth, D. J.,
2006. Using chlorofluorocarbons (CFCs) and sulphur
hexafluoride (SF6) to characterise groundwater movement and
residence time in a lowland Chalk catchment. Journal of Hydrol-
ogy, 330(1–2), 44–52, doi:10.1016/j.jhydrol.2006.04.011.
 GROUNDWATER DATING WITH ATMOSPHERIC HALOGENATED COMPOUNDS
Goode, D. J., 1996. Direct simulation of groundwater age.
Water Resources Research, 32(2), 289–296, doi:10.1029/
95wr03401.
Hall, B. D., Dutton, G. S., Mondeel, D. J., Nance, J. D., Rigby, M.,
Butler, J. H., Moore, F. L., Hurst, D. F., and Elkins, J. W., 2011.
Improving measurements of SF6 for the study of atmospheric
transport and emissions. Atmospheric Measurement Techniques,
4(11), 2441–2451, doi:10.5194/amt-4-2441-2011.
Harnisch, J., Borchers, R., Fabian, P., Gaggeler, H. W., and
Schotterer, U., 1996. Effect of natural tetrafluoromethane.
Nature, 384(6604), 32–32.
Harnisch, J., Frische, M., Borchers, R., Eisenhauer, A., and Jordan,
A., 2000. Natural fluorinated organics in fluorite and rocks. Geo-
physical Research Letters, 27(13), 1883–1886, doi:10.1029/
2000gl008488.
HATS. 2013. Halocarbons & Other Atmospheric Trace Species
Group (HATS). National Oceanic and Atmospheric Administra-
tion. Accessed on April 12, 2014 at http://www.esrl.noaa.gov/
gmd/hats/.
Heaton, T. H. E., and Vogel, J. C., 1981. “Excess air” in groundwa-
ter. Journal of Hydrology, 50, 201–216, doi:10.1016/0022-
1694(81)90070-6.
Ho, D. T., Schlosser, P., Smethie, W. M., and Simpson, H. J., 1998.
Variability in atmospheric chlorofluorocarbons (CCl3F and
CCl2F2) near a large urban area: Implications for groundwater
dating. Environmental Science and Technology, 32(16),
2377–2382, doi:10.1021/es980021h.
Ho, D. T., Ledwell, J. R., and Smethie, W. M., Jr., 2008. Use of
SF5CF3 for ocean tracer release experiments. Geophysical
Research Letters, 35(4), L04602, 5 p, doi:10.1029/2007gl032799.
Hofer, M., and Imboden, D. M., 1998. Simultaneous determination
of CFC-11, CFC-12, N2, and Ar in water. Analytical Chemistry,
70(4), 724–729, doi:10.1021/ac970499o.
Höhener, P., Werner, D., Balsiger, C., and Pasteris, G., 2003. World-
wide occurrence and fate of chlorofluorocarbons in groundwater.
Critical Reviews in Environmental Science and Technology,
33(1), 1–29, doi:10.1080/10643380390814433.
IAEA, 2006. Use of Chlorofluorcarbons in Hydrology:
A Guidebook. Vienna: International Atomic Energy Agency.
IAEA/WMO, 2006. Global Network of Isotopes in Precipitation. The
GNIP database. International Atomic Energy Agency/World
Meteorological Organization website Accessed April 12, 2014 at
http://www-naweb.iaea.org/napc/ih/IHS_resources_gnip.html.
IPCC. 2007. Climate Change 2007: Synthesis Report. In Interna-
tional Panel on Climate Change Fourth Assessment Report
(AR4): IPCC. Accessed on April 22 2014 at: http://www.ipcc.
ch/publications_and_data/publications_ipcc_fourth_assessmen-
t_report_synthesis_report.htm.
Ivy, D. J., Arnold, T., Harth, C. M., Steele, L. P., Mühle, J., Rigby,
M., Salameh, P. K., Leist, M., Krummel, P. B., Fraser, P. J.,
Weiss, R. F., and Prinn, R. G., 2012. Atmospheric histories and
growth trends of C4F10, C5F12, C6F14, C7F16 and C8F18. Atmo-
spheric Chemistry and Physics, 12(9), 4313–4325,
doi:10.5194/acp-12-4313-2012.
Johnson, J. E., 1999. Evaluation of a seawater equilibrator for ship-
board analysis of dissolved oceanic trace gases. Analytica
Chimica Acta, 395(1–2), 119–132, doi:10.1016/S0003-
2670(99)00361-X.
Johnston, C. T., Cook, P. G., Frape, S. K., Plummer, L. N.,
Busenberg, E., and Blackportm, R. J., 1998. Groundwater age
and nitrate distribution within a glacial aquifer beneath a thick
unsaturated zone. Ground Water, 36(1), 171–180.
Jurgens, B. C., Böhlke, J. K., and Eberts, S. M., 2012. TracerLPM
(Version 1): An Excel(R) Workbook for Interpreting
Groundwater Age Distributions from Environmental Tracer
Data. U.S. Geological Survey Techniques and Methods Report,
4-F3, 60 p.
315
Katz, B. G., Böhlke, J. K., and Hornsby, H. D., 2001. Timescales for
nitrate contamination of spring waters, northern Florida, USA.
Chemical Geology, 179(1–4), 167–186, doi:10.1016/s0009-
2541(01)00321-7.
Ko, M. K. W., Sze, N. D., Wang, W.-C., Shia, G., Goldman, A.,
Murcray, F. J., Murcray, D. G., and Rinsland, C. P., 1993. Atmo-
spheric sulfur hexafluoride: Sources, sinks and greenhouse
warming. Journal of Geophysical Research, [Atmospheres],
98(D6), 10499–10507, doi:10.1029/93jd00228.
Levin, I., Naegler, T., Heinz, R., Osusko, D., Cuevas, E., Engel, A.,
Ilmberger, J., Langenfelds, R. L., Neininger, B., Rohden, C. V.,
Steele, L. P., Weller, R., Worthy, D. E., and Zimov, S. A.,
2010. The global SF6 source inferred from long-term high preci-
sion atmospheric measurements and its comparison with emis-
sion inventories. Atmospheric Chemistry and Physics, 10(6),
2655–2662, doi:10.5194/acp-10-2655-2010.
Lindsey, B. D., Phillips, S. W., Donnelly, C. A., Speiran, G. K.,
Plummer, L. N., Böhlke, J. K., Focazio, M. J., Burton, W. C.,
and Busenberg, E., 2003. Residence times and nitrate transport
in groundwater discharging to streams in the Chesapeake Bay
Watershed. In U. S. Geological Survey Water-Resources Investi-
gations Report 2003–4035. Reston: U.S. Geological Survey.
Liu, J., Chen, Z., Wei, W., Zhang, Y., Li, Z., Liu, F., and Guo, H.,
2013. Using chlorofluorocarbons (CFCs) and tritium (3H) to
estimate groundwater age and flow velocity in Hohhot Basin,
China. Hydrological Processes:n/a–n/a. doi:10.1002/hyp.9659.
Loose, B., Stute, M., Alexander, P., and Smethie, W. M., 2009.
Design and deployment of a portable membrane equilibrator
for sampling aqueous dissolved gases. Water Resources
Research, 45(4), W00D34, doi:10.1029/2008wr006969.
Lovelock, J. E., 1958. A sensitive detector for gas chromatography.
Journal of Chromatography A, 1, 35–46, doi:10.1016/S0021-
9673(00)93398-3.
Lovelock, J. E., 1974. The electron capture detector: theory and
practice. Journal of Chromatography A, 99, 3–12,
doi:10.1016/S0021-9673(00)90840-9.
Lovley, D. R., and Woodward, J. C., 1992. Consumption of Freons
CFC-11 and CFC-12 by anaerobic sediments and soils. Environ-
mental Science and Technology, 26(5), 925–929, doi:10.1021/
es00029a009.
Maiss, M., and Brenninkmeijer, C. A. M., 1998. Atmospheric SF6:
trends, sources, and prospects. Environmental Science and Tech-
nology, 32(20), 3077–3086, doi:10.1021/es9802807.
Maiss, M., Steele, L. P., Francey, R. J., Fraser, P. J., Langenfelds,
R. L., Trivett, N. B. A., and Levin, I., 1996. Sulfur hexafluoride –
a powerful new atmospheric tracer. Atmospheric Environment,
30(10–11), 1621–1629, doi:10.1016/1352-2310(95)00425-4.
Manning, A. H., Clark, J. F., Diaz, S. H., Rademacher, L. K.,
Earman, S., and Niel Plummer, L., 2012. Evolution of ground-
water age in a mountain watershed over a period of thirteen
years. Journal of Hydrology, 460–461, 13–28, doi:10.1016/j.
jhydrol.2012.06.030.
McCarthy, R. L., Bower, F. A., and Jesson, J. P., 1977. The
fluorocarbon-ozone theory – I. Production and release – world
production and release of CCl3F and CCl2F2 (fluorocarbons
11 and 12) through 1975. Atmospheric Environment (1967),
11(6), 491–497, doi:10.1016/0004-6981(77)90065-8.
McMahon, P., Plummer, L., Böhlke, J., Shapiro, S., and Hinkle, S.,
2011. A comparison of recharge rates in aquifers of the United
States based on groundwater-age data. Hydrogeology Journal,
1–22, doi:10.1007/s10040-011-0722-5.
Miller, B. R., Weiss, R. F., Salameh, P. K., Tanhua, T., Greally,
B. R., Muhle, J., and Simmonds, P. G., 2008. Medusa:
a sample preconcentration and GC/MS detector system for in
situ measurements of atmospheric trace halocarbons, hydrocar-
bons, and sulfur compounds. Analytical Chemistry, 80(5),
1536–1545, doi:10.1021/ac702084k.
316 GROUNDWATER DATING WITH ATMOSPHERIC HALOGENATED COMPOUNDS
Montzka, S. A., and Reimann, S., 2010. Ozone-depleting sub-
stances (ODSs) and related chemicals. In Ajavon, A.-L. N.,
Newman, P. A., Pyle, J. A., and Ravishankara, A. R. (eds.), Sci-
entific Assessment of Ozone Depletion: 2010. Geneva: World
Meteorological Organization.
Montzka, S. A., Dlugokencky, E. J., and Butler, J. H., 2011.
Non-CO2 greenhouse gases and climate change. Nature,
476(7358), 43–50.
Mühle, J., Ganesan, A. L., Miller, B. R., Salameh, P. K., Harth,
C. M., Greally, B. R., Rigby, M., Porter, L. W., Steele, L. P.,
Trudinger, C. M., Krummel, P. B., O’Doherty, S., Fraser, P. J.,
Simmonds, P. G., Prinn, R. G., and Weiss, R. F., 2010. Perfluor-
ocarbons in the global atmosphere: tetrafluoromethane,
hexafluoroethane, and octafluoropropane. Atmospheric Chemis-
try and Physics, 10(11), 5145–5164, doi:10.5194/acp-10-5145-
2010.
Nelms, D. L., Harlow, G. E., Plummer, L. N., and Busenberg, E.,
2003. Aquifer susceptibility in Virginia, 1998–2000. In
U. S. Geological Survey Water-Resources Investigations Report
03–4278. Reston: U.S. Geological Survey.
Newman, B. D., Osenbrück, K., Aeschbach-Hertig, W., Kip
.
Solomon, D., Cook, P., Rózański, K., and Kipfer, R., 2010.
Dating of ‘young’ groundwaters using environmental tracers:
advantages, applications, and research needs. Isotopes in Envi-
ronmental and Health Studies, 46(3), 259–278, doi:10.1080/
10256016.2010.514339.
Nydal, R., and Lövseth, K., 1983. Tracing bomb 14C in the atmo-
sphere 1962–1980. Journal of Geophysical Research: Oceans,
88(C6), 3621–3642, doi:10.1029/JC088iC06p03621.
Ohta, T., Mahara, Y., Momoshima, N., Inoue, F., Shimada, J.,
Ikawa, R., and Taniguchi, M., 2009. Separation of dissolved
Kr from a water sample by means of a hollow fiber membrane.
Journal of Hydrology, 376(1–2), 152–158, doi:10.1016/j.
jhydrol.2009.07.022.
Oster, H., Sonntag, C., and Münnich, K. O., 1996. Groundwater age
dating with chlorofluorocarbons. Water Resources Research,
32(10), 2989–3001, doi:10.1029/96wr01775.
Plummer, L. N., 2005. Dating of young groundwater. In Aggarwal,
P. K., Gat, J. R., and Froehlich, K. F. O. (eds.), Isotopes in the
Water Cycle: Past, Present and Future of a Developing Science.
Dordrecht: Springer, pp. 193–220.
Plummer, L. N., and Friedman, L. C., 1999. Tracing and dating
young groundwater. USGS Fact Sheet 134–99. U.S. Geological
Survey, Reston VA. Accessed on April 22, 2014: http://pubs.
usgs.gov/fs/FS-134-99/pdf/fs-134-99.pdf.
Plummer, L. N., Michel, R. L., Thurman, E. M., and Glynn, P. D.,
1993. Environmental tracers for age dating young groundwater.
In Alley, W. (ed.), Regional Ground-Water Quality. New York:
Van Nostrand Reinhold, pp. 255–294.
Plummer, L. N., Rupert, M. G., Busenberg, E., and Schlosser, P.,
2000. Age of irrigation water in groundwater from the Eastern
Snake River Plain Aquifer, south-central Idaho. Ground Water,
28(2), 20.
Plummer, L. N., Busenberg, E., Eberts, S. M., Bexfield, L. M.,
Brown, C. J., Fahlquist, L. S., Katz, B. G., and Landon, M. K.,
2008. Low-level detections of halogenated volatile organic com-
pounds in groundwater: use in vulnerability assessments. Jour-
nal of Hydrologic Engineering, 13(11), 1049–1068,
doi:10.1061/(asce)1084-0699(2008)13:11(1049).
Prather, M. J., and Hsu, J., 2008. NF3, the greenhouse gas missing
from Kyoto. Geophysical Research Letters, 35(12), L12810,
doi:10.1029/2008gl034542.
Preston, S. P., 2003. Sulfur hexafluoride, MIT, and the atomic
bomb. Chemical Heritage, 2(2), 30–36.
Probst, P.C., Yokochi, R., Sturchio, N. C., 2007. Method for extrac-
tion of dissolved gases from groundwater for radiokrypton anal-
ysis. In Lippmann-Pipke, J., and Aeschbach-Hertig, W. (eds.),
Proceedings of the 4th Mini Conference on Noble Gases in the
Hydrosphere and in Natural Gas Reservoirs, GERMANY:
GFZ Potsdam.
Reilly, T. E., Plummer, L. N., Phillips, P. J., and Busenberg, E.,
1994. The use of simulation and multiple environmental tracers
to quantify groundwater flow in a shallow aquifer. Water
Resources Research, 30(2), 421–433, doi:10.1029/93wr02655.
Roberts, R. A., and Ramsey, P. J., 2013. Evaluation of fluorocarbon
emissions from the aluminum smelting process. In Essential
Readings in Light Metals. Hoboken: Wiley, pp. 1007–1014.
Santoro, M. A., 2000. Clarifying the SF5CF3 record. Science,
290(5493), 935–936, doi:10.1126/science.290.5493.935.
Schauffler, S. M., Heidt, L. E., Pollock, W. H., Gilpin, T. M.,
Vedder, J. F., Solomon, S., Lueb, R. A., and Atlas, E. L., 1993.
Measurements of halogenated organic compounds near the trop-
ical tropopause. Geophysical Research Letters, 20(22),
2567–2570, doi:10.1029/93gl02840.
Sebol, L. A., Robertson, W. D., Busenberg, E., Plummer, L. N.,
Ryan, M. C., and Schiff, S. L., 2007. Evidence of CFC degrada-
tion in groundwater under pyrite-oxidizing conditions. Journal
of Hydrology, 347(1–2), 1–12, doi:10.1016/j.
jhydrol.2007.08.009.
Seinfeld, J. H., and Pandis, S. N., 1998. The atmosphere. In Atmo-
spheric Chemistry and Physics. New York: Wiley, pp. 1–47.
Shapiro, S. D., Busenberg, E., Focazio, M. J., and Plummer, L. N.,
2004. Historical trends in occurrence and atmospheric inputs
of halogenated volatile organic compounds in untreated
groundwater used as a source of drinking water. Science of the
Total Environment, 321(1–3), 201–217, doi:10.1016/j.
scitotenv.2003.09.007.
Simmonds, P. G., Greally, B. R., Olivier, S., Nickless, G., Cooke,
K. M., and Dietz, R. N., 2002. The background atmospheric
concentrations of cyclic perfluorocarbon tracers determined
by negative ion-chemical ionization mass spectrometry. Atmo-
spheric Environment, 36(13), 2147–2156, doi:10.1016/S1352-
2310(02)00194-2.
Smith, M. J., Ho, D. T., Law, C. S., McGregor, J., Popinet, S., and
Schlosser, P., 2011. Uncertainties in gas exchange parameteriza-
tion during the SAGE dual-tracer experiment. Deep Sea
Research Part II: Topical Studies in Oceanography, 58(6),
869–881, doi:10.1016/j.dsr2.2010.10.025.
Solomon, D. K., Hunt, A., and Poreda, R. J., 1996. Source of radio-
genic helium 4 in shallow aquifers: implications for dating
young groundwater. Water Resources Research, 32(6),
1805–1813, doi:10.1029/96wr00600.
Sousa, S. R., and Bialkowski, S. E., 1997. Temperature-dependent
electron capture detector response to common alternative
fluorocarbons. Analytical Chemistry, 69(19), 3871–3878,
doi:10.1021/ac9703921.
Straume, A. G., Dietz, R. N., Koff, E. N. D., and Nodop, K., 1998.
Perfluorocarbon background concentrations in Europe. Atmo-
spheric Environment, 32(24), 4109–4122, doi:10.1016/S1352-
2310(98)00193-9.
Sturges, W. T., Wallington, T. J., Hurley, M. D., Shine, K. P., Sihra,
K., Engel, A., Oram, D. E., Penkett, S. A., Mulvaney, R., and
Brenninkmeijer, C. A. M., 2000. A potent greenhouse gas iden-
tified in the atmosphere: SF5CF3. Science, 289(5479), 611–613,
doi:10.1126/science.289.5479.611.
Sturges, W. T., Oram, D. E., Laube, J. C., Reeves, C. E., Newland,
M. J., Hogan, C., Martinerie, P., Witrant, E., Brenninkmeijer,
C. A. M., Schuck, T. J., and Fraser, P. J., 2012. Emissions halted
of the potent greenhouse gas SF5CF3. Atmospheric Chemistry
and Physics, 12(8), 3653–3658, doi:10.5194/acp-12-3653-
2012.
Stute, M., and Schlosser, P., 1993. Principles and applications of the
noble gas paleothermometer. In Swart, P. K., Lohmann, K. C.,
McKenzie, J., and Savin, S. (eds.), Climate Change in
 GROUNDWATER DATING WITH ATMOSPHERIC HALOGENATED COMPOUNDS
Continental Isotopic Records, Vol. 374. Geophys. Monogr. Ser.
Washington, DC: AGU.
Stute, M., Clark, J. F., Schlosser, P., Broecker, W. S., and Bonani,
G., 1995. A 30,000 year continental paleotemperature record
derived from noble gases dissolved in groundwater from the
San Juan Basin, New Mexico. Quaternary Research, 43(2),
209–220, doi:10.1006/qres.1995.1021.
Sun, T., Hall, C. M., and Castro, M. C., 2010. Statistical properties
of groundwater noble gas paleoclimate models: Are they robust
and unbiased estimators? Geochemistry, Geophysics,
Geosystems, 11(2), Q02002, doi:10.1029/2009gc002717.
Szabo, Z., Rice, D. E., Plummer, L. N., Busenberg, E., Drenkard, S.,
and Schlosser, P., 1996. Age dating of shallow groundwater with
Chlorofluorocarbons, Tritium/Helium: 3, and flow path analysis,
Southern New Jersey Coastal Plain. Water Resources Research,
32(4), 1023–1038, doi:10.1029/96wr00068.
Thompson, G. M., and Hayes, J. M., 1979. Trichlorofluoromethane
in groundwater – a possible tracer and indicator of groundwater
age. Water Resources Research, 15(3), 546–554, doi:10.1029/
WR015i003p00546.
Turnbull, J. C., Lehman, S. J., Miller, J. B., Sparks, R. J., Southon,
J. R., and Tans, P. P., 2007. A new high precision 14CO2 time
series for North American continental air. Journal of Geophysi-
cal Research, [Atmospheres], 112(D11), D11310, doi:10.1029/
2006jd008184.
von Rohden, C., Kreuzer, A., Chen, Z., and Aeschbach-Hertig, W.,
2010a. Accumulation of natural SF6 in the sedimentary aquifers
of the North China Plain as a restriction on groundwater dating.
Isotopes in Environmental and Health Studies, 46(3), 279–290,
doi:10.1080/10256016.2010.494771.
von Rohden, C., Kreuzer, A., Chen, Z., Kipfer, R., and Aeschbach-
Hertig, W., 2010b. Characterizing the recharge regime of the
strongly exploited aquifers of the North China Plain by environ-
mental tracers. Water Resources Research, 46(5), W05511,
doi:10.1029/2008wr007660.
Warner, M. J., and Weiss, R. F., 1985. Solubilities of chlorofluoro-
carbons 11 and 12 in water and seawater. Deep Sea Research
Part A: Oceanographic Research Papers, 32(12), 1485–1497,
doi:10.1016/0198-0149(85)90099-8.
Watson, T. B., Wilke, R., Dietz, R. N., Heiser, J., and Kalb, P., 2007.
The atmospheric background of perfluorocarbon compounds
used as tracers. Environmental Science and Technology,
41(20), 6909–6913, doi:10.1021/es070940k.
Weiss, R. F., 1970. The solubility of nitrogen, oxygen and argon in
water and seawater. Deep Sea Research and Oceanographic
Abstracts, 17(4), 721–735, doi:10.1016/0011-7471(70)90037-9.
Weiss, R. F., and Price, B. A., 1980. Nitrous oxide solubility in
water and seawater. Marine Chemistry, 8(4), 347–359,
doi:10.1016/0304-4203(80)90024-9.
317
Weiss, R. F., Mühle, J., Salameh, P. K., and Harth, C. M.,
2008. Nitrogen trifluoride in the global atmosphere. Geophysical
Research Letters, 35(20), L20821, doi:10.1029/2008gl035913.
Weissmann, G. S., Zhang, Y., LaBolle, E. M., and Fogg, G. E., 2002.
Dispersion of groundwater age in an alluvial aquifer system.
Water Resources Research, 38(10), 1198, doi:10.1029/
2001wr000907.
Wilson, R. D., and Mackay, D. M., 1996. SF6 as a conservative
tracer in saturated media with high intragranular porosity or high
organic carbon content. Ground Water, 34(2), 241–249.
Winger, K., Feichter, J., Kalinowski, M. B., Sartorius, H.,
and Schlosser, C., 2005. A new compilation of the
atmospheric 85-krypton inventories from 1945 to 2000 and its
evaluation in a global transport model. Journal of Environmental
Radioactivity, 80(2), 183–215, doi:10.1016/j.jenvrad.2004.
09.005.
WMO, 2010. Scientific assessment of ozone depletion: 2010. In
Ajavon, A.-L. N., Newman, P. A., Pyle, J. A., and Ravishankara,
A. R. (eds.), Scientific Assessment of Ozone Depletion: 2010.
Geneva: World Meteorological Organization.
Worton, D. R., Sturges, W. T., Gohar, L. K., Shine, K. P., Martinerie,
P., Oram, D. E., Humphrey, S. P., Begley, P., Gunn, L., Barnola,
J.-M., Schwander, J., and Mulvaney, R., 2007. Atmospheric
trends and radiative forcings of CF4 and C2F6 inferred from Firn
air. Environmental Science and Technology, 41(7), 2184–2189,
doi:10.1021/es061710t.
Zuber, A., 1986. Mathematical models for the interpretation of envi-
ronmental radioisotopes in groundwater systems. In Fritz, P., and
Fontes, J.-C. (eds.), Handbook of Environmental Isotope
Geochemistry. Amsterdam: Elsevier, pp. 1–59.
Zuber, A., and Rosanski, K., 2007. Groundwater dating with the aid
of tracers: strategies, pitfalls, and open problems. In: Lippmann-
Pipke, J., and Aeschbach-Hertig, W. (eds.) The 4th Mini Confer-
ence on Noble Gases in the Hydrosphere and in Natural Gas
Reservoirs. GERMANY: GFZ Potsdam.
Zuber, A., Weise, S. M., Osenbrück, K., Grabczak, J., and
.
Cizkowski, W., 1995. Age and recharge area of thermal waters
in Ladek Spa (Sudeten, Poland) deduced from environmental
isotope and noble gas data. Journal of Hydrology, 167(1–4),
327–349, doi:10.1016/0022-1694(94)02587-2.
Cross-references
Bomb Carbon
Groundwater Dating with Atmospheric Halogenated Compounds
Historical Development of Dating Methods
Noble Gas Mass Spectrometer
U–Th/He Dating
 H
HISTORICAL DEVELOPMENT OF DATING METHODS
James M. Mattinson
Department of Earth Science, University of California,
Santa Barbara, CA, USA
Synonyms
Absolute dating; Isotopic dating; Isotopic geochronology;
Radio-isotope dating; Radiometric dating
Definition
H. S. Williams (1893) originally proposed the term “geo-
chronology” for the study of the time scale of geological
events. Williams coined the term just 3 years prior to the
discovery of radioactivity, at a time when science-based
estimates for the age of the Earth and the age of major geo-
logic events were derived from various physical and
chemical models, and a range of geologic observations
and measurements. The discovery of radioactivity in
1896 by the French physicist, Henri Becquerel, led to the
development of dating methods primarily based on
the decay of long-lived, naturally occurring radioactive
“parent isotopes” to stable “daughter isotopes.” In modern
usage, “geochronology” is used more or less exclusively
to refer to dating geological materials using methods
based on radioactive decay.
Introduction
To quote Frank Richter’s paper on “Kelvin and the age of
the Earth” (1986, p. 395): “The true vastness of geologic
time, and the overthrow of the Biblical chronology that
would measure it in generations of man, is in my view
the most profound of geology’s many contributions.” In
fact, the problem of determining the age of the Earth is
also an ideal framework for a discussion of the history of
geochronology. And the discussion is a lively one, replete
J.W. Rink, J.W. Thompson (eds.), Encyclopedia of Scientific Dating Methods, DOI 10.1007/978-94-007-6304-3,
© Springer Science+Business Media Dordrecht 2015
with religious repression; battles among the titans of phys-
ics, biology, and geology; the dawn of a new scientific age
with the discovery of radioactivity; and finally, the emer-
gence of modern geochronology based on radioactive
decay systems. This contribution will begin with a brief
discussion of the influence of theological chronologies
for the age of the Earth, followed by the gradual emer-
gence of science-based models for the origin, nature, and
age of the Earth and solar system in the face of religious
persecution and, later, less lethal religious disapproval.
Next is a discussion of the great scientific battles on the
age of the Earth from the mid-1800s into the early 1900s
that pitted leading physicists such as Lord Kelvin against
the leading geologists of his day, and also against Charles
Darwin and his theory of origin of the species and natural
selection. Interesting subplots involve George Darwin
(second son of Charles Darwin) and also John Perry
(former protégé of, and assistant to Kelvin). Then came
the scientific revolution that began with the discovery of
radioactivity by Henri Becquerel in 1896, revolutionizing
physics and giving birth to geochronology in its modern
sense. Finally, over the past 100 years or so, “modern”
geochronology has developed and matured into an essen-
tial field for the study of Earth, moon, and solar system
origin and history.
Each section includes specific references within the text
as needed, and a list of more general references at the end
of the section.
Age of the earth from religious sources
Many ancient civilizations developed creation stories,
some including specific ideas about time. The ancient
Greeks, Hindus, Chaldeans, and Mayans believed in infi-
nite time, usually in the form of repeated cycles, each
one lasting millions of years. The longest cycle in the
Mayan “long count” calendar lasted ca. 6.3  107 year.
However, far more important in the western world over
320 HISTORICAL DEVELOPMENT OF DATING METHODS
the past 2,000 years or so were numerous calculations for
the date of creation of the Earth and humans, using as
a starting point Biblical sources such as the Book of
Genesis. Beginning in about 169 A.D., church scholars
calculated a series of very young ages for the Earth rang-
ing from ca. 6,000 to 9,000 years BP.
James Ussher (1581–1656), an Archbishop in the Irish
Anglican Church, published the most influential “age of
the Earth” in the 1650s. Ussher’s date of October 22,
4004 B.C. for the creation of the Earth was not the first
ca. 4000 B.C. calculation, but it was inserted as
a marginal note into the 1701 edition of the English Bible,
some 45 years after Ussher’s death. Insertion into the
Bible, even as a marginal note, gave the 4004 B.C. date
all the weight of inerrant Biblical truth to those who
believed in a literal, word for word interpretation of the
Bible. This had a profound impact on the emergence of
scientific inquiry. Long before Ussher’s date was embed-
ded in the Bible, holding, teaching, and/or publishing
views that conflicted in any way with church dogma could
lead to charges of heresy. However, the threat of persecu-
tion fluctuated, depending whether the Roman Catholic
Church had more serious issues to deal with at any given
time. For example, Copernicus (1473–1543) published
his book on the heliocentric theory of the solar system in
1543. He actually dedicated the book to the sitting Pope,
even though the heliocentric model was obviously in
direct conflict with the geocentric solar system/universe
of Genesis. However, at the time, the Roman Catholic
Church had other, more pressing issues to deal with.
The “Protestant Reformation” of Martin Luther began
in ca. 1517 with the printing and posting of Luther’s “dis-
putations.” The movement spread rapidly, and grew into
a more and more serious threat to the Roman Catholic
Church. In the 1530s, Henry VIII implemented the
“English Reformation” and also broke away from the
Roman Catholic Church. Little wonder then that Coperni-
can theory was a relatively low priority for Church inves-
tigation. As a result, the work of Copernicus, and those
who supported his work, such as Galileo, escaped formal
condemnation for many decades. By ca. 1545, only 2
years after the death of Copernicus, the Catholic Church
began a formal “Counter Reformation.” However, when
reformation movements continued to grow, even spread-
ing into Italy, the Roman Catholic Church took a more
extreme step, instituting a new inquisition, the “Roman
Inquisition,” in 1588. In the early years of the inquisition,
heliocentrism was still not the highest priority, but in 1616
the Church finally got around to Copernicus (73 years
after his death!) formally denouncing the heliocentric
model of Copernicus as “dangerous to faith” and banning
his book.
At the time of the denunciation and ban, Galileo already
had been publishing material in support of Copernicus for
several years, based in part on his own telescopic observa-
tions. Galileo was required to appear before a powerful
Cardinal who ordered him not to “hold or defend” Coper-
nican theory. Presumably, Galileo complied with this
order for several years. However, in 1623, an old friend
of Galileo’s became Pope Urban VIII, and Galileo was
emboldened to publish about Copernican theory in
a way that he thought might not directly violate the Cardi-
nal’s orders. In 1632, Galileo published a book on Coper-
nican theory, thinly disguised as a conversation between
fictional characters, one arguing for, and the other against
Copernican theory. One character, “Salviati” (obviously
Galileo), brilliantly presented evidence supporting Coper-
nican theory, including detailed astronomical observa-
tions, graphs, etc. The other character, the unfortunately
named “Simplicio” (representing the Church, perhaps
even the Pope himself), comes across as dim-witted,
befuddled by Salviati’s arguments, and able only to pre-
sent the standard church position in opposition to Coperni-
can theory. The one-sided nature of the “debate” was all
too obvious. Galileo was examined by the inquisition,
found guilty, forced to recant, his book was banned, and
he was placed under house arrest for the rest of his life.
It took just over 200 years for Galileo’s book to be
removed from the banned book list. Little wonder then,
that for many years, most observers of the natural world
carefully fitted their interpretations into the time frame of
Archbishop Ussher, either from true belief or self-preser-
vation. To quote Burchfield (1990, p. 5): “The question
to be answered by the natural philosopher was not how
long it had taken for geological forces to sculpture the face
of the Earth, but how they had accomplished their work in
the few millennia since Creation.”
General references (listed alphabetically): Albritton
(1980), Burchfield (1990), Dalrymple (1991), Holmes
(1913), Stanford Encyclopedia of Philosophy (http://
plato.stanford.edu/entries/copernicus), Stanford Encyclo-
pedia of Philosophy (http://plato.stanford.edu/entries/
galileo).
Age of the earth from early scientific methods:
pre-Kelvin
Eventually, scientists began to push more openly against
the constraints placed on them by an Earth only several
thousand years old. As the science of geology began to
emerge, in particular, the study of processes such as uplift,
erosion, and sedimentation, the concept of a far more
ancient Earth began to emerge. Initially, caution was still
required. For example, Benoit de Maillet (1656–1738),
French diplomat and naturalist, traveled widely as
a diplomat and made many observations of geology and
geography around the Mediterranean. He concluded that
the Earth must be greatly older than the Ussher Biblical
chronology suggested, perhaps on the order of 2 Ga. He
presented his theory, based on a theory of steady gradual
lowering of sea level over geologic time, in the form of fic-
tional conversations between a French missionary and an
Indian sage named “Telliamed” (de Maillet spelled back-
ward). Having been born only 14 years after the death of
Galileo, de Maillet would have been well aware of
Galileo’s persecution, so he took the additional precaution
 HISTORICAL DEVELOPMENT OF DATING METHODS
of arranging for his own book to be published in 1748,
10 years after his own death.
One of the most remarkable scientific figures of the
eighteenth century was Georges-Louis Leclerc
(1707–1788), later the “compte de Buffon,” courtesy of
King Louis XV of France. In 1749, just 1 year after the
posthumous publication of de Maillet’s theory, Buffon
published the first 3 of his 35 volumes on “Natural
History. . ..” Buffon, who would become one of the giants
of the Enlightenment, proposed a detailed theory for the
origin of the planets that was clearly at odds with Genesis.
However, he claimed that his theory was only “pure phil-
osophical speculation.” This, plus the fact that Buffon had
the strong support and protection of Louis XV, saved him
from any serious action by the Church. Of even more
interest from the perspective of geochronology are the sci-
entific experiments Buffon conducted to provide some
actual numerical estimates of the age of the Earth. In addi-
tion to his many other talents and enterprises, Buffon
owned an iron foundry. On the theory that the Earth had
cooled from an originally molten state, Buffon had his
workers make a series of spheres of graduated sizes, some
of pure iron, others mixtures of iron and silicates. He had
the spheres heated to just below the melting point, then
measured the time required for the surface of each sphere
to cool to the point where it could be touched with a bare
hand without burning the fingers, then the time for the sur-
face to reach room temperature. From these data, Buffon
extrapolated how long it would take for a sphere the size
of the Earth to cool. The “short version” of the results,
published in 1778, gave a result of 75 Ka for the Earth to
cool to its present temperature. “Long chronologies”
found in Buffon’s unpublished manuscripts give results
as high as 3 Ma. Why the large discrepancy? “Internal evi-
dence strongly suggests that Buffon favored the longer
version. Why, then, did he elect to publish the shorter?
. . .. The best guess is that he thought the lower figure
was about as much as the traffic would bear.” (Albritton,
1980, pp. 85–86).
With the emergence of “the Enlightenment,” and espe-
cially in non-Catholic countries, it became more accept-
able to challenge the Biblical chronologies. For example,
James Hutton (1726–1797) published a paper in 1785 that
would revolutionize geologic thinking. Hutton’s studies
convinced him that ordinary geologic processes acting
over vast amounts of time could explain all of his observa-
tions. Hutton saw no need for great Biblical catastrophes,
such as Noah’s flood. Working in Scotland, Hutton did not
have to fear an inquisition, but religious beliefs were still
strongly held by many. To paraphrase one of Hutton’s
most caustic critics: “. . .the suspicion of the high antiquity
of the globe has been fatal to Mosaic (i.e., Biblical) his-
tory, and consequently to religion and morality” (see
Burchfield, 1990, p. 7, emphasis added). Charles Lyell
(1797–1875) published his Principles of Geology in
1830, reinforcing and popularizing Hutton’s concept of
gradual changes over “inconceivably vast” periods of
time. Charles Darwin (1809–1882), a contemporary,
321
friend, and admirer of Lyell, published On the Origin
of Species in 1859, depending heavily on the availability
of great expanses of time for the operation of natural
selection.
General references: Albritton (1980), Burchfield
(1990), Dalrymple (1991), Holmes (1913).
Age of the earth from early scientific methods:
the Kelvin era
The writings of Hutton, Lyell, and, in particular, On the
Origin of Species by Darwin served as a major catalyst
for a great flowering of inventive approaches for estimat-
ing the age of the Earth (or the oceans, or the sun, or the
moon, or the Earth “as an abode fitted for life”) from the
mid-1800s to the early 1900s. Scientists applied physics,
chemistry, and geology to the “age of the Earth problem.”
The age estimates based on a wide range of approaches
varied over orders of magnitude and sparked several
decades of heated and sometimes bitter controversy. The
central figure was William Thomson (1824–1907), later
known as Lord Kelvin, courtesy of Queen Victoria’s
1892 elevation of Thomson to a “peer of the realm.”
Kelvin was widely regarded as the most brilliant phys-
icist of his era by his peers. Kelvin evidently strongly
agreed with this opinion, and he was particularly resistant
to changing any of his ideas, even in the face of new infor-
mation, which he did not always keep up with anyway:
“Kelvin’s failure to keep up with the literature even in
his own field was notorious. It led him frequently to repeat
experiments and discoveries already in the literature”
(Burchfield, 1990, p. 54). As with many other scientists
of his era, Kelvin had broad interests, including geology,
but geology’s lack of quantitative rigor annoyed him. He
particularly objected to “uniformitarianism,” as proposed
by Hutton and expanded by Lyell. The idea of terrestrial
conditions such as surface temperatures remaining largely
the same over unimaginably long periods of Earth history
was anathema to Kelvin. After all, Kelvin had discovered
the second law of thermodynamics. The Earth had to be
cooling, with a consequent slowing down of geologic pro-
cesses over time. Kelvin was also annoyed by Charles
Darwin’s demands for vast lengths of time for the process
of natural selection. Starting in ca. 1862, and extending for
over the 40 or so years, Kelvin fought a running battle with
geologists, biologists, and even fellow physicists. Starting
in the 1860s, Kelvin published a series of papers on the
age and energy source of the sun, the age and thermal his-
tory of the Earth, and the effects of tidal friction retarding
the rate of rotation of the Earth. All of these studies were
designed to place quantitative limits on the age of the
Earth as a planet, or “the age of the Earth as an abode fitted
for life.”
Kelvin’s work on the thermal history of the Earth is per-
haps the best known. His approach was similar to that of
Buffon, except that Kelvin used data from the Earth itself,
rather than from experiments on small spheres. Kelvin
assumed that the Earth initially was totally molten,
322 HISTORICAL DEVELOPMENT OF DATING METHODS
solidified completely from the core outward, then cooled
entirely by conduction to its present thermal state. He used
measured thermal gradients from wells and mines, mea-
sured thermal conductivities for a few common crustal
rock types (and assumed that these conductivities applied
to the entire interior of the Earth), and then applied the
mathematical approach to heat transfer published by Fou-
rier in 1822. Kelvin derived an “age” for the Earth (at least,
the time since the outermost crust solidified) of 98 Ma, but
allowed a range of 20–400 Ma, based on uncertainties of
his data. The ca. 100 Ma figure published by Kelvin in
1862 is widely quoted, but over the following decades,
Kelvin reduced both his estimate for the Earth’s age, and
the uncertainties. To quote Burchfield (1990, p. 43):
By 1868 he was convinced that sufficient evidence existed to
justify limiting the assumed duration of life on Earth, if not
the Earth’s total age, to no more than 100 million years.
In 1876, he was willing to accept an upper limit of only
50 million years for the Earth’s age, and in 1881 a limit of
20 to 50 million years. Finally, in 1897, he declared that the
Earth’s age was nearer 20 million than 40 million years,
and embraced Clarence King’s estimate of 24 million years
as the best available.
Also in 1862, Kelvin published estimated age limits of
10–500 Ma for the sun – quite similar to his original limits
for the age of the Earth of 20–400 Ma. Kelvin ruled out
chemical reactions as a source of energy for the sun and
proposed that the sun’s heat had been produced primarily
by gravitational accretion during its formation. Thus, he
viewed the sun, like the Earth, as a simple cooling body
with no internal source of energy. As Kelvin’s estimates
for the age of the Earth came down, he eagerly accepted
younger estimates of the sun’s age published by some of
his colleagues. As the assumptions of his original work
on the age of the Earth came under attack, he relied more
heavily on these estimates of the sun’s age to cling to his
original conclusions about the Earth.
One of the more fascinating approaches by Kelvin was
based on retardation of the Earth’s rotation by oceanic
tidal friction and the implications for the Earth’s equatorial
bulge. Kelvin pointed out that a totally molten Earth
would assume a polar flattening and equatorial bulge in
proportion to its rate of rotation. Kelvin then assumed that
when the Earth completely solidified, it would become
many times more rigid than steel and could no longer
respond to changes in the rate of rotation. Thus, Kelvin
reasoned, the equatorial bulge would be locked in at the
time of solidification. By knowing the rate of slowing of
the Earth’s rotation, one could theoretically calculate
when, in the past, the rate of rotation matched the suppos-
edly locked in (present) shape of the Earth. In fact, the pre-
sent shape of the Earth is consistent with its present rate of
rotation. Kelvin, recognizing that there were great uncer-
tainties in his approach, concluded that the data could only
limit the age of the Earth’s solidification to <1 Ga. What
he failed to recognize (or chose to ignore) was that
a “young” Earth was not the only explanation for the
agreement between the present shape of the Earth and
the present rate of rotation. The data were also consistent
with the present-day Earth being nonrigid – capable of
flowage and continuously adapting its shape to the
slowing rotation rate of the Earth.
One of the interesting “subplots” in the “age of the
Earth” debate was the involvement of George Darwin,
the second son of Charles Darwin. George Darwin, an
eminent and respected scientist in his own right, was
a professor of “astronomy and experimental philosophy”
at Cambridge for almost 30 years. Early in his career, he
had attracted the attention of Kelvin, who responded
enthusiastically to George’s first major paper on the
effects of geological change on the Earth’s rotation.
Kelvin encouraged George to analyze the tidal interac-
tions of the Earth-moon system, to follow up Kelvin’s
Earth-ocean tide study. The work led to George’s well-
known 1879 theory that the moon had spun off from the
Earth during an earlier era of more rapid rotation. The cal-
culations also gave a minimum age of 54–57 Ma for the
Earth/moon system to reach the present Earth/moon axial
rotation rate/orbital period, but Darwin pointed out that the
maximum age might be very much greater. As might be
expected, Darwin’s theory of lunar origin sparked consid-
erable interest by both the scientific community and the
general public. Unfortunately, popular versions tended to
be oversimplified: “Thus too little emphasis was given to
the fact that the figures 54 million and 57 million years
for the moon represented possible minima and in no way
limited the maximum age of the Earth-moon system.
Instead these figures passed into the literature as still
another proof that the Earth must be less than 100 million
years old” (Burchfield, 1990, p. 115). It is interesting that
George Darwin’s calculations were all based on a viscous,
nonrigid Earth. This stands in sharp contrast to Kelvin
who based both his cooling and tidal models on
a completely rigid Earth. Instead, Darwin’s model of
a viscous Earth was in accord with geological data that
were being used at about the same time to attack Kelvin’s
age(s) for the Earth.
General references: Albritton (1980), Burchfield
(1990), Dalrymple (1991), Holmes (1913).
Age of the earth from early scientific methods:
the geologists fight back
Some geologists (and a few physicists) began attacking
Kelvin’s assumptions. No one was willing to challenge
Kelvin’s prowess as a mathematician, but many believed
that Kelvin’s assumption of a rigid Earth, physically uni-
form throughout, was the Achilles heel of his model. For
example, the Reverend Osmond Fisher (1817–1914), an
early geophysicist from the long and rich tradition of theo-
logically trained naturalists, mounted an attack against
Kelvin. In “Physics of the Earth’s Crust,” published in
1881, Fisher summarized evidence for depression of the
Earth’s crust in Scandinavia by glacial ice-loading and
post-glacial rebound of the crust after the glaciers had
melted. It was obvious to Fisher that the geological
 HISTORICAL DEVELOPMENT OF DATING METHODS
evidence was completely incompatible with Kelvin’s rigid
Earth assumption, instead requiring the crust to float on
a viscous interior. Fisher continued his attacks in a series
of papers with titles such as “On depression of ice-loaded
lands,” published in 1882, and “Rigidity not to be relied on
in estimating the Earth’s age,” published in 1893.
Even more telling were the analyses of John Perry
(1850–1920). Perry was a brilliant mathematician and
also a former student, assistant, and collaborator of
Kelvin’s. Perry was obviously fond of Kelvin and thus
was reluctant to criticize Kelvin’s work on the age of the
Earth, making this another interesting “subplot” in the
controversy over age of the Earth: “Some of my friends
have blamed me severely for not publishing the above
document sooner. I was Lord Kelvin’s pupil, and am still
his affectionate pupil. . .. He has been uniformly kind to
me, and there have been times when he must have found
this difficult” (Perry, 1895). Affection and respect not-
withstanding, Perry was particularly concerned with
Kelvin’s simplistic assumptions that the Earth’s physical
properties, including thermal conductivity, were uniform
throughout and that the Earth was rigid, with no possibil-
ity for convection. Suppose “quasi-conductivity” (con-
duction plus some convection) allowed the transfer of
heat from the Earth’s interior to the base of the crust at
ten times the rate based on Kelvin’s measurements of con-
ductivity of typical upper crustal rocks? Perry (1895) dem-
onstrated that such an increase in heat transfer from the
Earth’s deep interior would increase Kelvin’s calculated
age for the Earth by a factor of 56. Perry’s goal was not
to calculate a specific age for the Earth but to show that
Kelvin greatly underestimated his upper limit for the age
of the Earth. Perry (1895, p. 227) includes a letter from
Kelvin, in which Kelvin more or less concedes this point:
“. . .I thought my range from 20 millions to 400 millions
was probably wide enough, but it is quite possible that
I should have put the superior limit a good deal higher,
perhaps 4000 instead of 400. . ..” However, Kelvin
quickly reverted to much younger age limits. Even after
the discovery of radioactivity, the production of heat from
radioactive decay, and publications of the first dates based
on radioactive decay, Kelvin clung tenaciously to his
earlier ideas.
Meanwhile, geologists had devised methods of their
own to calculate the age of the Earth. Particularly popular
were methods based on observed thicknesses of sedimen-
tary rock coupled with estimated rates of erosion and
sedimentation and methods based on ocean chemistry,
predominantly those based on sodium accumulation.
Excellent, detailed, and very readable accounts have been
provided by Albritton (1980), Burchfield (1990), and
Dalrymple (1991) and are very briefly noted here.
Between 1860 and 1917, some 32 papers by 21 different
authors estimated the ages of various parts of the sedimen-
tary record, based on erosion and sedimentation rates
(Dalrymple, 1991, Table 2.1). Results varied wildly from
ca. 3 Ma to ca. 15 Ga, but some efforts were remarkably
accurate. For example, John Goodchild in 1897 published
323
a detailed analysis of the Phanerozoic sedimentary record.
To quote Burchfield (1990, p. 148): “Goodchild’s
approach was eclectic. Borrowing freely from the methods
pioneered by Phillips, Reade, Geike, and others, he deter-
mined the duration of each separate geological period
using what he regarded as the most appropriate and reli-
able method for each. . .. In all, he determined the ages
of 15 geological periods since the beginning of the Cam-
brian, totaling just over 700 million years. . ..” This is
a stunning result, considering that our present age for the
beginning of the Cambrian is just over 540 Ma, less than
25 % below Goodchild’s figure. Goodchild also estimated
“at least an equal duration for the Precambrian” for an age
of the Earth >1.4 Ga. However, a significant number of
the other estimates based on sedimentation and erosion
did fall within Kelvin’s original 20–400 Ma age limits.
Between 1876 and 1931, some 12 papers by nine differ-
ent authors used the chemistry of the oceans to place limits
on the age of the Earth (Dalrymple, 1991, Table 2.1). Most
of the studies used methods based on sodium in the
oceans. Perhaps the best-known results and those most
widely quoted in elementary texts were by John Joly
(1857–1933). In 1899, Joly proposed a simple model:
(1) that the primordial oceans were freshwater, condensed
from the Earth’s atmosphere when surface temperatures
dropped below the boiling point of water soon after the
Earth’s crust solidified, and (2) that over the remainder
of the Earth’s history, the oceans gradually became more
and more salty by the addition of sodium weathered from
the continents and carried into the ocean by rivers. The
concentration of Na in the oceans was already well known,
the volume of the oceans had been estimated, and the Na
contents and discharge rates of most major rivers were
approximately known. Joly had only to divide the total
Na in the oceans by the annual Na flux of the rivers to esti-
mate the age of the oceans, and the time elapsed since the
Earth had cooled below the boiling point of water. Apply-
ing a few corrections and allowing for uncertainties in the
data, Joly calculated that the oceans had an age between
80 and 90 Ma. In 1900, Joly adjusted his estimate slightly
upward, to 90–100 Ma. Comparable calculations by other
authors using slightly different inputs arrived at more or
less identical results – ca. 100 Ma. Again, these estimates
were in agreement with Kelvin’s original age estimates for
the Earth and sun. Some geologists were ready to
accept an age for the Earth of ca. 100 Ma and do their best
to fit all of Earth history into this “modest” time frame.
However, Kelvin, Tate, King, and others kept reducing
their estimates (still based on calculations of heat loss),
limiting the age of the Earth and sun to only 10–40 Ma.
As mentioned earlier, Kelvin’s final word on the subject
in 1897 was that Clarence King’s 24 Ma age for the Earth,
published in 1893, was the best available.
However, the scientific world was about to change dra-
matically. In the few years encompassing Perry’s (1895)
critique of Kelvin’s assumptions, Goodchild’s 1897 esti-
mate of the age of the Phanerozoic, Kelvin’s 1897 “final
word,” and Joly’s 1899 age of the oceans, a spectacular
324 HISTORICAL DEVELOPMENT OF DATING METHODS
scientific revolution was beginning. The discovery of
x-rays by Roentgen in 1895 led directly to the discovery
of radioactivity by Becquerel in 1896 which led, within
a decade, to direct measurements of ages on minerals
and rocks based on radioactive decay systems. Modern
geochronology had been conceived.
General references: Albritton (1980), Burchfield
(1990), Dalrymple (1991), Holmes (1913).
The discovery of radioactivity and the birth
of geochronology
In the late 1800s there was a general impression in physics
that all the great discoveries had been made. Younger
physicists would have to be content with improving the
accuracy and precision of already known physical con-
stants. Then, late in 1895, only months after Perry had
published his critique of Kelvin (and exposed the fatal
flaw in Kelvin’s assumption about the Earth’s transfer
of heat), Wilhelm Roentgen (1845–1923) discovered
“X-rays,” penetrating rays emitted from a cathode
ray tube. X-rays became an overnight sensation –
“radiographs” showing the bones inside the hand of
a live human being must have seemed absolutely magical;
the general public was enthralled. The scientific commu-
nity also was invigorated by this completely unexpected
development. Well, most of the scientific community:
“Kelvin was not moved. He had regarded the original
announcement of Roentgen’s x-rays as a hoax” (Albritton,
1980, p. 204). To be fair, an 1896 X-ray image of Kelvin’s
own hand exists, so he soon accepted Roentgen’s discov-
ery. One of the many scientists who did pay close attention
to Roentgen’s discovery was physicist Henri Becquerel
(1852–1908). To quote from a summary by Mattinson
(2013a, p. 2):
Following in his father’s footsteps, Becquerel had conducted
research in phosphorescence, the property of some materials,
including some uranium compounds, to emit light after expo-
sure to sunlight or ultraviolet light. However, by 1895, the
43-year-old Becquerel evidently had been inactive in
research for about five years. The discovery in late 1895 by
Wilhelm Roentgen of ‘X rays’ emitted from a cathode-ray
apparatus spurred Becquerel back into action. Becquerel
began experiments on some of his phosphorescent uranium
samples to determine if they might emit not just visible light,
but also invisible penetrating rays similar to Roentgen’s
X-rays.
Becquerel placed photographic plates that he had
wrapped in heavy black cloth on a south-facing window-
sill and placed uranium samples on top of the covered pho-
tographic plates. After exposure of the uranium samples to
sunlight, Becquerel developed the plates to see if any rays
from the uranium had penetrated the black paper and
exposed the plate. Fortune smiled on Becquerel. First,
some of his previous phosphorescence work had been on
uranium compounds. Of all the elements, only uranium
and thorium and radioactive “daughter” elements pro-
duced by their decay generate strong penetrating radiation.
If Becquerel had not had U or Th samples available, all his
results would have been negative. Instead, he almost
immediately obtained positive results – the plates had
been exposed, producing fuzzy images the shape of the
samples placed on the plates. Second, Becquerel initially
assumed that activation by exposure to sunlight was nec-
essary for any penetrating rays to be produced, by analogy
with true phosphorescence. However, for reasons that
have never been clear (e.g., Badash, 1996), after a period
of dark, rainy weather, Becquerel developed some photo-
graphic plates, even though the uranium samples had not
been exposed to sunlight – they had been left in a closed
drawer, on top of the covered photographic plates. The
photographic plates showed the same degree of exposure
as before. Becquerel realized that the exposure was not
related to activation by sunlight, but was an inherent prop-
erty of the uranium samples. The new form of radiation,
initially referred to variously as “uranium-rays” or
“Becquerel-rays,” was later called “radioactivity” by
Marie and Pierre Curie. That name would stick.
Initially, Becquerel’s discovery was greatly
overshadowed by Roentgen’s discovery of X-rays. It
was not immediately apparent that Becquerel had discov-
ered an entirely different phenomenon, and scientists were
slow to enter the new field of radioactivity. Quinn (1995,
p. 143) in her excellent book on Marie Curie comments:
Becquerel himself seems to have concluded that the subject
was exhausted: after his initial papers in 1896, he published
only two in 1897 and none the following year. By early
1898, as science historian Lawrence Badash has noted, the
subject was ‘something of a dead horse.’ Other than
Becquerel’s papers, only four were devoted to the uranium
rays at the Academy in the year of their discovery. In contrast,
Gustave LeBon’s false claims for ‘black light’ were the sub-
ject of fourteen papers. And Roentgen’s X rays got the most
attention of all. Nearly one hundred papers were delivered at
the Academy in 1896 concerning X rays.
In this respect it seems particularly ironic that one of the
very few scientists who took any interest in confirming
some of Becquerel’s work was William Thomson (Lord
Kelvin), then 73 years old. The general lack of interest in
Becquerel’s discovery (certainly) and the fact that Kelvin,
who did show interest, had, for the previous few years
been somewhat of a mentor to Pierre Curie (likely) were
responsible for Marie Curie choosing to study radioactiv-
ity for her Ph.D. project, starting in 1898. Marie soon
found that the mineral pitchblende (uranium oxide) was
several times more radioactive than pure uranium metal.
This led her to discover two new radioactive elements,
radium (Ra) and polonium (Po), both decay products of
uranium. She then concentrated Ra for chemical charac-
terization and further study. Marie’s meticulous work
and exciting discoveries inspired others to enter the field:
In a series of brilliant papers in 1902, Ernest Rutherford and
Frederick Soddy established that radioactivity resulted from
the transformation of unstable elements into new forms.
A competing theory, that radioactive elements were some-
how absorbing then releasing energy from outside sources,
soon died out. Rutherford and Soddy derived the mathemat-
ical principles of radioactive decay and suggested that helium
 HISTORICAL DEVELOPMENT OF DATING METHODS
(He) was a stable daughter product of U decay. Rutherford
realized that the ratio He/U could be used to measure the ages
of geological materials and presented the first dates based on
radioactive decay in a 1904 lecture. (Mattinson, 2013b, p. 53)
Modern geochronology was born.
Meanwhile, in 1903, Pierre Curie and Albert Laborde
discovered that the radioactive decay of Ra produced
enormous amounts of heat, invalidating Kelvin’s assump-
tion of a simple cooling Earth with no internal heat source
(more on this later). The discovery inspired a flurry of let-
ters to Nature in 1903. W. E. Wilson, an independent “gen-
tleman-astronomer” (Letter to the Editor, July 9, 1903)
enthusiastically suggested that if the sun contained 3.6 g
of radium per cubic meter, its entire energy output could
be provided by the radioactive decay of radium (later, Joly
and Rutherford would point out that to support this con-
centration of radium, the sun would have to be made up
almost entirely of uranium!). George Darwin was more
nuanced in a September 16, 1903, letter: “Knowing as
we now do, that an atom of matter is capable of containing
an enormous store of energy in itself, I think we have no
right to assume that the sun is incapable of liberating
atomic energy to a degree at least comparable with that
which it would do if made of radium.” John Joly also
weighed in, commenting on the effects of heat from radio-
active decay in the sun and the Earth in an October 1,
1903, letter: “. . .it would appear that the estimates derived
from physical speculations are now subject to modifica-
tion in just the direction which geological data required.
The 100 million years which the doctrine of uniformity
requires may, in fact, yet be gladly accepted by the physi-
cist.” Note: Joly’s enthusiasm for methods based on radio-
active decay dimmed when the first He/U and Pb/U dates
indicated that Joly’s own method greatly underestimated
the age of the Earth.
The year 1903 was also the year that the Nobel Prize in
physics was awarded jointly to Henri Becquerel, Pierre
Curie, and Marie Curie for their work on radioactivity.
Radium, radiation, and the Curies quickly acquired super-
star status. In the geological community, the enthusiasm
for radium, along the lines of the Nature letters quoted
above, promoted a myopic focus on radiogenic heat pro-
duction as dealing the fatal blow to Kelvin’s four-decade
dominance of the age of the Earth debate. This view is still
promoted in many elementary texts, e.g., “Perhaps the
most durable myth in elementary Earth science texts is
the view that the discovery of radioactivity . . . and radio-
active heat generation . . . forever ended the late nineteenth
century controversy involving Lord Kelvin and the age of
the Earth” (Harrison, 1987). A most unfortunate side
effect was that John Perry’s devastating (and much more
correct) 1895 critique of Kelvin’s model sank into obscu-
rity and remained there until relatively recently. A modern
analysis of Kelvin’s approach to the age of the Earth dem-
onstrates that Perry had it right:
. . . even if Kelvin had included a reasonable radiogenic heat
production in his thermal evolution models, he would still
have found grounds for arguing that the age of the Earth
325
was on the order of 108 years. The essential missing process
is not really radiogenic heating at all but thermal convection,
which allows the surface flux to exploit the entire heat of the
Earth as opposed to simply that of a shallow conductive
boundary layer. (Richter, 1986, p. 397, emphasis added)
But Richter inexplicably makes no reference to Perry,
even though Perry’s contributions are thoroughly covered
in the 1975 first edition of Burchfield’s book, which Rich-
ter cites as a general reference for the age of the Earth con-
troversy. This oversight was pointed out by Harrison
(1987) and corrected in later papers by Richter and his col-
leagues. “If the scientific community of the day had
absorbed Perry’s message. . .. The realization that convec-
tion takes place within the Earth’s apparently solid interior
would have made it hard to sustain the “fixist” view of the
Earth—which held that it was physically impossible for
continents to move horizontally. That perception of fixity
exerted a powerful brake on geological progress in the first
half of the 20th century. . .” (England et al., 2007, p. 349).
Within a few years of the paper on radium heat produc-
tion by Curie and Laborde and the Nobel Prize for
Becquerel and the Curies, the first dates of geologic mate-
rials based the decay of uranium to helium were formally
published by Rutherford in 1906 and those based on the
decay of uranium to lead (the ultimate stable end-product
of the uranium “decay series”) were published by Bertrand
Boltwood in 1907, demonstrating that Paleozoic and Pre-
cambrian rocks ranged from hundreds of millions to bil-
lions of years old. Lord Kelvin died in 1907, but not
before reasserting his belief that gravity was the only pos-
sible source of the heat of the Earth and sun, and
suggesting “. . .that somehow ethereal waves may supply
the energy to the radium as it is giving out heat to the pon-
derable matter around it” (Burchfield, 1990, p. 165). This
was a version of an idea held for a time by Becquerel and
the Curies, the phosphorescence model, that the energy
released by radioactive decay had somehow been
absorbed from outside forces. With Kelvin’s death, one
might have imagined that the geological community
would have embraced the new field of geochronology
with great enthusiasm. This would be a mistake. First,
radioactivity had been discovered and studied by physi-
cists and chemists, who also presented the first dates based
on radioactive decay. Having been burned once by the
physicists (Kelvin et al.), geologists were wary.
Radioactive dating clearly had not taken geology by
storm. As Joseph Barrell succinctly summarized the situa-
tion in 1917: ‘After one experience with the fallibility of
physical argument notwithstanding its mathematical char-
acter, it would certainly be unwise for geologists to accept
unreservedly the new and larger measurements of time
given by radioactivity. There may be here, also, factors
undetected and unsuspected which vitiate the results’.
(Burchfield, 1990, p. 190)
Second, some geologists had a considerable intellectual
investment in their own earlier estimates of the age of the
Earth. For example, Joly’s 100 Ma age based on the salin-
ity of the oceans had gained him considerable attention
326 HISTORICAL DEVELOPMENT OF DATING METHODS
and prestige, especially after heat from radioactive decay
seemed to invalidate Kelvin’s most recent younger esti-
mates of the age of the Earth. Only a couple of years later,
ages based on radioactive decay published by Rutherford
and Boltwood appeared to demolish Joly’s 100 Ma figure.
Joly denied the validity of radiometric ages for more than
two decades. “It was not until 1930 that he showed any
sign of actually yielding to the accumulated mass of radio-
logical evidence, and like Kelvin he died without publicly
relinquishing his convictions” (Burchfield, 1990, p. 189).
By then, Rutherford and Boltwood had long since left the
field of geochronology, having returned to their main
interests in the physics of matter and radioactive decay,
and the chemistry of the U and Th decay series, and new
giant in the field of geochronology had emerged – Arthur
Holmes.
General references: Albritton (1980), Badash (1996),
Burchfield (1990), Dalrymple (1991), Holmes (1913),
Quinn (1995), Rhodes (1986).
Arthur Holmes and the growth of geochronology
Arthur Holmes (1890–1965) was a small boy when radio-
activity was discovered, and a teenager when the first
He/U and Pb/U dates were published. He grew up with
the excitement of the dawning of a new scientific era.
Holmes attended a “higher grade school” with a strong
science orientation, and an exceptional physics teacher
who used geological examples in many of his lectures.
In 1906 Holmes was mesmerized by an exchange of letters
in The Times of London. Kelvin initiated the exchange
after he read a report in The Times about new experiments
on radioactivity by Soddy and William Ramsay:
Kelvin wrote an angry letter to The Times criticizing their
work. . . .he also considered it added nothing new to the the-
ory of atoms that had been proposed by Democritus two and
a half thousand years ago! The irritation felt by the younger
generation by Kelvin’s obstinate refusal to accept the new
evidence was thinly veiled: ‘Lord Kelvin’s letter will of
course receive respectful attention. . .but it is also known that
his brilliant original mind has not submitted patiently to the
task of assimilating the work of others by the process of read-
ing’. (Quinn, 1995, pp. 11–12)
The bare-knuckles match went on for over a month,
with Kelvin on one side, and the young lions of radioactiv-
ity on the other: Ramsay, Rutherford, Soddy, and Robert
Strutt (the latter having narrowly missed becoming the
first to publish an age based on the accumulation of helium
from U decay). Holmes was hooked!
In 1907 Holmes started to study physics at the Royal
College of Science, London. In his second year, Holmes
renewed his interest in geology thanks to a brilliant and
dynamic young geology professor, William Watts. After
Holmes’ second year, he switched his emphasis to geol-
ogy. Another young professor at the Royal College of Sci-
ence was none other than Robert Strutt. Holmes
recognized a splendid opportunity to combine his interests
in geology and radioactivity and began a research project
developing improved Pb/U dating techniques under the
direction of Strutt. The timing was particularly fortuitous,
as Boltwood had returned to his primary research interest
on the chemistry of the uranium and thorium decay series,
leaving the newly born field of Pb/U geochronology wide
open. Almost immediately, Holmes made a series of
important contributions. He developed more sensitive
chemical and radiochemical techniques for the measure-
ment of U and Pb than those used by Boltwood. Whereas
Boltwood had been limited to U-rich ore minerals,
Holmes could analyze minerals much lower in U and Pb
such as zircon and even feldspar. Using an updated decay
constant for U, Holmes published a series of new Pb/U
analyses and dates and also recalculated Boltwood’s ear-
lier Pb/U dates. Arthur Holmes, the physicist, might have
stopped there, but Arthur Holmes, the geologist, investi-
gated the geological settings of all the samples dated and
produced the beginnings of a geologic time scale in
1911. Holmes’ results are remarkable in that they are all
in general agreement with modern results: Carboniferous
ca. 340 Ma, Devonian ca. 370 Ma, “pre-Carboniferous”
ca. 410 Ma, “Silurian or Ordovician” ca. 430 Ma, and Pre-
cambrian ca. 1,025–1,640 Ma. Holmes published a 196 p.
monograph in 1913 boldly titled “The Age of the Earth.”
The book provides a summary of early efforts to determine
the age of the Earth and gives a flavor of the commonly
bitter controversies that surrounded these efforts. As
discussed above, some of the same geologists who consid-
ered Lord Kelvin’s estimates of the age of the Earth to be
much too young now objected to minimum ages for the
Earth based on radioactive decay systems as being much
too old. Holmes hoped to use his book to convince the
geological community of the fundamental accuracy of
geochronology based on radioactive decay. Summarizing
all the available He/U and Pb/U dates, Holmes presented
an expanded version of his geologic time scale, from
Pleistocene to Precambrian. For Holmes, this was only
the beginning of a long and remarkable career devoted to
expanding and improving the dating of the geologic time
scale and determining the age of the Earth itself.
General references: Holmes (1913), Lewis (2000),
significant parts of this section are adapted with slight
modification from Mattinson (2013a, b).
The discovery of isotopes and early isotopic
research
As might be expected, the discoveries of x-rays, radioac-
tivity, and the radioactive transmutation of atoms into dif-
ferent forms had far-reaching effects, inspiring a great
outburst of research on the nature of matter itself. In
1911 Rutherford discovered the atomic nucleus by
bombarding thin metal foils with collimated beams of
alpha particles. Highly accurate measurements revealed
that the atomic weights of many elements were not whole
numbers. In particular, in 1913, T. W. Richards measured
the atomic weight of Pb produced by the decay of U and
found that it was significantly lower than the atomic
weight of “ordinary” Pb. Innovative instrumentation
 HISTORICAL DEVELOPMENT OF DATING METHODS
played an increasingly vital role. J. J. Thomson
(Rutherford’s professor) built a “positive ray” apparatus,
a simple antecedent to today’s mass spectrometers. Thom-
son discovered that neon had two different forms – one
with a mass of about 20 and another with a mass of about
22. Soddy coined the term “isotopes” for these atoms of
the same element but with different atomic masses. The
neutron was suggested by Rutherford in 1920 (later con-
firmed experimentally by Chadwick in 1932), fleshing
out the basic picture of the atomic nucleus. In 1920,
F. W. Aston built an advanced version of Thomson’s “pos-
itive ray” instrument. Aston’s “mass spectrograph” used
photographic plates to record the mass spectra of elements
and allowed semiquantitative measurements of isotopic
ratios. After confirming Thomson’s work on neon, Aston
went on to investigate the isotopic composition of many
elements over his career, including “ordinary” Pb from
Pb ore samples in 1927 and “radiogenic” Pb from a U
ore sample in 1929. Pb from the U ore unexpectedly
contained 207Pb. This discovery led to the inference that
U contained not just 238U, but also a minor, previously
unknown isotope that was the parent of 207Pb, i.e.,
235
U. J. L. Rose and R. K. Stranathan in 1936 realized that
the different decay rates for 238U and 235U required that
the 207Pb/206Pb ratio of Pb produced by uranium decay
increase with increasing age – an age could be determined
for a uranium-bearing mineral simply from the lead isoto-
pic composition alone. Thus the foundation for modern
isotope geology and geochronology was laid.
General references: Harper (1973), significant parts of
this section are adapted with slight modification from
Mattinson (2013a, b).
Alfred nier, isotopes, and mass spectrometry
Over the next several years, improvements in instrument
design resulted in the “mass spectrometer” in which pho-
tographic plates were replaced by collection and electronic
amplification of the small electrical currents produced by
the beams of positively charged ions of each isotope.
The technological advance allowed much more precise
and accurate isotopic measurements. Alfred Nier not only
made major contributions to mass spectrometer design but
also made several discoveries that were critical to
advances in geochronology. Nier in 1938 found that
“common lead,” e.g., lead from lead ores, had significant
variations in isotopic compositions. Earlier measurements
of the mean atomic mass of lead from a variety ore sam-
ples gave essentially identical results, leading to the
assumption that all common leads must have ca. identical
isotopic compositions. Nier demonstrated that relative to
the non-radiogenic isotope, 204Pb, 206Pb, and 208Pb tended
to vary in a highly correlated way that fortuitously resulted
in little or no variation in mean atomic mass. Nier further
suggested that common lead must consist of a mixture of
primeval lead of fixed isotopic composition dating to the
time of the Earth’s formation, plus radiogenic lead gener-
ated by the decay of uranium and thorium after the Earth’s
327
formation. This was a conceptual leap crucial to all subse-
quent estimations of the age of the Earth. In 1939 Nier
turned his attention to uranium and lead in uranium ores,
making a remarkably accurate measurement of the
238 235
U/ U and 234U/238U ratios of uranium. Nier then
combined the uranium isotope data with the lead isotope
data from the same samples to calculate a remarkably
accurate decay constant for 235U. In 1941, Nier and his
collaborators reported a 207Pb/206Pb age of ca. 2.6 Ga for
a monazite from Manitoba, Canada. This was the first iso-
topic age determination for the antiquity of the Precam-
brian, and a rigorous minimum age for the Earth. Nier
and his colleagues also reported lead isotope measure-
ments for a wide array of lead ores. These data attracted
the attention of E. K. Gerling, Arthur Holmes, and F. G.
Houtermans all of whom realized that Nier’s isotopic data
for lead ores could be used to estimate the age of the Earth.
Gerling, Holmes, and Houtermans calculated ages for the
Earth ranging from ca. 3.0 to 3.9 Ga on the (incorrect)
assumption that the ore with the lowest 206Pb/204Pb and
207
Pb/204Pb ratios approximated the isotopic composition
of primeval lead. At this stage in the evolution of geochro-
nology, isotope ratios were being measured by more or
less modern techniques, but the amounts of uranium and
lead were still measured by classical chemical methods.
This was about to change dramatically.
General references: Harper (1973), significant parts of
this section are adapted with slight modification from
Mattinson (2013a, b).
The discovery of nuclear fission and the separation
of isotopes
Meanwhile, other momentous discoveries in nuclear
physics were underway which would have profound
implications for geochronology, both directly and indi-
rectly. In 1933, only months before Marie Curie’s death,
her daughter and son-in-law, Irene and Frederic Joliot-
Curie, bombarded aluminum with alpha rays. The alumi-
num captured alpha particles and emitted neutrons, pro-
ducing a radioactive isotope of phosphorous. The
phosphorous in turn decayed with a short half-life to
a stable isotope of silicon. The Joliot-Curies had artifi-
cially transformed a stable element into a different, radio-
active element. Many labs began bombarding a wide
range of elements with various particles, including neu-
trons. Irene Joliot-Curie and Pavel Savitch, a visitor to
her lab, bombarded uranium with neutrons and produced
a radioactive product that appeared to have the properties
of lanthanum. Conventional wisdom at the time dictated
that products of irradiation should be within a few atomic
mass units (amu) of the target element. Since lanthanum is
ca. 100 amu lighter than uranium, the Joliot-Curie-Savitch
result seemed highly unlikely, in fact impossible, even to
Joliot-Curie and Savitch themselves. Had they been
bolder, they might have realized that they had induced
nuclear fission. The Germans Otto Hahn and Fritz
Strassmann, who also doubted the original result,
328 HISTORICAL DEVELOPMENT OF DATING METHODS
duplicated the experiments. They eventually positively
identified barium (also about 100 amu lighter than ura-
nium), plus several other elements, as products of the neu-
tron irradiation – clear evidence that the atom had been
split. The discovery of fission triggered another period of
intense research, and with the onset of World War II, the
“Manhattan Project,” and the development of nuclear
weapons. An essential part of the Manhattan Project was
development of isotope separation techniques to concen-
trate 235U, the isotope of uranium susceptible to “induced
fission” by neutron irradiation. The ability to separate iso-
topes would open the door to a major breakthrough in geo-
chronology, the “isotope dilution method” for precise and
accurate measurements of very small amounts of a wide
range of elements.
General references: Harper (1973), Quinn (1995),
Rhodes (1986), significant parts of this section are adapted
with slight modification from Mattinson (2013a, b).
The isotope dilution method and the
development of geochronology
After World War II many scientists recognized that the dis-
coveries of military research could be applied to a wide
range of research problems in physics, chemistry, and
geology. One such scientist was Harrison Brown, who
headed a group at the University of Chicago studying trace
elements in meteorites. In addition to his interests in mete-
orite chemistry in general, Brown was familiar with the
recent work on the age of the Earth using Pb isotopes.
He thought that iron meteorites might contain Pb but not
U, thus providing a true primordial isotopic composition
of Pb from the early solar system. Key members of the
group included Mark Ingram, a physicist and mass spec-
trometer expert, and graduate students Clair “Pat”
Patterson and George Tilton. The group developed new,
much more sensitive mass spectrometric techniques for
measuring the isotopic compositions of small amounts of
lead in meteorites and minerals from terrestrial samples
(e.g., the mass spectrometry methods they developed
reduced the amount of Pb needed for analysis by a factor
of ca. 1,000 compared with Nier’s earlier work). Brown’s
group also perfected the isotope dilution (ID) technique
for accurately measuring low concentrations of Pb, Th,
and U. In the ID technique, a known amount of a highly
purified isotope of the element of interest is added to
a sample, prior to any chemical separations. For example,
a known amount of highly pure 235U is added to a sample
containing natural U, which is over 99 % 238U. After sep-
arating U from all of the other elements in the sample, the
ratio of 235U/238U is measured by mass spectrometry, and
the amount of 238U in the sample can be calculated with
high accuracy. The same approach is used for Pb, using
a “tracer” or “spike” of a highly purified isotope of lead.
The work led by Tilton on minerals such as zircon, sphene
(titanite), apatite, and several other minerals from terres-
trial granite was an immediate success, and fully published
after some delay owing to “top secret” classification of the
very existence of purified 235U (Tilton et al. 1955). The Pb
isotopic work on meteorites went more slowly owing to low
Pb levels in most meteorite samples and high levels of Pb
contamination in the environment. Patterson continued with
the meteorite work, overcame the challenges of Pb contam-
ination, and determined a ca. 4.55 Ga age for meteorites
and the Earth (Patterson et al., 1955; Patterson, 1956), quite
compatible with modern measurements. The long quest to
determine the age of the Earth had finally succeeded.
General references: Harper (1973), Lewis (2000), sig-
nificant parts of this section are adapted with slight modi-
fication from Mattinson (2013a, b).
50 Years of refinement, expansion, and growth
of geochronology
At about the same time that Brown’s group developed the
next generation of U–Th–Pb geochronology, similar tech-
niques using advanced chemical separation, mass spec-
trometry, and isotope dilution were applied to the Rb–Sr
and K–Ar parent-daughter systems. Rb and K were dis-
covered to be “weakly radioactive” by 1906 but could
only be exploited for geochronology by about the 1950s
when instrumentation and methods had sufficiently
improved. Also developed in the early 1950s was the car-
bon-14 dating method. By the mid-1960s these techniques
were all well established.
The fifty years between 1963 and the present have seen, quite
literally, an explosion in the amount of geochronological
research, the diversity of methods, the quality of the analyses,
and the sophistication of data interpretation. Major driving
forces have included development of new instrumentation,
analytical techniques, data analysis techniques, and, especially
in the last decade, laboratory collaboration at national and inter-
national levels. . .. depending on how one counts, close to 40
different dating methods are now available, spanning age ranges
from months to billions of years. (Mattinson, 2013a, pp. 5, 10)
Any detailed discussion of all these methods is far
beyond the scope of this historical review. Interested
readers can find more breadth and detail in the following
sources: (1) general information, Allègre (2008), Dickin
(2005), Faure and Mensing (2005); (2) recent publica-
tions, mostly on state-of-the-art techniques plus some
history, Amelin and Ireland (2013), Condon and
Schmitz (2013), Corfu (2013), Mattinson (2013a, b),
Nemchin et al. (2013), Schmitz and Kuiper (2013), and
Schoene et al. (2013); and cross-references.
Bibliography
Albritton, C. C., 1980. The Abyss of Time: Changing Conceptions of
the Earth’s Antiquity After the Sixteenth Century. San Francisco:
Freeman, Cooper and Company.
Allègre, C. J., 2008. Isotope Geology. Cambridge: Cambridge
University Press.
Amelin, Y., and Ireland, T. R., 2013. Dating the oldest rocks and
minerals in the solar system. Elements, 9, 39–44.
Badash, L., 1996. The discovery of radioactivity. Physics Today, 49,
21–26.
Burchfield, J. D., 1990. Lord Kelvin and the Age of the Earth.
Chicago: The University of Chicago Press. 1975.
 HOMINID EVOLUTION TIMESCALE
Condon, D. J., and Schmitz, M. D., 2013. One hundred years of
isotope geochronology, and counting. Elements, 9, 15–17.
Corfu, F., 2013. A century of U-Pb geochronology: the long quest
towards concordance. Geological Society of America Bulletin,
125, 33–47.
Dalrymple, G. B., 1991. The Age of the Earth. Stanford: Stanford
University Press.
Dickin, A. P., 2005. Radiogenic Isotope Geology, 2nd edn. Cam-
bridge: Cambridge University Press.
England, P. C., Molnar, P., and Richter, F. M., 2007. Kelvin, Perry,
and the age of the earth: had scientists better appreciated one of
Kelvin’s contemporary critics, the theory of continental drift might
have been accepted decades earlier. American Scientist, 95,
342–349.
Faure, G., and Mensing, T. M., 2005. Isotopes: Principles and
Applications, 3rd edn. Hoboken: Wiley.
Harper, C. T., 1973. Geochronology: Radiometric Dating of Rocks
and Minerals. Stroudsburg: Dowden, Hutchinson & Ross.
Harrison, T. M., 1987. Kelvin and the age of the earth. Journal of
Geology, 95, 725–727.
Holmes, A., 1913. The Age of the Earth. London/New York: Harper
& Brothers.
Lewis, C., 2000. The Dating Game: One Man’s Search for the Age
of the Earth. New York: Cambridge University Press.
Mattinson, J., 2013a. The geochronology revolution. In
Bickford, M. E. (ed.), The Web of Geological Sciences:
Advances, Impacts, and Interactions. Denver: Geological Soci-
ety of America. Geological Society of America Special Paper
500, pp. 303–320.
Mattinson, J., 2013b. Revolution and evolution: 100 years of U-Pb
geochronology. Elements, 8, 53–57.
Nemchin, A. A., Horstwood, M. S. A., and Whitehouse, M. J., 2013.
High- spatial-resolution geochronology. Elements, 9, 31–37.
Patterson, C., 1956. Age of meteorites and the earth. Geochimica et
Cosmochimica Acta, 10, 230–237.
Patterson, C., Tilton, G. R., and Inghram, M., 1955. Age of the
earth. Science, 121, 69–75.
Perry, J., 1895. On the age of the earth. Nature, 51, 224–227.
Quinn, S., 1995. Marie Curie a Life. Cambridge, MA: Da Capo Press.
Rhodes, R., 1986. The Making of the Atomic Bomb. New York:
Simon and Schuster.
Richter, F. M., 1986. Kelvin and the age of the earth. Journal of
Geology, 94, 395–401.
Schmitz, M. D., and Kuiper, K. F., 2013. High-precision geochro-
nology. Elements, 9, 25–30.
Schoene, B., Condon, D. J., Morgan, L., and McLean, N., 2013.
Precision and accuracy in geochronology. Elements, 9, 19–24.
Tilton, G. R., Patterson, C., Brown, H., Inghram, M., Hayden, R.,
Hess, D., and Larsen, E., 1955. Isotopic composition and distri-
bution of lead, uranium, and thorium in a Precambrian granite.
Geological Society of America Bulletin, 66, 1131–1148.
Williams, H. S., 1893. The elements of the geological time-scale.
Journal of Geology, 1, 283–295.
Cross-references
Ar–Ar and K–Ar Dating
Fission Track Dating and Thermochronology
Lu-Hf Dating: The Lu-Hf Isotope System
Rhenium–Osmium Dating (Meteorites)
Rubidium–Strontium Dating, Hydrothermal Events
Sedimentary Rocks (Rb-Sr Geochronology)
Terrestrial Cosmogenic Nuclide Dating
Uranium–Lead Dating
U-Series Dating
U–Th/He Dating
Zircon
329
HOMINID EVOLUTION TIMESCALE
Antoine Balzeau
Department of Prehistory, UMR 7194, CNRS, Musée
National d’Histoire Naturelle, Paris, France
Definition
The hominid evolution timescale has been established
thanks to primate genomic sequencing (Schrago and
Voloch, 2013). It is now generally accepted that the diver-
gence between our closest relatives, bonobos and chimpan-
zees (Pan lineage), and modern humans dates back to at
least 7.0 Ma (Figure 1) based on genetic data from extant
primate species. However, discoveries of fossil hominins
have also been crucial to the debate on the theoretical
aspects and for the acceptation of the timing of the appear-
ance of the hominin lineage. These fossils are still debated,
but Sahelanthropus tchadensis, ca. 7 Ma, and Orrorin
tugenensis, ca. 6 Ma, may represent one of the earliest evi-
dences of the hominin lineage. Unfortunately, contempo-
rary evidence of early specimens of the Pan lineage is not
yet documented. The oldest fossils attributed to Pan are
only 500 Ka and the divergence between bonobos (Pan
paniscus) and chimpanzees (Pan troglodytes) would have
occurred around 2 Ma based on genetic evidence. A discus-
sion exists on the use of the terms hominid and hominin, but
the general agreement, now, is to use the term hominin to
refer to all the member of the human clade, i.e., to speci-
mens that are bipedal and that share anatomical specificities
related to this mode of locomotion. Between 6 Ma and
1.8 Ma, more fossil hominins are known. They belong to
at least three genera (Ardipithecus, Australopithecus, and
Paranthropus and maybe Kenyanthropus) and 13 species
and were mostly unearthed in South and East Africa but also
in Central Africa (Chad for Sahelanthropus tchadensis and
Australopithecus bahrelghazali). Another major event on
the hominid timeline was the appearance of the genus Homo
around 2.5–2 Ma that coincides with the first hominin
migration out of Africa. However, the definition of the
genus Homo is controversial or at least problematic, as its
ranges of anatomical and behavioral variations have been
widely expanded and encompass those now known for
other hominins. An important event of divergence between
several species (Homo heidelbergensis and possibly Homo
rhodesiensis emerging from Homo erectus sensu lato and
then evolving into Homo neanderthalensis and sapiens)
has occurred in Africa and possibly in Eurasia around
500 ka. That led in particular to the appearance of the Homo
sapiens species 200 ka years ago. Homo neanderthalensis is
important for cultural and historical reasons (first discovery
of fossil hominin in Europe, questions on a possible other
humanity, etc.), but also because it is the species with the
largest fossil record in relation to their early use of burials.
Finally, our species has cohabited with several other species
(at least Homo erectus, Homo neanderthalensis, and Homo
floresiensis) until very recent times, only several ka com-
pared to 7 Ma of hominid evolution. Recent developments
330 HYDROCARBONS/RHENIUM–OSMIUM (RE–OS): ORGANIC-RICH SEDIMENTARY ROCKS
Hominid Evolution Timescale, Figure 1 Chronological distribution of hominin species and representation of selected species.
in paleogenetics (e.g., Meyer et al., 2012) have offered new
information about the recent history of the human lineage,
but these data need to be integrated in the wider range of
existing paleoanthropological data (Henke and Tattersall,
2007) to be interpreted.
Bibliography
Henke, W., and Tattersall, I., 2007. Handbook of Paleoanthropol-
ogy. New York: Springer.
Meyer, M., Kircher, M., Gansauge, M. T., et al., 2012. A high-
coverage genome sequence from an archaic Denisovan individ-
ual. Science, 338, 222.
Schrago, C. G., and Voloch, C. M., 2013. The precision of the hom-
inid timescale estimated by relaxed clock methods. Journal of
Evolutionary Biology, 26, 746.
Cross-references
Electron Spin Resonance (ESR) Dating of Fossil Tooth Enamel
Molecular Clocks
Molecular Clocks, Human Evolution
Skeletal Remains (14C)
HYDROCARBONS/RHENIUM–OSMIUM (RE–OS):
ORGANIC-RICH SEDIMENTARY ROCKS
David Selby1, Vivien M. Cumming1,2, Alan D. Rooney2
and Alexander J. Finlay3
1
Department of Earth Sciences, University of Durham,
Durham, UK
2
Department of Earth and Planetary Sciences, Harvard
University, Cambridge, MA, USA
3
Origin Analytical LTD, Welshpool, Powys, UK
Definition
Utilization of the rhenium–osmium (Re–Os) radioisotope
geochronometer to (1) determine the depositional age of
organic-rich rocks (ORR), (2) establish the Os isotope
(187Os/188Os) composition of sedimentary successions,
and (3) date the timing of hydrocarbon (oil) generation
related to thermal maturation of an organic-rich sedimentary
rock.
 HYDROCARBONS/RHENIUM–OSMIUM (RE–OS): ORGANIC-RICH SEDIMENTARY ROCKS
Introduction
The first application of the Re–Os isotope
geochronometer to ORR demonstrated that the Re–Os
data, although imprecise, could yield the depositional
age of sedimentary successions (Ravizza and Turekian,
1989), a remarkable outcome given the analytical chal-
lenges and data uncertainties at the time. Today, utilizing
new analytical and mass spectrometry techniques (see
below) and the most accurate 187Re decay constant
(1.666  10 11 year 1; Smoliar et al., 1996) has permitted
the determination of precise (in some cases 1 % uncer-
tainty) depositional ages for ORR that possess 0.5 wt%
total organic carbon (TOC) (e.g., Cohen et al., 1999; Selby
and Creaser, 2005a; Kendall et al., 2009a; Xu et al., 2009;
Georgiev et al., 2011; Cumming et al., 2012). These stud-
ies have provided the foundation to apply the Re–Os
organic-rich sedimentary rock geochronometer to yield
ages throughout geological time for stratigraphic bound-
ary intervals and major Earth events (e.g., rise of oxygen,
Proterozoic glaciations etc.; Selby and Creaser, 2005a;
Selby, 2007; Anbar et al., 2007; Xu et al., 2009; Yang
et al., 2009; Rooney et al., 2011).
Rhenium–osmium geochronology of ORR utilizes the
isochron method, which yields the initial Os isotope com-
position (187Os/188Os) for the sampled unit, taken to repre-
sent the 187Os/188Os composition of the water column at
the time of deposition (Ravizza and Turekian, 1989;
Cohen et al., 1999; Peucker-Ehrenbrink and Ravizza,
2000, 2012).
In addition to ORR, hydrocarbon products of such
matured rocks, for example, crude oil and bitumen, have
been shown to also be enriched in both Re and Os at the
ppb and ppt level, respectively (Barre et al., 1995; Wood-
land et al., 2001; Selby and Creaser, 2005b; Selby et al.,
2007; Finlay et al., 2011). The hydrocarbon Re–Os iso-
tope data from several petroleum systems has shown that
Re–Os systematics of hydrocarbons can provide the timing
of petroleum generation from an oil source rock (Selby et al.,
2005; Selby and Creaser, 2005b; Finlay et al., 2011), with
the initial 187Os/188Os value derived from Re–Os hydrocar-
bon data interpreted to reflect the 187Os/188Os inherited
from the source rock at the time of petroleum expulsion
(Selby and Creaser, 2005b; Rooney et al., 2012).
Methodology: analytical protocol and sampling
Since the pioneering application of the Re–Os chronometer
to dating of ORS (Ravizza and Turkeian, 1989), significant
analytical advancements in digestion methods, chemical
purification, and mass spectrometry have permitted the
determination of precise Re–Os data. Current methods
employ analysis of an aliquot of ground organic-rich rock
digested in either inverse aqua regia (3:6 ml mix of concen-
trated HCl:HNO3) or 8 ml of 4 N H2SO4 with 0.25 g/g CrO3
along with a known amount of isotope tracer solution
(185Re + 190Os) in a Carius tube at 220 C for 48 h
(Cohen et al., 1999; Selby and Creaser, 2003). The use of
331
the CrVI–H2SO4 digestion method has been shown to limit
the release of detrital Re and Os from organic-rich rock
matrices and as a result reduce the scatter about the line of
an isochron (Selby and Creaser, 2003; Kendall et al.,
2004). The inverse aqua regia medium is used for the diges-
tion of hydrocarbons, such as asphaltene and bitumen.
Osmium from the acidic solution is typically isolated and
purified by solvent extraction (CCl4 or CHCl3) and
microdistillation (Cohen et al., 1999; Selby and Creaser,
2003; Selby et al., 2007). The Os extracted acidic solution
is then used to isolate and purify the Re fraction of the sam-
ple. This is achieved by solvent extraction methods and
anion chromatography (Selby and Creaser, 2003; Chen
et al., 2010). The isotope compositions of the purified Re
and Os fractions are commonly determined by negative
thermal ionization mass spectrometry (Creaser et al., 2002
and references therein).
Re–Os organic-rich rock geochronology requires
samples that have not been significantly affected by
postdepositional processes such as weathering (Peucker-
Ehrenbrink and Hannigan, 2000), metamorphism (above
greenschist grade; Kendall et al., 2004), and hydrothermal
fluid flow (Kendall et al., 2009b; Rooney et al., 2011), in
order to minimize any postdepositional mobilization of
Re and Os (Kendall et al., 2009a). The application of the
isochron method for geochronology requires two principal
criteria: (1) the sample set possesses the same or near-
identical initial 187Os/188Os, but has varied parent/daugh-
ter ratios (187Re/188Os), and (2) the isotope systematics is
not disturbed. At this time, there is no proxy to determine
if a section will have both homogeneous initial
187
Os/188Os and variable 187Re/188Os (cf. Turgeon et al.,
2007; Cumming et al., 2012). A review of sampling proto-
cols employed in various studies suggests intervals rang-
ing from a few centimeters to several meters can be used
to produce precise ages, with at least 20 g of powdered
sample required to achieve sample Re–Os homogeneity
(cf. Creaser et al., 2002; Kendall et al., 2009a).
Hydrocarbon Re–Os geochronology analysis utilizes
either the asphaltene isolated from crude oil or bitumen
(Selby et al., 2007). To date, sampling from wells across
an oil field has provided Re–Os data sets that have given
valuable age information to determine the timing of hydro-
carbon generation (Selby et al., 2005; Selby and Creaser,
2005b; Finlay et al., 2011; Lillis and Selby, 2013).
Organic-rich rocks: rhenium–osmium
geochronology and systematics
Rhenium and Os are organophilic, enabling the uptake of
hydrogenous Re and Os into organic matter during depo-
sition of ORS (Ravizza and Turekian, 1989; Cohen
et al., 1999; Selby and Creaser, 2003). The Re and Os
concentrations in marine and lacustrine ORS range from
average upper-crustal abundances (0.2–1 ppb Re and
30–50 ppt Os; e.g., Peucker-Ehrenbrink and Jahn, 2001)
to several hundred ppb/ppt of Re and Os, respectively
(e.g., Cohen et al., 1999; Creaser et al., 2008;
332 HYDROCARBONS/RHENIUM–OSMIUM (RE–OS): ORGANIC-RICH SEDIMENTARY ROCKS
Rooney et al., 2010; Baioumy et al., 2011; Georgiev et al.,
2011; Rooney et al., 2011; Poirier and Hillaire-Marcel,
2011; Cumming et al., 2012). Although organophilic, Re
and Os abundances do not always correlate to TOC despite
the majority of these metals being held within the kerogen
(Rooney et al., 2010, 2012; Cumming et al., 2012).
From the first Re–Os ORR study (Ravizza and
Turekian, 1989), numerous studies have reported accurate
and precise Re–Os depositional ages of organic-rich
marine and lacustrine sedimentary units (e.g., Cohen
et al., 1999; Creaser et al., 2002; Turgeon et al., 2007;
Xu et al., 2009; Rooney et al., 2011; Cumming et al.,
2012; ). Studies also indicate that hydrocarbon maturation
and/or low-grade metamorphism does not appreciably dis-
turb the Re–Os systematics (Kendall et al., 2004; Yang
et al., 2009; Rooney et al., 2010, 2011, 2012). In contrast
to studies yielding highly precise Re–Os ages, the Re–Os
data for the Aptian/Albian, Cenomanian/Turonian, and
Frasnian–Famennian stage boundaries give accurate, but
imprecise ages (Turgeon et al., 2007; Selby et al., 2009).
This imprecision is attributed to the limited spread in
187
Re/188Os ratios, highlighting the need for variation in
187
Re/188Os ratios but no variation in initial 187Os/188Os
to produce precise Re–Os dates. The lack of spread is con-
sidered to be due to exhaustion of the Re reservoir due to
increasing anoxia leading up to the Frasnian–Famennian
boundary (Turgeon et al., 2007). However, no correlation
is evident between redox conditions and 187Re/188Os
values of ORS (Selby et al., 2009). Alternatively, organic
matter type may control Re–Os fractionation in ORS
(Cumming et al., 2012; Harris, 2012).
Rhenium–osmium hydrocarbon geochronology
and systematics
Petroleum exploration is often hampered by the lack of
spatial and temporal controls on hydrocarbon formation.
Traditionally, dating the timing of hydrocarbon generation
involves basin modeling using estimated parameters (e.g.,
Ruble et al., 2001). Therefore, knowing the absolute age
of petroleum generation provides vital constraints for elu-
cidating petroleum maturation and migration.
A pioneering study (Selby et al., 2005) provided evidence
of the ability of the Re–Os hydrocarbon geochronometer
by yielding an accurate age for bitumen (374.9  9 Ma)
from the Polaris Mississippi-Valley-type deposit, Canada,
that was in agreement with paleomagnetic and Rb–Sr
sphalerite age constraints for mineralization. Further to
this, the Re–Os geochronometer has also been applied to
the tar sands of the Western Canadian Sedimentary Basin
and provided an accurate age (112  5.3 Ma) for the
timing of hydrocarbon generation, suggesting that Re–Os
hydrocarbon geochronology records the age of oil genera-
tion (Selby and Creaser, 2005b). A subsequent Re–Os
study of oils from the UK Atlantic margin oil fields pro-
vided a Re–Os generation age of 68  13 Ma that agrees
with basin models and Ar–Ar ages for K-feldspars bearing
oil-rich fluid inclusions and demonstrated that the Re–Os
hydrocarbon systematics is unaffected by biodegradation
(Finlay et al., 2011).
Both Re and Os in hydrocarbons have been shown to be
held in the heavy molecular fraction of oil (asphaltene)
and suggested to reside within metalloporphyrins similar
to Ni and Vor heteroatomic ligands due to their abundance
in asphaltene. The asphaltene fraction is shown to possess
the same isotopic compositions to that of the whole oil.
This study advanced the analytical application of this tool
as the separation of asphaltene for Re–Os analysis allows
much higher abundances to be analyzed, yielding more
precise Re–Os data (Selby et al., 2007).
Oil-to-source correlation is most commonly carried out
by using carbon isotopes and/or biomarkers, but these
methods can be hampered by biodegradation as they rely
on the light fractions of oil that are removed by biodegra-
dation (Peters et al., 2005). The initial 187Os/188Os value
derived from the Re–Os hydrocarbon data is interpreted
to reflect the 187Os/188Os inherited from the source rock
at the time of petroleum expulsion (Selby and Creaser,
2005b; Selby et al., 2005, 2007; Finlay et al., 2011,
2012). The reliability of Os as an oil-to-source fingerprint-
ing tool was further corroborated experimentally (Rooney
et al., 2012). This involved artificial maturation of the
source rocks coupled with Re–Os analysis of the original
rocks and the products derived from the hydrous pyrolysis
experiments in order to assess transfer of Re and Os from
source to oil. The 187Os/188Os compositions in bitumens
were found to be the same as their source rock in both
units studied, providing strong evidence that Os isotopes
can be used as powerful oil-source correlation tools.
Although the petroleum Re–Os systematics is not
appreciably disturbed, biodegradation studies have indi-
cated that the systematics can be affected by the interaction
of mantle fluids (Finlay et al., 2011) and TSR (Lillis and
Selby, 2013). These studies demonstrated the importance
of Os in understanding crustal-scale fluid dynamics, oil
migration pathways, and Re–Os hydrocarbon systematics.
Summary
Rhenium–osmium analysis of marine and/or lacustrine
ORR permits the age of deposition of the sediment to be
established. This method also yields the osmium isotope
composition of the water column associated with the
deposition of the sediment. The Re–Os systematics of
ORR is not disturbed by processes associated with oil gen-
eration and/or low-grade metamorphism. Rhenium–
osmium analysis of oil generated and migrated from
ORR has been shown to yield the timing of hydrocarbon
maturation. Further the petroleum Os isotope data have
been shown to be a potentially powerful tool to aid in fin-
gerprinting an oil-to-source ORR.
Bibliography
Anbar, A. D., Duan, Y., Lyons, T. W., Arnold, G. L., Kendall, B.,
Creaser, R. A., Kaufman, A. J., Gordon, G. W., Scott, C., Garvin,
J., and Buick, R., 2007. A whiff of oxygen before the Great
Oxidation Event? Science, 317, 1903–1906.
 HYDROCARBONS/RHENIUM–OSMIUM (RE–OS): ORGANIC-RICH SEDIMENTARY ROCKS
Baioumy, H. M., Eglinton, L. B., and Peucker-Ehrenbrink, B., 2011.
Rhenium-osmium isotope and platinum group element system-
atics of marine vs. non-marine organic-rich sediments and coals
from Egypt. Chemical Geology, 285, 70–81.
Barre, A. B., Prinzhofer, A., and Allegre, C. J., 1995. Osmium iso-
topes in the organic matter of crude oil and asphaltenes. Terra
Abstracts, 7, 199.
Chen, Z. H., Li, C., Qu, W. J., Wang, D. H., Zhang, J. J., and Du,
A. D., 2010. Research and preliminary application in
metallogenic chronology of Re-Os isotope system in graphite
samples. Acta Petrologica Sinica, 26, 3411–3417.
Cohen, A. S., Coe, A. L., Bartlett, J. M., and Hawkesworth, C. J.,
1999. Precise Re-Os ages of organic-rich mudrocks and the Os
isotope composition of Jurassic seawater. Earth and Planetary
Science Letters, 167, 159–173.
Creaser, R. A., Sannigrahi, P., Chacko, T., and Selby, D., 2002. Fur-
ther evaluation of the Re-Os geochronometer in organic-rich
sedimentary rocks: a test of hydrocarbon maturation effects in
the Exshaw Formation, Western Canada Sedimentary Basin.
Geochimica et Cosmochimica Acta, 66, 3441–3452.
Creaser, R., Szatmari, P., and Milani, E. J., 2008. Extending Re–Os
shale geochronology to lacustrine depositional systems: a case
study from the major hydrocarbon source rocks of the Brazilian
Mesozoic marginal basins. In Proceedings of the 33rd Interna-
tional Geological Congress, Oslo (Abstract).
Cumming, V. M., Selby, D., and Lillis, P. G., 2012. Re-Os geochro-
nology of lacustrine organic-rich sedimentary rocks: insights
into direct depositional dating of lacustrine successions. Re-Os
systematics and paleocontinental weathering. Earth and Plane-
tary Science Letters, 359–360, 194–205.
Finlay, A. J., Selby, D., Osborne, M. J., and Finucane, D., 2010. Fault-
charged mantle-fluid contamination of United Kingdom North
Sea oils: insights from Re-Os isotopes. Geology, 38, 979–982.
Finlay, A. J., Selby, D., and Osborne, M. J., 2011. Re-Os geochro-
nology and fingerprinting of United Kingdom Atlantic margin
oil: temporal implications for regional petroleum systems. Geol-
ogy, 39, 475–478.
Finlay, A. J., Selby, D., and Osborne, M. J., 2012. Petroleum source
rock identification of United Kingdom Atlantic Margin oil fields
and the Western Canadian Oil Sands using Platinum, Palladium,
Osmium and Rhenium: implications for global petroleum sys-
tems. Earth and Planetary Science Letters, 313–314, 95–104.
Georgiev, S., Stein, H. J., Hannah, J. L., Bingen, B., Weiss, H. M.,
and Piasecki, S., 2011. Hot acidic late Permian seas stifle life
in record time. Earth and Planetary Science Letters, 310,
389–400.
Harris, N.B., Mnich, C.A., Selby, D., Korn, D., 2012. Minor and
Trace Element and Re-Os Chemistry of the Upper Devonian
Woodford Shale, Permian Basin, West Texas: Insights into Metal
Abundance and Basin Processes. Chemical Geology,
356,76–93. http://dx.doi.org/10.1016/j.chemgeo.2013.07.018
Kendall, B. S., Creaser, R. A., Ross, G. M., and Selby, D., 2004.
Constraints on the timing of Marinoan ‘Snowball Earth’ glacia-
tion by 187Re – 187Os dating of a Neoproterozoic post-glacial
black shale in Western Canada. Earth and Planetary Science
Letters, 222, 729–740.
Kendall, B., Creaser, R. A., and Selby, D., 2009a. 187Re-187Os geo-
chronology of Precambrian organic-rich sedimentary rocks.
London: Geological Society, Special publication number 326,
pp. 85–107.
Kendall, B., Creaser, R. A., Gordon, G. W., and Anbar, A. D.,
2009b. Re-Os and Mo isotope systematics of black shales from
the Middle Proterozoic Velkerri and Wollogorang Formations,
McArthur Basin, northern Australia. Geochimica et
Cosmochimica Acta, 73, 2534–2558.
Lillis, P. G., and Selby, D., 2013. Evaluation of the rhenium-osmium
geochronometer in the phosphoria petroleum system, Bighorn
333
Basin of Wyoming and Montana, USA. Geochimica
Cosmochimica Acta, 118, 312–330.
Peters, K. E., Walters, C. C., and Moldowan, J. M., 2005. The
Biomarker Guide volume 1: Biomarkers and Isotopes in the
Environment and Human History. Cambridge: Cambridge
University Press.
Peucker-Ehrenbrink, B., and Hannigan, R. E., 2000. Effects of black
shale weathering on the mobility of rhenium and platinum group
elements. Geology, 28, 475–478.
Peucker-Ehrenbrink, B., and Jahn, B., 2001. Rhenium-osmium iso-
tope systematics and platinum group element concentrations:
loess and the upper continental crust. Geochemistry, Geophysics,
Geosystems, 2, 1061–1083.
Peucker-Ehrenbrink, B., and Ravizza, G., 2000. The marine
osmium isotope record. Terra Nova, 12, 205–219.
Peucker-Ehrenbrink, B., and Ravizza, G., 2012. Osmium isotope
stratigraphy. In Gradstein, F. M., Ogg, J. G., Schmitz, M., and
Ogg, G. (eds.), The Geological Timescale, 1. Oxford: Elsevier,
pp. 145–166.
Poirier, A., and Hillaire-Marcel, C., 2011. Improved Os-isotope
stratigraphy of the Arctic Ocean. Geophysical Research Letters,
38, L14607.
Ravizza, G., and Turekian, K. K., 1989. Application of the
187
Re-187Os system to black shale geochronometry. Geochimica
et Cosmochimica Acta, 53, 3257–3262.
Ravizza, G., and Turekian, K. K., 1992. The osmium isotopic com-
position of organic-rich marine sediments. Earth and Planetary
Science Letters, 110, 1–6.
Ravizza, G., Turekian, G. G., and Hay, B. J., 1991. The geochemis-
try of rhenium and osmium in recent sediments from the Black
Sea. Geochimica et Cosmochimica Acta, 55, 3741–3752.
Rooney, A. D., Selby, D., Houzay, J.-P., and Renne, P. R., 2010.
Re-Os geochronology of a Mesoproterozoic sedimentary succes-
sion, Taoudeni basin, Mauritania: implications for basin-wide
correlations and Re-Os organic-rich sediments systematics.
Earth and Planetary Science Letters, 289, 486–496.
Rooney, A. D., Chew, D. M., and Selby, D., 2011. Re-Os geochro-
nology of the Neoproterozoic-Cambrian Dalradian Supergroup
of Scotland and Ireland: implications for Neoproterozoic stratig-
raphy, glaciations and Re-Os systematics. Precambrian
Research, 185, 202–214.
Rooney, A. D., Selby, D., Lewan, M. D., Lillis, P. G., and Houzay, J.-P.,
2012. Evaluating Re–Os systematics in organic-rich sedimentary
rocks in response to petroleum generation using hydrous pyrolysis
experiments. Geochimica et Cosmochimica Acta, 77, 275–291.
Ruble, T. E., Lewan, M. D., and Philp, R. P., 2001. New insights on
the Green River Petroleum System in the Uinta Basin from
hydrous pyrolysis experiments. AAPG Bulletin, 85, 1333–1371.
Selby, D., 2007. Direct Rhenium-Osmium age of the Oxfordian-
Kimmeridgian boundary, Staffin bay, Isle of Skye, U.K., and the
Late Jurassic time scale. Norwegian Journal of Geology, 87, 9.
Selby, D., and Creaser, R. A., 2003. Re-Os geochronology of
organic rich sediments: an evaluation of organic matter analysis
methods. Chemical Geology, 200, 225–240.
Selby, D., and Creaser, R. A., 2005a. Direct radiometric dating of
the Devonian-Mississippian time-scale boundary using the Re-
Os black shale geochronometer. Geology, 33, 545–548.
Selby, D., and Creaser, R. A., 2005b. Direct radiometric dating of
hydrocarbon deposits using rhenium-osmium isotopes. Science,
308, 1293–1295.
Selby, D., Creaser, R. A., Dewing, K., and Fowler, M., 2005. Evalua-
tion of bitumen as a Re-187-Os-187 geochronometer for hydrocar-
bon maturation and migration: a test case from the Polaris MVT
deposit, Canada. Earth and Planetary Science Letters, 235, 1–15.
Selby, D., Creaser, R. A., and Fowler, M. G., 2007. Re-Os elemental
and isotope systematics in crude oils. Geochimica et
Cosmochimica Acta, 71, 378–386.
334 HYDROTHERMAL ORES (THERMOCHRONOLOGY)
Selby, D., Mutterlose, J., and Condon, D. J., 2009. U-Pb and Re-Os
geochronology of the Aptian/Albian and Cenomanian/Turonian
stage boundaries: Implications for timescale calibration, osmium
isotope seawater composition and Re-Os systematics in organic-
rich sediments. Chemical Geology, 265, 394–409.
Smoliar, M. I., Walker, R. J., and Morgan, J. W., 1996. Re-Os iso-
tope constraints on the age of Group IIA, IIIA, IVA, and IVB iron
meteorites. Science, 271, 1099–1102.
Turgeon, S. C., Creaser, R. A., and Algeo, T. J., 2007. Re-Os depo-
sitional ages and seawater Os estimates for the Frasnian-
Famennian boundary: Implications for weathering rates, land
plant evolution, and extinction mechanisms. Earth and Plane-
tary Science Letters, 261, 649–661.
Woodland, S., Ottley, C., Pearson, D. G., and Swarbrick, R. E.,
2001. Microwave digestion of oils for analysis of platinum group
and rare earth elements by ICP-MS. Special Publications of the
Royal Society of Chemistry, 267, 17–24.
Xu, G., Hannah, J. L., Stein, H. J., Bingen, B., Yang, G.,
Zimmerman, A., Weitschat, W., Mørk, A., and Weiss, H. M.,
2009. Re–Os geochronology of Arctic black shales to evaluate
the Anisian–Ladinian boundary and global faunal correlations.
Earth and Planetary Science Letters, 288, 581–587.
Yang, G., Hannah, J. L., Zimmerman, A., Stein, H. J., and Bekker,
A., 2009. Re-Os depositional age for Archean carbonaceous
slates from the southwestern Superior Province: challenges and
insights. Earth and Planetary Science Letters, 280, 83–92.
HYDROTHERMAL ORES (THERMOCHRONOLOGY)
István Márton
Stockwork GeoConsulting Ltd, Huedin, Cluj, Romania
Synonyms
Hydrothermal mineral deposits; Hydrothermal
mineralization
Definition
Hydrothermal ores are metal-rich mineral aggregates pre-
cipitated directly from hot aqueous solutions circulating
through the upper part of the crust. The terms “ore” and
“mineral deposit” are economic geology expressions that
refer to a mass of rock which contains metals at economi-
cally extractable grade. The applicability of these terms
can fluctuate as a function of metal market prices, avail-
ability of subsidy, or technological conditions.
Introduction
Understanding of the origin of hydrothermal ore deposits
requires accurate knowledge about the source and interac-
tion of five important components: heat, fluids, ligands,
metals, and permeability. The combination of these factors
might provide the conditions necessary for the formation
of hydrothermal mineral deposits, including the
(i) presence of hot aqueous solution to dissolve and trans-
port minerals, (ii) presence of interconnected openings in
the rock to allow the aqueous solutions to move, (iii) avail-
ability of sites for the deposits, and (iv) geochemical
reactions that will result in deposition (Barnes, 1997).
Hydrothermal fluids can be derived from multiple fluid
sources in Earth’s crust, and these sources and deposi-
tional processes have given rise to a variety of genetic
classifications of hydrothermal ore deposits (Pirajno,
1992). Magmatic–hydrothermal fluids originate from
magmas as they cool and crystallize at various levels of
the crust and give rise to porphyry-related Cu–Mo–Au
deposits (e.g., Sillitoe, 2010), skarn- and greisen-related
Sn–W deposits (e.g., Meinert et al., 2005), intrusion-
linked iron oxide Cu–Au deposits (e.g., Groves
et al., 2010), Pb–Zn–Au–Ag-rich replacement, and
epithermal Au–Ag ores (e.g., Hedenquist et al., 2000;
Simmons et al., 2005). Metamorphic fluids with aqua–
carbonic compositions (H2O + CO2), liberated during pro-
grade mineral reactions, are generally associated with oro-
genic or shear-hosted Au deposits (Goldfarb et al., 2005).
Saline connate fluids formed during diagenesis have been
implicated in the formation of stratiform sedimentary
rock-hosted Cu and Mississippi Valley-type Pb–Zn ores
(e.g., Hitzman et al., 2005; Leach et al., 2005). Seawater
circulating through the oceanic crust in the vicinity of
active ridges and vented onto sea floor as black smokers
forms Cu–Zn-dominated volcanic massive sulfide
(VMS) or Zn–Pb-dominated sedimentary exhalative
(SEDEX) deposits (e.g., Franklin et al., 2005). Circulation
of near-surface meteoric waters can dissolve, transport,
and redeposit metals, giving rise to different sedimentary
rock-hosted deposits such as a variety of uranium ores
(e.g., Cuney, 2010). The establishment of distinct and
well-based genetic hypotheses for several distinct hydro-
thermal ore deposits remains complicated or poorly under-
stood. For example, multiple fluid sources were proposed
for the well-studied Carlin-type Au deposits, including the
involvement of magmatic fluid (Sillitoe and Bonham,
1990), metamorphic and meteoric sources (Muntean
et al., 2011), and a model involving deeply sourced mag-
matic and metamorphic fluids combined with convecting
meteoric water resulting in the dissolution and transport
of gold from different locations in the crust (Cline
et al., 2005).
Accurate knowledge about the sources of heat that exert
a major control on fluid circulation and metal solubility is
a key factor to understanding the formation of hydrother-
mal mineralization. Therefore, low-temperature, radio-
metric thermochronological methods can potentially
provide a unique insight into the mechanisms of ore gene-
sis in the upper crust. Thermochronology can constrain the
location, timing, and duration of mineralizing processes. It
also can elucidate the relative roles of different heat
sources such as metamorphism, magmatism, the burial
of sediments, or the juxtaposition of high-temperature iso-
therms against cooler rocks, via rock uplift and exhuma-
tion. Furthermore, there is increasing academic and
industry interest in improving our understanding of post-
mineralization processes, which preserve or destroy ore
deposits and their overlying rocks. Low-temperature
thermochronology quantifies the cooling, and potential
exhumation histories of rocks, and therefore has useful
 HYDROTHERMAL ORES (THERMOCHRONOLOGY) 335

applications for ore preservation studies, especially when (U–Th)/He dating of hydrothermal ore deposits has
it is combined with other thermochronometric techniques. gained more and more focus with the development of
A comprehensive review of the applications of new (U–Th)/He chronometers which have closure tem-
thermochronology to the study of porphyry ore deposits peratures overlapping with hydrothermal conditions and
has been provided by McInnes et al. (2005). This entry which use minerals commonly occurring in hydrothermal
presents descriptions and examples of thermochro- ore paragenesis. As examples, concordant (U–Th)/He
nological methods that have been applied to reveal various were obtained for epidote and garnet from skarn mineral-
types of hydrothermal ore depositional environments. ization (Nicolescu and Reiners, 2005), for specular hema-
tite (Lippolt et al., 1993), for fluorite from the La Azul
fluorspar deposit (Pi et al., 2005), and for titanite
Dating hydrothermal events (Reiners and Farley, 1999) and magnetite (Blackburn
Temperature-sensitive radiometric dating techniques can et al., 2007), respectively.
be utilized very efficiently for deciphering the timing
and duration of mineralization processes. Additionally,
Mapping thermal footprints related to
because each low-temperature thermochronometer char-
hydrothermal fluid flow
acterizes a different range of temperatures, the application
of multiple thermochronologic techniques is invaluable in One of the main challenges facing exploration geologists
constraining the time–temperature history of ore deposi- investigating subsurface hydrothermal mineralization is
tion. When thermochronometers are combined with the identification of the far-field expressions (footprints)
geochronometers (e.g., U–Pb dating, Re–Os dating, or of hydrothermal systems (Kelley et al., 2006). Tradition-
Rb–Sr dating), they can be utilized to the best advantage ally, research on the footprints related to hydrothermal
and can elucidate many aspects of ore deposit genesis, fluid flow focuses on mineralogical alteration products
including intrusive emplacement, ore deposition, alter- (i.e., visual alteration) and on the patterns of primary or
ation processes, overprinting or retrograde hydrothermal secondary lithogeochemical dispersion halos, which are
events, erosion, and weathering. the result of progressive fluid–rock interaction. However,
Ar–Ar dating is regularly used to determine the timing often these proximal footprints are subtle or are partially
of mineralization processes, as K-bearing minerals often restricted to main fluid flux zones, which make it difficult
form as part of the ore mineral paragenesis or as alteration to identify and map out concealed hydrothermal ores
minerals. Biotite, K-feldspar (adularia), muscovite, and (Figure 1). A commonly neglected aspect of hydrothermal
alunite are commonly dated from various hydrothermal fluid flow is the resultant heat transfer to rock within and
deposit types, including porphyry Cu–Mo–Au,
epithermal Au–Ag, orogenic or shear-hosted Au, or Mis-
sissippi Valley-type Pb–Zn mineralization systems. The
Ar–Ar step-heating technique can yield both the mini-
mum age of host-rock crystallization (high-temperature
steps) and the timing of younger thermal events (lower-
temperature steps) on a single K-feldspar. It is therefore
very effective to determine the timing of retrograde hydro-
thermal heating events and to estimate the cooling rates of
the mineralizing systems. Recently, there has been increas-
ing interest in dating sulfide minerals where K is present in
mineral or fluid inclusions, as sulfides generally have very
low diffusivity for Ar, and hence protects the inclusions
from resetting. Promising examples include concordant
Ar–Ar ages that have been obtained for fluid inclusion-rich
sphalerites from Fankou Pb–Zn deposit (Qiu and Jiang,
2007) and for mica-bearing pyrites from Archean Kanowna
Belle Au deposit (Philips and Miller, 2006).
Fission-track thermochronology can be used for abso-
lute dating; however, this is only possible for samples that
have cooled rapidly and have remained undisturbed at, or
close to, the surface since formation. Due to complicated
thermal histories and because of the relative lack of apatite
in hydrothermal ore paragenesis, fission-track analysis is
rarely used for direct dating of hydrothermal systems.
More commonly, apparent ages are obtained which are Hydrothermal Ores (Thermochronology), Figure 1 Idealized
used for the reconstruction of the postmineral thermal representation of hydrothermal and thermal footprint zones
history (cooling, exhumation, preservation). related to single fluid flow conduits in hydrothermal ore systems.
336 HYDROTHERMAL ORES (THERMOCHRONOLOGY)
around the fluid conduit. Heating is a product of advection
(flow of heated fluid mass) and diffusion (conduction of
heat through a fluid or rock mass). A thermal halo should
develop around all hydrothermal deposits; the size of the
halo being a function of the duration of hydrothermal flow,
the magnitude of the fluid flux, the temperature difference
between the thermal fluids and the country rock, and the
size, shape, and density of the fluid conduits (Cathles
et al., 2006). Recent studies demonstrate that these thermal
haloes can be detected at distances up to several hundred
meters around hydrothermal mineralization when study-
ing stable isotope fractionation or the rate of annealing
of fission tracks within the wall rock minerals (Arehart
and Donelick, 2006; Hickey et al., 2007; Barker
et al., 2013). The extent of fission-track resetting is a
nonlinear function of the duration and magnitude of the
heating. As the magnitude and duration of conductive
heating decreases outward from the hydrothermal fluid
conduit, fission tracks close to the conduit will be reset
to a greater extent than those farther away. The pattern of
fission-track resetting will largely reflect the period
and temperature of the fluid flow event that induced the
conductive heating pulse.
Dating by apatite fission tracks (AFT; which reset as apa-
tite is heated above 60  C; Donelick et al., 2005), in combi-
nation with the apatite (U–Th)/He dating method (closure
temperature 70  C; Ehlers and Farley, 2003), is the most
widely utilized thermochronometer for mapping
low-temperature thermal fronts associated with channelized
fluid flow. Studies of sedimentary rock-hosted gold
deposits in the northern Carlin trend (USA) suggest that
single-conduit focusing of 200–240  C ore fluids over a rel-
atively short duration (104 years) is capable of completely
resetting AFT ages at a distance of 325 m, while a 10 %
change in AFT age relative to background occurs at a dis-
tance of 800 m (Hickey et al., 2014). In systems where
conduits are close enough for conductive thermal fronts to
interact (double-sided slab heating), the 10 % resetting and
total resetting fronts are able to completely propagate across
Hydrothermal Ores (Thermochronology), Figure 2 Model illustrating various possible fluids (blue arrows), metals (gray arrows),
ligands (green arrows), and heat (red arrows; A magmatic, B burial, C high geothermal gradient) (sources associated with the formation
of sedimentary rock-hosted gold mineralization from Eastern Rhodopes, Bulgaria (after Márton 2009)).
slabs 1,100 m (10 % change in AFTages relative to back-
ground) and 700 m (complete resetting of AFT ages) wide,
respectively (Hickey et al., 2014).
Genetic aspects of hydrothermal mineralization:
possible heat sources
Since the stability of metal–ligand complexes is primarily
controlled by temperature and generally enhanced by ele-
vated temperatures, it is clear that the role of heat is critical
during the leaching, transport, and deposition of metals by
hydrothermal fluids. Generally, strong thermal gradients
present the ideal conditions for the generation of high-
grade magmatic hydrothermal ores, whereas more diffuse
mineralization haloes would be expected for more slowly
cooled igneous intrusions. Moreover, many authors have
discussed the significance of thermally driven fluid flow
during the formation of basin-related hydrothermal ore
deposits. Therefore, the application of thermal history
modeling to investigate possible heat sources and their
gradients represents a powerful tool for understanding
genetic aspects of ore deposition.
As illustrated by a case study from Eastern Rhodopes,
Bulgaria (Figure 2), Márton (2009) demonstrated that suffi-
cient heat accumulated from multiple sources to sustain an
ore-deposit-forming hydrothermal system. The relative
roles of different heat sources resulting from metamor-
phism, magmatism, the burial of sediments in environment
with high geothermal gradients, or the juxtaposition of
high-temperature isotherms against cooler rocks via rock
uplift and exhumation were quantified using multiple
thermochronometers and applying appropriate sampling
profiles which consider the various geological processes.
Preservation and exhumation of hydrothermal
ore deposits
Hydrothermal ore deposits that form in the shallow crust
(<5 km) are frequently obliterated by erosion and
reworking, particularly if they form in environments that
 HYDROTHERMAL ORES (THERMOCHRONOLOGY) 337

Hydrothermal Ores (Thermochronology), Figure 3 Relationships between emplacement depth, erosion history (upper panels), and
time–temperature evolution (lower panels) during the exhumation history of epithermal Au deposits (red dots). (a) The area
experienced continuous exhumation since hypogene ore deposit formation, and the ore deposit was eroded in the past making the
area unfavorable for present-day exploration, although potential supergene/placer ores might form in related river catchment zones.
(b) After ore formation the early history of the area includes a burial period and subsequent exhumation. This scenario renders
the area favorable for present-day ore exploration.

are being exhumed (Wilkinson and Kesler, 2007). The
probability of exposure of a hypogene subsurface ore is
clearly a function of the depth at which it formed and the
rate at which it is exhumed toward the surface. Therefore,
concealment of shallow, epithermal ore deposits by post-
mineralization sedimentation represents a convenient
way to preserve them for a longer time (Figure 3). Thermal
history modeling provides a means to quantify the timing,
magnitude, and duration of heating and cooling events that
occurred during the tectonic evolution of an area and is
useful to establish a plausible exhumation history. In this
context, low-temperature thermochronological data can
be used as an exploration tool for the identification of shal-
low ore deposits in the crusts. Favorable targets can be
represented by time–temperature results that corroborate
ore emplacement depth, exhumation rate, and
age-distribution characteristics.
Márton et al. (2010) integrated the results of 40Ar/39Ar
step-heating analyses of hornblende, muscovite, biotite,
and adularia, along with zircon and apatite fission-track
analyses from upper and lower plate rocks of the
Kesebir–Kardamos metamorphic core complex from the
Eastern Rhodopes, Bulgaria, where Oligocene epithermal
sedimentary rock-hosted Au mineralization was formed in
upper plate position in close spatial relationship to detach-
ment faults. The resulting thermal history models for the
upper plate show that ore formation was followed by con-
tinuous sedimentation until 33–30 Ma, which may
338 HYDROTHERMAL ORES (THERMOCHRONOLOGY)
account for the fact that the mineralized zones were pre-
served, and the exhumation of these deposits suddenly
started with the onset of horst–graben formation at
33–30 Ma. The steep, post-mineralization normal faults
played a key role during the exhumation of the deposit
representing the bounding structures of the exhumed gold
deposits. Therefore, structural mapping should focus not
only on the low angle detachment faults, which might be
genetically related to ore formation, but also on post-
mineralization faults, which exhumed the deposits.
LA–ICP-MS apatite U–Pb dating and fission-track
thermochronology were also applied to determine the
preservation potential of magnetite–apatite deposits in
the Luzong and Ningwu volcanic basins, eastern China
(Liu et al., 2014). Results suggest that although the
metallogenic conditions of both basins were similarly
favorable for the formation of the of magnetite–apatite
deposits, the Luzong basin had better preservation poten-
tial and therefore more promise for development of the
magnetite–apatite deposits.
Quantifying fault displacement in mineralized
zones
The identification of postmineral faults and the recon-
struction of fault displacement have major implications
for the exploration of ore deposits, as it can reveal the
tilting history of an explored area and allow identification
of missing (faulted) portions of an ore deposit.
Under thermal steady-state conditions, where isotherms
are fixed relative to the surface, changes in time–tempera-
ture profiles from a certain sampling profile reflect the
movement of a sample with respect to the surface. In this
situation, low-temperature thermochronometers can be
used to constrain patterns of footwall denudation associ-
ated with fault-array evolution. Samples collected along
fault-perpendicular footwall transects can be used to con-
struct age versus elevation profiles that may constrain the
timing of fault displacement as well as the magnitude and
rate of footwall denudation. Vertical fault displacement
and erosion must be sufficient to exhume the partial reten-
tion zone; otherwise, the fault displacement will not be
resolved. Samples at higher structural levels pass through
this temperature zone earlier than those located at lower
levels, resulting in an age–elevation profile in which
apparent cooling ages decrease linearly with elevation.
For a purely vertical transect, the slope of the age–eleva-
tion profile defines the rate of footwall denudation and
may constrain the timing of fault activity.
A well-documented example of postmineral fault dis-
placement was provided at the Chuquicamata Cu–Mo
deposit, where a significant portion of the western side
of the ore body was dislocated by post-mineralization
motion along the West Fault, a major N–S strike-slip fea-
ture that cuts through Northern Chile (McInnes
et al., 1999). U–Th/He and fission-track data collected
on both sides of the West Fault documented a clear offset,
revealing the vertical component of displacement.
Younger ages in the hanging wall (to the west) indicate
that it has been exhumed relative to the footwall, and there
has been 600 m of vertical displacement along the West
Fault. The results suggest that the missing portion of the
Chuquicamata Cu–Mo deposit was not yet eroded during
tectonic uplift and denudation and should still be present
at an elevation of 3,600 m, 25–35 km along the strike
(sinistral) of the current open pit at Chuquicamata.
Summary
Thermochronologic data provides important constraints
on the role of different processes in the formation, preser-
vation, and exhumation of hydrothermal ore deposits.
Thermochronologic results integrated in time–tempera-
ture paths permit an assessment of the timing, magnitude,
and duration of hydrothermal mineralization events in
relation to magmatic, metamorphic, and sedimentary pro-
cesses. The source of heat and the thermal-gradient foot-
prints related to hydrothermal systems can be mapped
using diverse thermochronometers. Resulting models of
thermal history offer a quantitative framework for the
evaluation of ore preservation potential of an area and
for the economic assessment of different tectonic units
from a mineralized belt.
Bibliography
Arehart, G. B., and Donelick, R. A., 2006. Thermal and isotopic
profiling of the pipeline hydrothermal system: application to
exploration for Carlin-type gold deposits. Journal of Geochemi-
cal Exploration, 91, 27–40.
Barker, S., Dipple, G., Hickey, K., Lepore, W., and Vaughan, J.,
2013. Applying stable isotopes to mineral exploration: teaching
an old dog new tricks. Economic Geology, 108, 1–9.
Barnes, H. L., 1997. Geochemistry of Hydrothermal Ore Deposits,
3rd edn. New York: Wiley, p. 972.
Blackburn, T. J., Stockli, D., and Walker, D., 2007. Magnetite
(U–Th)/He dating and its application to the geochronology
of intermediate to mafic volcanic rocks. Earth and Planetary
Science Letters, 259, 360–371.
Cathles, L. M., Chen, D. F., and Nicholson, B. F., 2006.
A dimensionless vent number characterizing the thermal impact
of fluid discharge through planar and cylindrical vents with par-
ticular application to seafloor gas vents crystallizing hydrate.
Journal of Geophysical Research, 111, B10205.
Cline, J. S., Hofstra, A. H., Muntean, J. L., Tosdal, R. M., and
Hickey, K. A., 2005. Carlin-type gold deposits in Nevada: criti-
cal geologic characteristics and viable models. In Hedenquist,
J. W., Thompson, J. F. H., Goldfarb, R. J., and Richards, J. P.
(eds.), Economic Geology 100th Anniversary Volume. Littleton:
Society of Economic Geologists, pp. 451–484.
Cuney, M., 2010. Evolution of uranium fractionation processes
through time: driving the secular variation of uranium deposit
types. Economic Geology, 105, 553–569.
Donelick, R. A., O’Sullivan, P. B., and Ketcham, R. A., 2005. Apa-
tite fission-track analysis. Reviews in Mineralogy and Geochem-
istry, 58, 49–94.
Ehlers, T. A., and Farley, K. A., 2003. Apatite (U–Th)/He
thermochronometry: methods and applications to problems in
tectonic and surface processes. Earth and Planetary Science
Letters, 206, 1–14.
Franklin, J. M., Gibson, H. L., Galley, A. G., and Jonasson, I. R.,
2005. Volcanogenic massive sulfide deposits. In Hedenquist,
 HYDROTHERMAL ORES (THERMOCHRONOLOGY)
J. W., Thompson, J. F. H., Goldfarb, R. J., and Richards, J. P.
(eds.), Economic Geology 100th Anniversary Volume. Littleton:
Society of Economic Geologists, pp. 523–560.
Goldfarb, R. J., Baker, T., Dube, B., Groves, D. I., Hart, C. J. R., and
Gosselin, P., 2005. Distribution, character and genesis of gold
deposits in metamorphic terranes. In Hedenquist, J. W., Thomp-
son, J. F. H., Goldfarb, R. J., and Richards, J. P. (eds.), Economic
Geology 100th Anniversary Volume. Littleton: Society of
Economic Geologists, pp. 407–450.
Groves, I. D., Bierlein, F. P., Meinert, L. D., and Hitzman, M. W.,
2010. Iron oxide copper-gold (IOCG) deposits through earth his-
tory: implications for origin, lithospheric setting, and distinction
from other epigenetic iron oxide deposits. Economic Geology,
105, 641–654.
Hedenquist, J. W., Arribas, A. R., and Gonzales-Urien, E., 2000.
Exploration for epithermal gold deposits. Reviews in Economic
Geology, 13, 245–277.
Hickey, K. A., Tosdal, R., Dipple, G., Donelick, R., and Arehart, R.,
2007. The thermal footprint of Carlin-type Au-deposits –
defining the far-field extent of hydrothermal flow using
low-temperature thermochronology (abst.). In Andrew, C. J.
(ed.), Digging Deeper. Proceedings of the Ninth Biennial
Meeting of the Society for Geology Applied to Mineral Deposits,
Dublin, Ireland, 20–23 August 2007, Vol. 2, pp. 1251–1254.
Hickey, K. A., Barker, S. L. L., Dipple, G. M., Arehart, G. B., and
Donelick, R. A., 2014. The brevity of hydrothermal fluid flow
revealed by thermal haloes around giant gold deposits: implica-
tions for Carlin-type gold systems. Economic Geology, 109,
1461–1487.
Hitzman, M. W., Kirkham, R., Broughton, D., Thorson, J., and
Selley, D., 2005. The sediment-hosted stratiform copper ore
system. In Hedenquist, J. W., Thompson, J. F. H., Goldfarb, R. J.,
and Richards, J. P. (eds.), Economic Geology 100th Anniversary
Volume. Littleton: Society of Economic Geologists, pp. 609–642.
Kelley, D. L., Kelley, K. D., Coker, W. B., Caughlin, B., and
Doherty, M. E., 2006. Beyond the obvious limits of ore deposits:
the use of mineralogical, geochemical, and biological features
for the remote detection of mineralization. Economic Geology,
101, 729–752.
Leach, L. D., Sangster, D. F., Kelley, K. D., Large, R. R., Garven,
G., Allen, C. R., Gutzmer, J., and Walters, S., 2005. Sediment-
hosted lead-zinc deposits: a global perspective. In Hedenquist,
J. W., Thompson, J. F. H., Goldfarb, R. J., and Richards, J. P.
(eds.), Economic Geology 100th Anniversary Volume. Littleton:
Society of Economic Geologists, pp. 523–560.
Lippolt, J. J., Wernicke, R. S., and Boschmann, W., 1993. 4He dif-
fusion in specular hematite. Physics and Chemistry of Minerals,
20, 415–418.
Liu, W., Zhang, J., Sun, T., and Wang, J., 2014. Application of apa-
tite U–Pb and fission-track double dating to determine the pres-
ervation potential of magnetite–apatite deposits in the Luzong
and Ningwu volcanic basins, eastern China. Journal of Geo-
chemical Exploration, 138, 22–32.
Márton, I., 2009. Formation, Preservation and Exhumation of Sed-
imentary Rock-Hosted Gold Deposits in the Eastern Rhodopes,
Bulgaria. University of Geneva Press, Terre & Environment,
Geneva, Switzerland, Vol. 84, p. 134.
Márton, I., Moritz, R., and Spikings, R., 2010. Application of
low-temperature thermochronology to hydrothermal ore
deposits: formation, preservation and exhumation of epithermal
gold systems from the Eastern Rhodopes, Bulgaria.
Tectonophysics, 483, 240–254.
McInnes, B. I. A., Farley, K. A., Sillitoe, R. H., and Kohn, B. P.,
1999. Application of apatite U-He thermochronology to the
determination of the sense and amount of vertical fault
339
displacement at the Chuquicamata porphyry copper deposit,
Chile. Economic Geology, 94, 937–945.
McInnes, B. I. A., Evans, N. J., Fu, F. Q., and Garwin, S., 2005.
Application of thermochronology to hydrothermal ore deposits.
Reviews in Mineralogy and Geochemistry, 58, 467–498.
Meinert, L. D., Dipple, G. M., and Nicolescu, S., 2005. World skarn
deposits. In Hedenquist, J. W., Thompson, J. F. H., Goldfarb,
R. J., and Richards, J. P. (eds.), Economic Geology 100th Anni-
versary Volume. Littleton: Society of Economic Geologists,
pp. 299–336.
Muntean, J. L., Cline, J. S., Simon, A. C., and Longo, A. A., 2011.
Magmatic–hydrothermal origin of Nevada’s Carlin-type gold
deposits. Nature Geoscience, 4, 122–127.
Nicolescu, S., and Reiners, P. W., 2005. (U–Th)/He dating of epi-
dote and andradite garnet. Geochimica et Cosmochimica Acta,
69, A26.
Philips, D., and Miller, J. M., 2006. 40Ar/39Ar dating of mica-
bearing pyrite from thermally overprinted Archean gold
deposits. Geology, 34, 397–400.
Pi, T., Sole, J., and Taran, Y., 2005. (U–Th)/He dating of fluorite:
application to the La Azul fluorspar deposit in the Taxco mining
district, Mexico. Mineralium Deposita, 39, 976–982.
Pirajno, F., 1992. Hydrothermal Mineral Deposits. Principles and
Fundamental Concepts for the Exploration Geologist. Berlin:
Springer-Verlag, p. 709.
Qiu, H. N., and Jiang, Y. D., 2007. Sphalerite 40Ar/39Ar progressive
crushing and stepwise heating techniques. Earth and Planetary
Science Letters, 256, 224–232.
Reiners, P. W., and Farley, K. A., 1999. He diffusion and (U–Th)/He
thermochronometry of titanite. Geochimica et Cosmochimica
Acta, 63, 3845–3859.
Sillitoe, R. H., 2010. Porphyry copper systems. Economic Geology,
105, 3–41.
Sillitoe, R. H., and Bonham, H. F., 1990. Sediment-hosted gold
deposits: distal products of magmatic–hydrothermal systems.
Geology, 18, 157–161.
Simmons, S. F., White, N. C., and John, D. A., 2005. Geological
characteristics of epithermal precious and base metal deposits.
In Hedenquist, J. W., Thompson, J. F. H., Goldfarb, R. J., and
Richards, J. P. (eds.), Economic Geology 100th Anniversary Vol-
ume. Littleton: Society of Economic Geologists, pp. 485–522.
Wilkinson, B. H., and Kesler, S. E., 2007. Tectonism and exhuma-
tion of convergent margin orogens: insights from ore deposits.
Journal of Geology, 115, 611–627.
Cross-references
Apatite
Ar–Ar and K–Ar Dating
Exhumation (Thermochronology)
Fault Zone (Thermochronology)
Fission Track Dating and Thermochronology
Metamorphic Terranes (K–Ar/40Ar/39Ar)
Minerals (40Ar–39Ar)
Potassium–Argon (Argon–Argon), Structural Fabrics
Rubidium–Strontium Dating, Hydrothermal Events
Thermochronology, Detrital Zircon
Thermochronology, Landform Evolution
Uranium–Lead Dating
Uranium–Lead, Ore Deposits
U–Th/He Dating
Volcanic Rocks (Ar/Ar)
Volcanogenic Sedimentary Rocks (K/Ar, 40Ar/39Ar)
Zircon
 I
ICE CORES
Anders Svensson
Niels Bohr Institute, Ice and Climate Research, University
of Copenhagen, Copenhagen, Denmark
Introduction
Ice cores from Antarctica, from Greenland, and from
a number of smaller glaciers around the world yield
a wealth of information on past climates and environments.
Ice cores offer unique records on past temperatures, atmo-
spheric composition (including greenhouse gases), volca-
nism, solar activity, dustiness, and biomass burning,
among others. In Antarctica, ice cores extend back more
than 800,000 years before present (Jouzel et al., 2007),
whereas Greenland ice cores cover the last 130,000 years
(Dahl-Jensen et al., 2013). A few ice cores from high-
elevation glaciers in the Himalayas (Thompson
et al., 1997) and in the Andes (Thompson et al., 1998)
extend back into the last glacial period, whereas most other
ice cores from lower latitudes are constrained to the Holo-
cene period (last 11,700 years). In order to make proper
interpretation of ice core records, it is essential to establish
accurate and precise ice core chronologies that assign an
age to each depth segment of the core.
Under favorable conditions, the annual signal in ice
cores can be traced thousands of years back in time. The
temporal resolution of ice cores is generally high com-
pared to that of other paleoarchives, such as marine and
lacustrine sediments, stalagmites, and tree rings
(Ruddiman, 2008). The annual accumulation on glaciers
ranges from a few centimeters of ice per year in high-
elevation areas of Antarctica to several meters of ice at
costal sites and at lower latitudes. In the upper part of an
ice sheet, snow compacts to incompressible ice in
a transition zone known as the firn. Due to the continuous
J.W. Rink, J.W. Thompson (eds.), Encyclopedia of Scientific Dating Methods, DOI 10.1007/978-94-007-6304-3,
© Springer Science+Business Media Dordrecht 2015
accumulation of snow, annual layers become buried in
the ice sheet with time, and due to a general flow of ice
toward ice margins, annual layers are thinned with depth
(Figure 1). Close to the base of an ice sheet and close to
ice margins, the stratigraphy is often disturbed as the ice
is pushed over bedrock undulations.
Ice core dating methods
Ice core chronologies are based on a number of different
techniques that include annual layer counting, use of
well-dated reference horizons, glaciological modeling,
and “orbital tuning” where ice core records are linked
to well-known temporal variations of the Earth’s orbit.
Often several techniques are combined to produce the
optimal ice core chronology (Cuffey and Paterson,
2010). In regions with significant surface melt, such as
low-latitude glaciers or coastal regions, ice core dating
may be particularly challenging as the stratigraphy will
be disturbed to some degree.
Layer counting
An annual signal in ice cores may be preserved in the iso-
topic composition of the ice itself or in ice impurities such
as terrestrial dust or ionic species. The isotopic ratios
18
O/16O and 2H/1H of the most abundant stable water iso-
topes in glacier ice are related primarily to the local tem-
perature at the time of precipitation (Dansgaard
et al., 1969). They therefore vary over the seasons and
preserve a signal of the annual cycle (Figure 2). The iso-
topic rations are normalized with respect to a standard
and reported as d18O for oxygen and d2H (or dD) for
hydrogen. At high-accumulation sites, the annual signal
of those isotopes may be traced several thousands of
years back in time (Vinther et al., 2006). At lower accu-
mulation sites, however, diffusion of water molecules in
the firn reduces or completely erases the seasonal varia-
tion (Johnsen et al., 2000). Furthermore, layer thinning
342 ICE CORES

Ice Cores, Figure 1 Schematic cross section of a large glacier such as the Greenland ice sheet. Thick arrows indicate the main ice flow
pattern, and horizontal lines represent descending annual layers that are stretched and thinned over time due to ice flow. The most
well-preserved ice core profiles are obtained from the central part of the ice sheet where the ice flow is mostly vertical, and there is no
melting.

Ice Cores, Figure 2 A five meter section from a Central Greenland ice core showing annual layers in the stable water isotopes (d18O).
High d18O values are summer layers, and low values occur during winter. The ice core is dated by annual layer counting from the
surface, and the shown depth interval covers the time period 1946–1955 AD. At high-accumulation sites, annual layers can be
identified in ice cores water isotopes thousands of years back in time (Figure from www.iceandclimate.nbi.ku.dk).

caused by ice flow eventually makes the water isotopic
seasonal cycle disappear with depth.
Ice core impurities, such as terrestrial dust and sea salt,
often express a seasonal cycle in ice cores. For example, at
present, Greenland dust generally shows a maximum dur-
ing springtime, whereas sea salt peaks during winter
(Kuramoto et al., 2011). The dust annual cycle may even
be observed directly in the visual stratigraphy of the ice
core (Alley et al., 1997; Figure 3). Because dust particles
are fixed in the ice matrix, they are less vulnerable to dif-
fusion than water isotopes and ionic species, and they,
therefore, preserve annual layers much further back in
time. Extensive high-resolution dust and chemical impu-
rity datasets from several deep Greenland ice cores have
allowed for multi-proxy layer counting and tracing of
annual layers well into the last glacial period (Andersen
et al., 2006).
The strength of the layer-counted time scales lies in the
very well-constrained event durations that allow precise
determination of the sequence of events and of the exact
event duration of past climate change (Steffensen
et al., 2008). On the other hand, the counting error of
counted time scales increases with age making the
absolute error fairly large compared to other dating
methods. Recently, automated methods for annual layer
counting in ice cores have become available (Winstrup
et al., 2012), which may play a more important role in
future ice core chronologies. Annual layer counting has
been applied to ice cores worldwide, but Central Green-
land and coastal (high-accumulation) sites in Antarctica
are the only sites where annual layer counting is feasible
in the early Holocene and in the last glacial period.
Stratigraphic linking
Numerous well-dated volcanic eruptions have left traces
in ice cores, sometimes in the form of ash particles or
“tephra” (Abbott and Davies, 2012; Narcisi et al., 2012)
but more often in the form of elevated acidity levels
(mostly sulfuric acid) (Figure 4). For example, the well-
known 1815 AD Tambora eruption (Indonesia) has been
identified as an acidity spike in many ice cores. During
the last glacial period, the North Atlantic Ash Zones
I and II provide important tephra marker horizons for
Greenland at 12 and 38 ka before present, respectively,
whereas Mt. Berlin is a central reference tephra layer in
 ICE CORES 343

Ice Cores, Figure 3 The annual layering of the Greenland ice sheet is visible to the naked eye. This example shows 5  1.65 m
sections of visual stratigraphy from 2050 m depth in the NGRIP ice core from Central Greenland. At this depth, the ice is some
38 thousand years (ka) old, and the shown sections cover some 275 years of precipitation. The image is obtained with indirect
illumination of the ice core so that black areas correspond to clear ice, whereas white layers contain elevated amounts of dust
(Data from Svensson et al., 2005).

Ice Cores, Figure 4 Example of a visible tephra layer from the Icelandic Katla volcanic in the Greenland NGRIP ice core. The layer is
from 1895.24 m depth and has an age of about 29 ka. (a) The visible tephra horizon in the ice core. (b) The associated sulfate peak in
the ice. (c) Basaltic glass shards from the eruption as found in the ice core (Figure modified from Abbott and Davies, 2012).

Antarctica at 92 thousand years before present (ka BP).
Besides providing information on the timing and magni-
tude of past eruptions, ice core acidity spikes are also
widely used for synchronization of ice cores. All of the
deep Greenland ice cores are thus synchronized through-
out the last glacial cycle based on numerous volcanic
markers (Vinther et al., 2006; Rasmussen et al., 2008),
and within Antarctica ice cores are being linked in
a similar way (Severi et al., 2007; Parrenin et al., 2012).
A number of different approaches have been taken to
synchronize ice cores from Greenland with those of Ant-
arctica whereby climate change can be studied on
a global scale. The air bubbles in ice cores from both hemi-
spheres share a global signal of well-mixed atmospheric
gases. For example, methane concentrations show great
variability over the last glacial period and can be applied
for bipolar ice core matching (Blunier and Brook, 2001;
EPICA Community Members, 2006). Also the global sig-
nal of oxygen isotopes of air trapped in ice cores can be
applied to match ice cores from the two hemispheres
(Bender et al., 1994; Capron et al., 2010). Bipolar volcanic
matching of ice cores is feasible when large low-latitude
eruptions inject great quantities of acid into the strato-
sphere. Over the last two millennia, where ice cores are
344
precisely dated, some 50 bipolar volcanic events have thus
been identified (Sigl et al., 2013). Greenland and Antarctic
ice cores are also synchronized at the time of the largest
volcanic eruption of the last million years, the Toba erup-
tion (Indonesia) occurring some 74 ka before present
(Svensson et al., 2013). Furthermore, the global signal of
cosmogenic 10Be in ice cores has been applied to match
Greenland and Antarctic ice cores at the Laschamp geo-
magnetic excursion that occurred about 41 ka ago
(Raisbeck et al., 2007).
Orbital tuning
On time scales of tens of thousands of years, the Earth’s
climate is modulated by orbital variations that influence
the solar insolation distribution on Earth. Several param-
eters measured in ice cores are shown to vary with the
orbital parameters that are known with high accuracy
(Laskar et al., 2004). The older section of Antarctic ice
cores is dated by tuning the ice core record variability
to that of the orbital parameters applying so-called orbital
tuning (Figure 5). The ice core records expressing vari-
ability on orbital time scales include the oxygen isotope
ratio of air bubbles (Sowers et al., 1989; Landais
et al., 2010), the O2/N2 ratio of air bubbles (Bender
et al., 1994; Kawamura et al., 2007), and the total air con-
tent of the ice (Raynaud et al., 2007). Unfortunately, the
linking between orbital forcing and the ice core records
is rather poorly constrained. Therefore, the use of orbital
tuning for ice core dating implies uncertainties of thou-
sands of years (Bazin et al., 2013).
Glaciological modeling
In the deeper part of the large ice sheets and at locations
where annual layers are not available, ice core age scales
are based on modeling. To model an ice core time scale,
Ice Cores, Figure 5 Example of orbital tuning of the Antarctic TALDICE ice core for the past 250 ka. Top: stable water isotopes of the
ice (d18Oice) with labeling of interglacial periods (MIS5e and MIS7e). Middle: stable water isotopes of the gas record (d18Oatm) with
colors referring to different datasets. Bottom: the Earth’s orbital precession parameter that partly controls the d18Oatm and that is used
for tuning the time scale (Figure from Bazin et al., 2013).
ICE CORES
it is needed to reconstruct (1) past accumulation rates
and (2) the thinning of annual layers with depth. In polar
regions, accumulation rates are to first order linked to tem-
peratures that may be estimated from temperature proxies,
such as stable water isotopes (d18O and dD). However,
many other parameters influence the accumulation as
well, for example, past ice sheet elevation changes,
“upstream effects” (i.e., due to ice flow, the ice in the
deeper part of an ice core must be traced back to its origin
at the surface), and changes in the precipitation source(s).
The ice flow may be simulated by simple one- or
two-dimensional approaches (Johnsen and Dansgaard,
1992) or by sophisticated 3D models that take into account
many of the physical properties of the ice sheet, such as ice
temperatures, ice impurity content, ice crystal texture, and
basal melting (Ritz et al., 2001). Generally, many of those
properties are poorly known, and modeled ice core chro-
nologies are often constrained by other dating methods,
such as age markers or orbital tuning. The most recent
ice core age models use inverse methods and combine
records from several ice cores and their stratigraphic links
to produce a consistent chronology common for all the
cores (Bazin et al., 2013).
Radiometric and other dating methods
Currently, radiometric methods have rather limited appli-
cations for ice core dating. Atmospheric testing of
nuclear weapons has produced reference horizons in
1950–1960 that can be detected as elevated beta-activity
of ice. Young ice can be dated by 210Pb or 32Si counting,
but the precision cannot compete with that of annual
layer counting. Due to a significant in situ production,
ice cores can generally not be dated by 14C (Petrenko
et al., 2013) unless the ice contains sufficient biogenic
material (Jenk et al., 2009). As a consequence, ice core
 ICE CORES
time scales and 14C calibration curves may not be
entirely compatible during the last glacial period
(Muscheler et al., 2008). The cosmogenic isotopes of
10
Be and 36Cl can be measured in ice, but due to long
half-life times and poorly constrained production rates,
they mostly have applications for age estimation of very
old ice, such as basal ice. The Matuyama-Brunhes (M-B)
geomagnetic reversal observed as a 10Be minimum at
776 ka is applied as a reference marker to constrain ages
of the oldest Antarctic ice cores (Raisbeck et al., 2006).
Recently, developments have been made toward dating
Antarctic ice cores based on the accumulation of 234U in
the ice matrix from recoil during 238U decay out of dust
bound within the ice (Aciego et al., 2011). Also 81Kr
may have a potential for dating of old ice, but its abun-
dance in glacier ice is very low (Jiang et al., 2012). The
age of basal ice containing material from bedrock (silty
ice) can be estimated by a number of alternative methods
such as 10Be/36Cl isotope ratios, optically stimulated lumi-
nescence, amino acid racemization, and maximum likeli-
hood estimates for the branch length of the invertebrate
cytochrome oxidase subunit I sequences (Willerslev
et al., 2007).
Gas record chronologies
An important issue for ice core time scales is that of dat-
ing the air bubbles enclosed in ice cores. The snow or firn
in the upper part of an ice sheet is generally permeable to
air, and air bubbles are only permanently enclosed in the
ice at the so-called lock-in depth that occurs typically at
50–100 m depth when snow compacts to ice. Because
of the mixing of air in the upper part of the ice sheet,
the age of air in an ice core sample is always younger
than that of the encapsulating ice (Schwander
et al., 1993). This difference in age between the air bub-
bles and surrounding ice is site specific and depends
mainly on past accumulation rates and temperatures
(Goujon et al., 2003). The gas-ice age difference varies
from a few decades at present-day high-accumulation
sites to several millennia at low-accumulation sites dur-
ing glacial periods (Loulergue et al., 2007). The age dif-
ference can be constrained by modeling and by isotopic
measurements of air molecules that respond to tempera-
ture gradients in the snow during bubble formation
(Severinghaus and Brook, 1999). Because it is important
to link the climate signal of the ice to that of the atmo-
spheric composition contained in the air bubbles, there
are generally two age scales associated with an ice core:
one for the ice and one for the gas records.
Greenland and antarctic ice core chronologies
Numerous chronologies have been made available for
Greenland ice cores over the last decades (Southon, 2004).
Today, all of the deep Greenland ice cores (DYE-3, GRIP,
GISP2, NGRIP, NEEM, Camp Century, Renland; see refer-
ences in Johnsen et al. (2001)) are on a common time scale
named the Greenland Ice Core Chronology 2005 (GICC05).
345
This time scale is based on layer counting in several ice
cores and covers the last 60,000 years (Svensson et al., 2008).
For the most recent 8000 years, GICC05 is based on
counting of annual layers in the water isotopic signal
(Vinther et al., 2006), whereas dating of the older part is
based on layer counting in dust and other impurities in the
ice (Andersen et al., 2006; Rasmussen et al., 2006). The
GICC05 time scale is associated with a counting uncertainty
that provides an error estimate. The absolute error is around
1 % in the Holocene and about 5 % in the last glacial period.
GICC05 has been extended with an ice flow model to cover
the 123,000 time span of the Greenland ice cores (Wolff
et al., 2010). The Greenland ice cores are internally synchro-
nized using common volcanic reference horizons
(Rasmussen et al., 2008), and tephra is used to link the
Greenland ice cores to other paleoarchives (Abbott and
Davies, 2012; Blockley et al., 2012).
For Antarctica, individual time scales have traditionally
been developed for each of the deep ice cores (Byrd,
Vostok, EDC, Dome Fuji, EDML, TALDICE, Law Dome,
Taylor Dome, see references in Stenni et al. (2010)). The
time scales are mostly based on ice flow models that are
constrained by stratigraphic makers and orbital tuning
(Kawamura et al., 2007; Parrenin et al., 2007). Recently,
inverse methods have been applied to integrate several
Antarctic ice core chronologies with those of Greenland
(Lemieux-Dudon et al., 2010). The most recent ice core
time scale for Antarctica is the Antarctic Ice Core Chro-
nology 2012 (AICC2012) that applies to several deep Ant-
arctic ice cores (Bazin et al., 2013; Veres et al., 2013). For
the last 60,000 years, AICC2012 is synchronized to the
Greenland ice cores through their global gas concentration
records and through bipolar volcanic and geomagnetic
events. For the older part, AICC2012 is constrained by
independently dated tephra layers, by orbital tuning, and
by the B-M geomagnetic boundary. For the last glacial
cycle, the AICC2012 error estimate is similar to that of
Greenland time scales, and beyond that the accuracy is
within 5000 years back to the M-B boundary. It is
expected that more Antarctic ice cores will be integrated
with the AICC2012 time scale and that a new layer-
counted time scale will become available from the recently
recovered WAIS ice core reaching back to 68 ka (WAIS
Divide Project Members, 2013).
Bibliography
Abbott, P. M., and Davies, S. M., 2012. Volcanism and the Green-
land ice-cores: the tephra record. Earth-Science Reviews, 115,
173–191.
Aciego, S., Bourdon, B., Schwander, J., Baur, H., and Forieri, A.,
2011. Toward a radiometric ice clock: uranium ages of the Dome
C ice core. Quaternary Science Reviews, 30, 2389–2397.
Alley, R. B., Shuman, C. A., Meese, D. A., Gow, A. J., Taylor,
K. C., Cuffey, K. M., Fitzpatrick, J. J., Grootes, P. M., Zielinski,
G. A., Ram, M., Spinelli, G., and Elder, B., 1997. Visual-
stratigraphic dating of the GISP2 ice core: Basic, reproducibility,
and application. Journal of Geophysical Research, 102,
26367–26382.
346 ICE CORES
Andersen, K. K., Svensson, A., Rasmussen, S. O., Steffensen, J. P.,
Johnsen, S. J., Bigler, M., Röthlisberger, R., Ruth, U., Siggaard-
Andersen, M.-L., Dahl-Jensen, D., Vinther, B. M., and Clausen,
H. B., 2006. The Greenland Ice Core Chronology 2005, 15–42
ka. Part 1: constructing the time scale. Quaternary Science
Reviews, 25, 3246–3257.
Bazin, L., Landais, A., Lemieux-Dudon, B., Kele, H. T. M., Veres,
D., Parrenin, F., Martinerie, P., Ritz, C., Capron, E., Lipenkov,
V., Loutre, M.-F., Raynaud, D., Vinther, B., Svensson, A., Ras-
mussen, S. O., Severi, M., Blunier, T., Leuenberger, M., Fischer,
H., Masson-Delmotte, V., Chappellaz, J., and Wolff, E., 2013.
An optimized multi-proxy, multi-site Antarctic ice and gas
orbital chronology (AICC2012): 120–800 ka. Climate of the
Past, 9, 1715–1731.
Bender, M., Sowers, T., Dickson, M.-L., Orchardo, J., Grootes, P.,
Mayewski, P. A., and Meese, D. A., 1994. Climate correlations
between Greenland and Antarctica during the last 100,000 years.
Nature, 372, 663–666.
Blockley, S. P. E., Lane, C. S., Hardiman, M., Rasmussen, S. O.,
Seierstad, I. K., Steffensen, J. P., Svensson, A., Lotter, A. F.,
Turney, C. S. M., and Ramsey, C. B., 2012. Synchronisation of
palaeoenvironmental records over the last 60,000 years, and an
extended INTIMATE(1) event stratigraphy to 48,000 b2k. Qua-
ternary Science Reviews, 36, 2–10.
Blunier, T., and Brook, E. J., 2001. Timing of millennial-scale cli-
mate change in Antarctica and Greenland during the last glacial
period. Science, 291, 109–112.
Capron, E., Landais, A., Lemieux-Dudon, B., Schilt, A., Masson-
Delmotte, V., Buiron, D., Chappellaz, J., Dahl-Jensen, D.,
Johnsen, S., Leuenberger, M., Loulergue, L., and Oerter, H.,
2010. Synchronising EDML and NorthGRIP ice cores using
delta O-18 of atmospheric oxygen (delta O-18(atm)) and CH4
measurements over MIS5 (80–123 kyr). Quaternary Science
Reviews, 29, 222–234.
Cuffey, K. M., and Paterson, W. S. B., 2010. The Physics of Gla-
ciers. Amsterdam: Elsevier.
Dahl-Jensen, D., Albert, M. R., Aldahan, A., Azuma, N., Balslev-
Clausen, D., Baumgartner, M., Berggren, A. M., Bigler, M.,
Binder, T., Blunier, T., Bourgeois, J. C., Brook, E. J., Buchardt,
S. L., Buizert, C., Capron, E., Chappellaz, J., Chung, J., Clausen,
H. B., Cvijanovic, I., Davies, S. M., Ditlevsen, P., Eicher, O.,
Fischer, H., Fisher, D. A., Fleet, L. G., Gfeller, G., Gkinis, V.,
Gogineni, S., Goto-Azuma, K., Grinsted, A., Gudlaugsdottir,
H., Guillevic, M., Hansen, S. B., Hansson, M., Hirabayashi,
M., Hong, S., Hur, S. D., Huybrechts, P., Hvidberg, C. S., Iizuka,
Y., Jenk, T., Johnsen, S. J., Jones, T. R., Jouzel, J., Karlsson,
N. B., Kawamura, K., Keegan, K., Kettner, E., Kipfstuhl, S.,
Kjaer, H. A., Koutnik, M., Kuramoto, T., Kohler, P., Laepple,
T., Landais, A., Langen, P. L., Larsen, L. B., Leuenberger, D.,
Leuenberger, M., Leuschen, C., Li, J., Lipenkov, V., Martinerie,
P., Maselli, O. J., Masson-Delmotte, V., McConnell, J. R., Miller,
H., Mini, O., Miyamoto, A., Montagnat-Rentier, M., Mulvaney,
R., Muscheler, R., Orsi, A. J., Paden, J., Panton, C., Pattyn, F.,
Petit, J. R., Pol, K., Popp, T., Possnert, G., Prie, F., Prokopiou,
M., Quiquet, A., Rasmussen, S. O., Raynaud, D., Ren, J.,
Reutenauer, C., Ritz, C., Rockmann, T., Rosen, J. L., Rubino,
M., Rybak, O., Samyn, D., Sapart, C. J., Schilt, A., Schmidt,
A. M. Z., Schwander, J., Schupbach, S., Seierstad, I.,
Severinghaus, J. P., et al., 2013. Eemian interglacial
reconstructed from a Greenland folded ice core. Nature, 493,
489–494.
Dansgaard, W., Johnsen, S. J., Moller, J., and Langway, C. C., Jr.,
1969. One thousand centuries of climatic record from Camp
Century on the Greenland ice sheet. Science, 166, 377–381.
EPICA Community Members, 2006. One-to-one coupling of glacial
climate variability in Greenland and Antarctica. Nature, 444,
195–198.
Goujon, C., Barnola, J.-M., and Ritz, C., 2003. Modeling the densi-
fication of polar firn including heat diffusion: Application to
close-off characteristics and gas isotopic fractionation for Ant-
arctica and Greenland sites. Journal of Geophysical Research,
108, 4792, doi:10.1029/2002JD003319.
Jenk, T. M., Szidat, S., Bolius, D., Sigl, M., Gaggeler, H. W.,
Wacker, L., Ruff, M., Barbante, C., Boutron, C. F., and
Schwikowski, M., 2009. A novel radiocarbon dating technique
applied to an ice core from the Alps indicating late Pleistocene
ages. Journal of Geophysical Research-Atmospheres, 114.
Jiang, W., Bailey, K., Lu, Z. T., Mueller, P., O’Connor, T. P., Cheng,
C. F., Hu, S. M., Purtschert, R., Sturchio, N. C., Sun, Y. R., Wil-
liams, W. D., and Yang, G. M., 2012. An atom counter for mea-
suring Kr-81 and Kr-85 in environmental samples. Geochimica
Et Cosmochimica Acta, 91, 1–6.
Johnsen, S. J., and Dansgaard, W., 1992. On flow model dating of
stable isotope records from Greenland ice cores. In Bard, E.,
and Broecker, W. S. (eds.), The Last Deglaciation: Absolute
and Radiocarbon Chronologies. New York: Springer-Verlag.
NATO ASI Series, Vol. I, pp. 13–24.
Johnsen, S. J., Clausen, H. B., Cuffey, K. M., Hoffmann, G.,
Schwander, J., and Creyts, T., 2000. Diffusion of stable isotopes
in polar firn and ice: The isotope effect in firn diffusion. In
Hondoh, T. (ed.), Physics of Ice Core Records. Sapporo: Hok-
kaido University Press, pp. 121–140.
Johnsen, S. J., DahlJensen, D., Gundestrup, N., Steffensen, J. P.,
Clausen, H. B., Miller, H., Masson-Delmotte, V.,
Sveinbjornsdottir, A. E., and White, J., 2001. Oxygen isotope
and palaeotemperature records from six Greenland ice-core sta-
tions: Camp Century, Dye-3, GRIP, GISP2, Renland and
NorthGRIP. Journal of Quaternary Science, 16, 299–307.
Jouzel, J., Masson-Delmotte, V., Cattani, O., Dreyfus, G., Falourd,
S., Hoffmann, G., Minster, B., Nouet, J., Barnola, J. M.,
Chappellaz, J., Fischer, H., Gallet, J. C., Johnsen, S.,
Leuenberger, M., Loulergue, L., Luethi, D., Oerter, H., Parrenin,
F., Raisbeck, G., Raynaud, D., Schilt, A., Schwander, J., Selmo,
E., Souchez, R., Spahni, R., Stauffer, B., Steffensen, J. P., Stenni,
B., Stocker, T. F., Tison, J. L., Werner, M., and Wolff, E. W.,
2007. Orbital and millennial Antarctic climate variability over
the past 800,000 years. Science, 317, 793–796.
Kawamura, K., Parrenin, F., Lisiecki, L., Uemura, R., Vimeux, F.,
Severinghaus, J. P., Hutterli, M. A., Nakazawa, T., Aoki, S.,
Jouzel, J., Raymo, M. E., Matsumoto, K., Nakata, H., Motoyama,
H., Fujita, S., Goto-Azuma, K., Fujii, Y., and Watanabe, O., 2007.
Northern Hemisphere forcing of climatic cycles in Antarctica over
the past 360,000 years. Nature, 448, 912–U4.
Kuramoto, T., Goto-Azuma, K., Hirabayashi, M., Miyake, T.,
Motoyama, H., Dahl-Jensen, D., and Steffensen, J. P., 2011. Sea-
sonal variations of snow chemistry at NEEM, Greenland. Annals
of Glaciology, 52, 193–200.
Landais, A., Dreyfus, G., Capron, E., Masson-Delmotte, V.,
Sanchez-Goni, M. F., Desprat, S., Hoffmann, G., Jouzel, J.,
Leuenberger, M., and Johnsen, S., 2010. What drives the millen-
nial and orbital variations of delta O-18(atm)? Quaternary Sci-
ence Reviews, 29, 235–246.
Laskar, J., Robutel, P., Joutel, F., Gastineau, M., Correia, A. C. M.,
and Levrard, B., 2004. A long-term numerical solution for the
insolation quantities of the Earth. Astronomy & Astrophysics,
428, 261–285.
Lemieux-Dudon, B., Blayo, E., Petit, J. R., Waelbroeck, C.,
Svensson, A., Ritz, C., Barnola, J. M., Narcisi, B. M., and
Parrenin, F., 2010. Consistent dating for Antarctic and Green-
land ice cores. Quaternary Science Reviews, 29, 8–20.
Loulergue, L., Parrenin, F., Blunier, T., Barnola, J.-M., Spahni, R.,
Schilt, A., Raisbeck, G., and Chappellaz, J., 2007. New con-
straints on the gas age-ice age difference along the EPICA ice
cores, 0–50 kyr. Climate of the Past, 3, 527–540.
 ICE CORES
Muscheler, R., Kromer, B., Björck, S., Svensson, A., Friedrich, M.,
Kaiser, K. F., and Southon, J., 2008. Tree rings and ice cores
reveal 14C calibration uncertainties during the Younger Dryas.
Nature Geoscience, 1, 263–267.
Narcisi, B., Petit, J. R., Delmonte, B., Scarchilli, C., and Stenni, B.,
2012. A 16,000-yr tephra framework for the Antarctic ice sheet:
a contribution from the new Tabs Dome core. Quaternary Sci-
ence Reviews, 49, 52–63.
Parrenin, F., Barnola, J.-M., Beer, J., Blunier, T., Castellano, E.,
Chappellaz, J., Dreyfus, G., Fischer, H., Fujita, S., Jouzel, J.,
Kawamura, K., Lemieux-Dudon, B., Loulergue, L., Masson-
Delmotte, V., Narcisi, B., Petit, J.-R., Raisbeck, G., Raynaud,
D., Ruth, U., Schwander, J., Severi, M., Spahni, R., Steffensen,
J. P., Svensson, A., Udisti, R., Waelbroeck, C., and Wolff, E.,
2007. The EDC3 chronology for the EPICA Dome C ice core.
Climate of the Past, 3, 485–497.
Parrenin, F., Petit, J. R., Masson-Delmotte, V., Wolff, E., Basile-
Doelsch, I., Jouzel, J., Lipenkov, V., Rasmussen, S. O.,
Schwander, J., Severi, M., Udisti, R., Veres, D., and Vinther,
B. M., 2012. Volcanic synchronisation between the EPICA
Dome C and Vostok ice cores (Antarctica) 0–145 kyr
BP. Climate of the Past, 8, 1031–1045.
Petrenko, V. V., Severinghaus, J. P., Smith, A. M., Riedel, K.,
Baggenstos, D., Harth, C., Orsi, A., Hua, Q., Franz, P., Takeshita,
Y., Brailsford, G. W., Weiss, R. F., Buizert, C., Dickson, A., and
Schaefer, H., 2013. High-precision C-14 measurements demon-
strate production of in situ cosmogenic (CH4)–C-14 and rapid
loss of in situ cosmogenic (CO)–C-14 in shallow Greenland firn.
Earth and Planetary Science Letters, 365, 190–197.
Raisbeck, G. M., Yiou, F., Cattani, O., and Jouzel, J., 2006. 10Be
evidence for the Matuyama–Brunhes geomagnetic reversal in
the EPICA Dome C ice core. Nature, 444, 82–84.
Raisbeck, G. M., Yiou, F., Jouzel, J., and Stocker, T. F., 2007. Direct
north–south synchronization of abrupt climate change records in
ice cores using beryllium 10. Climate of the Past, 3, 541–547.
Rasmussen, S. O., Andersen, K. K., Svensson, A. M., Steffensen,
J. P., Vinther, B. M., Clausen, H. B., Siggaard-Andersen, M.-
L., Johnsen, S. J., Larsen, L. B., Dahl-Jensen, D., Bigler, M.,
Röthlisberger, R., Fischer, H., Goto-Azuma, K., Hansson,
M. E., and Ruth, U., 2006. A new Greenland ice core chronology
for the last glacial termination. Journal of Geophysical
Research, 111, D06102, doi:10.1029/2005JD006079.
Rasmussen, S. O., Seierstad, I. K., Andersen, K. K., Bigler, M.,
Dahl-Jensen, D., and Johnsen, S. J., 2008. Synchronization of
the NGRIP, GRIP, and GISP2 ice cores across MIS 2 and
palaeoclimatic implications. Quaternary Science Reviews, 27,
18–28.
Raynaud, D., Lipenkov, V., Lemieux-Dudon, B., Duval, P., Loutre,
M. F., and Lhomme, N., 2007. The local insolation signature of
air content in Antarctic ice. A new step toward an absolute dating
of ice records. Earth and Planetary Science Letters, 261, 337–349.
Ritz, C., Rommelaere, V., and Dumas, C., 2001. Modeling the evo-
lution of Antarctic ice sheet over the last 420,000 years: Implica-
tions for altitude changes in the Vostok region. Journal of
Geophysical Research, 106, 31943–31964.
Ruddiman, W. F., 2008. Earth’s Climate, Past and Future. New
York: W. H. Freeman.
Schwander, J., Barnola, J.-M., Andrié, C., Leuenberger, M., Ludin,
A., Raynaud, D., and Stauffer, B., 1993. The age of the air in the
firn and the ice at Summit, Greenland. Journal of Geophysical
Research, 98, 2831–2838.
Severi, M., Becagli, S., Castellano, E., Morganti, A., Traversi, R.,
Udisti, R., Ruth, U., Fischer, H., Huybrechts, P., Wolff, E.,
Parrenin, F., Kaufmann, P., Lambert, F., and Steffensen, J. P.,
2007. Synchronisation of the EDML and EDC ice cores for the
last 52 kyr by volcanic signature matching. Climate of the Past,
3, 367–374.
347
Severinghaus, J. P., and Brook, E. J., 1999. Abrupt climate change at
the end of the last glacial period inferred from trapped air in polar
ice. Science, 286, 930–934.
Sigl, M., McConnell, J. R., Layman, L., Maselli, O., McGwire, K.,
et al., 2013. A new bipolar ice core record of volcanism from
WAIS Divide and NEEM and implications for climate forcing
of the last 2000 years. Journal of Geophysical Research-Atmo-
spheres, 118, 1151–1169, 10.1029/2012jd018603.
Southon, J., 2004. A radiocarbon perspective on Greenland ice-core
chronologies: can we use ice cores for 14C calibration? Radio-
carbon, 46, 1239–1259.
Sowers, T. A., Bender, M. L., and Raynaud, D., 1989. Elemental
and isotopic composition of occluded O2 and N2 in polar ice.
Journal of Geophysical Research, 94, 5137–5150.
Steffensen, J. P., Andersen, K. K., Bigler, M., Clausen, H. B., Dahl-
Jensen, D., Fischer, H., Goto-Azuma, K., Hansson, M., Johnsen,
S. J., Jouzel, J., Masson-Delmotte, V., Popp, T., Rasmussen,
S. O., Röthlisberger, R., Ruth, U., Stauffer, B., Siggaard-
Andersen, M.-L., Sveinbjörnsdóttir, Á. E., Svensson, A., and
White, J. W. C., 2008. High-resolution Greenland ice core data
show abrupt climate change happens in few years. Science,
321, 680–684.
Stenni, B., Buiron, D., Frezzotti, M., Albani, S., Barbante, C., Bard,
E., Barnola, J. M., Baroni, M., Baumgartner, M., Bonazza, M.,
Capron, E., Castellano, E., Chappellaz, J., Delmonte, B.,
Falourd, S., Genoni, L., Iacumin, P., Jouzel, J., Kipfstuhl, S.,
Landais, A., Lemieux-Dudon, B., Maggi, V., Masson-Delmotte,
V., Mazzola, C., Minster, B., Montagnat, M., Mulvaney, R.,
Narcisi, B., Oerter, H., Parrenin, F., Petit, J. R., Ritz, C.,
Scarchilli, C., Schilt, A., Schupbach, S., Schwander, J., Selmo,
E., Severi, M., Stocker, T. F., and Udisti, R., 2010. Expression
of the bipolar see-saw in Antarctic climate records during the last
deglaciation. Nature Geoscience, 4, 46–49.
Svensson, A., Nielsen, S. W., Kipfstuhl, S., Johnsen, S. J.,
Steffensen, J. P., Bigler, M., Ruth, U., and Röthlisberger, R.,
2005. Visual stratigraphy of the North Greenland Ice Core Pro-
ject (NorthGRIP) ice core during the last glacial period. Journal
of Geophysical Research, 110, D02108, doi:10.1029/
2004JD005134.
Svensson, A., Andersen, K. K., Bigler, M., Clausen, H. B., Dahl-
Jensen, D., Davies, S. M., Johnsen, S. J., Muscheler, R.,
Parrenin, F., Rasmussen, S. O., Röthlisberger, R., Seierstad, I.,
Steffensen, J. P., and Vinther, B. M., 2008. A 60 000 year Green-
land stratigraphic ice core chronology. Climate of the Past, 4,
47–57.
Svensson, A., Bigler, M., Blunier, T., Clausen, H. B., Dahl-Jensen,
D., Fischer, H., Fujita, S., Goto-Azuma, K., Johnsen, S. J.,
Kawamura, K., Kipfstuhl, S., Kohno, M., Parrenin, F., Popp,
T., Rasmussen, S. O., Schwander, J., Seierstad, I., Severi, M.,
Steffensen, J. P., Udisti, R., Uemura, R., Vallelonga, P., Vinther,
B. M., Wegner, A., Wilhelms, F., and Winstrup, M., 2013. Direct
linking of Greenland and Antarctic ice cores at the Toba eruption
(74 ka BP). Climate of the Past, 9, 749–766.
Thompson, L. G., Yao, T., Davis, M. E., Henderson, K. A., Mosley-
Thompson, E., Lin, P.-N., Beer, J., Synal, H.-A., Cole-Dai, J.,
and Bolzan, J. F., 1997. Tropical climate instability: the last gla-
cial cycle from a Qinghai-Tibetan ice core. Science, 276,
1821–1825.
Thompson, L. C., Davis, M. E., Mosley-Thompson, E., Sowers,
T. A., Henderson, K. A., Zagorodnov, V. S., Lun, P.-N.,
Mikhalenko, V. N., Campen, R. K., Bolzan, J. F., Cole-Dai, J.,
and Francou, B., 1998. A 25.000-year tropical climate history
from Bolivian ice cores. Science, 282, 1858–1864.
Veres, D., Bazin, L., Landais, A., Kele, H. T. M., Lemieux-Dudon,
B., Parrenin, F., Martinerie, P., Blayo, E., Blunier, T., Capron, E.,
Chappellaz, J., Rasmussen, S. O., Severi, M., Svensson, A.,
Vinther, B., and Wolff, E. W., 2013. The Antarctic ice core
348 IMPACT GLASS (FISSION TRACKS)
chronology (AICC2012): an optimized multi-parameter and
multi-site dating approach for the last 120 thousand years.
Climate of the Past, 9, 1733–1748.
Vinther, B. M., Clausen, H. B., Johnsen, S. J., Rasmussen, S. O.,
Andersen, K. K., Buchardt, S. L., Dahl-Jensen, D., Seierstad,
I. K., Siggaard-Andersen, M.-L., Steffensen, J. P., Svensson,
A. M., Olsen, J., and Heinemeier, J., 2006. A synchronized dat-
ing of three Greenland ice cores throughout the Holocene. Jour-
nal of Geophysical Research, 111, D13102, doi:10.1029/
2005JD006921.
WAIS Divide Project Members, 2013. Onset of deglacial warming
in West Antarctica driven by local orbital forcing. Nature, 500.
Willerslev, E., Cappellini, E., Boomsma, W., Nielsen, R.,
Hebsgaard, M. B., Brand, T. B., Hofreiter, M., Bunce, M.,
Poinar, H. N., Dahl-Jensen, D., Johnsen, S., Steffensen, J. P.,
Bennike, O., Schwenninger, J. L., Nathan, R., Armitage, S., de
Hoog, C. J., Alfimov, V., Christl, M., Beer, J., Muscheler, R.,
Barker, J., Sharp, M., Penkman, K. E. H., Haile, J., Taberlet, P.,
Gilbert, M. T. P., Casoli, A., Campani, E., and Collins, M. J.,
2007. Ancient biomolecules from deep ice cores reveal
a forested Southern Greenland. Science, 317, 111–114.
Winstrup, M., Svensson, A. M., Rasmussen, S. O., Winther, O.,
Steig, E. J., and Axelrod, A. E., 2012. An automated approach
for annual layer counting in ice cores. Climate of the Past, 8,
1881–1895.
Wolff, E. W., Chappellaz, J., Blunier, T., Rasmussen, S. O., and
Svensson, A., 2010. Millennial-scale variability during the last
glacial: the ice core record. Quaternary Science Reviews, 29,
2828–2838.
IMPACT GLASS (FISSION TRACKS)
Giulio Bigazzi1 and Maria Laura Balestrieri2
1
CNR- Institute of Geosciences and Earth Resources,
Pisa, Italy
2
CNR- Institute of Geosciences and Earth Resources,
UOS Florence, Italy
Definition
Glass formed by melting of a siliceous rock exposed on
the Earth surface due to the impact of a meteorite with
the ground. Fission-track dating is a radiometric dating
technique used for dating those glasses, and it is based
on the damage (tracks) produced by the decay by fission
of 238U.
Introduction
Innumerable extraterrestrial bodies reach our planet every
day. Most of them, due to their high speed, fully vaporize
when they cross the atmosphere or their size is strongly
reduced. So, these meteorites do not produce relevant
effects. Only the largest bodies have probability to reach
the Earth surface with a kinetic energy sufficient to pro-
duce recognizable impact structures. Luckily, large-sized
meteorites are rare in comparison with the high number
of those that reach the terrestrial atmosphere. For this rea-
son impact craters, which are evidence of a collision of
a heavy body occurred sometime in the past, are relatively
rare. The fall of a large meteorite may have catastrophic
effects, such as long-lasting climate changes. It is known
that the abrupt ecosystem changes with extinction of many
species, which represent the Cretaceous/Tertiary bound-
ary, might be correlated to such an event. Therefore, it is
an important task to study the chronology of significant
impacts in order to establish possible correlation with
environmental changes and to investigate on their period-
icity, if any.
During an impact, due to transferring of a high amount
of kinetic energy to the ground, target rocks suffer high
shock-wave pressures, and their temperatures grow
instantaneously up to 1,000  C. This phenomenon,
named shock-wave metamorphism, involves various
effects – such as vaporization, melting, and fracturing –
depending on the kinetic energy of the extraterrestrial
body and on the original setting of the rock in relation
to the impact. When the duration and amount of temper-
ature are sufficient to provoke the total annealing of
preexisting tracks in a mineral phase containing U as
trace element, the fission tracks that one can etch and
count are those formed after the impact, and the resulting
age is the age of the impact itself. Thus, the fission-track
technique may be an efficient method for the age determi-
nation of impacts, providing that a suitable mineral phase
is present in the target rocks. For the lower thermal stabil-
ity of fission tracks and for its abundance, apatite is com-
monly preferred, but also sphene and zircon have been
used. A crucial point is to establish whether the total
annealing of preexisting tracks really occurred. There-
fore, track length measurements are needed in order to
compare their distribution with those of rocks that suf-
fered fast cooling, such as volcanic rocks.
When siliceous rocks are present in the target area of
a meteorite, the impact produces melting of these rocks.
The newly formed glass starts track recording as soon as
it cools, and its age is the age of the impact. Glasses are
frequently found in impact structures, and most of these
structures were dated using glass, that is, an ideal material
(homogeneous distribution of uranium, easy etchability)
for fission-track dating.
Impact structures
Ages determined on impact structures by various authors
since the early studies of the fission-track method are
shown in Table 1. Known terrestrial impact locations are
shown in Figure 1. Ages are distributed over a long time
span, from few 1,000 years up to about 300 Ma. Com-
monly, ages measured by different authors are in agree-
ment, except in a few cases, especially those studies in
the 1960s, probably due to uncertainties on the calibration
of the fission-track system. Only in the early 1980s the
problem of the calibration of the fission-track system
was tackled, and international age standards were identi-
fied, in order to allow comparison of measurements from
different laboratories. The stability of fission tracks in
glasses is rather poor: a certain amount of partial annealing
 IMPACT GLASS (FISSION TRACKS) 349

Impact Glass (Fission Tracks), Table 1 Summary of impact structures dated using the fission-track method (Modified after Wagner
and Van den Haute (1992))

Structure Material Age (Ma) Correction techn. Reference

1. Henbury Glass 0.0042  0.0019 Storzer and Wagner (1977)

2. Wabar Glass 0.0064  0.0026 Storzer (1971)
0.0064  0.0025 Storzer and Wagner (1977)
3. Köfels Glass 0.008  0.006 Storzer et al. (1971)
0.0089  0.0029 Storzer and Wagner (1977)
4. Darwin Glass 0.37  0.76 Fleischer et al. (1969a)
0.72  0.04 Gentner et al. (1969)
0.71  0.01 Gentner et al. (1973)
0.74  0.04 Size Storzer and Wagner (1977)
5. Bosumtwi Glass 1.04  0.20 Gentner et al. (1967)
0.32  0.19¸ 1.48  0.47 Fleischer et al. (1965)
Fleischer and Price (1967)
1.03  0.02 Storzer (1971)
1.04  0.11 Size Storzer and Wagner (1977)
6. Zhamanshin Glass 1.07  0.05 Plateau Storzer and Wagner (1977)
1.07  0.06 Plateau Storzer and Wagner (1979)
0.81  0.16 Florenski et al. (1979)
7. Aouelloul Glass 0.16  0.05, 0.50  0.16 Fleischer et al. (1965)
Fleischer and Price (1967)
0.46  0.61 Wagner (1966)
3.25  0.50 Size Storzer and Wagner (1977)
8. Elgygytgyn Glass 4.52  0.11 Plateau Storzer and Wagner (1979)
9. Ries Glass 15.2  1.6 Fleischer et al. (1965)
14.8  0.7 Wagner (1966)
14.6  0.6 Gentner et al. (1967)
14.6  0.6 Storzer and Gentner (1970)
14.7  0.3 Plateau Storzer and Wagner (1977)
Sphene 14.8  0.4 Wagner (1977)
Apatite 14.7  0.4 Wagner (1977)
10. Haughton Apatite 22.4  1.4 Omar et al. (1987)
11. LDG Glass 37.5  2.4 Fleischer et al. (1965)
32.1  3.0 Wagner (1966)
26.6  1.7 Gentner et al. (1969)
28.5  2.3 Storzer (1971)
29.4  0.5 Storzer and Wagner (1977)
28.5  0.8 Bigazzi and De Michele (1996)
12. Popigai Glass 38.9  2.2 Size Storzer and Wagner (1977)
30.5  1.2 Plateau Storzer and Wagner (1977)
13. Manson Apatite 61  9 Hartung et al. (1986)
14. Mistastin Glass 39.6  4.4 Size Storzer and Wagner (1977)
15. Mien Glass 92  6 Size Storzer and Wagner (1977)
16. Boltysh 100  10 Size Komarov and Raikhlin (1976)
17. Goses Bluff Apatite + Zircon 134  6 Milton and Sutter (1987)
18. Manicouagan Glass 208  25 Fleischer et al. (1969b)
200  30 Size Storzer (1971)
Size Storzer and Wagner (1977)
19. Rochechouart Glass 210  100 Storzer (1971)
206  39 Size Storzer and Wagner (1977)
Apatite 198  25 Storzer and Wagner (1977)
20. Clearwater Glass 290  30 Fleischer et al. (1969b)
287  43 Size Storzer (1971)

occurs also at ambient temperature. For this reason an age
on glass is considered a minimum age, even if a technique
for correcting thermally lowered fission-track ages –
namely, the size correction method or the plateau method –
is applied (Storzer and Wagner, 1969; Storzer and
Poupeau, 1973).
Commonly the Ar40/Ar39 dating method is preferred to
fission tracks, substantially for its higher precision, con-
sidering that the main component of the experimental
error of a fission-track age is related to the number of
spontaneous tracks one can count under a microscope.
However, in several cases the 40Ar/39Ar method failed
350 IMPACT GLASS (FISSION TRACKS)

Impact Glass (Fission Tracks), Figure 1 Location of the terrestrial impact structures dated with the fission-track method (numbers
are referred to Table 1). The tektite terrestrial (red) and oceanic (pink) strewn fields with their mean ages are also shown. The dashed
strewn field represents the extension of the Australasian s.f. southward to Antarctica, as suggested by recent discovery of
microtektite trapped in sediments of the Transantarctic Mountains in northern Victoria Land (Redrawn with modifications after
Wagner and Van den Haute (1992)).

and only fission tracks yielded a reliable age. A significant
example is the Libyan Desert Glass (LDG) shown in
Table 1 and Figure 1. Several authors tried to apply the
40
Ar/39Ar method on it but only using fission tracks
a reliable age for this glass was obtained (Bigazzi and
De Michele, 1996 and references therein). This was possi-
ble because, due to the presence of inclusions, the identifi-
cation and counting of spontaneous tracks can be
performed without uncertainties and for the relatively
high areal density – more than 10,000 tracks for cm2
(Figure 2a). LDG is a magnificent very pure quartz glass
(about 98 %) (Figure 2b), known since the antiquity, used
for tool making during prehistoric times as well as for jew-
elry in the ancient Egypt. LDG is found in blocks and frag-
ments of various sizes in the Great Sand Sea of the Libyan
Desert in Egyptian territory. It is distributed mostly in
a relatively large area of the desert, northward in direction
of the Mediterranean Sea. Systematic exploration of the
Libyan Desert, since the expeditions of Patrick Clayton,
in the 1920s and 1930s of the last century, has shown that
spreading of LDG is substantially due to anthropic trans-
portation, the original location being in a rather restricted
area – few tens of kilometers – relatively close to the Lib-
yan border. LDGs lie on the Cretaceous Nubian basement,
along the “corridors,” north–south directed – located in
between the sand dunes. Origin of LDG was a puzzling
issue and various hypotheses were made – sedimentary
glass and volcanic glass possibly extraterrestrial – but
the most reliable is that LDG is due to melting of the desert
sand caused by an impact. However, although various cra-
ters have been considered, no correlation of LDG with
a specific structure has been found.
Tektites
Tektites are other puzzling natural glasses in our planet
that caused long debate since they were described for the
first time in 1774 by Count František Josef Kinský as
“chrysolites and topazes” and later on, with a scientific
approach, by Prof. Josef Mayer at the 1786 meeting of
the Czech Academy of Sciences. Tektites have been dis-
covered in four main areas on the Earth’s surface – named
strewn fields (Figure 1) – where mysterious glasses are
spread, showing an “alien” chemical composition, unlike
the sediments where they are found. They have various
shapes depending on the history they experienced after
they formed – such as reworking of the original sedi-
ments, long-distance transportation, and redeposition in
younger deposits – meteoric sculpturation. In Figure 3,
samples of the Central European tektites – named
Moldavites – are shown, their lengths being between
about 3 and 6 cm. Except for the European tektites whose
whole strewn field is on ground, the remaining three
strewn fields extend in the adjacent Atlantic, Indian, and
Pacific oceans where microtektites were discovered in
the marine sediments.
 IMPACT GLASS (FISSION TRACKS)
Impact Glass (Fission Tracks), Figure 2 (a) Fission tracks etched
for 240 s in 20 % HF at 40  C on a polished surface of Libyan
Desert Glass; (b) a 2,140 g sample of Libyan Desert Glass
“sculptured” and incrusted by the Sahara sand.
Sizes of tektites are relatively small; those of Central
Europe do not exceed 300 g of weight and those exceeding
100 g are very rare. Only in Southeastern Asia signifi-
cantly larger samples were found among the “Muong
Nong”-type tektites. Nowadays, the most accepted
hypothesis is that tektites have a terrestrial origin and are
produced by impacts of heavy extraterrestrial bodies.
The high-energy shock wave preceding the meteorite
vaporizes the siliceous target rocks, which are ejected in
the high atmosphere, and they solidify and fall at distances
of 100 and sometimes 1,000 km. In this scenario, compar-
ison of ages of tektites with the age of impact craters, their
“potential source,” may provide evidence for the above
described mechanism.
The four strewn fields have well-distinct ages. The
oldest tektites are those of the Central American strewn
field (age 34.6  0.6 Ma, Storzer and Wagner, 1977) which
extends in Texas (Bediasites) and Georgia (Georgiaites) –
in the continent – and in the Caribbean Sea. Although var-
ious impact structures of the American continent have been
studied, no correlation with these tektites was found. The
351
Impact Glass (Fission Tracks), Figure 3 Samples of Moldavite
from different deposits of Bohemia and Moravia (After Bouška,
1994).
next strewn field is the Central European one. Moldavites
are found in deposits of various ages, since Middle Mio-
cene, Pliocene, and Pleistocene in South Bohemia and
Moravia up to Holocene in Elba riverbanks. The fission-
track age of Moldavite, 14.7  0.4 Ma (Wagner, 1966;
Storzer and Wagner, 1977), is the same determined by the
same authors for the Ries crater (Table 1) located in Bayer,
Germany. This represents an important confirmation of the
terrestrial origin of tektites and of the formation model
described above. Moldavite is an excellent glass, with rela-
tively high-track densities, without defects that might pro-
duce “spurious” tracks, so fission-track identification
during counting is unequivocal. In addition, many age
determinations obtained by several authors using also K-
Ar and 40Ar/39Ar are available. For these reasons it has been
distributed as reference material for fission-track dating in
glass (Balestrieri et al., 1998). The Ivory Coast tektites are
significantly younger (1.08  0.10 Ma, Storzer and Wag-
ner, 1977), confirming the rarity of the event responsible
for the formation of these glasses. Also in case of this
strewn field, an associated impact structure was identified,
i.e., the Bosumtwi crater located in Ghana (1.04 
0.11 Ma, Table 1). The youngest strewn field, the
352 IMPACT GLASS (FISSION TRACKS)
Impact Glass (Fission Tracks), Figure 4 Up: microtektite from the Transantarctic Mountain trapped in fine-grained detritus. Bottom:
two 238U fission tracks etched for 120 s in 20 % HF at 40  C in the polished section of a microtektite grain at a magnification of 100
(left) and of 500.
Australasian, is the largest one. It extends in Southeastern
Asia and Australia and in the Indian and Pacific Ocean.
These tektites have been named Indochinites, Javanites,
Philippinites, Billitonites, and Australites, according to the
area where they were found. Their age (0.71  0.07 Ma,
Storzer and Wagner, 1977; 0.693  0.025 Ma, Storzer
and Wagner, 1980; 0.77  0.08 Ma, Bigazzi and De
Michele, 1996) is in agreement with the age of the Darwin
crater in Tasmania (0.74  0.04 Ma, Table 1), but this crater
is relatively small (ca. 1 km), so it is highly unlikely that it
could have been the source of such extended strewn field.
All the known Asian craters, potential sources of the Aus-
tralasian strewn field, have been considered, but none show
an age consistent with the tektites. In addition, it has been
also hypothesized that this large strewn field may have been
produced by more than one impact: Storzer and Wagner
(1980) determined for some Australian tektites ages older
(0.830  0.028 Ma) than those of Southeastern Asia cited
above. Therefore, this topic needs further investigation.
Recent research has shown that the Australasian strewn
field might be even more extended than expected.
Microtektites were recently discovered on the Victoria
Land Transantarctic Mountains (Figure 4). They were
found trapped in the fine-grained local detritus accumu-
lated in weathering pits of the glacially eroded summits
of Miocene age known as Transantarctic Mountain micro-
meteorite traps (Folco et al., 2008). Petrographic studies,
major and trace element compositions, Sr and Nd isotopic
compositions, and a low-resolution, yet Quaternary,
40
Ar/39Ar age of 0.76  0.98 Ma suggested that these
microtektites most likely represent a major southern
extension of the Australasian tektite strewn field. In order
to better constrain the formation age of Transantarctic
Mountain microtektites, their fission-track age was deter-
mined. The obtained fission-track plateau age of 0.85 
0.17 Ma, although affected by a relatively large experi-
mental error due to the low number of counted spontane-
ous tracks owing to the small surface suitable for the
analysis, is well consistent with the ages determined for
the Australasian tektites.
Summary and conclusions
Impacts of large meteoritic bodies with the Earth’s surface
ground in presence of siliceous rocks may produce impact
glasses through melting. By dating these glasses, one
obtains the age of the related impact events. Since the
early days of the fission-track method, various attempts
have been successfully made in order to date impact
glasses that usually are ideal material where to apply this
 INDEX FOSSIL
technique as most of them show a relatively high number
of tracks that can be unequivocally identified. Although
the fission-track method usually relies lower precision
ages in comparison with other radiometric methods, its
application supplied precious information for correlating –
or for excluding correlation – of impactites with specific
craters and for reconstructing tektite strewn fields. Impres-
sive examples of the performance of fission tracks in this
field are those quoted above of the Libyan Desert Glass
and of the Antarctic microtektites. This method produced
unique efficacious data where other techniques failed.
Bibliography
Balestrieri, M. L., Bigazzi, G., Bouska, V. J., Labrin, E., Hadler,
J. C., Kitada, N., Osorio, A., Popeau, G., Wadatsumi, K., and
Zúñiga, A., 1998. Potential glass age standards for fission-track
dating: an overview. In Van den Haute, P., and De Corte, F.
(eds.), Advances in Fission-Track Geochronology. Dordrecht:
Kluwer Academic, pp. 287–304.
Bigazzi, G., and De Michele, V., 1996. New fission-track age deter-
minations on impact glasses. Meteoritics & Planetary Science,
31, 234–236.
Bouška, V., 1994. Moldavites. The Czech Tektites. Prague:
Stylizace.
Fleischer, R. L., and Price, P. B., 1967. Ages of impact glasses from
the Ashanti and Aouelloul craters: a correction. Geochimica et
Cosmochimica Acta, 31, 2451–2452.
Fleischer, R. L., Price, P. B., and Walker, R. M., 1965. On the simul-
taneous origin of tektites and other natural glasses. Geochimica
et Cosmochimica Acta, 29, 161–166.
Fleischer, R. L., Price, P. B., Viertl, J. R. M., and Woods, R. T.,
1969a. Ages of Darwin glass, Macedon glass, and Far Eastern
tektites. Geochimica et Cosmochimica Acta, 33, 1071–1074.
Fleischer, R. L., Viertl, J. R. M., and Price, P. B., 1969b. Age of the
manicouagan and clearwater lakes craters. Geochimica et
Cosmochimica Acta, 33, 523–527.
Florenski, P. V., Perelygin, V. P., Bazhenov, M. L., Lhagvasuren,
D. D., and Stetzenko, S. G., 1979. Complex determination of
the age of the meteoritic crater Zhamanshin (Russia). Astronom-
icheskii Vestnik, 13, 178–186.
Folco, L., Rochette, P., Perchiazzi, N., D’Orazio, M., Laurenzi, M.,
and Tiepolo, M., 2008. Microtektites from Victoria Land
Transantarctic Mountains. Geology, 36, 291–294.
Folco, L., Bigazzi, G., D’Orazio, M., and Balestrieri, M. L., 2011.
Fission track age of Transantarctic Mountain microtektites.
Geochimica et Cosmochimica Acta, 75, 2356–2360.
Gentner, W., Kleinmann, B., and Wagner, G. A., 1967. New K-Ar
and fission track ages of impact glasses and tektites. Earth and
Planetary Science Letters, 2, 83–86.
Gentner, W., Storzer, D., and Wagner, G. A., 1969. New fission
track ages of tektites and related glasses. Geochimica et
Cosmochimica Acta, 33, 1075–1081.
Gentner, W., Kirsten, T., Storzer, D., and Wagner, G. A., 1973. K-Ar
and fission track dating of Darwin Krater glass. Earth and Plan-
etary Science Letters, 20, 204–210.
Hartung, J. B., Izett, G. A., Naeser, C. W., Kunk, M. J., and Sutter,
J. F., 1986. The Manson, Iowa, impact structure and the Creta-
ceous-Tertiary boundary event. Earth and Planetary Science
Letters, 17, 313–314.
Komarov, A. N., and Raikhlin, A. J., 1976. Comparison of fission-
track and potassium-argon dating on impactites. Academy Nauk
SSSR (Geology), 228, 35–38.
353
Milton, D. J., and Sutter, D. F., 1987. Revised age for the Gosses
Bluff impact structure, Northern Territory, Australia, based on
40
Ar-39Ar dating. Meteoritics, 22, 281–289.
Omar, G., Johnson, K. R., Hickey, L. J., Robertson, P. B., Dawson,
M. R., and Barnosky, C. W., 1987. Fission-track dating of
Haughton Astrobleme and included biota, Devon Island, Can-
ada. Science, 237, 1603–1605.
Storzer, D., 1971. Fission track dating of some impact craters in the
range between 6000 y and 300 my. Meteoritics, 6, 319.
Storzer, D., and Gentner, W., 1970. Spaltspurenaltern von
Riesgläsern, Moldaviten und Bentoniten. Jahresbericht und
Mitteilungen des Oberrheinischen Geologischen Vereins, 52,
97–111.
Storzer, D., and Poupeau, G., 1973. Géochronologie – age-plateaux
de minéraux et verres par la méthode des traces de fission.
Comptes Rendus de l'Académie des Sciences, 275, 137–139.
Storzer, D., and Wagner, G. A., 1969. Correction of thermally
lowered fission track age of tektites. Fission track dating meteor-
ite impacts. Earth and Planetary Science Letters, 5, 463–468.
Storzer, D., and Wagner, G. A., 1977. Fission track dating meteorite
impacts. Meteoritics, 12, 368–369.
Storzer, D., and Wagner, G. A., 1979. Fission track dating of
Elgygytgyn, Popigai and Zhamanshin impact craters: no sources
for Australasian or North-American tektites. Meteoritics, 14,
541–542.
Storzer, D., and Wagner, G. A., 1980. Australites older than
Indochinites. Naturwissenschaften, 67, 90.
Storzer, D., Horn, P., and Kleinmann, B., 1971. The age and the ori-
gin of Kfels structure, Austria. Earth and Planetary Science Let-
ters, 12, 238–244.
Wagner, G. A., 1966. Altersbestimmungen an Tektiten und anderen
natürlichen Gläsern mittels Spuren der spontanen Kernspaltung
des Uran238 (“fission track” – methode). Zeitschrift für
Naturforschung, 31, 733–745.
Wagner, G. A., 1977. Spalspurendatierung an Apatit und Titanit aus
dem Ries: Ein Beitrag zum Alter und zur Wärmegeschichte.
Geologica Bavarica, 75, 349–354.
Wagner, G., and Van den Haute, P., 1992. Fission-Track Dating.
Dordrecht: Kluwer Academic.
INDEX FOSSIL
Peter Harries
School of Geosciences, University of South Florida,
Tampa, FL, USA
Synonyms
Guide fossils; Range fossils; Zone fossils
Definition
A primary goal of the Earth sciences is to obtain as highly
resolved correlations between different regions as possi-
ble. One of the most useful tools in this pursuit is the
numerous fossils contained in most sedimentary units.
The concept of using index fossils was initiated by the
work of William Smith in the early 1800s; he used the
fossil content of the units he examined as a critical
354 ISUA SUPRACRUSTAL BELT, WEST GREENLAND: GEOCHRONOLOGY
component in the development of the first geologic map of
the United Kingdom (Smith, 1815). His observations on
the distribution of fossils through time was formalized in
his principle (or law) of faunal succession and served as
the basis for determining the relative age of the rocks
containing specific fossils. Building on Smith’s concept,
later workers used the temporal intervals defined by index
fossils to represent range zones that form the basis of
biostratigraphy, the science of correlation using fossils.
Thus, index fossils form among the best components used
in establishing a relative chronology and using that chro-
nology to correlate globally.
Because all organisms evolve, this affords the opportu-
nity to subdivide geologic time into a series of intervals
characterized by individual, morphologically distinct spe-
cies (the so-called morphospecies) into three intervals: an
interval prior to their origination, one equivalent to their
total temporal range, and another interval subsequent to
their extinction. Therefore, any fossil species has the poten-
tial to be an index fossil. The best index fossils, however,
tend to have the following characteristics: (1) rapid evolu-
tion so that the temporal intervals represented are short;
(2) wide geographic distribution allowing correlation over
a broad area; (3) virtually instantaneous range expansion
following origination and simultaneous extinction decreas-
ing the likelihood that these boundaries are time transgres-
sive; (4) distinctive morphologic characters allowing
differentiation from related taxa; (5) mineralized skeletal
elements that are preserved in the fossil record; and (6) suf-
ficient abundance affording the likelihood of finding their
remains. Few taxa display all these characteristics, but that
does not preclude their biostratigraphic utility.
Throughout the Phanerozoic, a broad range of different
organisms has been used as index fossils. Given the criteria
outlined above, the marine organisms that have proven
most effective for defining short intervals are generally
those that had planktic or nektic life habits. These float-
ing/free-swimming organisms include both microinver-
tebrates (e.g., planktic foraminifers, radiolarians, and
coccolithophores) and macroinvertebrates (e.g., ammo-
noids, graptolites). However, given the importance of deter-
mining the relative age of sedimentary units, representatives
of virtually all phyla with a fossil record have been used
including bivalves, brachiopods, corals, echinoids, gastro-
pods, and trilobites (see Kauffman and Hazel (1977) for
a more detailed treatment). Within the terrestrial realm, bio-
stratigraphy has largely been based on pollen and spores as
well as mammals (the so-called land mammal ages).
Bibliography
Kauffman, E. G., and Hazel, J. F. (eds.), 1977. Concepts and
Methods of Biostratigraphy. Stroudsburg: Dowden, Hutchinson
& Ross, p. 658.
Smith, W., 1815. A Delineation of the Strata of England and Wales
with a Part of Scotland: Exhibiting the Collieries and Mines, the
Marshes and Fen Lands Originally Overflowed by the Sea and
the Varieties of Soil According to the Variations in the Substrata.
London: John Cary.
ISUA SUPRACRUSTAL BELT, WEST GREENLAND:
GEOCHRONOLOGY
Vickie C. Bennett1 and Allen P. Nutman2
1
Research School of Earth Sciences, The Australian
National University, Canberra, ACT, Australia
2
School of Earth & Environmental Sciences, University of
Wollongong, Wollongong, NSW, Australia
Synonyms
Isua greenstone belt
Definition
The 35-km-long by up to 4-km-wide Isua supracrustal belt
(ISB) lying within the Isukasia terrane, southern West
Greenland (Figure 1), consists of ca. 3,700 and 3,800 mil-
lion years old (Ma) volcanic and sedimentary rocks that
are among the oldest on the Earth. The ISB is the most sig-
nificant and highly studied region for understanding early
Earth chemistry and environments. This is because of its
great antiquity and excellent exposures of a diverse range
of rock types, including pillow basalts and banded iron
formations (BIF).
Introduction
The ISB belt has been recognized for more than 40 years
to be a significant resource for revealing early Earth his-
tory, as it contains the most extensive and least deformed
sequences of Eoarchean (>3,600 Ma) ancient rocks
deposited on Earth’s surface. Although all rocks
>3,600 Ma have undergone some degree of heating
(metamorphism) and deformation due to later tectonic
events, portions of the ISB rocks contain the world’s least
heated and least deformed oldest rocks and therefore more
faithfully record their early history and processes of
formation. The ISB is dominated by meta-volcanic
amphibolites of island arc tholeiite and picrite geochemical
affinity, some preserving pillow structures with lesser
amounts of clastic and chemical sediments including
quartz-magnetite banded iron formation, metachert,
dolomites, and some felsic schists and pelites of volcano-
sedimentary origin. The ISB lies within a larger complex
of Eoarchean rocks composed mainly of quartzo-
feldspathic orthogneisses derived from granitic rocks
(Figure 1).
The term Isua or Isua supracrustal belt is sometimes
generically (and incorrectly) applied to all Eoarchean
(>3,600 Ma) rocks from the >3,000 km2 ancient terranes
of West Greenland; however, these terms should only be
reserved for this specific region and the term Itsaq gneiss
complex (Itsaq is the Greenlandic work for ancient) as
defined in Nutman et al. (1996), used when referring to
the entire assemblage of >3,600 Ma rocks in the region.
 ISUA SUPRACRUSTAL BELT, WEST GREENLAND: GEOCHRONOLOGY

Isua Supracrustal Belt, West Greenland: Geochronology, Figure 1 Summary geological map of the Isua area, Nuuk region, West Greenland. The arcuate Isua supracrustal
belt is composed of two temporally distinct terranes, both containing a range of volcanic and metasedimentary rocks. The northern terrane has rocks with ages of ca. 3,700
million years old and the southern terrane with rock ages of ca. 3,800 million years old (constrained by >30 zircon U/Pb age determinations – see Nutman and Friend 2009).
The legend lists the rock types present in each terrane.
355
356 ISUA SUPRACRUSTAL BELT, WEST GREENLAND: GEOCHRONOLOGY
History of geochronology
The first mapping of the Isua region in 1967 was done by
geologists employed by prospecting companies, with the
first visit by academic geologists (S. Moorbath and
V. McGregor) in 1971. Following this visit Moorbath
and coworkers at Oxford University produced the first
radiometric age determination (Moorbath et al., 1972;
~3,695 Ma banded iron formation indicated on Figure 1)
for the ISB using the whole rock rubidium-strontium iso-
chron dating method. This early work established an age
for the ISB of 3,700 Ma, making these the oldest volca-
nic and sedimentary rocks then known. However, owing
to the limitations of the method, this age had large uncer-
tainties (>100 myr). Other attempts to date the age of
the ISB using Pb-Pb, rubidium-strontium, and samarium-
neodymium isochrons also confirmed the great antiquity
of these rocks, but again with large uncertainties
(a detailed review of the history of the investigation of
the ISB is provided in Nutman and Friend, 2009). U-Pb
zircon geochronology (Baadsgaard, 1976; Michard-Vitrac
et al., 1977; first zircon dating site indicated on Figure 1)
also confirmed the great antiquity of the rocks, but with
much improved precision. Because variations in the ages
of adjacent rock units could not be resolved with this level
of precision, the whole of the ISB was regarded until the
mid-1990s as a single assemblage formed at more or less
the same time.
A breakthrough in geochronology, particularly for
determining the ages of Archean rocks arrived in
mid-1980s, with the first development of in situ methods
for U-Pb dating of zircon crystals using the Sensitive High
Resolution Ion MicroProbe (SHRIMP), designed and built
at the Australian National University. With this instrument
25- mm-wide sites in single zircon crystals, extracted from
rocks, could be individually dated with a precision on the
207
Pb/206Pb age commonly better than 10 Ma. The first
SHRIMP U-Pb zircon dating at Isua (Compston
et al., 1986) used the same sample as Baadsgaard and
achieved a far superior pooled precision of 2 Myr (2s)
for the entire dated zircon population. As detailed U-Pb
SHRIMP geochronology became increasingly available
for ISB rocks (e.g., Nutman et al., 1997), it could be dem-
onstrated that the ISB contained two unrelated packages of
volcanic rocks differing in age by about 100 Ma. Com-
bined with detailed mapping (available now as 1:20,000
scale maps in Nutman and Friend (2009)), this showed
that the ISB contains two unrelated sequences of rocks.
The northern supracrustal terrane has ages of ca. 3,700 Ma
and the southern terrane has ages of ca. 3,800 Ma
(summarized in Figure 1). The juxtaposition of these two
separate regions occurred by 3,660 Ma, as determined by
the age of the oldest unit common to both terranes.
Intensive U-Pb dating efforts by various groups using
both solution single zircon crystal (e.g., Crowley, 2003)
and in situ large ion probe methods, with now more than
30 ages available for the ISB as compiled in Nutman and
Friend (2009), have confirmed the earlier findings and
extended to >3,900 Ma the ages of individual detrital zir-
con crystals within sediments. However, despite intensive
searches, so far there is no U-Pb zircon evidence for
>3,900 Ma rocks in the ISB or surrounding areas.
Current investigations
After more than 40 years of investigation, the ISB is still
the prime natural resource for the study of Earth’s early
history and is visited regularly by teams of researchers
from institutions worldwide. Examples of past work have
included using the presence of basalts with pillow struc-
tures and the identification of water-deposited chemical
sediments to demonstrate the presence of liquid water on
the surface of the Earth at 3,800 Ma. The geochemical
characteristics of the least altered volcanic rocks in both
the northern and southern ISB terranes have chemical
affinities with those of younger subduction-related rocks.
This has been used to argue for the formation of ancient
crust in convergent margin, plate tectonic processes on
the early Earth (see review of chemistry of Isua volcanic
rocks in Polat et al. (2011)). Ongoing research by various
groups is focused on using a range of chemical and isoto-
pic signatures preserved in these oldest sediments and vol-
canic rocks to reveal the chemistry of the early ocean and
atmospheres, to trace the chemical development of the
continental crust and mantle reservoirs, and to search for
evidence of early life.
Bibliography
Baadsgaard, H., 1976. Further U-Pb dates on zircons from the early
Precambrian rocks of the Godthaabsfjord area, West Greenland.
Earth and Planetary Sciences Letters, 33, 261–267.
Compston, W., Kinny, P. D., Williams, I. S., and Foster, J. J., 1986.
The age and lead loss behaviour of zircons from the Isua
supracrustal belt as determined by ion microprobe. Earth and
Planetary Science Letters, 80, 71–81.
Crowley, J. L., 2003. U-Pb geochronology of 3810–3630 Ma gran-
itoid rocks south of the Isua greenstone belt, southern West
Greenland. Precambrian Research, 126, 235–257.
Michard-Vitrac, A., Lancelot, J., Allegre, C. J., and Moorbath, S.,
1977. U-Pb ages on zircons from early Precambrian rocks of
West Greenland and the Minnesota river valley. Earth and Plan-
etary Science Letters, 35, 449–453.
Moorbath, S., O’Nions, R. K., Pankhurst, R. J., Gale, N. H., and
McGregor, V. R., 1972. Further rubidium-strontium age determi-
nations on the very early Precambrian rocks of the Godthåb dis-
trict: West Greenland. Nature, 240, 78–82.
Nutman, A. P., and Friend, C. R. L., 2009. New 1:20,000 scale geo-
logical maps, synthesis and history of investigations of the Isua
supracrustal belt and adjacent orthogneisses, southern West
Greenland: a glimpse of Eoarchean crust formation and orogeny.
Precambrian Research, 172, 189–211.
Nutman, A. P., McGregor, V. R., Friend, C. R. L., Bennett, V. C., and
Kinny, P. D., 1996. The Itsaq Gneiss Complex of southern West
Greenland; the world’s most extensive record of early crustal
evolution (3900–3600 Ma). Precambrian Research, 78, 1–39.
 ISUA SUPRACRUSTAL BELT, WEST GREENLAND: GEOCHRONOLOGY 357

Nutman, A. P., Bennett, V. C., Friend, C. R. L., and Rosing, M. T., Cross-references
1997. 3710 and 3790 Ma volcanic sequences in the Isua
(Greenland) supracrustal belt; structural and Nd isotope implica- Acasta Gneiss Complex
tions. Chemical Geology, 141, 271–287. Rb–Sr Dating
Polat, A., Appel, P. W. U., and Fryer, B. J., 2011. An overview of the Sm–Nd Dating
geochemistry of Eoarchean to Mesoarchean ultramafic to mafic Uranium–Lead, Zircon
volcanic rocks, SW Greenland: implications for mantle deple- Zircon
tion and petrogenetic processes at subduction zones in the early
Earth. Gondwana Research, 20, 255–283.
 J
JACK HILLS ZIRCON
Simon A. Wilde
School of Applied Geology, Curtin University, Perth,
Australia
Definition
The Jack Hills are located 800 km north of Perth, Western
Australia, within the Archean Narryer Terrane. They
became internationally known as the site of the world’s
oldest crystals (Compston and Pidgeon, 1986; Wilde
et al., 2001; Mojzsis et al., 2001).
Initial investigation at Mt. Narryer (50 km southwest
of Jack Hills) led to the discovery of a granulite-facies
quartzite containing detrital zircons with an age
4,150 Ma (Froude et al., 1983). Subsequent work on
a greenschist facies conglomerate from Jack Hills (W74
site on Eranondoo Hill) led to the identification of two
even older zircon grains with ages of 4,276  12 Ma
(Compston and Pidgeon, 1986). A reinvestigation of
zircons from that site identified one grain that recorded
a 207Pb/206Pb age of 4,404  8 Ma (Wilde et al., 2001),
the oldest age obtained from terrestrial material.
Jack Hills’ zircons have continued to provide important
information on the early Earth. This has included oxygen,
lithium, and lutetium-hafnium isotopic data and titanium-
in-zircon thermometry, as well as studies at the atomic
level on lead distribution. Initial studies of oxygen
isotopes identified elevated d18 oxygen values, implying
zircon growth in rocks previously subjected to interaction
with surface waters (Wilde et al., 2001; Mojzsis et al.,
2001), leading to the hypothesis of a cool early Earth
(Valley et al., 2002).
More controversial is whether the oxygen data require
the presence of oceans and continents in the Hadean and
whether plate tectonics was operative at this time
J.W. Rink, J.W. Thompson (eds.), Encyclopedia of Scientific Dating Methods, DOI 10.1007/978-94-007-6304-3,
© Springer Science+Business Media Dordrecht 2015
(Hopkins et al., 2008). This is compounded by the lack
of direct evidence on the nature of the protolith of the
ancient detrital zircons. However, lutetium-hafnium data
are consistent with the protolith being an intermediate to
mafic igneous rock, with the possibility that it was formed
from consolidation of a magma ocean at 4.45 Ga (Kemp
et al., 2010).
Bibliography
Compston, W., and Pidgeon, R. T., 1986. Jack Hills, evidence of
more very old detrital zircons in Western Australia. Nature,
321, 766–769.
Froude, D. O., Ireland, T. R., Kinny, P. D., Williams, I. S.,
Compston, W., Williams, I. R., and Myers, J. S., 1983. Ion
microprobe identification of 4100–4200 Myr-old terrestrial
zircons. Nature, 304, 616–618.
Hopkins, M., Harrison, T. M., and Manning, C. E., 2008. Low heat
flow inferred from >4 Gyr zircons suggests Hadean plate bound-
ary interactions. Nature, 456, 493–496.
Kemp, A. I. S., Wilde, S. A., Hawkesworth, C. J., Coath, C. D.,
Nemchin, A., Pidgeon, R. T., Vervoort, J. D., and DuFrane,
S. A., 2010. Hadean crustal evolution revisited: new constraints
from Pb-Hf isotope systematics of the Jack Hills zircons. Earth
and Planetary Science Letters, 296, 45–56.
Mojzsis, S. J., Harrison, T. M., and Pidgeon, R. T., 2001. Oxygen-
isotope evidence from ancient zircons for liquid water at the
Earth’s surface 4300 Myr ago. Nature, 409, 178–181.
Valley, J. W., Peck, W. H., King, E. M., and Wilde, S. A., 2002.
A cool early Earth. Geology, 30, 351–354.
Wilde, S. A., Valley, J. W., Peck, W. H., and Graham, C. M., 2001.
Evidence from detrital zircons for the existence of continental
crust and oceans on the Earth 4.4 Gyr ago. Nature, 409,
175–178.
Cross-references
Acasta Gneiss Complex
Age of the Earth
Lu-Hf Dating: The Lu-Hf Isotope System
Uranium–Lead, Detrital Zircon
Zircon
 K
KIMBERLITES (K-Ar/Ar-Ar)
David Phillips
School of Earth Sciences, The University of Melbourne,
Parkville, VIC, Australia
Synonyms
40
Ar/39Ar geochronology; Conventional K–Ar dating;
K–Ar dating; Kimberlite clan rocks
Definitions
Potassium-argon (K–Ar) dating: Scientific dating method
based on the natural decay of 40K to 40Ar, where the ratio
of the parent isotope, 40K, relative to the daughter product
40
Ar (radiogenic 40Ar or 40Ar*) is used to calculate the age
of a sample.
40
Ar/39Ar (Argon-Argon) geochronology: A variation of
the K–Ar method, where the 40K content is determined
indirectly by neutron activation of 39K to 39Ar; here, the
40
Ar* to 39Ar ratio is proportional to age, with sample ages
calculated relative to co-irradiated standards of known
age.
Excess argon: 40Ar, aside from atmospheric 40Ar, that is
incorporated into rocks and minerals by processes other
than in situ radioactive decay of 40K. Excess argon results
in 40Ar to 36Ar >;295.5 and may be identified from iso-
chron plots.
Inherited argon: Radiogenic argon (40Ar*) included in
rocks and minerals through contamination by older mate-
rial or retention of 40Ar* from a previous event, e.g., feld-
spar xenocrysts in extrusive volcanic rocks that are not
isotopically reset by the magmatism and retain some pro-
portion of the 40Ar* generated between the times of
xenocryst crystallization and magma eruption.
J.W. Rink, J.W. Thompson (eds.), Encyclopedia of Scientific Dating Methods, DOI 10.1007/978-94-007-6304-3,
© Springer Science+Business Media Dordrecht 2015
Extraneous argon: A sac term for excess or inherited
argon.
Kimberlites (Definitions modified from Mitchell (1986)):
Small volume, volatile-rich, potassic ultramafic igneous
rocks that usually occur as diatremes, but also as dykes
and sills and sometimes contain diamonds. Kimberlites
typically have an inequigranular texture due to abundant
crustal and mantle fragments set in a fine-grained matrix.
Two varieties are recognized:
i. Group I kimberlites: CO2-rich, potassic ultramafic
igneous rocks with an inequigranular texture.
Megacryst/macrocryst minerals include olivine,
ilmenite, pyrope garnet, phlogopite, enstatite, and
chromite. Matrix minerals include olivine, phlogopite,
perovskite, spinel, diopside, monticellite, apatite,
calcite, and serpentine.
ii. Orangeites (or Group II kimberlites): H2O-rich,
ultrapotassic, peralkaline igneous rocks, often with
an inequigranular texture. Megacryst/macrocryst min-
erals include phlogopite, olivine, pyrope garnet,
enstatite, and chromite. Matrix minerals typically
include abundant phlogopite and lesser clinopyroxene,
spinel, perovskite, apatite, REE-phosphates, K-Ba-
titanates, rutile, and ilmenite. More evolved orangeites
may contain matrix sanidine, K-richterite, and
Zr-silicates.
Lamproites (Definition modified from Mitchell (1995)):
Small volume, peralkaline, ultrapotassic igneous rocks
that occur as diatremes, cinder cones, dykes, and sills.
Megacryst/macrocryst minerals include olivine, pyrope
garnet, phlogopite, enstatite, and chromite. Minerals typi-
cal of lamproites include leucite, titanian phlogopite phe-
nocrysts, titanian tetraferriphlogopite, Ti-K-richterite,
diopside, sanidine, K-Ba-titanates (e.g., priderite), and
K-Zr-silicates (e.g., wadeite).
362 KIMBERLITES (K-Ar/Ar-Ar)
Introduction
Kimberlites and related rocks are volumetrically insignif-
icant igneous rocks but have received considerable scien-
tific attention because they are a primary host rock for
diamonds and contain a large range of xenoliths (rock
fragments) from the underlying crust and mantle. Kimber-
lites and related rocks have proven difficult to date accu-
rately, due to their propensity to alteration and the almost
ubiquitous presence of crustal and mantle xenolithic mate-
rial. Over the years, several scientific dating methods have
been applied to kimberlite geochronology, including
the K-Ar, 40Ar/39Ar, U-Pb, Rb-Sr, and fission track
methods. Currently, the most widely applied techniques
involve Rb-Sr dating of phlogopite and U-Pb dating
of perovskite methods (see reviews by Allsopp
et al., 1989 and Heaman et al., 2003). The K-Ar and
40
Ar/39Ar techniques (for information on these methods,
see McDougall and Harrison (1999)) are less commonly
employed, due to the common presence of extraneous
argon (excess or inherited argon) in kimberlitic minerals
(e.g., phlogopite), which results in anomalously old ages.
Nonetheless, the 40Ar/39Ar dating method can provide use-
ful kimberlite and related rock age information, provided
that samples are carefully chosen and appropriate experi-
mental procedures are employed to test for extraneous
argon (Phillips et al., 1998, 1999; Phillips et al., 2012a).
Historical development
The earliest attempts to date kimberlitic rocks using the
K–Ar dating method focused on kimberlites located in
Kentucky and Kansas. These studies targeted large phlog-
opite grains (megacrysts), which yielded variable age
results that were difficult to interpret (Zartman et al.,
1967; Brookins, 1969; Brookins and Naeser, 1971). Many
subsequent attempts to date kimberlites using conven-
tional K–Ar and 40Ar/39Ar dating methods also utilized
large phlogopite macrocrysts or phlogopite-bearing man-
tle xenoliths (e.g., Macintyre and Dawson, 1976; Kaneoka
and Aoki, 1978). Although reliable K–Ar ages consistent
with stratigraphic or other isotopic evidence were reported
for some kimberlitic phlogopite samples, most coarse
phlogopite grains produced anomalously old K–Ar or
40
Ar/39Ar ages as well as disturbed 40Ar–39Ar step-
heating spectra (see reviews by Phillips et al., 1999;
Phillips, 2012).
The anomalously old ages are consistent with crystalli-
zation of phlogopite macrocrysts/megacrysts prior to
kimberlite eruption, and are usually attributed to the pres-
ence of inherited argon, although some authors also
invoked the presence of excess argon to explain some
results (see reviews by Phillips, 1991; Hopp et al., 2008;
Phillips, 2012). Allsopp and Roddick (1984) conducted
the first systematic study to compare 40Ar/39Ar results
from large phlogopite macrocrysts and groundmass
phlogopite grains extracted from the Dokolwayo orangeite
in Swaziland. This study obtained older apparent 40Ar/39Ar
ages from large phlogopite macrocrysts, but concordant
(plateau) age results for groundmass phlogopite crystallized
from the orangeite magma. Consequently, more recent stud-
ies have focused on matrix phlogopite grains, which, if
unaltered, can provide reliable age results for kimberlite
magmatism (see Phillips et al., 1998, 1999; O’Brien et al.,
2007; Phillips et al., 2012a).
Key concepts
Sample selection and characterization
Phlogopite mica is the preferred mineral for
K-Ar and 40Ar/39Ar dating of kimberlites and related rocks,
due to its common occurrence and relatively high K content
(~8 wt%). Phlogopite occurs in these rocks as primary and
secondary grains in entrained mantle xenoliths, as
megacrysts and macrocrysts (crystal fragments) and
as groundmass crystallites of the kimberlite and related rock
melts (Mitchell, 1986, 1995). Orangeites and lamproites
also contain other potassium-bearing minerals that may be
suitable for K-Ar or 40Ar/39Ar dating, such as sanidine,
K-richterite, K–Ba-titanates, and K–Zr-silicates (Jaques
et al., 1984; Phillips et al., 2012a); however, few studies
have been conducted on these phases.
Careful sample selection and characterization is key to
obtaining reliable 40Ar/39Ar ages for kimberlites and
related rocks because of the presence of extraneous argon
in phlogopite xenocrysts and phenocrysts. Therefore, pet-
rographic characterization of kimberlite and related sam-
ples is a minimum requirement for identifying samples
hosting suitable matrix phlogopite (or other K-bearing
magmatic minerals). The best material for 40Ar/39Ar geo-
chronology is fresh groundmass phlogopite that is suffi-
ciently coarse grained (>100 mm) to allow routine
mineral separation; these grains are usually restricted to
magmatic kimberlite and related rock occurrences (e.g.,
dykes and sills). Volcaniclastic samples are generally
unsuitable, due to fine grain sizes, heterogeneous mineral
populations, and susceptibility to alteration.
Where present, magmatic (i.e., matrix) micas in Group
I kimberlites belong to the kinoshitalite compositional
series. Orangeites and lamproites typically contain abun-
dant, euhedral tetraferriphlogopite groundmass crystals
(Figure 1). In thin section, matrix phlogopite grains may
be distinguished from xenocrysts/phenocrysts by their
finer grain size and the common presence of inclusions
of other groundmass minerals such as spinel (see Figure 1).
In cases where matrix phlogopite is uncommon (e.g., many
Group I kimberlites), more effort is usually required to
locate rare matrix minerals or phlogopite associated with
reaction rims on garnets or other xenocrysts (Phillips
et al., 2012b). Once phlogopite mineral separates are
obtained, it is desirable to analyze a number of single
 KIMBERLITES (K-Ar/Ar-Ar) 363

phlogopite grains using the 40Ar/39Ar step-heating method

to assess age reliability (see below).

Kimberlites (K-Ar/Ar-Ar), Figure 1 Photomicrograph showing
euhedral, matrix phlogopite grains in a magmatic kimberlite.
Note the presence of other matrix mineral (e.g., spinel)
inclusions in the phlogopite grains, confirming a magmatic
paragenesis. The field of view of the photomicrograph is 2 mm.
40
Ar/39Ar geochronology of kimberlites and
related rocks
As noted, the attainment of reliable 40Ar/39Ar age data
for kimberlites and related rocks is dependent on the
selection of suitable, unaltered, magmatic minerals such
as matrix phlogopite. Despite the fine grain size of
matrix phlogopite grains (<200 mm), the advent of
microanalytical 40Ar/39Ar laser-probe dating techniques
(e.g., Phillips et al., 1998, 1999) permits step-heating
analyses of single groundmass phlogopite grains (or aliquots
of a few grains). This approach avoids problems associated
with bulk mineral separates, such as contamination by
phlogopite xenocrysts and phenocrysts hosting extraneous
argon, and also allows assessment of other potential prob-
lems such as thermal- or alteration-induced argon loss and
recoil effects (see Hall (2013) for a discussion of recoil
effects). An example of an 40Ar/39Ar experiment conducted
on single matrix tetraferriphlogopite grains (~200 mm in size)
from the Voorspoed orangeite (South Africa) is shown in
Figure 2 (see Phillips et al., 1998 for details). In this case,

Kimberlites (K-Ar/Ar-Ar), Figure 2 40Ar/39Ar step-heating age spectra obtained on individual matrix tetraferriphlogopite grains from
the Voorspoed kimberlite, South Africa (Modified from Phillips et al., 1998). Aside from a few discordant low temperature results,
most steps show concordant results, averaging 131.5  1.6 Ma (2s). Note: each color represents a separate step-heating experiment
on a single phlogopite grain; the boxes represent individual step-heating analyses, with the width indicating the relative proportion
of 39Ar released and bar heights indicating plus/minus one sigma uncertainties.
364 KIMBERLITES (K-Ar/Ar-Ar)
three phlogopite grains were individually step-heated
in 3–6 increments. Aside from minor discordance asso-
ciated with the lowest temperature steps, all three grains
show reproducible results, with a weighted mean age
131.5  1.6 Ma (2s). This and other examples (see Phillips
et al., 1999) demonstrate the high precision capability of the
40
Ar/39Ar laser-probe method, provided that suitable sample
material is available.
Bibliography
Allsopp, H. L., and Roddick, J. C., 1984. Rb–Sr and 40Ar–39Ar age
determinations on phlogopite micas from the pre-Lebombo
Group Dokolwayo kimberlite pipe. In Erlank, A. J. (ed.), Petro-
genesis of the Volcanic Rocks of the Karoo Province. Johannes-
burg: The Geological Society of South Africa. Special
publication number 33, pp. 267–271.
Allsopp, H. L., Bristow, J. W., Smith, C. B., Brown, R., Gleadow,
A. J. W., Kramers, J. D., and Garvie, O. G., 1989. A summary
of radiometric dating methods applicable to kimberlite and
related rocks. In Ross, J. (ed.), Proceedings of the Fourth Inter-
national Kimberlite Conference, Vol. 1, Kimberlites and Related
Rocks: Their Composition, Occurrence, Origin and Emplace-
ment. Oxford: Blackwell. Geological Society of Australia spe-
cial publication number 14, pp. 343–357.
Brookins, D. G., 1969. The significance of K-Ar dates on altered
kimberlitic phlogopite from Riley County, Kansas. Journal of
Geology, 77(1), 102–107.
Brookins, D. G., and Naeser, C. W., 1971. Age of emplacement of
Riley County, Kansas, kimberlites and possible minimum age
for the Dakota sandstone. Geological Society of America Bulle-
tin, 82, 1723–1726.
Hall, C. M., 2013. Direct measurement of recoil effects on 40Ar/39Ar
standards. In Jourdan, F., Mark, D. F., and Verati, C. (eds.), Advances
in 40Ar/39Ar Dating: From Archaeology to Planetary Sciences.
London: Geological Society. Special publication number 378.
Heaman, L. M., Kjarsgaard, B. A., and Creaser, R. A., 2003. The
timing of kimberlite magmatism in North America: implications
for global kimberlite genesis and diamond exploration. Lithos,
71, 153–184.
Hopp, J., Trieloff, M., Brey, G. P., Woodland, A. B., Simon,
N. S. C., Wijbrans, J. R., Siebel, W., and Reitter, E., 2008.
40
Ar/39Ar-ages of phlogopite in mantle xenoliths from South
African kimberlites: evidence for metasomatic mantle impregna-
tion during the Kibaran orogenic cycle. Lithos, 106, 351–364.
Jaques, A. L., Smith, C. B., Gregory, G. P., Ferguson, J., Webb,
A. W., Fanning, C. M., Black, L. P., and Pidgeon, R. T., 1984.
The age of diamond-bearing pipes and associated leucite
lamproites of the West Kimberley region, Western Australia.
BMR Journal of Australian Geology and Geophysics, 9, 1–7.
Kaneoka, I., and Aoki, K., 1978. 40Ar/39Ar analyses of phlogopite
nodules and phlogopite-bearing peridotites in South African
kimberlites. Earth and Planetary Science Letters, 40, 119–129.
MacIntyre, R. M., and Dawson, J. B., 1976. Age significance of
some South African kimberlites. In Fourth European Collo-
quium of Geochronology, Cosmochronology and Isotope,
Isotope Geology, Amsterdam, Abstract, 66.
McDougall, I., and Harrison, T. M., 1999. Geochronology and
Thermochronology by the 40Ar/39Ar Method, 2nd edn.
New York: Oxford University Press, 269 pp.
Mitchell, R. H., 1986. Kimberlites: Mineralogy, Geochemistry and
Petrology. New York: Plenum, 442 pp.
Mitchell, R. H., 1995. Kimberlites, Orangeites, and Related Rocks.
New York: Plenum, 410 pp.
O’Brien, H., Phillips, D., and Spencer, R., 2007. Isotopic ages of
Lentiira – Kuhmo – Kostomuksha olivine lamproite – Group
II kimberlites. Bulletin of the Geological Society of Finland,
79, 203–215.
Phillips, D., 1991. Argon isotope and halogen chemistry of phlogo-
pite from South African kimberlites: a combined step-heating,
laser probe, electron microprobe and TEM study. Chemical
Geology, 87, 71–98.
Phillips, D., 2012. Comment on “New Ar-Ar ages of southern
Indian kimberlites and a lamproite and their geochemical evolu-
tion” by Osborne et al. [Precambrian Res. 189 (2011) 91–103].
Precambrian Research, 208–211, 49–52.
Phillips, D., Machin, K. J., Kiviets, G. B., Fourie, L. F., Roberts,
M. A., and Skinner, E. M. W., 1998. A petrographic and
40
Ar/39Ar geochronological study of the Voorspoed
kimberlite, South Africa: implications for the origin of
Group II kimberlite magmatism. South African Journal of
Geology, 101, 299–306.
Phillips, D., Kiviets, G. B., Barton, E. S., Smith, C. B., Viljoen,
K. S., and Fourie, L. F., 1999. 40Ar/39Ar dating of kimberlites
and related rocks: problems and solutions. In Gurney, J. J., Gur-
ney, J. L., Pacsoe, M. D., and Richardson, S. H. (eds.), Proceed-
ings of the Seventh International Kimberlite Conference, Cape
Town, Red Roof Design, Vol. 2, pp. 677–688.
Phillips, D., Clarke, W., and Jaques, A. L., 2012a. New 40Ar/39Ar
ages for the West Kimberley lamproites and implications for
Australian plate geodynamics. In 10th International Kimberlite
Conference, Abstract 2012, February, Bangalore.
Phillips, D., Giuliani, A., Jelsma, H., and Joy, S., 2012b. 40Ar/39Ar
analyses of kelyphite: a new approach for dating kimberlites and
related rocks. In 10th International Kimberlite Conference,
Abstract 2012, February, Bangalore.
Zartman, R. E., Brock, M. R., Heyl, A. V., and Thomas, H. H., 1967.
K-Ar and Rb-Sr ages of some alkalic intrusive rocks from central
and eastern United States. American Journal of Science, 265,
848–870.
Cross-references
Ar–Ar and K–Ar Dating
Fission Track Dating and Thermochronology
Potassium–Argon (Argon–Argon), Structural Fabrics
Rb–Sr Geochronology (Igneous Rocks)
Minerals (40Ar–39Ar)
Rb–Sr Dating
Uranium–Lead Dating
Uranium–Lead, Igneous Rocks
Uranium–Lead, Rubidium-Strontium, Kimberlite
Uranium–Lead, Zircon
Volcanogenic Sedimentary Rocks (K/Ar, 40Ar/39Ar)
Zircon
 L
LACUSTRINE ENVIRONMENTS (14C)
Susan Zimmerman1 and Amy Myrbo2
1
Center for Accelerator Mass Spectrometry, Lawrence
Livermore National Laboratory, Livermore, CA, USA
2
LacCore, Department of Earth Sciences, University of
Minnesota, Minneapolis, MN, USA
Synonyms
Radiocarbon
Definition
Lacustrine. Having to do with any part of a lake system.
Introduction
Lakes are extremely powerful archives of climate and
environmental information, because they record changes
in the landscape around them, and the records they pre-
serve are thus directly relevant to the activities and envi-
ronments of human societies (e.g., Adams et al., 2008).
Lakes commonly have very high sedimentation rates
(5–20 years/cm) and preserve biological, geochemical,
and sedimentary proxies and therefore may contain highly
detailed records of past change. However, lakes are also
dynamic sedimentary systems, and steady, continuous
sedimentation may be disrupted by floods, lake-level
changes, slope failures, and seismic shaking (e.g., Smith
et al., 2013). Along with sedimentary fabric, radiometric
dating is the best way to identify missing or repeated sec-
tions (Grimm, 2011) and to identify and quantify changes
in sedimentation rate.
Radiocarbon dating is the technique most often used to
build age models for proxy records of paleoenvironmental
change over the last
40,000 years. Identifiable terrestrial
organic plant macrofossils are the gold-standard material
for dating, because they are most closely tied to the
J.W. Rink, J.W. Thompson (eds.), Encyclopedia of Scientific Dating Methods, DOI 10.1007/978-94-007-6304-3,
© Springer Science+Business Media Dordrecht 2015
coeval atmospheric 14C content and least susceptible to
postdepositional chemical alteration (Staff et al., 2011).
The frequent paucity of terrestrial macrofossils, however,
means that often other materials must be used. These
materials should be carefully evaluated in comparison to
independent age information.
The high throughput and small sample size of modern
accelerator mass spectrometer (AMS) measurements of
14
C has been a boon to the development of lacustrine
age models, permitting them to potentially match the cen-
tennial to decadal resolution of many proxy records.
Unfortunately, radiocarbon analyses remain expensive
relative to most proxy measurements, and the analytical
precision is often degraded by calibration from 14C years
to calendar years, which is required to account for changes
in atmospheric 14C content through time. Tremendous
sustained community effort has produced a robust, precise
international calibration curve covering the entire range of
14
C dating from 0 to 50 ka (IntCal13, Reimer et al., 2013).
The advent of user-friendly computer programs for appli-
cation of Bayesian statistics to 14C-based age-depth
models (Bronk Ramsey, 1995; Blaauw and Christen,
2011; Bronk Ramsey and Lee, 2013) in combination with
the calibration curves enables the production of robust and
relatively high-precision radiocarbon-based age models.
Sources of carbon to lakes
Carbon in lakes may be brought as dissolved or particulate
carbon in surface water, groundwater, or springs, as
organic molecules or particles recycled within the lake
water, biota, and sediments, and through gas exchange
with the atmosphere at the lake surface (Abbott and
Stafford, 1996; Jull et al., 2013; Alberic et al., 2013). In
areas with ancient carbonate bedrock or old carbonate-rich
surface material (e.g., lake sediments or glacial till), the
dissolved carbon in surface waters is frequently depleted
in or devoid of 14C, and the flux of carbon to the lake water
366 LACUSTRINE ENVIRONMENTS (14C)
from other sources may overwhelm the rate of gas exchange
with the atmosphere, producing the hard-water, or reservoir,
effect (Deevey et al., 1954; Broecker and Walton, 1959).
In addition, even where the bedrock does not contain
carbonate minerals, groundwater that has been isolated
from the atmosphere may have undergone significant
radioactive decay of 14C, the lacustrine equivalent of
14
C-depleted “old” deep water in the ocean. The oxidation
of organic material deposited in a lake may also introduce
old carbon to the lake water, as dissolved organic carbon
(DOC). Finally, deeply circulating hot springs may intro-
duce 14C-dead geothermal fluids into a lake system
(Oremland et al., 1987). Although this source is easy to
rule out in many places, in tectonically active regions it
may be a highly variable effect, as the flow rate from
springs may change suddenly and drastically in response
to tectonic forces. In lakes affected by any of these pro-
cesses, the dissolved inorganic carbon (DIC) pool in the
lake water, from which aquatic plants and animals and
inorganic carbonates get their carbon, may be hundreds
or thousands of years out of equilibrium with the
coexisting atmosphere and variable through time
(Deevey et al., 1954).
Sample selection
The desire for radiocarbon-based age models to give the
calendar year of events recorded in lake sediments sets
several requirements for choosing appropriate materials
for dating:
1. The 14C content of the material should be that of the
atmosphere in the year of its formation.
2. The material should be deposited in the lake sediment
as soon as possible after isolation from the atmosphere
(i.e., death of plant/animal).
3. The C-isotope composition of the material should not
have changed after its deposition.
Thus, the best material is identifiable terrestrial plant
material: needles, leaves, twigs, seeds, etc., and charcoal
from these materials, along with insect parts that are iden-
tifiably terrestrial. In general, the best assurance of the
nearly immediate deposition of the material is to select
delicate macrofossils, such as leaves, which are unlikely
to survive storage or reworking. However, it is quite com-
mon for lake sediments to be lacking in sufficient terres-
trial material, especially in landscapes with sparse
vegetation such as the Arctic (Saulnier-Talbot
et al., 2009) and dry lands, Plazcek et al. (2006) as well
as in cores from the depocenter of large lakes (Shanahan
et al., 2013). The lack of sufficient high-quality datable
material is a problem that increases with the need for large
numbers of dates to build high-resolution age models. As
a result, many investigators are left choosing between less
desirable alternatives, each with its own drawbacks.
Dating of pollen concentrated from sediments has been
proposed by several authors, using a number of different
methods to separate the pollen from other organic or
inorganic sources of carbon (Brown et al., 1992; Mensing
and Southon, 1999). A recent development is the
application of flow cytometry, developed to sort cells for
biological applications, to the sorting of pollen
(or microcharcoal) from a sediment sample (Tennant
et al., 2013). Flow cytometry can separate millions of
grains in a few hours, making the sorting of samples large
enough for 14C analysis straightforward, especially in set-
tings with abundant sources of large pollen (e.g., Pinus).
However, comparisons between pollen and other kinds
of dates show mixed results (Newnham et al., 2007;
Neulieb et al., 2013). With the easier application of pollen
dating potentially enabled by flow cytometry, it may be
that the main sources of difficulty (non-pollen matter,
reworked pollen, etc.) can be identified and overcome in
suitable cases.
Aquatic plant parts (Marty and Myrbo in press) and
shells are secondary alternatives to terrestrial material, as
they are produced from dissolved inorganic carbon
(DIC) in the lake water, which may be older than the coe-
val atmosphere for reasons listed in the previous section.
Plant material is preferred to shells because contaminating
carbonate and organic acids are removed by chemical
pretreatment (see “Radiocarbon Dating” entry), and it is
as resistant to postdepositional alteration as terrestrial
plant material. Carbonates require an additional caution,
as calcium carbonate minerals (e.g., aragonite, calcite)
are quite soluble and thus are easily altered after deposi-
tion. Mineralogical and microscopic examination of shells
is important, especially in shells expected to be more than
20,000 years old, where even small amounts of younger
carbon can have a large effect on the apparent age
(Figure 1).
In addition to the reservoir effect (making ages too old)
and the possibility of alteration, carbonates exposed by
lake-level drop may adsorb young carbon from the atmo-
sphere or shallow groundwater. This “modern carbon”
effect (Figure 2) has the opposite influence as the reservoir
effect, making dates too young; thus the true magnitude of
each effect may be masked by the other (Plazcek
et al., 2006; Zimmerman et al., 2012).
With these considerations in mind, carefully selected
and prepared aquatic plants and shells (Pigati et al.,
2010; Myrbo et al., 2011; Marty and Myrbo, in press)
can yield robust 14C analyses. However, it is still critical
to remember that aquatic materials may hide a variable
lake reservoir effect, when interpreting their 14C ages as
dates. Pairing the dates on aquatic material to a few terres-
trial 14C dates or other known ages at the same level in the
sediment sequence may provide some measure of confi-
dence and a basis for estimating a reservoir correction.
In lakes with little else to date, bulk sediment measure-
ments should be used only with extremely close attention
to the multiple processes affecting the cycling of carbon in
the particular lake and only when measurements at several
depths in the sediment column show a relatively consistent
offset with independent age markers such as terrestrial
macrofossils or well-dated tephras (Blaauw et al., 2011).
 LACUSTRINE ENVIRONMENTS (14C) 367

Lacustrine Environments (14C), Figure 1 Effect of small amounts of modern 14C on increasingly old materials; with only 1 % modern
carbon, all infinite-age materials will appear to be 38 ka. (Inset) By 40 ka <1 % of the original 14C remains.

Lacustrine Environments (14C), Figure 2 Progressive dissolution experiment on late Pleistocene ostracodes (Zimmerman
et al., 2012). 14C dating of CO2 fractions evolved during reaction of ostracodes with acid shows large component of young 14C on
outer surfaces. The true age of the ostracodes is the age of the last step or older; dotted line shows the integrated 14C age of
ostracodes from the same sample (measured separately).

All of the sources of solid and dissolved carbon discussed
in the previous section may influence the carbon isotope
chemistry of the sediment, and additional sources of old
carbon may be region or lake specific and difficult to iden-
tify (e.g., Colman et al., 1996; Grimm et al., 2009). Most
importantly, the kind and amount of organic matter, the
degree and frequency of oxygenation, and rise and fall of
lake level (natural or anthropogenic) all affect the carbon
isotope composition of the bulk sediment, and all are
likely to be changing through time.
A word about outliers
Radiocarbon ages that “look wrong” in the context of the
whole dataset should be carefully examined, and there
should be some basis for rejection of outliers that are
368 LACUSTRINE ENVIRONMENTS (14C)

Lacustrine Environments (14C), Figure 3 Probability density function produced by calibration of a 14C age of 2565  35 years before
present (left axis); wiggles in the calibration curve (diagonal line) produce windows of ages with high and low probability (bottom
axis), rather than an equal likelihood of any age within the calibrated range. Figure made with Calib 6.0 (Stuiver et al., 2005 ref).

<3-sigma different from the expected age. A few common
causes of suspicious dates may include the following.
Samples taken from too close to the core edge, top, or bot-
tom, from a zone of disturbance, or from the core catcher
may be too young or too old and can be rejected (and
avoided in the future). Unidentified plant parts that are
too old may be aquatic and indicate that the water had a
reservoir age at the time the plant grew (regardless of the
modern situation). Charcoal or wood that is too old may
have been stored on the landscape before deposition or
be from the interior of a long-lived tree.
Aside from coring disturbance, it is relatively difficult
to make a radiocarbon sample too young. Small
(<0.3 mg) or old samples (>20,000 years, depending on
sample size and precision required) contain small amounts
of original 14C and therefore are more sensitive to contam-
ination by small amounts of modern carbon. This may be
from contamination by carbonate or microbiological
growth (Wohlfarth et al., 1998) or simply from exposure
to the atmosphere. Concurrent preparation of matrix-
matched, similarly sized background materials is a critical
tool for monitoring these effects.
Building an age model
A key consideration when building an age model for
lacustrine sequences is the time resolution it provides for
interpretation of the proxy records. The high sedimenta-
tion rates in many lakes provide an unparalleled view of
paleoenvironmental change, but the number of dates in
the accompanying age model is often an order of magni-
tude or even two lower than the proxy measurements. This
mismatch results in a situation where decadal or centen-
nial interpretations and correlations of proxies are made,
when only millennial-scale time resolution is available.
Radiocarbon ages are reported, like most isotopic mea-
surements, as values with some analytical uncertainty that
is normally distributed about the measurement value; typ-
ical 14C measurements made by AMS are reported with a
1-sigma analytical uncertainty of 20–40 years. The addi-
tional step of calibration to calendar years introduces a
nonuniform probability to 14C-based calendar ages, how-
ever, because of plateaus and reversals in the slope of the
calibration curve (Reimer et al., 2013). That is, within
the calibrated range of a 14C measurement, some calendar
years will be much more likely than others, and many 14C
ages have multiple ranges, separated by calendar years
with zero probability (Figure 3).
The information in the probability density function is
not captured by simple linear or spline fits to a series of
age-depth points nor by the practice of reporting a central
or highest-probability age and the oldest and youngest
ends of the ranges. In the usual lacustrine age model of
 LACUSTRINE ENVIRONMENTS (14C) 369

Lacustrine Environments (14C), Figure 4 Age model for a lake core built using the Bayesian-based program Bacon (Blaauw and
Christen, 2011). Purple shapes represent probability density functions for 14C dates; green ellipses are independently dated tephras.
Dotted lines enclose envelope of 95 % confidence interval, with darker gray shading for depths with higher confidence intervals.
Additional dates between 120 and 300 cm depth would greatly improve the precision of the age model.

5 or 10 dates or more, it becomes impossible to express the
precision with which the timing of a shift or event in the
proxy records is known, unless it coincides in depth with
a radiocarbon age measurement.
Programs such as OxCal (Bronk Ramsey, 1995) and
Bacon (Blaauw and Christen, 2011) use Bayesian
methods to take advantage of prior information (priors)
about a sediment sequence, including stratigraphic order,
the relative timing of events, and the location of hiatuses
in the sequence, as well as the fact that the probability of
any calendar year within the calibrated range varies
(Bronk Ramsey, 2009). Many thousands or millions of
iterations of Monte Carlo analysis are used to define the
probability of any depth in the sediment being equivalent
to a particular calendar year (Figure 4). This is especially
powerful for defining the precision of an age model
370 LACUSTRINE ENVIRONMENTS (14C)
between dated horizons and for choosing locations in the
sediment sequence where additional dates will provide
the most valuable information.
Summary
Advances in AMS sample preparation and measurement
and Bayesian modeling software enable construction of
robust, relatively high-precision and high-accuracy radio-
carbon age models to support the high-resolution, multi-
proxy lake records that have proliferated recently. Funda-
mentally, radiocarbon chronologies require an investment
in a sufficient number of ages to support the time resolu-
tion of the proxy-record interpretations. Furthermore, they
must include acknowledgment of the geological uncer-
tainties inherent to all sample types. While identifiable ter-
restrial macrofossils remain the preferred material for
radiocarbon dating, reliable age models can be based on
other materials, where careful consideration of pre- and
post-depositional effects on the carbon is combined with
application of as many chronological techniques as
possible.
Bibliography
Abbott, M. B., and Stafford, T. W., Jr., 1996. Radiocarbon geochem-
istry of modern and ancient Arctic lake systems, Baffin Island,
Canada. Quaternary Research, 45, 300–311.
Adams, K. D., Goebel, T., Graf, K., Smith, G. M., Camp, A. J.,
Briggs, R. W., and Rhode, D., 2008. Late Pleistocene and early
Holocene lake-level fluctuations in the Lahontan basin, Nevada:
Implications for the distribution of archaeological sites.
Geoarchaeology, 23, 608–643.
Alberic, P., Jezequel, D., Bergonzini, L., Chapron, E., Viollier, E.,
Massault, M., and Michard, G., 2013. Carbon cycling and
organic radiocarbon reservoir effect in a meromictic crater lake
(Lac Pavin, Puy-de-Dome, France). Radiocarbon, 55,
1029–1042.
Blaauw, M., and Christen, J. A., 2011. Flexible paleoclimate
age-depth models using an autoregressive gamma process.
Bayesian Analysis, 6, 457–474.
Blaauw, M., van Geel, B., Kristen, I., Plessen, B., Lyaruu, A.,
Engstrom, D. R., van der Plicht, J., and Verschuren, D., 2011.
High-resolution 14C dating of a 25,000-year lake-sediment
record from equatorial East Africa. Quaternary Science Reviews,
30, 3043–3059.
Bronk Ramsey, C., 1995. Radiocarbon calibration and analysis of
stratigraphy: the OxCal program. Radiocarbon, 37, 425–430.
Bronk Ramsey, C., 2009. Bayesian analysis of radiocarbon dates.
Radiocarbon, 51, 337–360.
Bronk Ramsey, C., and Lee, S., 2013. Recent and planned develop-
ments of the program OxCal. Radiocarbon, 55, 720–730.
Broecker, W. S., and Walton, A., 1959. The geochemistry of C14 in
fresh-water systems. Geochimica et Cosmochimica Acta,
16, 15–38.
Brown, T. A., Farwell, G. W., Grootes, P. M., and Schmidt, F. H.,
1992. Radiocarbon AMS dating of pollen extracted from peat
samples. Radiocarbon, 34, 550–556.
Colman, S. M., Jones, G. A., Rubin, M., King, J. W., Peck, J. A., and
Orem, W. H., 1996. AMS radiocarbon analyses from Lake
Baikal, Siberia: challenges of dating sediments from a large,
oligotrophic lake. Quaternary Science Reviews, 15, 669–684.
Deevey, E. S., Jr., Gross, M. S., Hutchinson, G. E., and Kraybill,
H. L., 1954. The natural C14 contents of materials from
hard-water lakes. Proceedings of the National Academy of
Science, 40, 285–288.
Grimm, E. C., Maher, L. J., Jr., and Nelson, D. M., 2009. The mag-
nitude of error in conventional bulk-sediment radiocarbon dates
from central North America. Quaternary Research, 72,
301–308.
Grimm, E. C., 2011. High-resolution age model based on AMS
radiocarbon ages for Kettle Lake, North Dakota, USA. Radio-
carbon, 53, 39–53.
Jull, A. J. T., Burr, G. S., Zhou, W., Cheng, P., Song, S. H., Leonard,
A. G., Cheng, L., and An, Z. S., 2013. 14C measurements of
dissolved inorganic and organic carbon in Qinghai Lake and
inflowing rivers (NE Tibet, Qinghai Plateau), China. Radiocar-
bon, 56, 1115–1127.
Marty, J. E., and Myrbo, A., in press. Radiocarbon dating suitability
of aquatic plant macrofossils. Journal of Paleolimnology, doi:
10.1007/210933-014-9796-0.
Mensing, S. A., and Southon, J. R., 1999. A simple method to sep-
arate pollen for AMS radiocarbon dating and its application to
lacustrine and marine sediments. Radiocarbon, 41, 1–8.
Myrbo, A., Morrison, A., and McEwan, R., 2011. Tool for micro-
scopic identification (TMI). http://tmi.laccore.umn.edu.
Accessed on 28 Sept 2014.
Neulieb, T., Levac, E., Southon, J., Lewis, M., Pendea, I. F., and
Chmura, G. L., 2013. Potential pitfalls of pollen dating.
Radiocarbon, 55, 1142–1155.
Newnham, R. M., Vandergoes, M. J., Garnett, M. H., Lowe, D. J.,
Prior, C., and Almond, P. C., 2007. Test of AMS 14C dating
of pollen concentrates using tephrochronology. Journal of
Quaternary Science, 22, 37–51.
Oremland, R. S., Miller, L. G., and Whiticar, M. J., 1987. Sources
and flux of natural gases from Mono Lake, California.
Geochimica et Cosmochimica Acta, 51, 2915–2929.
Pigati, J. S., Rech, J. A., and Nekola, J. C., 2010. Radiocarbon
dating of small terrestrial gastropod shells in North America.
Quaternary Geochronology, 5, 519–532.
Plazcek, C., Quade, J., and Patchett, P. J., 2006. Geochronology
and stratigraphy of late Pleistocene lake cycles on the
southern Bolivian Altiplano: implications for causes of tropical
climate change. Geological Society of America Bulletin, 118,
515–532.
Reimer, P. J., Bard, E., Bayliss, A., Beck, J. W., Blackwell, P. G.,
Bronk Ramsey, C., Buck, C. E., Cheng, H., Edwards, R. L.,
Friedrich, M., Grootes, P. M., Guilderson, T. P., Haflidason, H.,
Hajdas, I., Hatte, C., Heaton, T. J., Hoffmann, D. L., Hogg,
A. G., Hughen, K. A., Kaiser, K. F., Kromer, B., Manning,
S. W., Niu, M., Reimer, R. W., Richards, D. A., Scott, E. M.,
Southon, J. R., Staff, R. A., Turney, C. S. M., and van der Plicht,
J., 2013. IntCal13 and Marine13 radiocarbon age calibration
curves 0–50,000 years cal BP. Radiocarbon, 55, 1869–1887.
Saulnier-Talbot, E., Pienitz, R., and Stafford, T. W., Jr., 2009.
Establishing Holocene sediment core chronologies for northern
Ungava lakes, Canada, using humic acids (AMS 14C) and
210Pb. Quaternary Geochronology, 4, 278–287.
Shanahan, T. M., Peck, J. A., McKay, N. P., Heil, C. W., Jr., King, J.,
Forman, S. L., Hoffmann, D. L., Richards, D. A., Overpeck,
J. T., and Scholz, C. A., 2013. Age models for long lacustrine
sediment records using multiple dating approaches – An exam-
ple from Lake Bosumtwi, Ghana. Quaternary Geochronology,
15, 47–60.
Smith, S. B., Karlin, R. E., Kent, G. M., Seitz, G. G., and
Driscoll, N. W., 2013. Holocene subaqueous paleoseismology of
Lake Tahoe. Geological Society of America Bulletin, 125,
691–708.
Staff, R. A., Bronk Ramsey, C., Bryant, C. L., Brock, F., Payne, R. L.,
Schlolaut, G., Marshall, M. H., Brauer, A., Lamb, H. F., Tarasov, P.,
Yokoyama, Y., Haraguchi, T., Gotanda, K., Yonenobu, H.,
 LASER ABLATION INDUCTIVELY COUPLED MASS SPECTROMETER (LA ICP-MS)
Nakagawa, T., and Suigetsu 2006 Project Members, 2011. New
14C determinations from Lake Suigetsu, Japan: 12,000 to 0 cal
BP. Radiocarbon, 53, 511–528.
Stuiver, M., Reimer, P. J., and Reimer, R. W., 2005. CALIB 5.0.
[WWW program and documentation]. http://calib.qub.ac.uk/
calib/
Tennant, R. K., Jones, R. T., Brock, F., Cook, C., Turney, C. S. M.,
Love, J., and Lee, R., 2013. A new flow cytometry method
enabling rapid purification of fossil pollen from terrestrial sedi-
ments for AMS radiocarbon dating. Journal of Quaternary Sci-
ence, doi:10.1002/jqs.2606.
Wohlfarth, B., Skog, G., Possnert, G., and Holmquist, B., 1998. Pit-
falls in the AMS radiocarbon-dating of terrestrial macrofossils.
Journal of Quaternary Science, 13, 137–145.
Zimmerman, S. R. H., Steponaitis, E., Hemming, S. R., and
Zermeno, P., 2012. Potential for accurate and precise radiocar-
bon ages in deglacial-age lacustrine carbonates. Quaternary
Geochronology, 13, 81–91.
LASER ABLATION INDUCTIVELY COUPLED MASS
SPECTROMETER (LA ICP-MS)
Paul Sylvester
Department of Geosciences, Texas Tech University,
Lubbock, TX, USA
Definition
Laser ablation inductively coupled plasma mass spec-
trometer (LA-ICP-MS) is an in situ analytical instrument
that is used for determining the elemental and isotopic
composition of micrometer-sized areas of solid materials
such as minerals, glasses, metal alloys, bones, teeth, cal-
careous shell and wood, and fluids trapped as inclusions
within solids (Sylvester, 2008). A laser ablation system
is connected to an inductively coupled plasma mass spec-
trometer. Laser pulses remove small volumes of the sam-
ple, typically from pits that are 30–50 mm wide and
deep, during 1–2 min of ablation. The ablated aerosols
are transported in a carrier gas stream to the ICP and
converted to ions, which can be separated based on their
mass to charge ratios by a mass spectrometer and mea-
sured with a detector. The ICP is hot ionized argon gas
produced at the tip of a glass torch by an intense electro-
magnetic field generated by an oscillating current in a
water-cooled copper coil wrapped around the glass torch
(Linge and Jarvis, 2009). Most commonly, the mass spec-
trometer consists of a quadrupole mass filter or a magnetic
sector field mass analyzer incorporating single or multiple
collectors (MC). The detector in single-collector instru-
ments is typically a secondary electron multiplier, whereas
multiple Faraday cups can be used in MC devices. The
moderate capital cost, robust design, multielement capa-
bility, short analysis times, minimal sample preparation
requirements, and good precision and accuracy of
LA-ICP-MS compared to competing techniques for in situ
micrometer-scale, chemical analysis have led to its
371
widespread use in university, government, and industry
research laboratories worldwide.
For analysis, the sample surface is commonly prepared as
a thin polished slice or slab, and imaged using optical light
or scanning electron microscopy, in order to select areas
appropriate for laser sampling. The sample is ablated in an
airtight sample cell flushed with helium gas to transport
the ablated aerosols through plastic tubing to the ICP. Abla-
tion is usually carried out using a nanosecond pulsed laser,
emitting ultraviolet light, which is absorbed more
completely than infrared light by many minerals and other
materials, particularly those that are transparent or poor in
iron. Most modern systems are based on a neodymium-
doped yttrium aluminum garnet (Nd:YAG) solid-state laser
using the frequency-quintupled 213 nm output or an
argon-fluoride (ArF) excimer gas laser using its fundamental
193 nm wavelength. Recently, femtosecond pulsed lasers
have been shown to produce aerosols with smaller particle
sizes and negligible sample melting at the ablation site,
improving analytical performance (Shaheen et al., 2012).
LA-ICP-MS is used for scientific dating largely by
measuring the ages of minerals containing uranium or tho-
rium, which decay to radiogenic isotopes of lead. Zircon is
the mineral dated most commonly using the technique,
although U-Th-Pb ages for other minerals, such as mona-
zite, baddeleyite, allanite, apatite, rutile, and titanite are
increasingly reported. Minerals with ages from as young
as about a million years to those that comprise the oldest
known material found in meteorites can be dated by
LA-ICP-MS using the U-Th-Pb method. Ages are com-
monly reported with uncertainties of 1 % or less, but
interlaboratory comparisons have documented biases of
about 2 % (Košler et al., 2013). The major source of error
in U-Th-Pb dating by LA-ICP-MS is inaccurate correction
for laser-induced fractionation of more-volatile Pb relative
to more-refractory U and Th. U-Th/Pb fractionation is
matrix dependent and is affected by variations in laser
sampling as a function of mineral composition, color,
crystallinity, and fracturing. Various procedures are
employed to limit U-Th/Pb fractionation including
matching calibration standards to the matrix of the
unknowns, ablating with lasers emitting deep ultraviolet
(193 nm) light or femtosecond pulses, moving the sample
beneath the laser beam (referred to as raster or scan mode)
to limit the depth of ablation, and various mathematical
corrections.
Bibliography
Košler, J., Sláma, J., Belousova, E., Corfu, F., Gehrels, G. E.,
Gerdes, A., Horstwood, M. S. A., Sircombe, K. N., Sylvester,
P. J., Tiepolo, M., Whitehouse, M. J., and Woodhead, J. D.,
2013. U-Pb detrital zircon analysis – results of an inter-
laboratory comparison. Geostandards and Geoanalytical
Research, 37, 243–259.
Linge, K. L., and Jarvis, K. E., 2009. Quadrupole ICP-MS: intro-
duction to instrumentation, measurement techniques and analyt-
ical capabilities. Geostandards and Geoanalytical Research,
33, 445–467.
372 LICHENOMETRY

Shaheen, M. E., Gagnon, J. E., and Fryer, B. J., 2012. Femtosecond

(fs) lasers coupled with modern ICP-MS instruments provide
new and improved potential for in situ elemental and isotopic
analyses in the geosciences. Chemical Geology, 330–331,
260–273.
Sylvester, P. J., (ed) 2008. Laser ablation-ICP-Mass Spectrometry in
the earth sciences, current practices and outstanding issues. Min-
eralogical Association of Canada Short Course Series, 40, 348p.

Cross-references
Mass Spectrometry
Secondary Ion Mass Spectrometry (SIMS)
Uranium–Lead Dating
Zircon

Lichenometry, Figure 1 Edge of large Lecidea atrobrunnea in

LICHENOMETRY Sierra Nevada of California. Brown and tan are fungal cells; black
apothecia reproductive cells contain hidden white spore cases;
William B. Bull underlying black algae mat is visible along the right edge.
Department of Geosciences, University of Arizona,
Tucson, AZ, USA

Definition
Lichenometry dates when a young rock surface was cre-
ated by measuring the sizes of lichens whose growth rate
is known.

Introduction
Lichens are an ancient type of symbiotic growth where
algae create food for fungi, that provide shelter for the
algae. Their great variety includes delicate foliose lichens
growing on trees and rocks. Lichenometry uses only the
crustose types (Figures 1 and 2) that can adhere tightly
to rock substrates for a 1,000 years. Figure 2 shows the
type of block for dating a block. It has a range of thallus
sizes, but only the largest lichen is measured, the assump-
tion being that it most closely approximates the rockfall
arrival time.
Recent geomorphic events are dated using the sizes of
crustose lichens growing on blocks. Times of landslides,
advance and retreat of glaciers, earthquakes, floods, cli-
mate change, fires, slush avalanches, and human construc-
tion disturbance can all be dated with lichens.
This chapter summarizes the considerable promise of
using lichens for accurate, precise dating. First is how to
select a field study site and calibrate rates of lichen growth.
Then the emphasis is on precision, and diversity, of dating
when many measurements allow definition of peaks in
a lichen-size distribution.
Lichenometry started in the 1950s (Beschel, 1950,
1957, 1959) to study glaciation – at the same time that
radiocarbon dating began. But now lichenometry has been
virtually abandoned in favor of geochemical methods
such as radiocarbon dating of organic matter, optically
stimulated luminescence of quartz and feldspar, and
cosmogenic radionuclide dating of quartz crystals. The
traditional lichenometry method (Locke et al., 1979;
Lichenometry, Figure 2 This 1.8 m long block has many nice
Acarospora chlorophana. South Fork Kings River, Sierra Nevada,
California. The lichen on the right side seems to be the largest; its
thallus has acceptable quality for measurement. The handle on
the small day pack is 11 cm wide.
Innes, 1985) uses only the five largest lichens on moraines
left by a retreating glacier. The histogram lichen-size
peaks used here are measures of central tendency, not
statistical outliers. My New Zealand-California-Sweden
database contains ~80,000 lichen-size measurements.
Detailed annual measurements of the entire circumfer-
ence of lichens by botanists describe highly erratic growth
of crustose lichens on rock surfaces (Bradwell, 2010;
Matthews and Trenbirth, 2011). No outward growth may
occur for several years, or only one side of a thallus grows.
Some presume that dating with lichens is truly hopeless
(Loso and Doak, 2006). But the circular shape of large
crustose lichens is indicative of uniform outward growth
 LICHENOMETRY
over the long term. I avoid erratic growth tendencies by
using digital calipers to measure maximum lichen growth.
This is the longest axis of the largest lichen on many rock
surfaces of interest, such as each 0.5–2 m rockfall block
on hillside talus or a sequence of adjacent 4 m2 areas of
beveled rock surface exposed by recent retreat of a glacier.
Calibration of lichen growth rate
Initial calibration of lichen growth rate in California was
done in AD 1994 in Yosemite National Park. The long axis
of the largest lichen was measured on blocks exposed by
construction of a trail to Nevada Falls in AD 1940 and on
a AD 1908 landslide that fell off a granitic dome named Lib-
erty Cap. Blocks for an AD 1872 coseismic landslide were
identified by cross-dating annual growth rings of adjacent
damaged trees. These three calibration points in Figure 3
span only 68 years, not sufficient for dating old events.
A prehistoric rock avalanche called “The Slide” (Huber
et al., 2002) stopped abruptly in a forest of alpine hemlock
trees. Tree-ring analysis dated landslide damage to the for-
est as occurring between the end of the growing season in
September AD 1739 and before the start of renewed
growth in June AD 1740. This fourth point in the calibra-
tion Figure 3 falls on the same line as the other three
points, nicely defining a long-term linear style of lichen
growth. Regression of mean lichen size in millimeters,
D, and calendric years, t, describes uniform Rhizocarpon
subgenus Rhizocarpon growth as
D ¼ 189:43  0:095t ð1Þ
Uniform-phase growth is a constant 9.5 mm per cen-
tury. Lichens growing on substrates exposed in the year
AD 1 would have a mean size of 189 mm in AD 1994.
Calibration of three other lichen species allowed dating
of geomorphic processes in diverse climatic settings. Sub-
sequent addition of five new calibration points has not
changed Eq. (1).
Site selection
Site selection determines what lichenometry will date.
Markedly different geomorphic processes occur in differ-
ent parts of the Figure 4 hillslope. The flume-like features
on the left side of the view are debris-flow channels. They
have created debris cones with the passage of time.
Lichens on debris-flow levee boulders could be used to
date individual events and assess frequency of storms
capable of creating such events.
Such processes do not occur on the small patches of
talus in the lower-right part of the Figure 4 view. Most of
these blocks are detached from their source outcrops by
earthquakes. Not much happens between times of seismic
shaking, other than freeze-thaw cycles. There are no trees
to be uprooted by windstorms, and water flowing off the
outcrops just infiltrates into highly porous talus.
The site-evaluation routine for either site is as follows.
Hike up the hill to see which calibrated lichens are present:
Acarospora chlorophana, Lecanora sierrae, Lecidea
373
Lichenometry, Figure 3 Calibration of Rhizocarpon subgenus
Rhizocarpon growth rate in Yosemite National Park, California.
atrobrunnea, and Rhizocarpon subgenus Rhizocarpon.
After finding the largest lichen on a block, assess if its
thallus quality meets our standards. How smooth and com-
plete are the thallus edges at where we will measure each
optimal lichen size to 0.01 mm with digital calipers? Qual-
ity ratings of 1 through 4 are assigned to every measure-
ment. Evaluate both transported blocks and the source
outcrop joint surfaces from which blocks were derived.
We seek a dataset of several hundred measurements;
less than 200 is fine if the size range is only 10–50 mm.
The Figure 2 view of Acarospora chlorophana is the type
of block we prefer. It has a range of thallus sizes; much
better than having only one lichen to measure. Very nice,
but in this case reconnaissance up the talus slope revealed
few blocks with no usable lichens.
Much time is spent finding sites that have many lichens
of suitable quality. Surprisingly, some California talus
slopes may be completely barren – lichens! Only first-
generation crustose lichens growing on fairly smooth rock
surfaces should be measured. Rough surfaces, such as
coarsely crystalline rocks, provide sizes that are smaller
than the same age lichens on ideal substrates such as
smooth quartzite joint planes. The finite lifespan of
a lichen may be less than the age of the ancient substrate
it is growing on. Second- or third-generation lichen
growth never dates the exposure time of when blocks were
deposited and colonization by lichens was first initiated.
These older surfaces typically have a confusing array of
small to large lichens that have grown together.
Precise, accurate lichenometry dating
Historical earthquake rockfalls can be used to test the
lichenometry method as proposed by Bull and Brandon
(1998). The Strawberry Peak site yields typical results
(Figure 5). A first impression may suggest noise instead
of actual rockfall events. It looks like too many closely
spaced, narrow peaks in the 15–30 mm size range.
374 LICHENOMETRY

Lichenometry, Figure 4 Diverse granitic mountainside along Lone Pine Creek, Sierra Nevada, California. Debris flows at left; rockfall
block talus at right. Two-lane dirt road in lower left for scale (Image courtesy of Goggle Earth).

Lichenometry, Figure 5 Distribution of lichen sizes measured in 2013 on rockfall blocks northwest of Strawberry Peak cliffs, central San
Gabriel Mountains, Southern California. 0.5 mm class interval. Latitude is 34.28687 N. Longitude is 118.12007 W. Altitude is 1,560 m.

We need to compare the dates implied by these lichen-size
peaks with dates of historical earthquakes (Table 1).
Each peak nicely records the time of a historical earth-
quakes; so Figure 5 data noise is minimal. These were
large earthquakes – Mw magnitude >6.8. The AD 1857
Fort Tejon event was Mw 7.9 and is recorded by the dom-
inant 29 mm lichen-size peak. It is southern California’s
largest earthquake. Its epicenter was fairly distant, but
~9 m of right-lateral slip occurred along much of the
350 km long surface rupture that passed only 30 km to
the north of Strawberry Peak (http://www.sjvgeology.
org/geology/tejon_earthquake.html).
The lack of noise in the Figure 5 distribution of
lichen sizes and low earthquake dating error in Table 1
 LICHENOMETRY 375

Lichenometry, Table 1 Accuracy of dating historical California earthquakes using the lichen-size peaks of Figure 5

Earthquake name (s) Date, AD Distance to epicenter Lichen-size peak, mm Peak Date AD Dating error, years

Wrightwood, 1812.9 40 37.75 1813.3 0.4

Santa Barbara 160
Fort Tejon 1857.0 170 29.25 1858.2 1.2
Hayward, 1866.8 540 27.75 1866.1 0.7
San Francisco
Lone Pine 1872.9 250 26.75 1871.9 1.0
24.75 1882.0
Laguna Salada 1891.3 370 23.00 1892.1 0.8
San Jacinto 1900.0 170 21.00 1901.9 1.9
San Francisco 1906.3 550 19.75 1908.5 2.2
Imperial Valley 1915.5 270 18.25 1916.4 0.9
Long Beach 1933.2 60 15.75 1929.7 3.5

Lichenometry, Figure 6 Comparison of dendrochronology and lichenometry dating of prehistoric earthquakes on the Alpine fault,
New Zealand. From Bull, 2007, p. 234. n is the number dates for an event.

underscore how few nonseismic rockfalls are included
in this lichen-size dataset. Careful site selection was essen-
tial in order to avoid dating several geomorphic processes.
The broader 37 mm peak records the two big earth-
quakes of December 1812. The Wrightwood event was
on the plate-bounding San Andreas fault to the east, and
the Santa Barbara earthquake to the west occurred
offshore. Both had strong foreshocks and aftershocks.
The AD 1882 date for the 24.75 mm lichen-size peak
does not match the time of a known earthquake. This peak
occurs at many sites; being regional implies a seismic-
shaking origin. Historic earthquake records can be incom-
plete in both New Zealand and California (Bull, 2007).
Even a Mw magnitude 7.5 San Andreas fault event was
missed until recognized recently by Toppozada and
Borchardt (1998).
Four event epicenters (Table 1) were at least 250 km
away. Recording the strong San Francisco earthquake of
1906 is not surprising. That event was felt even farther
away, 150 km to the east according to Ellsworth (1990).
A mean lichenometry dating error of about 2 years is the
same accuracy as for other studies in New Zealand,
California, and Sweden (Bull et al., 1995; Bull, 1996a;
Bull, 1996b, 2003). Bull-Brandon lichenometry dating has
a conservative 5-year estimate of dating precision. Such
consistent quality can only be cross-checked by still better
dating methods – dendrochronology and historical records.
Calibration site ages should be known to the year.
New Zealand tree-ring analyses and lichenometry pre-
cisely dated three prehistorical earthquakes. Trees were
killed, and marked suppressions of annual growth rings
occur in New Zealand cedar (Libocedrus bidwillii) at the
seismically sensitive sites. One is a boggy swamp
and the other is a cliffy ridgecrest directly above the 45
dipping plate-bounding Alpine fault. The times of these
forest disturbance events were synchronous (2 years at
the 95 % confidence level) with times of regional rockfall
events dated by lichenometry using Rhizocarpon
subgenus Rhizocarpon (Figure 6).
Lichen-kill events
Univariate scattergrams detect lichen-kill events caused by
climate change or fire. A scattergram shows the sequence
of lichen-size measurements made while slowly traversing
across a bouldery deposit. Uniformly dispersed talus
accumulating during frequent events has a scattergram with
constant density of measurements and range of lichen sizes
across the entire traverse. There is a systematic decrease in
density abundance for the largest lichen sizes. This decrease
is attributed to the removal of old lichens by death or burial
by incoming younger blocks. In contrast, a snowkill or
firekill event is revealed by a scattergram that has an abrupt
decrease in density with increasing lichen size. A transition
to markedly lower density marks the approximate time of
the lichen-kill event.
376 LICHENOMETRY

Lichenometry, Figure 7 Univariate scattergram of Rhizocarpon subgenus Rhizocarpon size measurements at the Cattle Gully site,
South Island of New Zealand, reveals two lichen-size measurement domains. Measurement densities are different above and below
the 47 mm line; the time of a firekill event that killed most lichens; n ¼ 1,388. Bull and Brandon (1998, Figure 12C).

Dating a fire event
A New Zealand talus site called Cattle Gully highlights
a firekill event. The scattergram (Figure 7) shows two den-
sity domains; the transition at 47 mm dates to about
AD 1700. Dating precision is lower for scattergrams than
for the Figure 5 histogram. Pieces of charred, rot-resistant
totara wood on the Cattle Gully hillslopes suggest that fire
may have killed many lichens, perhaps as newly arrived
Maoris used fires for clearing travel routes and flushing
game (McGlone, 1979).
Dating climate change
Times of colder climate in the high Sierra Nevada of
California occurred during a recent cooler global climate
known as the Little Ice Age. Valley glaciers became larger
and many snowfields became permanent, killing lichens
after ~3 years of burial. Lichens would recolonize these
surfaces as snowbanks melted.
A lichenometry transect across the Chickenfoot glacial
moraine of latest Pleistocene age records two of the Sierra
Nevada snowkill events of the past 1,000 years (Figure 8).
These data show largest lichen maximum diameters on
consecutive blocks. A 200 mm Lecidea atrobrunnea
would date back to about AD 1070, well before the onset
of the Little Ice Age.
The density distribution of the first 200 lichen-size
measurements is similar to a height of 150 mm. Few
lichens were found above this size. Establishment of
a perennial snowbank would kill all lichens except those
on blocks that rose above the snow. It appears that the
Little Ice Age began here shortly before 150 mm time.
A pulse of persistent snow cover ended and recolonization
began at about AD 1290. The even distribution of lichen
sizes below the 150 mm horizontal line indicates that this
part of the measurement transect was not subject to further
snowkills.
An article by Miller et al. (2012) about causes of the
Little Ice Age postulates that cooling was caused by epi-
sodes of SO2-rich volcanic eruptions that also affected
oceanic circulation. When did it begin? End? “summer
cold and ice growth began abruptly between AD 1275
and AD 1300.” The Miller et al. data suggest that the Little
Ice Age ended by AD 1850.
The next 220 lichen-size measurements along the
Figure 8 transect are much smaller; none are larger than
76 mm and most are smaller than 20 mm. A second
snowkill event ended at 20 mm time (AD 1853  10
years). The many lichen sizes between 5 and 20 mm sug-
gest that renewed additional colonization did not occur on
all blocks at the same time. The thickest, more persistent
snow took longer to melt. Other Sierra Nevada sites with
different topographic settings might reveal times of
multiyear snowfields between AD 1290 and AD 1850.
Discussion
The crustose lichens used in these studies seem to have
a preferred geographic distribution and abundance. Most
are found in cooler latitudes, or above 1,000 m altitude
in mid-latitude and low latitude mountains. This may
restrict areas where lichenometry can be used. Fortunately,
rocky mountain ranges are common. Detailed information
about how landscapes evolve is a major virtue of
 LICHENOMETRY 377

Lichenometry, Figure 8 Univariate scattergram of Lecidea atrobrunnea at the Chickenfoot moraine lichenometry site, altitude of
3,540 m, in the Sierra Nevada of California. Two domains of anomalous decrease in density of lichen sizes that suggest times when
snowkill events ended. n ¼ 481. Bull and Brandon (1998, Figure 12B).

lichenometry, but only for the past 1,000 years for most
studies. Precise surface-exposure dating of fluvial, glacial,
and mass-movement landforms can be used to address
landscape change rates and the frequency of geomorphic
hazards to humans. Landslides, floods, and seismic shak-
ing can be deadly.
Reconnaissance and measurement field time are the
main costs. Fieldwork is a mix of working at sites of
known age to calibrate and further test rates of lichen
growth for several species of a genus, and collecting lon-
gest-axis lichen size data at other sites to better understand
the history and extent of a geomorphic process. After cal-
ibration of a lichen growth rate is complete, the cost per
lichenometry age determination is quite low, and one does
not have to wait months or years for expensive laboratory
work to be completed. The more expensive aspect of
lichenometry is the field time spent in finding good sites
and collecting representative samples of lichen sizes.
Botanists wonder why geologists assume that crustose
lichens grow in a predictable manner that can be defined
with an equation, when their year-to-year measurements
suggest nonsensical confusion. Geologists would like bot-
anists to tell them why each crustose lichen does not vary
in growth rate throughout the varied climate of a lofty
600 km long mountain range such as the Southern Alps
or the Sierra Nevada.
The lichenometry method of dating summarized here is
precise and accurate. Lichenometry is a surface-exposure
type of age determination, so it dates the actual time of
an event. The vastly more popular radiocarbon dating
method estimates the time of formation of organic matter
created before or after the event of interest, such as rupture
of layered sedimentary deposits.
Suggested innovative studies
Human-induced global warming has reached the point
where record high temperatures greatly outnumber record
lows and storms seem ever more powerful and prolonged.
Lichenometry can be used to assess recent retreat of gla-
cial ice that leaves behind moraines of lichen-covered
rocks. How much deglaciation is the result of natural cli-
mate variation? Does acceleration of montane glacier
retreat in places such as the Andes, Himalayas, and
Europe match the trend of an ever-increasing carbon diox-
ide in the Earth’s atmosphere? What is the pace of retreat
of the Greenland ice cap? Have any recent lichen snowkill
events occurred, indicative of climate variability? Arctic
lakes have diminished and some are now gone (Smol
and Douglas, 2007). We should date ages and rates
of retreat of now-exposed rocky shorelines, comparing
small ponds with large lakes. The timespan for these
studies should start before the industrial revolution
at ~ AD 1750–1800.
Southern California earthquake studies have focused
on radiocarbon dating of surface ruptures by the plate-
bounding San Andreas fault (Biasi and Weldon, 2009).
We know little about other prehistoric earthquakes that
occurred between the San Andreas fault events of AD
1812 and AD 1680. How frequently did strong regional
378
seismic shaking occur in Southern California during
recent prehistoric times? Seismic shaking decreases with
distance away from an earthquake epicenter. So, mapping
the decrease in coseismic rockfall abundance should be
a useful proxy for seismic shaking. Lichenometry sug-
gests that 13 strong regional seismic-shaking events
occurred between AD 1680 and AD 1812. Four are
recorded by the 41, 44, 48, and 50 mm lichen-size peaks
of Figure 5 of this article. How did the intensity of seismic
shaking for these recent prehistoric events compare with
the San Andreas fault events of 1680, 1812, and 1857?
Lichen-size data from sites in three mountain ranges could
be used to make maps depicting regional seismic shaking
to compare with maps for seismic shaking caused by the
1812 and 1857 earthquakes.
Bibliography
Beschel, R. E., 1950. Flechten aus Altersmaastab rezenter, moränen
(Lichens as a yardstick of age of late-glacial moraines).
Zeitschrift für Gletscherkunde und Glazialgeologie, 1, 152–161.
Beschel, R. E., 1957. A project to use lichens as indicators of
climate and time. Arctic, 10, 60.
Beschel, R. E., 1959. Lichenometrical studies in West Greenland.
Arctic, 11, 254.
Biasi, G. P., and Weldon, R. J., 2009. San Andreas fault rupture sce-
narios from multiple paleoseismic record; string pearls. Bulletin
of the Seismological Society of America, 99, 472–498.
Bradwell, T., 2010. Studies on the growth of Rhizocarpon
geographicum in NW Scotland, and some implications for
lichenometry. Geografiska Annaler, 92A(1), 41–52.
Bull, W. B., 1996a. Prehistorical earthquakes on the Alpine fault,
New Zealand. Journal of Geophysical Research, Solid Earth
Special Section, Paleoseismology, 101(B3), 6037–6050.
Bull, W. B., 1996b. Dating San Andreas fault earthquakes with
lichenometry. Geology, 24, 111–114.
Bull, W. B., 2003. Lichenometry dating of coseismic changes to
a New Zealand landslide complex. Annals of Geophysics, 46,
1155–1167.
Bull, W. B., 2007. Tectonic Geomorphology of Mountains, A New
Approach to Paleoseismology. Oxford: Blackwell Publishing,
p. 320.
Bull, W. B., and Brandon, M. T., 1998. Lichen dating of earthquake-
generated regional rockfall events, Southern Alps, New Zealand.
Geological Society of America Bulletin, 110, 60–84.
Bull, W. B., Schlyter, P., and Brogaard, S., 1995. Lichenometric
analysis of the Kärkerieppe slush-avalanche fan, Kärkevagge,
Sweden. Geografiska Annaler, 77A, 231–240.
Ellsworth, W. L., 1990. Earthquake history, 1769–1989. In Wallace,
R. E., (ed.), The San Andreas Fault System. U.S. Geological
Survey Professional Paper 1515, pp. 153–187.
Huber, N. K., Phillips, W. M., and Bull, W. B., 2002. The Slide,
Yosemite National Park, CA, Yosemite Association. Yosemite,
64(3), 2–4.
Innes, J. L., 1985. Lichenometry. Progress in Physical Geography,
9, 187–254.
Locke, W. W. III., Andrews, J. T., and Webber, P. J., 1979. A Manual
for Lichenometry. British Geomorphological Research Group
Technical Bulletin 26, 47pp.
Loso, M. G., and Doak, D. F., 2006. The biology behind
lichenometric dating curves. Oecologia, 147, 223–229.
Matthews, J. A., and Trenbirth, H. E., 2011. Growth rate of a very
large crustose lichen (Rhizocarpon subgenus) and its implica-
tions for lichenometry. Geografiska Annaler, 93, 27–39.
LUCY
McGlone, M. S., 1979. The Polynesian settlement of New Zealand
in relation to environmental and biotic changes. In Rudge, M. R.,
(ed.), Moas, Mammals, and Climate in the Ecological History of
New Zealand. New Zealand Journal of Ecology, 12, 15–129.
Miller, G. H., et al., 2012. Abrupt onset of the Little Ice Age trig-
gered by volcanism and sustained by sea-ice/ocean feedbacks.
Geophysical Research Letters, doi:10.1029/2011GL050168.
Smol, J. P., and Douglas, M. S. V., 2007. Crossing the final ecolog-
ical threshold in high Arctic ponds. Proceedings of the National
Academy of Sciences, 104, 12395–12397.
Toppozada, T. R., and Borchardt, G., 1998. Re-evaluation of the 1836
“Hayward Fault” and the 1838 San Andreas Fault Earthquakes.
Bulletin of the Seismological Society of America, 88(1), 140–159.
LUCY
William H. Kimbel
Institute of Human Origins, School of Human
Evolution and Social Change, Arizona State University,
Tempe, AZ, USA
Definition
Lucy is the colloquial name given to a presumed female
partial skeleton of the Pliocene hominin species Australo-
pithecus afarensis. The specimen, cataloged as Afar
Locality (A.L.) 288–1, was found by Donald Johanson
in November 1974, at the Hadar site, in the west-central
part of the Afar Rift Valley, Ethiopia. Pyroclastic sedi-
ments in the Hadar Formation at the Hadar site have been
dated by K/Ar and 40Ar/39Ar methods to between 3.4 Ma
and 3.0 Ma, which brackets the younger half of the known
temporal range of A. afarensis (Kimbel and Delezene,
2009). More than 375 specimens of A. afarensis have been
recovered from fluviolacustrine sediments at Hadar
through 2012. The species is also definitively represented
in similar-age deposits of several other Afar basin sites
(including Dikika, Ledi-Geraru, Woranzo-Mille, and
Maka) and in the Koobi Fora Formation, at Koobi Fora,
Kenya. The oldest known fossils of the species, including
the type specimen (a mandible, LH-4), occur in the Upper
Laetolil Beds (3.5–3.8 Ma), at Laetoli, northern Tanzania.
Lucy and other specimens of A. afarensis epitomize the
mosaic emergence of human anatomy during the early
evolution of our lineage (Kimbel and Delezene, 2009).
The cranium and jaws retain many features from the com-
mon ancestor we shared with chimpanzees, including
a small brain (350–550 cc) and projecting snout, but the
relatively low-crowned canine teeth wore down from their
tips as in modern humans. While Lucy’s hip, knee, and
foot were adapted to a balanced two-legged stride, with
fully extended knee and powerful push-off at the great
toe, she had long, powerful arms and curved fingers, rem-
iniscent of the tree-climbing anatomy of the great apes.
These anatomical contrasts have been fodder for intense
scholarly debate regarding the adaptations of our early
ancestors (Ward, 2002).
 Lu-Hf DATING: THE Lu-Hf ISOTOPE SYSTEM
Bibliography
Kimbel, W. H., and Delezene, L., 2009. Lucy redux: a review of
research on Australopithecus afarensis. Yearbook of Physical
Anthropology, 52, 2–48.
Ward, C. V., 2002. Interpreting the posture and locomotion of
Australopithecus afarensis: where do we stand? Yearbook of
Physical Anthropology, 45, 185–215.
Lu-Hf DATING: THE Lu-Hf ISOTOPE SYSTEM
Jeff Vervoort
School of the Environment, Washington State University,
Pullman, WA, USA
Definition
The radiogenic isotope applications of the Lu-Hf system
utilize variations in 176Hf resulting from the radioactive
decay of 176Lu to 176Hf and variations in the Lu/Hf ratios
in rocks and minerals. It is used both in geochronology,
primarily in rocks containing a high Lu/Hf phase, and in
tracer isotopic applications, to track the chemical evolu-
tion of the Earth and other solar system materials and to
constrain the sources of magmas and sediments.
Introduction
The Lu-Hf isotope system, with applications to geo- and
cosmochemistry, was first investigated in the early 1980s
(Patchett and Tatsumoto, 1980a; Patchett and Tatsumoto,
1980b; Patchett and Tatsumoto, 1980c, 1981; Patchett,
1983b) following the successful implementation of the
Rb-Sr and Sm-Nd isotope systems several years earlier.
There are some obvious similarities between the Lu-Hf
and Sm-Nd isotope systems and, as a result, they have
long been used in concert in a wide range of studies
(e.g., Patchett, 1983b; Patchett and Chauvel, 1984; Salters
and Hart, 1991; Beard and Johnson, 1993; Johnson and
Beard, 1993; Blichert-Toft and Albarède, 1997; Vervoort
and Blichert-Toft, 1999; Vervoort et al., 1999). In these
two systems all elements are lithophile and refractory with
high condensation temperatures. Because of these charac-
teristics it has long been assumed that their abundances in
the Earth can be approximated by chondritic meteorites
(see discussion below). In addition, all elements in these
systems behave incompatibly during melting and are
concentrated in the melt over the residual solid. In both
systems, the daughter element (e.g., Nd, Hf) is more
incompatible than the parent, leading to lower Sm/Nd
and Lu/Hf ratios in the melt than the residual solid. As
a result, the 143Nd/144Nd and 176Hf/177Hf ratios in mag-
matic rocks are highly correlated (e.g., Patchett, 1983a;
Salters and Hart, 1991; Johnson and Beard, 1993;
Vervoort and Blichert-Toft, 1999; Chauvel et al., 2008).
Both Lu and Hf are highly immobile and insoluble and,
as is the case with the Sm-Nd system, are thought to
be resistant to perturbations and retain their isotopic
379
information through significant degrees of alteration and
metamorphism.
In the case of the Lu-Hf isotope system, the parent is the
heaviest of the rare-earth elements (REEs) and has broad
geochemical similarities with all other +3 valence REEs.
The daughter element, however, is not an REE, but a high-
field-strength element (HFSE) with a +4 valence and, as
such, can behave much differently than the REEs. These
differences can confer some advantages to the Lu-Hf iso-
tope system for both geochronology and tracer isotopic
work as will be discussed below.
Although the Lu/Hf ratio is fractionated during mag-
matic processes, the degree of this fractionation, as is the
case for Sm-Nd, is not very large. This results in small var-
iations in Lu/Hf ratios between not only rocks within
a comagmatic suite but also for most mineral phases
within individual rocks. This lack of parent/daughter var-
iation severely limits the application of the Lu-Hf chro-
nometer, much as it does for the Sm-Nd isotope system.
Therefore, whole-rock Lu/Hf isochrons of truly
comagmatic suites of rocks either have too limited varia-
tion in Lu/Hf ratios to provide precise ages or, if there
is a significant Lu/Hf variation, chances are the suite of
rocks used in creating the isochron was not truly
comagmatic. There are a few phases, however, with high
affinity for Lu, which make them highly useful in geo-
chronology in both magmatic (e.g., apatite) and metamor-
phic (e.g., garnet, lawsonite) rocks. In addition, some
other phases (particularly phases with Zr or Ti as stoichio-
metric constituents) have strong affinities for Hf (e.g., zir-
con, baddeleyite, rutile), which make them useful in tracer
isotopic studies – especially when used in conjunction
with the U-Pb geochronologic information from these
mineral phases.
Although the Lu-Hf system was introduced to the geo-
chemical and cosmochemical communities in the 1980s,
analytical challenges limited the widespread use of this
technique until the late 1990s. The reason for this is that
Hf has a very high first ionization potential (6.83 eV, Lide,
2003), which prevented easy and routine analysis of Hf
isotopes using thermal ionization mass spectrometry
(TIMS). As a result there were only a few practitioners
using this system in the 1980s and much of the 1990s.
These difficult analyses utilized specialized techniques
and required considerable amounts of Hf for the analyses
to be successful. This all changed with the advent and
development of the multi-collector inductively coupled
plasma mass spectrometer (MC-ICPMS), which com-
bined the efficient ionization of Hf using an ICP ionization
source with the high precision analysis afforded by multi-
collector mass spectrometry (Walder and Freedman, 1992;
Blichert-Toft et al., 1997). This revolutionized the use of
the Lu-Hf isotope system for geochronology and tracer
isotope applications by allowing the routine analysis not
only of Hf isotopes in general but also of much smaller
amounts of Hf. This made several important things
possible. First, it allowed for the more precise and
accurate determination of the 176Lu decay constant
380 Lu-Hf DATING: THE Lu-Hf ISOTOPE SYSTEM
(Scherer et al., 2001; Söderlund et al., 2004). Second,
the new MC-ICPMS instrumentation also facilitated the
analysis of high Lu-Hf, but Hf-poor, phases like garnet
or apatite, which made the use of these phases for geochro-
nology possible. Third, it also made it possible to analyze
Hf-poor rocks such as chondritic meteorites, which
allowed more accurate constraints to be placed on the
chondritic Lu-Hf value (e.g., Bouvier et al., 2008).
Finally, MC-ICPMS technology has made it possible to
determine the Lu-Hf isotope composition in very small,
Hf-rich phases such as zircon. This latter application, par-
ticularly when done in conjunction with U-Pb geochronol-
ogy, has enjoyed an explosion of use in recent years (e.g.,
Amelin et al., 2000; Woodhead et al., 2004; Xie
et al., 2008; Yuan et al., 2008; Gerdes and Zeh, 2009;
Kemp et al., 2010; Fisher et al., 2014).
Background
Lutetium is the heaviest of the rare-earth elements (REEs)
and shares a +3 valence with the lanthanides as a whole.
It also has an odd atomic number (odd number of protons;
Z ¼ 71), which, in part, makes it the least abundant of the
REEs (67 ppb; McDonough and Sun, 1995). Lu has two
isotopes (Table 1): 175Lu is stable, is an odd-even nuclide,
and is the most abundant isotope (97.4 %); 176Lu is an
odd-odd nuclide, has low abundance (2.6 %), and is radio-
active, decaying primarily to 176Hf by beta decay. A small
percentage (<3 %) of 175Lu is reported to decay to 176Yb
by electron capture (Dixon et al., 1954) although Amelin
and Davis (2005) establish an upper limit of 0.9 % based
on combined electron capture and positron emissions.
Hafnium is the next higher element from Lu on the peri-
odic table (Z ¼ 72) and is a high-field-strength element
(HFSE) sharing geochemical similarities with other Group
IVB elements, most notably Zr and Ti. It is relatively
under-abundant in the Earth (0.28 ppm, McDonough and
Sun, 1995) but enriched in the continental crust
(5.3 ppm, Rudnick and Gao, 2003). It has six isotopes,
five of which are stable (Table 1). One isotope, 176Hf
(5.20 %), is radiogenic and produced, in part, by beta
decay of 176Lu and, together with 176Lu, provides the basis
of the Lu-Hf chronometer. By convention the reference
stable isotope for the Lu-Hf system is 177Hf (18.60 %).
Lu-Hf Dating: The Lu-Hf Isotope System, Table 1 Isotopic composition of Lu and Hf
175 176
Lu Lu
176
Ratio to Lu 37.666 1.0
Atomic % 97.41 2.59
Atomic weight 174.94079 175.94269
Atomic weight Lu 174.96670 g/mol
174 176
Hf Hf
177
Ratio to Hf 0.0086743 0.282160
Atomic % 0.1613 5.2474
Atomic weight 173.940065 175.94142
Atomic weight Hf 178.485367
In order to determine accurate Lu and Hf concentra-
tions and, most importantly, the precise 176Lu/177Hf ratios
required for geochronology and calculating accurate ini-
tial ratios, the “isotopic dilution” technique is required.
This technique involves the addition of measured amounts
of tracers of known isotopic composition (highly enriched
in 176Lu and 180Hf, typically, with some labs using 178Hf).
These tracers, or “spikes,” are generally mixed and pre-
cisely calibrated so that evaporative loss will not affect
the precise determination of the parent/daughter ratios.
These spikes are added to the sample solution following
or during dissolution, equilibrated with the solution, and
become part of the isotopic mixture of the sample. Ulti-
mately the 176Lu/177Hf and 180Hf/177Hf ratios of the sam-
ples are measured to determine Lu and Hf concentrations
and precise Lu/Hf ratios.
In order to correct for mass dependent fractionation
during mass spectrometry (i.e., mass bias), the convention
for the Hf isotope system is to normalize to
179
Hf/177Hf ¼ 0.7325 (Patchett and Tatsumoto, 1980b),
the value that has been used since the early days of Hf iso-
tope analyses. Mass bias for Lu isotopic analyses is com-
plicated by virtue of the fact that Lu has only two
isotopes, making internal normalization impossible. In
the early TIMS era of Lu-Hf isotope measurements, the
small degree of mass bias (~0.1 % per amu) was not as
significant for the less stringent requirements of the Lu
isotope dilution analysis. With the advent of the
MC-ICPMS analyses, however, mass fractionation that
is over an order of magnitude larger than TIMS necessi-
tated an accurate mass bias correction. The current
approach is to use a different element with at least one
invariant isotope ratio and which has a similar mass bias
response, such as Yb, to correct for mass bias in the Lu iso-
topic measurement. Because of the similarity of behavior
of the REEs (particularly for adjacent REEs), it is often
difficult to quantitatively remove Yb from Lu. The current
practice employed in most labs is to remove the majority
of the Yb (recall that Yb is an even element and so is much
more abundant than the odd numbered Lu) and use the Yb
isotope measurements to determine the mass bias and then
apply this to the Lu measurement. This approach is
somewhat complicated by the uncertain isotopic composi-
tion of Yb (e.g., Chu et al., 2002; Segal et al., 2003;
177 178 179 180
Hf Hf Hf Hf
1.0 1.467170 0.7325 1.886660
18.5972 27.2852 13.6224 35.0865
176.943233 177.94371 178.945827 179.946561
 Lu-Hf DATING: THE Lu-Hf ISOTOPE SYSTEM
Vervoort et al., 2004; Amelin and Davis, 2005) and the
fact that 176Yb is a relatively abundant isotope (12.70 %)
with the same mass as the 176Lu spike. The interference
of 176Yb on the spike isotope, 176Lu, requires that most of
the Yb be removed from the sample. In detail there are
probably small differences in the mass bias of Yb and Lu,
and there are approaches to quantify these differences
(e.g., White et al., 2000). Despite all of these challenges,
176
Lu/177Hf ratios in rock samples can probably be deter-
mined to ~0.2 % accuracy (Vervoort et al., 2004).
The Hf solution standard employed throughout the his-
tory of Hf isotope analysis has been JMC-475, a Hf metal
obtained from Johnson Matthey Corporation (Patchett and
Tatsumoto, 1980b) with a currently accepted 176Hf/177Hf
value of 0.282160 (Vervoort and Blichert-Toft, 1999).
The JMC-475 Hf standard solution is available from the
author on request.
Decay system fundamentals
Lutetium-176 decays to 176Hf by beta decay with a half-
life of 37.12 Ga:
176
Lu ! 176
Hf þ b :
Only a small percentage (~1 %) of 176Hf in rocks and
minerals is radiogenic and produced in this way; the
majority of 176Hf is “common” Hf, produced primarily
during nucleosynthesis. The radiogenic decay equation
for the Lu-Hf isotope system describes how radiogenic
176
Hf in a rock or mineral will evolve over time:

low Lu/Hf phases in the rock (baddeleyite, zircon, biotite)
176
Hf ðpÞ ¼176 Hf ðtÞ þ176 LuðpÞ  elt  1 :
This expression shows that the 176Hf present in a given
sample today is a function of (1) the 176Hf in that sample at
some time, t, in the past; (2) the amount of radiogenic
176
Lu in that material; (3) the amount of time elapsed since
formation; and (4) the decay constant. In practice, we ref-
erence these isotopes to 177Hf, a stable isotope of Hf. Thus
the equation above becomes:


176
Hf =177 Hf ðpÞ ¼ 176 Hf =177 Hf ðtÞ þ 176 Lu=177 Hf ðpÞ  elt  1 :
Therefore, the material with high Lu/Hf ratios, such as
garnet, will generate proportionally more 176Hf over time
and evolve to higher 176Hf/177Hf ratios. Conversely, the
material with low Lu/Hf ratios, such as zircon, will pro-
duce proportionally little 176Hf with time and so its
176
Hf/177Hf ratios will increase only very slowly.
An important part of using these radiogenic equations
for geochronology and for determining precise initial iso-
topic compositions is not only measuring the present-day
ratios accurately but also knowing the decay constant pre-
cisely and accurately. The 176Lu decay constant has been
a source of uncertainty, however, and this has limited the
full application of the Lu-Hf isotope system. The first
determination of the 176Lu decay constant for use in geo-
chemistry and cosmochemistry was performed by Patchett
and Tatsumoto (1980a) who determined an isochron on
381
a collection of basaltic and cumulate eucrite meteorites.
Using a presumed age of 4.55 Ga for the differentiation
event for the eucrites, they determined a value for l176Lu
of 1.962  0.081 * 1011 year1 with a corresponding
half-life of 35.3 Ga. This value was later revised by Patchett
(1983b) to l176Lu ¼ 1.94  0.07 * 10 11year 1 and a half-
life of 35.7  1.4 Ga. This revised value was similar to
a decay constant value of 1.93  0.03 * 1011 year1 deter-
mined by physical counting experiments (Sguigna
et al., 1982). The agreement between these results led, in
part, to the use of the l176Lu value of 1.94  0.07 * 1011
year1 throughout the 1980s and much of the 1990s, with
later studies (e.g., Blichert-Toft and Albarède, 1997 and
subsequent papers) using the Sguigna et al.’s (1982) values.
This early history of 176Lu decay constant determinations is
reviewed more thoroughly by Begemann et al. (2001).
The value determined for l176Lu changed dramatically
following age comparison studies by Scherer et al. (2001)
and Söderlund et al. (2004). Scherer et al. (2001)
suggested a l176Lu value of 1.865  0.015 * 1011 year1
based on the age comparisons of 4 samples from diverse
localities and with different ages. This included two peg-
matites from Norway with ages ~1.1 and 0.9 Ga, an
apatite-xenotime bearing gneiss from New York State with
an age of ~1.0 Ga, and a carbonatite from South Africa with
an age of ~2.1 Ga. The U-Pb crystallization ages for these
samples were based on the measurements of gadolinite,
xenotime, baddeleyite, and apatite in these rocks; the
Lu-Hf isochron was determined using the high Lu/Hf
ratios of gadolinite, xenotime, and apatite combined with
or the whole-rock compositions.
In a similar study, Söderlund et al. (2004) determined
a l176Lu value of 1.867  0.008 * 1011 year1 based on
age comparisons using two Proterozoic dolerite dikes
from Sweden. The ages of these rocks were determined
using baddeleyite U-Pb geochronology; apatite was the
high Lu/Hf phase providing the spread in the Lu-Hf iso-
chron. Additional determinations by Scherer et al. (2003)
agreed with the slightly faster decay of Söderlund et al.’s
(2004) value. As a result of this agreement, the isotopic
community is now using the l176Lu value of 1.867 
0.008 * 1011 year1. The half-life corresponding to this
value is 37.12 Ga. The revision in the value for the 176Lu
decay constant from the Patchett (1983b) and Sguigna
et al.’s (1982) values has two important consequences
for Lu-Hf geochronology and isotope geochemistry. The
first, and most obvious of these, is the change in calculated
Lu-Hf ages. Using the Scherer/Söderlund decay constant
results in ages that are 4 % older than ages determined
with the Patchett/Sguigna value. The second consequence
is that this lowers the calculated initial 176Hf/177Hf of
rocks. The degree of this effect is a function of age but also
of the 176Lu/177Hf ratios. For Archean and older samples,
the lowering of the calculated initial 176Hf/177Hf for
the Earth’s oldest rocks has very significant implications
for the record of the Hf isotope evolution of the Earth
(e.g., Scherer et al., 2001, 2007).
382 Lu-Hf DATING: THE Lu-Hf ISOTOPE SYSTEM
Although the Söderlund/Scherer decay constant determi-
nations, which used terrestrial samples in their age compar-
isons, have converged on a well constrained value of 1.867
 0.008 * 1011 year1, age comparison determinations
based on meteorites continue to yield a faster 176Lu decay
rate. Blichert-Toft et al. (2002) reanalyzed a suite of basaltic
and cumulate eucrites and concluded that their regression
was consistent with the 1.93  0.03 * 1011 year1 value
of Sguigna et al. (1982) and not the slower decay rates deter-
mined from terrestrial age comparisons. As was true with
the earlier work of Patchett and Tatsumoto (1980a), how-
ever, the basaltic and cumulate eucrites plot at opposite
ends of the isochron with the basaltic eucrites at the low
Lu-Hf end and the cumulate eucrites at the high Lu-Hf
end, which may indicate these are not truly
cogenetic. Further, these samples appear to have different
formation ages with the cumulate eucrites arguably younger
(Mittlefehldt et al., 1998). In another study using meteorite
age comparison, Bizzarro et al. (2003) determined a l176Lu
value of 1.983  0.033 * 1011 year1, even faster than the
Patchett/Sguigna values. The Bizzarro et al. (2003) study
included analyses of ordinary chondrites, carbonaceous
chondrites, and basaltic eucrites and used a common age
of 4.56 Ga. As was true with the eucrite-based determina-
tion, this study included a diverse suite of meteorites of
potentially different ages and metamorphic histories,
and therefore, the decay constant determination may be
affected in some way by the use of samples that do not
strictly meet the requirements for an isochron. These
caveats notwithstanding, there are indications that meteorite
isochrons yield a faster decay constant. One possible expla-
nation for this apparent difference between the terrestrial
and meteorite determinations is the presence of excess
176
Hf in meteorites. This may have been produced by
excitation of the long-lived radioactive isotope 176Lu to a
short-lived 176Lu isomer early in the history of the solar
system that then decays quickly to 176Hf, thereby producing
the excess 176Hf (Albarède et al., 2006; Thrane et al., 2010).
The exact mechanism of how this might have occurred is
still a matter of debate (Albarède et al., 2006; Thrane
et al., 2010).
The Lu-Hf isochron method
All Lu-Hf geochronology uses the “isochron” method to
determine absolute ages. Unlike some chronometers such
as U-Pb in zircon where there is little, if any, daughter iso-
tope at the time of mineral growth, in the Lu-Hf system,
there is invariably an abundance of daughter isotopes pre-
sent (i.e., “common Hf”). For this reason, it is not possible
to determine the age of a single mineral sample by measur-
ing the accumulation of daughter isotopes in the sample.
Thus all Lu-Hf geochronometry relies on the isochron
approach, which utilizes coexisting rocks and minerals to
determine an age. Important assumptions in this approach
include the following: (1) All the minerals formed at the
same time (i.e., have the same age); (2) all samples are
cogenetic (have the same isotopic composition at the time
.304
Garnet Lu-Hf isochron
.300 4
Hf/ 177Hf isochron
.296
.292
3
.288
176
.284
1 2
.280
0 1 2 3 4
176 Lu/ 177Hf isochron
Lu-Hf Dating: The Lu-Hf Isotope System, Figure 1 The
isochron method applied to garnet Lu-Hf geochronology.
(1) uniform Hf isotopic composition at time t ¼ 0; (2) phases in
rock acquire different Lu/Hf ratios (garnets with high Lu/Hf), but
all with the same 176Hf/177Hf; (3) Hf isotopic compositions of the
different components evolve with time as a function of their
Lu/Hf ratios; (4) if the rock has remained a closed system with
respect to Lu and Hf, all components will plot along a line
(isochron) that is a function of time.
of formation); and (3) the sample being dated has been
a closed system since the time of mineral growth.
The systematics of the Lu-Hf isochron method, using
garnet geochronology as an example, is shown in Figure 1:
(1) Prior to mineral formation, whether in a magma or
metamorphic protolith, the bulk rock has a low Lu/Hf ratio
and (at least approximately) a uniform Hf isotopic compo-
sition at the hand sample scale. (2) During crystallization,
or, in this example, metamorphism, different phases form
with different Lu/Hf ratios, but all with the same initial
176
Hf/177Hf. (3) Over time, the Hf isotopic compositions
of the different components evolve as a function of their
Lu/Hf ratios with high Lu/Hf components evolving more
quickly to elevated 176Hf/177Hf and low Lu/Hf compo-
nents evolving more slowly. (4) At some later time, if
the rock has been a closed system with respect to Lu and
Hf, all components will have compositions that plot along
a line that is a function of time. Regression of a line
through these points yields a slope from which an age
can be calculated using the formula:
Age ¼ lnðslope þ 1Þ=l176 Lu :
In detail, many of the assumptions of the garnet iso-
chron method are not strictly met. In most cases the rock
is certainly not in perfect isotopic equilibrium (i.e., homo-
geneous 176Hf/177Hf) prior to metamorphism. In addition,
the time different phases in the rock form may not be pre-
cisely the same, especially in the case of prolonged garnet
growth (e.g., Kohn, 2009). These can be serious obstacles
if there is a small range of Lu/Hf and 176Hf/177Hf
 Lu-Hf DATING: THE Lu-Hf ISOTOPE SYSTEM
variations in samples but are often overcome in isochrons
with high Lu/Hf phases and large amounts of radiogenic
Hf ingrowth.
Geochronologic applications
Whole-rock Lu-Hf isochrons
Because of the limited fractionation in Lu/Hf that occurs
during magmatic processes, there is rarely a sufficiently
large range in Lu/Hf ratios in suites of rocks that are truly
cogenetic to generate an isochron with enough precision
to be useful (Patchett, 1983b). If there is a large variation
in Lu/Hf in a suite of samples, it is likely these samples
are probably not strictly cogenetic. Examples of the
former may be represented by the eucrite isochrons com-
bining both basaltic and cumulate eucrites (Patchett,
1983b; Blichert-Toft et al., 2002) and in the collective
chondrite-eucrite isochron of Bizzarro et al. (2003). In
practice, therefore, Lu-Hf whole-rock isochrons are rarely
employed for geochronology, especially for terrestrial
samples.
Apatite Lu-Hf geochronology
There is potential in some magmatic rocks, however,
to crystallize a high Lu/Hf phase (e.g., apatite, magmatic
garnet) that can then be used in geochronology. The mag-
matic rocks that have the most promise for useful
geochronometry are those that contain apatite as a crystal-
lizing phase. As demonstrated by the decay constant work
of Scherer et al. (2001) and Söderlund et al. (2004) and
the pioneering work applying Lu-Hf geochronology to
phosphates by Barfod et al. (2003), apatite generally has
very high Lu/Hf ratios and can be used in generating mean-
ingful isochrons. In addition, these phases are likely in iso-
topic equilibrium with the magma at the time of their
formation and likely formed at the same time as the other
phases in the rock. Furthermore, apatite can occur in mafic
rocks that generally lack zircon and that may not contain
baddeleyite. The potential of Lu-Hf apatite geochronology
for providing important geochronologic information for
some rocks that are notoriously difficult to date is high,
and this should be a growth area in the future.
Garnet Lu-Hf geochronology
The most successful geochronologic application of the
Lu-Hf system has been garnet geochronology. This tech-
nique shares the same principles as Sm-Nd garnet geo-
chronology in that it utilizes garnet’s generally high
affinity for HREEs which results in elevated Sm/Nd ratios
and (generally higher) Lu/Hf ratios in the garnets as they
grow. The potential of the garnet Lu-Hf technique was
first demonstrated by a number of papers in the late
1990s (Duchene et al., 1997; Scherer et al., 1997; Blichert-
Toft et al., 1999). Following this early proof-of-concept
work, there have been a large number of studies in
a wide variety of metamorphic rocks: garnet-bearing
pelitic schists, para- and orthogneisses, amphibolites,
eclogites, and granulites (Scherer et al., 2000;
383
Anczkiewicz et al., 2004; Cheng et al., 2008; Zirakparvar
et al., 2010; Herwartz et al., 2011; Kelly et al., 2011; Wells
et al., 2012; Smit et al., 2013; Bird et al., 2013). Reliable
garnet Lu-Hf ages have been reported for rocks as old as
Archean (e.g., Smit et al., 2013) and as young as Miocene
(e.g., Zirakparvar et al., 2011) and everything in between.
Although there are many parallels between Lu-Hf and
Sm-Nd garnet geochronology, there are important differ-
ences between these two systems. First, the decay rate of
the parent isotope in the Lu-Hf system, 176Lu, is ~3 times
faster than 147Sm in the Sm-Nd system. This results in
a higher rate of ingrowth in the radiogenic daughter,
176
Hf, compared to 143Nd. Second, owing to the different
geochemical behavior between a REE and a HFSE, there
is generally more fractionation in the Lu/Hf ratios com-
pared to Sm/Nd, which is especially manifest in phases
like garnet and apatite.
Third, Lu is partitioned into garnet most strongly of all
the REEs, while Hf is incorporated into garnet only
weakly (Table 2). This often results in pronounced
Lu-Hf zoning in the garnet with high Lu/Hf ratios in the
core, due to the strong incorporation of Lu during early
garnet growth, and with lower Lu/Hf ratios toward the
rim due to the depletion of Lu in the garnet feeding zones.
Because the very high Lu-Hf in the cores will exert
a strong control on the slope of the regression, it has been
suggested that Lu/Hf ages will be biased toward the early
stages of garnet growth (Lapen et al., 2003; Skora
et al., 2006). In this way, we would expect slightly older
Lu-Hf garnet ages compared with Sm-Nd ages in
a slowly growing garnet (Lapen et al., 2003; Skora
et al., 2006).
Fourth, there are significant differences in the apparent
closure temperature ranges in garnet between the two isoto-
pic systems with garnets closing for Lu-Hf at a higher tem-
perature than for the Sm-Nd system. Debate continues on
closure temperature estimates of the Sm-Nd chronometer
(e.g., Ganguly et al., 1998; Scherer et al., 2000; Van Orman
et al., 2002; Dutch and Hand, 2009; Smit et al., 2013), but
appear to be in the range of ~700–750  C for garnets
1 mm diameter and larger (Van Orman et al., 2002; Smit
et al., 2013) and significantly higher (perhaps 100  C or
more) for Hf. This can lead to differences between the gar-
net Lu-Hf and Sm-Nd ages with the Lu-Hf ages systemati-
cally older (e.g., Scherer et al., 2000). In some granulite
terranes that may have remained buried at mid- to lower-
crustal levels for a prolonged time during and following
garnet growth, these age differences can be large – as much
as 50–70 m.y. (Smit et al., 2013; Vervoort et al., 2013).
Experimental work also indicates significantly slower dif-
fusion for Hf in garnets than the REEs (Ganguly
et al., 2010), which collectively have similar diffusion rates.
One possible consequence of the slower diffusion of Hf
than Lu is that the garnet may effectively close to Hf diffu-
sion before Lu, leading to changing Lu/Hf ratios, but static
Hf isotopic compositions (Ganguly et al., 2010). This could
potentially result in “isochron rotation” and garnet Lu-Hf
ages that are artificially too old.
384 Lu-Hf DATING: THE Lu-Hf ISOTOPE SYSTEM

Lu-Hf Dating: The Lu-Hf Isotope System, Table 2 Melt-solid partition coefficients for Lu and Hf in various minerals

Mineral D(Lu) D(Hf) Melt References

Olivine 0.0015 0.01 Basalt McKenzie and O’Nions, 1991

Clinopyroxene 0.233 0.28 Basalt McKenzie and O’Nions, 1991
Clinopyroxene 0.223 0.623 Basaltic Hauri et al., 1994
Clinopyroxene 0.256 0.433 Alk. basalt Zack and Brumm, 1998
Plagioclase 0.025 0.01 Basalt McKenzie and O’Nions, 1991
Orthopyroxene 0.22 0.06 Thol. basalt Green et al., 2000
Ilmenite 0.084 0.38 Alk. basalt Zack and Brumm, 1998
Rutile 0.0124 4.98 Tonalite Foley et al., 2000
Amphibole 2.1 0.54 Andesite Bacon and Druitt, 1988
Garnet 5.5 0.23 Basalt McKenzie and O’Nions, 1991
Garnet 7.1 0.24 Basalt Johnson, 1998
Garnet 57 0.57 Andesite Hauri et al., 1994
Garnet 23.5 3.3 Dacite Irving and Frey, 1978
Apatite 13.8 0.73 Andesite Bacon and Druitt, 1988
Apatite 21.5 0.878 Dacite Fujimaki, 1986
Zircon 196 958 Andesite Fujimaki, 1986
Zircon 689 971 Dacite Fujimaki, 1986

The partition coefficient, D, is defined as the ratio of the concentration of the element in the solid compared to melt. Values are from GERM
database (http://earthref.org/KDD/) compiled from the original references listed above

Finally, the presence of accessory phases (e.g., zircon,
apatite, monazite, allanite), especially as inclusions in
the garnets, presents a special challenge (e.g., DeWolf
et al., 1996; Prince et al., 2000; Scherer et al., 2000; Thoni,
2002; Anczkiewicz and Thirlwall, 2003), and each of
these minerals affects the Lu-Hf and Sm-Nd system in dif-
ferent ways. For the Lu-Hf system, zircons provide the
biggest challenge. Zircons are extraordinarily enriched in
Hf (~10,000 ppm Hf in zircon) relative to garnet
(1 ppm) and, therefore, can easily overwhelm Hf con-
centration budget if present as inclusions in the garnets,
thus lowering Lu/Hf ratios of the garnet dissolutions. Fur-
thermore, inclusions of preexisting zircons, which are not
likely to be re-equilibrated prior to metamorphic growth,
can significantly lower the measured 176Hf/177Hf of the
garnet sample since these zircons will have a less radio-
genic Hf isotopic composition than the bulk rock. If these
zircons are much older than the bulk rock, this difference
can be significant. The incorporation of zircon Hf not
in isotopic equilibrium with the protolith at the time of
metamorphism in the analyses, either in garnet or as
a dispersed accessory phase in the rock matrix, can result
in erroneous ages (e.g., Scherer et al., 2000). Fortunately,
what makes zircon resistant to isotopic equilibration dur-
ing metamorphism also makes it resistant to dissolution,
and thus zircon inclusions can be largely avoided
with a nonaggressive dissolution protocol (e.g., Lagos
et al., 2007; Cheng et al., 2008). Very high 176Lu/177Hf
ratios (generally  1) in the garnet analyses is an indica-
tion that zircon (very low 176Lu/177Hf ratios, usually
0.002 to 0.001) has not been significantly dissolved. Apa-
tite inclusions (very high 176Lu/177Hf ratios) in the garnet
can also potentially have an effect on an isochron if they
either are not in isotopic equilibrium with the garnet at
the time of metamorphic growth or are formed or closed
to Lu-Hf diffusion at a different time than the garnet.
The effect of apatite inclusions on garnet Lu-Hf geochro-
nology, however, has not yet been examined in the litera-
ture. For the Sm-Nd isotope system, the inclusions that
are most problematic are phases enriched in REEs and
with low Sm/Nd ratios such as monazite and allanite.
The effect of these inclusions is generally to drastically
reduce the spread in the isochron. If present within the gar-
nets and incorporated in the dissolution, the garnets will
have low Sm/Nd ratios – often as low as the whole rock –
thus eliminating the possibility of producing a meaningful
isochron.
Lawsonite Lu-Hf geochronology
A recent addition to the Lu-Hf geochronology toolbox is
the use of the mineral lawsonite to date subduction zone
processes (Mulcahy et al., 2009; Vitale Brovarone and
Herwartz, 2013). Depending on its paragenesis, lawsonite
can have sufficiently elevated Lu/Hf ratios to produce
meaningful isochrons. Lawsonite is an important index
mineral formed during low temperature-high pressure
subduction zone metamoprhism and potentially provides
one of the few tools we have available to date subduction
zone processes. Initial work with this geochronologic tool
has shown that when lawsonite forms on the retrograde
path from the breakdown of garnet, it generally has ele-
vated Lu/Hf ratios and can yield robust Lu-Hf ages. When
the lawsonite forms from the breakdown of zeolites on
a prograde path, however, it may not develop elevated
Lu/Hf ratios and, therefore, may not provide useful
Lu-Hf ages (Mulcahy et al., in review). The full potential
of this geochronologic tool is just now being explored.
 Lu-Hf DATING: THE Lu-Hf ISOTOPE SYSTEM
Hf radiogenic isotope geochemistry
As mentioned above, both Lu and Hf are lithophile and
refractory elements with high condensation temperatures.
Because of these characteristics, it has long been assumed
that the abundance of these elements in the bulk Earth can
be approximated by chondritic meteorites. It has also been
long assumed that the isotopic composition of a refractory
element such as Hf was homogeneous in the solar nebula
and can also be approximated by chondritic meteorites.
This is known as the Chondritic Uniform Reservoir
model, or CHUR. Recent work, primarily based on the
short-lived 146Sm-142Nd isotopic system, has questioned
whether the Earth is strictly chondritic in composition
(see discussion in Sm-Nd section). Regardless of whether
the Earth is strictly chondritic, the CHUR model provides
an important reference value for a plausible bulk Earth
composition. The Lu-Hf and Sm-Nd systems are used in
tandem for planetary evolutionary studies for the impor-
tant reason that these are the only two such isotopic sys-
tems in which all elements are lithophile and refractory.
The bulk silicate earth and CHUR
Because of the analytical challenges in measuring Hf iso-
tope compositions in Hf-poor (~0.10 ppm; McDonough
and Sun, 1995) chondrites accurately and precisely, the
early Lu-Hf CHUR values had considerable uncertainty.
There have been two approaches used to determine
a CHUR value for the Lu-Hf and Sm-Nd isotope systems.
For Sm-Nd, because there is very little fractionation
between Sm and Nd in chondrites and there are no phases
with low Sm/Nd ratios, it has not been possible to deter-
mine a chondritic (Solar System) initial value for this sys-
tem. Instead, the approach has been to take an average of
representative present-day 143Nd/144Nd and 147Sm/144Nd
values of undifferentiated chondrites. This works because
the total Sm-Nd variation in chondrites is modest. Using
this approach, Jacobsen and Wasserburg (1980, 1984)
determined present-day CHUR values of 143Nd/144Nd ¼
0.512638 and 147Sm/144Nd ¼ 0.1966, which corresponded
to the composition of two CI chondrites, Allende and
Murchison.
The situation for the Lu-Hf isotope system was initially
different because of the difficulty in measuring Hf pre-
cisely in the Hf-poor chondrites and the large apparent
spread in 176Lu/177Hf and 176Hf/177Hf in chondrites,
which makes determining a mean chondritic value more
ambiguous. In order to compare this system to the more
established Sm-Nd isotope system, Patchett and
Tatsumoto (1981) based the Hf CHUR values on the same
CI chondrites used to define the Sm-Nd isotope system,
Allende and Murchison. Using this approach, they arrived
at present-day values of 176Hf/177Hf ¼ 0.282860 and
176
Lu/177Hf ¼ 0.0334. The 176Hf/177Hf value was later
adjusted to 0.282830 (Vervoort and Patchett, 1996) to
account for the high 176Hf/177Hf bias on the TIMS in Den-
ver, the instrument on which all the early pioneering Hf
isotope work was performed.
385
With the advent of the MC-ICPMS and its application to
measuring Hf isotopes came the ability to analyze Hf-poor
chondrites more precisely. In order to constrain the Lu-Hf
CHUR values, Blichert-Toft and Albarede (1997) analyzed
25 chondritic meteorites including carbonaceous, ordinary,
and enstatite chondrites. Because the measured values of
these samples had a wide spread in 176Hf/177Hf and
176
Lu/177Hf (~18 %) and did not define an isochron, they
took the weighted average of the 25 chondrites to arrive at
present-day values of 176Hf/177Hf ¼ 0.282772 and
176
Lu/177Hf ¼ 0.0332. The 176Hf/177Hf values were signif-
icantly lower by 0.000058 from the adjusted Patchett and
Tatsumoto values (Vervoort and Patchett, 1996).
In order to address the problem of the spread in Lu/Hf
ratios and the scatter about a reference isochron, Bouvier
et al. (2008) analyzed only unequilibrated chondrites
(lower metamorphic grade, petrologic types 1–3) includ-
ing 12 carbonaceous chondrites, 13 ordinary chondrites,
and 3 enstatite chondrites. The resulting data had a much
smaller spread in Lu/Hf ratios (only 3 %) which allowed
a more precise determination of the present-day CHUR
values: 176Hf/177Hf ¼ 0.282785  11 and 176Lu/177Hf ¼
0.0336  1. These are the values currently in use by the
geochemical and cosmochemical communities. Bouvier
et al. (2008) also measured the Sm-Nd isotopic composi-
tions of the same solutions on which the Lu-Hf values
were determined and arrived at new CHUR values for
the Sm-Nd isotope system of 143Nd/144Nd ¼ 0.512630
and 147Sm/144Nd ¼ 0.1960. Although these are very close
to the values of Jacobsen and Wasserburg (1980, 1984),
the Bouvier et al.’s (2008) values are more appropriate
to use in studies reporting both Hf and Nd data relative
to the chondritic reference because they represent values
determined on the same set of chondrites.
As is true for the Sm-Nd system, Hf isotope data are
often expressed in epsilon values, which are variations in
parts per 10,000 from the CHUR reference. Expressed in
terms of present-day compositions, this expression is:

eHf ð0Þ ¼ 176 Hf =177 Hf ðsample, 0Þ


 176 Hf =177 Hf ðchur, 0Þ Þ= 176 Hf =177 Hf ðchur, 0Þ 104 :
Expressed in terms of compositions at any time in the
past, t, this expression is:

eHf ðtÞ ¼ 176 Hf =177 Hf ðsample, tÞ


 176 Hf =177 Hf ðchur, tÞ Þ= 176 Hf =177 Hf ðchur, tÞ 104 :
Frequently, when values are calculated back in time,
they are expressed as “initial” values, signifying the for-
mation age of that rock or mineral. (It is important to note
that the CHUR value also needs to be calculated back to its
composition at time, t, from the present-day values.)
Lu-Hf tracers of geologic processes
The current paradigm for the evolution of the Earth holds
that the crust was formed from silicate melts extracted in
386 Lu-Hf DATING: THE Lu-Hf ISOTOPE SYSTEM

some way from the mantle. The crustal rocks resulting a 0.2835
from this process are enriched in incompatible elements Hf isotope evolution
0.2830 De
(hence the term enriched crust), and the mantle left ple
from this extraction of crust is depleted in incompatible ted
0.2825 Ma
elements (hence the term depleted mantle). As ntl
CH ee
0.2820 UR vo
a consequence of this differentiation occurring throughout

Hf/ 177Hf
(Bu lut
lk S ion
the history of the Earth, including at its very beginning, the 0.2815 ilic
ate
Earth’s different reservoirs have evolved following differ- Ea

176
Crusta rth
ent paths. The crustal reservoirs are enriched in incompat- 0.2810 l evolu ?)
tion
ible elements but have lower Lu/Hf and Sm/Nd ratios than 0.2805
bulk Earth, and so their 176Hf/177Hf and 143Nd/144Nd Zircon evolution
evolve more slowly. Conversely, the depleted mantle com- 0.2800
ponents, while depleted in incompatible elements, have
higher Lu/Hf ratios and Sm/Nd ratios, and so their 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
176
Hf/177Hf and 143Nd/144Nd compositions evolve more Age (Ga)
quickly than bulk Earth (Figure 2a). In terms of epsilon
values relative to bulk Earth (i.e., CHUR), depleted mantle b 20
Depleted
Mantle evo Hf isotope evolution
components evolve toward positive epsilon values, and lution
enriched crustal values evolve toward negative epsilon 0
CHUR (Bulk Silicate Earth?)
values (Figure 2b). This is a fundamental feature in both
the Hf and Nd isotopic records of Earth’s evolution and −20 n
has been used to great utility in a wide range of crust/man- lutio
al evo
rust
εHf
tle petrogenetic/geochemical studies. C
−40
An important component in all radiogenic isotopes is ion
o lut
time; time is an inherent aspect of Hf isotope evolution n ev
−60 co
and therefore Hf isotope compositions can be used to qual- Zir
itatively assess the time of separation from different refer-
ence reservoirs. These time estimates are called “model −80
ages” and, although not ages in the strict sense, are still
a useful way to represent and visualize data with 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
a temporal framework. Age (Ga)
The concept of a model age in the Lu-Hf isotope system Lu-Hf Dating: The Lu-Hf Isotope System, Figure 2 Schematic
follows that of Sm-Nd model ages (DePaolo, 1981) and is diagram showing the hypothetical Hf isotope evolution of
illustrated in Figure 3a. The 176Hf/177Hf isotopic composi- different terrestrial reservoirs following a major differentiation
tion of a sample is measured today, and the 176Lu/177Hf event at 3.8 Ga: Chondritic Uniform Reservoir (CHUR) broadly
ratio of that sample is used to calculate its isotopic compo- taken as representative of bulk silicate Earth (BSE), depleted
mantle evolution from depletion at 3.8 Ga to its composition
sition back in time until it intersects with a reservoir of today, evolution of an enriched crustal reservoir
interest such as the depleted mantle. The time, (176Lu/177Hf ¼ 0.015), and evolution of a typical zircon formed at
corresponding to this intersection, is a Hf model age – in 3.8 Ga (176Lu/177Hf ¼ 0.001). (a) Evolution of 176Hf/177Hf.
this case, the Hf depleted mantle model age HfT(DM). (b) The same diagram in epsilon Hf notation.
This model age has also been termed a crust-formation
age, HfT(CR), signifying the average time since separation
from the depleted mantle or time of residence in a crustal
reservoir (DePaolo, 1981). evolved crustal sources, or some mixture of the two. This
Depleted mantle model ages have been used to identify is identical to the approach used for Sm-Nd model ages in
the sources of magmatic rocks; if the Hf TDM values are sediments with the caveat that Hf model ages are often
older than the crystallization ages, then the interpretation more variable than Nd model ages because of the presence
is that these rocks were derived from older sources or or absence of Hf-rich (and unradiogenic) zircons in the
sources that had a crustal prehistory – sources that had sediments (e.g., Vervoort et al., 1999; Prytulak
separated from the mantle some time in the past. In this et al., 2006).
way it is simply another means of conveying the Hf isoto- A recent adaptation of Hf model ages – and one that has
pic compositions in the context of time – in much the same seen an explosion of use in recent years – is the use of Hf
way as an epsilon Hf isotope diagram places the Hf iso- model ages of zircons. The way this works is shown in
tope compositions in a time context. Model ages are also Figure 3b. The Hf isotopic composition is measured in
applied to sediments to identify the provenance of their the zircon and the zircon’s Lu/Hf ratio is used to calculate
sources: more juvenile mantle-derived sources, older back to its crystallization age. At that point, the growth
 Lu-Hf DATING: THE Lu-Hf ISOTOPE SYSTEM 387

a 20
Deplete
trajectory changes, assuming a source 176Lu/177Hf ratio.
d Mantle wr Hf model ages As was the case with the other model ages, the intersection
Evolutio
n
10 with the depleted mantle reference line yields the zircon
Hf depleted mantle model age. This technique has recently
CHUR (Bulk Silicate Earth?)
0 been applied to detrital zircon studies because of the facil-
= 0.0
25 ity of acquiring U-Pb crystallization ages and Lu-Hf iso-
177 Hf ratio tope compositions on the same zircon grain by laser
−10
εHf

176 Lu/
ablation-inductively coupled plasma mass spectrometry
−20 .01
(LA-ICPMS).
=0 There are several aspects of this approach, however,
io
rat that result in considerable uncertainty in the zircon Hf
−30 7 Hf
17
176 L
u/
TCHUR (.025) TCHUR (.01) TDM
model “ages.” Because the context of the rock has been
−40
lost, the Lu/Hf ratio of the zircon’s precursor rock has to
0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 be estimated, and this value can vary widely from
Age (Ga) a mafic precursor (high Lu/Hf) to a felsic precursor (low
Lu/Hf). The further the composition of a zircon is from
b 20
Depleted the depleted mantle (or other) reference, the greater the
Mantle Evo zircon Hf model ages
10
lution uncertainty in the model age. Another limitation of this
CHUR (Bulk Silicate Earth?) technique is that the model ages rely on the intersection
0 of the zircon + precursor rock trajectory with a depleted
−10
mantle reference curve. The depleted mantle, however, is
not a discrete entity with a discrete isotopic composition.
1
−20 .00 It is, in fact, typically defined by its most depleted, rather
εHf

0
i o=
−30 rat
7 Hf
17
/
−40 6 Lu
17
−50
−60
0 0.5 1.0 1.5 2.0
Age (Ga)
Lu-Hf Dating: The Lu-Hf Isotope System, Figure 3 (a)
Schematic diagram showing the calculation of Hf model ages for
two hypothetical whole-rock (wr) samples that formed from
a depleted mantle reservoir at ~2.7 Ga: a mafic sample with
a high Lu-Hf ratio (176Lu/177Hf ¼ 0.025) and a negative present-
day epsilon Hf value of approx. 10 and a felsic sample with
a low Lu-Hf ratio (176Lu/177Hf ¼ 0.010) and a highly negative
present-day epsilon Hf value of ~ 38. Also shown in this
diagram are the depleted mantle model ages (TDM) and the
CHUR model ages (TCHUR). Note the different trajectories for
the two samples and the differences in TCHUR. (b) Schematic
diagram showing the calculation of zircon Hf model ages. In this
example the zircon has a present-day epsilon Hf value of ~ 55
and an initial epsilon Hf value of ~ 5 calculated at
a crystallization age of ~2.2 Ga. Calculation of depleted mantle
model ages (TDM) requires an assumption of the 176Lu/177Hf of
the source rock. If the source rock is more evolved
(176Lu/177Hf ¼ 0.01 in this case), the TDM is ~2.85 Ga. If the
source rock is more mafic (176Lu/177Hf ¼ 0.025 in this case), the
TDM is much older, ~3.45 Ga. As can be seen from this diagram,
the uncertainty in the TDM increases dramatically the further
the initial Hf value is away from the depleted mantle
reference line. Further, unrecognized Pb loss in the zircon,
which would yield anomalously young crystallization ages,
exacerbates the difference between crystallization and
depleted mantle ages. See text for further discussion of
zircon Hf model ages.
than average, compositions (Vervoort and Blichert-Toft,
1999). Finally, it is not possible to determine the true ages
176
Lu/ 177Hf 176
Lu/ 177Hf of some detrital zircons precisely, due to the potential for
= 0.01 = 0.025
TDM (F)
ancient Pb loss and other factors. If ancient Pb loss exists
TDM (M)
in these grains, this would greatly exacerbate the differ-
ences between the crystallization and the Hf model ages
2.5 3.0 3.5 with the former being too young and the latter being too
old. In total, all of these factors contribute to very large
uncertainties in Hf model “ages.” It is tempting to treat
these model ages, particularly for detrital zircons, as
values with quantifiable age information rather than
model-dependent estimates. It is important to recognize,
however, that these model ages are not ages in any true
sense of the word, but rather “models” with large uncer-
tainties. While they provide some useful constraints on
the timing and nature of crust forming events, it is imper-
ative to understand the limitations of such “age”
information.
Conclusions
The Lu-Hf isotope system is an important chronometer in
geochemistry and cosmochemistry. Its utility as
a chronometer in whole-rock samples, however, is limited
by the relatively small degrees of Lu-Hf fractionation dur-
ing magmatic processes. In rocks with high Lu-Hf phases,
the Lu-Hf isotope system can be a powerful chronometer;
this is especially true for garnet-bearing metamorphic
rocks and magmatic rocks with apatite as a crystallizing
phase.
There now appears to be widespread agreement on
a value for the 176Lu decay constant of 1.867  0.008
1011 Ga (Scherer et al., 2001; Söderlund et al., 2004).
388 Lu-Hf DATING: THE Lu-Hf ISOTOPE SYSTEM
This value has important implications not only for Lu-Hf
geochronology but also for early Earth differentiation
and evolution based on calculated initial Hf isotope
values. Similarly, there is currently widespread acceptance
of the Bouvier et al. (2008) CHUR values of
176
Hf/177Hf ¼ 0.282785 and 176Lu/177Hf ¼ 0.0336.
Corresponding to these are the Sm-Nd CHUR values of
143
Nd/144Nd ¼ 0.512630 and 147Sm/144Nd ¼ 0.1960.
Notwithstanding discussions of whether or not the Earth
is strictly chondritic, these values provide a self-consistent
baseline on which to reference the isotopic evolution of
the Earth.
Tracer isotopic applications provide equal utility of the
Lu-Hf isotopic system and, in conjunction with indepen-
dent age information, can be used to determine the Hf iso-
topic composition of rocks and minerals back into time.
This information is useful in constraining both the evolu-
tion and source regions of igneous systems as well as the
provenance of sedimentary rocks. By far the most heavily
utilized mineral for tracer isotope studies is zircon, which
provides a unique opportunity for determining both age
and Hf isotope compositions in individual crystals (and
even separate growth zones within a single crystals), such
as in the case of inheritance or overgrowths. For this rea-
son, the integrated U-Pb and Hf isotope information from
zircon has seen a rapid increase in use in recent years and
will likely continue. Many of these studies often report zir-
con Hf model ages. While these provide useful qualitative,
model-based data, these are not in any way truly ages and
should not be interpreted as containing quantitative chro-
nologic information.
In summary, with the full implementation of techniques
and applications of the Lu-Hf isotope system following
the advent of MC-ICPMS technology, Lu-Hf geochronol-
ogy and tracer isotope work has finally taken its place at
the table with the other main players in geochronology
and radiogenic isotope geochemistry.
Bibliography
Albarède, F., Scherer, E. E., Blichert-Toft, J., Rosing, M.,
Simionovici, A., and Bizzarro, M., 2006. Gamma-ray irradiation
in the early solar system and the conundrum of the 176Lu decay
constant. Geochimica et Cosmochimica Acta, 70, 1261–1270.
Amelin, Y., and Davis, W. J., 2005. Geochemical test for branching
decay of 176Lu. Geochimica et Cosmochimica Acta, 69, 465–473.
Amelin, Y., Lee, D.-C., and Halliday, A. N., 2000. Early-middle
Archaean crustal evolution deduced from Lu-Hf and U-Pb isoto-
pic studies of single zircon grains. Geochimica et Cosmochimica
Acta, 64, 4205–4225.
Anczkiewicz, R., and Thirlwall, M. F., 2003. Improving precision of
Sm-Nd garnet dating by H2SO4 leaching – a simple solution to
the phosphate inclusion problem. In Vance, D., Mueller, W.,
and Villa, I. M. (eds.), Geochronology: Linking the Isotopic
Record with Petrology and Textures. London: Geological Soci-
ety, pp. 83–91.
Anczkiewicz, R., Platt, J., Thirlwall, M. F., and Wakabayashi, J.,
2004. Franciscan subduction off to a slow start: evidence from
high-precision Lu-Hf garnet ages on high grade blocks. Earth
and Planetary Science Letters, 225, 147–161.
Bacon, C. R., and Druitt, T. H., 1988. Compositional evolution of
the zoned calcalkaline magma chamber of Mount Mazama, Cra-
ter Lake, Oregon. Contributions to Mineralogy and Petrology,
98, 224–256.
Barfod, G. H., Otero, O., and Albarède, F., 2003. Phosphate Lu–Hf
geochronology. Chemical Geology, 200, 241–253.
Beard, B. L., and Johnson, C. M., 1993. Hf isotope composition of
late Cenozoic basaltic rocks from northwestern Colorado, USA:
new constraints on mantle enrichment processes. Earth and
Planetary Science Letters, 119, 495–509.
Begemann, F., Ludwig, K. R., Lugmair, G. W., Min, K., Nyquist,
L. E., et al., 2001. Call for an improved set of decay constants
for geochronological use. Geochimia et Cosmochimica Acta,
65, 111–121.
Bird, A. F., Thirlwall, M. F., Strachan, R. A., and Manning, C. J.,
2013. Lu–Hf and Sm–Nd dating of metamorphic garnet: evi-
dence for multiple accretion events during the Caledonian orog-
eny in Scotland. Journal of the Geological Society of London,
170, 301–317.
Bizzarro, M., Baker, J. A., Haack, H., Ulfbeck, D., and Rosing, M.,
2003. Early history of Earth’s crust-mantle system inferred from
hafnium isotopes in chondrites. Nature, 421, 931–933.
Blichert-Toft, J., and Albarède, F., 1997. The Lu-Hf isotope geo-
chemistry of chondrites and the evolution of the mantle-crust
system. Earth and Planetary Science Letters, 148, 243–258.
Blichert-Toft, J., Chauvel, C., and Albarède, F., 1997. Separation of
Hf and Lu for high precision isotope analysis of rock samples by
magnetic sector-multiple collector ICP-MS. Contributions to
Mineralogy and Petrology, 127, 248–260.
Blichert-Toft, J., Albarede, F., and Kornprobst, J., 1999. Lu-Hf isotope
systematics of garnet pyroxenites from Beni Bousera, Morocco:
implications for basalt origin. Science, 283, 1303–1306.
Blichert-Toft, J., Boyet, M., Telouk, P., and Albarede, F., 2002.
147
Sm-143Nd and 176Lu-176Hf in eucrites and the differentiation
of the HED parent body. Earth and Planetary Science Letters,
204, 167–181.
Bouvier, A., Vervoort, J. D., and Patchett, P. J., 2008. The Lu–Hf
and Sm–Nd isotopic composition of CHUR: constraints from
unequilibrated chondrites and implications for the bulk composi-
tion of terrestrial planets. Earth and Planetary Science Letters,
273, 48–57.
Chauvel, C., Lewin, E., Carpentier, M., Arndt, N., and Marini, J.-C.,
2008. Role of recycled oceanic basalt and sediment in generating
the Hf–Nd mantle array. Nature Geoscience, 1, 64–67.
Cheng, H., King, R., Nakamura, E., Vervoort, J., and Zhou, Z.,
2008. Coupled Lu-Hf and Sm-Nd geochronology constrains pro-
grade garnet growth in ultrahigh-pressure eclogites from the
Dabie orogen. Journal of Metamorphic Geology, 26, 741–758.
Chu, N. C., Taynor, R. N., Chavagnac, V., Nesbitt, R. W., Boella,
R. M., Milton, J. A., et al., 2002. Hf isotope ratio analysis using
multi-collector inductively coupled plasma mass spectrometry:
an evaluation of isobaric interference corrections. Journal of
Analytical Atomic Spectrometry, 17, 1567–1574.
DePaolo, D. J., 1981. Neodymium isotopes in the Colorado Front
Range and crust mantle evolution in the Proterozoic. Nature,
291, 193–196.
DeWolf, C. P., Zeissler, C. J., Halliday, A. N., Mezger, K., and
Essene, E. J., 1996. The role of inclusions in U-Pb and Sm-Nd
garnet geochronology: stepwise dissolution experiments and
trace uranium mapping by fission track analysis. Geochimica
et Cosmochimica Acta, 60, 121–134.
Dixon, D., McNair, A., and Curran, S. C., 1954. The natural radio-
activity of lutetium. Philosophical Magazine, 45, 683–694.
Duchene, S., Blichert-Toft, J., Luais, B., Télouk, P., Lardeaux,
J.-M., and Albarède, F., 1997. The Lu-Hf dating of garnets and
the ages of the Alpine high-pressure metamorphism. Nature,
387, 586–589.
 Lu-Hf DATING: THE Lu-Hf ISOTOPE SYSTEM
Dutch, R., and Hand, M., 2009. Retention of Sm–Nd isotopic ages
in garnets subjected to high-grade thermal reworking: implica-
tions for diffusion rates of major and rare earth elements and
the Sm–Nd closure temperature in garnet. Contributions to
Mineralogy and Petrology, 159, 93–112.
Fisher, C. M., Vervoort, J. D., and DuFrane, S. A., 2014. Accurate
Hf isotope determinations of complex zircons using the “laser
ablation split stream” method. Geochemistry, Geophysics,
Geosystems, 15, 121–139.
Foley, S. F., Barth, M. G., and Jenner, G. A., 2000. Rutile/melt parti-
tion coefficients for trace elements and an assessment of the influ-
ence of rutile on the trace element characteristics of subduction
zone magmas. Geochimica et Cosmochimica Acta, 64, 933–938.
Fujimaki, H., 1986. Partition coefficinets of Hf, Zr, and REE
between zircon, apatite, and liquid. Contributions to Mineralogy
and Petrology, 94, 42–45.
Ganguly, J., Tirone, M., and Hervig, R. L., 1998. Diffusion kinetics
of samarium and neodymium in garnet, and a method for deter-
mining cooling rates of rocks. Science, 281, 805–807.
Ganguly, J., Bloch, E., and Motoo, I., 2010. Experimental diffusion
kinetics of geochronological systems and interpretations of min-
eral ages in terrestrial rocks and meteorites. Geological Society
of America Abstract Program, 42, 340.
Gerdes, A., and Zeh, A., 2009. Zircon formation versus zircon alter-
ation – new insights from combined U–Pb and Lu–Hf in-situ
LA-ICP-MS analyses, and consequences for the interpretation
of Archean zircon from the Central Zone of the Limpopo Belt.
Chemical Geology, 261, 230–243.
Green, T., Blundy, J., Adam, J., and Yaxley, G., 2000. SIMS deter-
mination of trace element partition coefficients between garnet,
clinopyroxene and hydrous basaltic liquids at 2–7.5 Gpa and
1080–1200C. Lithos, 53, 165–187.
Hauri, E. H., Wagner, T. P., and Grove, T. L., 1994. Experimental
and natural partitioning of Th, U, Pb and other trace elements
between garnet, clinopyroxene and basaltic melts. Chemical
Geology, 117, 149–166.
Herwartz, D., Nagel, T. J., Münker, C., Scherer, E. E., and
Froitzheim, N., 2011. Tracing two orogenic cycles in one
eclogite sample by Lu–Hf garnet chronometry. Nature Geosci-
ence, 4, 178–183.
Irving, A. J., and Frey, F. A., 1978. Distribution of trace elements
between garnet megacrysts and host volcanic liquids of
kimberlitic to rhyolitic composition. Geochimica et
Cosmochimica Acta, 42, 771–787.
Jacobsen, S. B., and Wasserburg, G. J., 1980. Sm-Nd isotopic evo-
lution of chondrites. Earth and Planetary Science Letters,
50, 139–155.
Jacobsen, S. B., and Wasserburg, G. L., 1984. Sm-Nd isotopic
evolution of chondrites and achondrites II. Earth and Planetary
Science Letters, 67, 137–150.
Johnson, K. T. M., 1998. Experimental determination of partition
coefficients for rare earth and high-field-strength elements
between clinopyroxene, garnet, and basaltic melt at high pres-
sures. Contributions to Mineralogy and Petrology, 133, 60–68.
Johnson, C. M., and Beard, B. L., 1993. Evidence from hafnium iso-
topes for ancient sub-oceanic mantle beneath the Rio Grande rift.
Nature, 362, 441–444.
Kelly, E. D., Carlson, W. D., and Connelly, J. N., 2011. Implications
of garnet resorption for the Lu–Hf garnet geochronometer: an
example from the contact aureole of the Makhavinekh Lake Plu-
ton, Labrador. Journal of Metamorphic Geology, 29, 901–916.
Kemp, A. I. S., Wilde, S. A., Hawkesworth, C. J., Coates, C. D.,
Nemchin, A., et al., 2010. Hadean crustal evolution revisited:
new constraints from Pb–Hf isotope systematics of the Jack Hills
zircons. Earth and Planetary Science Letters, 296, 45–56.
Kohn, M. J., 2009. Models of garnet differential geochronology.
Geochimica et Cosmochimica Acta, 73, 170–182.
389
Lagos, M., Scherer, E. E., Tomaschek, F., Münker, C., Keiter, M.,
et al., 2007. High precision Lu–Hf geochronology of Eocene
eclogite-facies rocks from Syros, Cyclades, Greece. Chemical
Geology, 243, 16–35.
Lapen, T. J., Johnson, C. M., Baumgartner, L. P., Mahlen, N. J.,
Beard, B. L., and Amato, J. M., 2003. Burial rates during pro-
grade metamorphism of an ultra-high-pressure terrane: an exam-
ple from Lago di Cignana, western Alps, Italy. Earth and
Planetary Science Letters, 2003, 57–72.
Lide, D. R. (ed.), 2003. Section 10, Atomic, Molecular, and Optical
Physics; Ionization Potentials of Atoms and Atomic Ions. Boca
Raton: CRC Press.
McDonough, W. F., and Sun, S.-S., 1995. The composition of the
Earth. Chemical Geology, 120, 223–253.
McKenzie, D., and O’Nions, R. K., 1991. Partial melt distributions
from inversion of rare earth element concentrations. Journal of
Petrology, 32, 1021–1091.
Mittlefehldt, D. W., McCoy, T. J., Goodrich, C. A., and Kracher, A.,
1998. Non-chondritic meteorites from asteroidal bodies. In
Papike, J. J. (ed.), Planetary Materials. Washington, DC: Miner-
alogical Society of America, pp. 1–195.
Mulcahy, S. R., Vervoort, J. D., Renne, P. R., (in review). Dating
subduction zone metamorphism with combined garnet and
lawsonite Lu-Hf geochronology. Journal of Metamorphic
Geology.
Mulcahy, S., King, R. L., and Vervoort, J., 2009. Lawsonite Lu-Hf
geochronology: a new geochronometer for subduction zone pro-
cesses. Geology, 37, 987–990.
Patchett, P. J., 1983a. Hafnium isotope results from mid-ocean
ridges and Kerguelen. Lithos, 16, 47–51.
Patchett, P. J., 1983b. Importance of the Lu-Hf isotope system in
studies of planetary chronology and chemical evolution.
Geochimica et Cosmochimica Acta, 47, 81–91.
Patchett, J., and Chauvel, C., 1984. The mean life of continents is
currently not constrained by Nd and Hf isotopes. Geophysical
Research Letters, 11, 151–153.
Patchett, P. J., and Tastumoto, M., 1980. Hafnium isotope variations
in oceanic basalts. Geophysical Research Letters, 7, 1077–1080.
Patchett, P. J., and Tatsumoto, M., 1980a. Lu-Hf total-rock isochron
for the eucrite meteorites. Nature, 288, 571–574.
Patchett, P. J., and Tatsumoto, M., 1980b. A routine high-precision
method for Lu-Hf isotope geochemistry and chronology. Contri-
butions to Mineralogy and Petrology, 75, 263–267.
Patchett, P. J., and Tatsumoto, M., 1981. Lu/Hf in chondrites and
definition of a chondritic hafnium growth curve. Lunar and
Planetary Science, 12, 822–824.
Prince, C. I., Kosler, J., Vance, D., and Günther, D., 2000. Compar-
ison of laser ablation ICP-MS and isotope dilution REE analy-
sis – implications for Sm-Nd garnet geochronology. Chemical
Geology, 168, 255–274.
Prytulak, J., Vervoort, J. D., Plank, T., and Yu, C., 2006. Astoria
fan sediments, DSDP Site 174, Cascadia Basin: Hf-Nd-Pb
constraints on provenance and outburst flooding. Chemical
Geology, 233, 276–292.
Rudnick, R. L., and Gao, S., 2003. Composition of the continental
crust. In Rudnick, R. L., Heinrich D. H and Karl K. T. (eds.),
Treatise on Geochemistry. Oxford: Elsevier-Pergamon, 3, 1–64.
Salters, V. J. M., and Hart, S. R., 1991. The mantle sources of ocean
ridges, islands and arcs: the Hf-isotope connection. Earth and
Planetary Science Letters, 104, 364–380.
Scherer, E. E., Cameron, K. L., Johnson, C. M., Beard, B. L.,
Barovich, K. M., and Collerson, K. D., 1997. Lu-Hf geochronol-
ogy applied to dating Cenozoic events affecting lower crustal
xenoliths from Kilbourne Hole, New Mexico. Chemical Geol-
ogy, 142, 63–78.
Scherer, E. E., Cameron, K. L., and Blichert-Toft, J., 2000. Lu–Hf
garnet geochronology: closure temperature relative to the
390 LUMINESCENCE DATING
Sm–Nd system and the effects of trace mineral inclusions.
Geochimica et Cosmochimica Acta, 64, 3413–3432.
Scherer, E., Münker, C., and Mezger, K., 2001. Calibration of the
lutetium–hafnium clock. Science, 293, 683–686.
Scherer, E. E., Mezger, K., and Münker, C., 2003. The 176Lu decay
constant discrepancy: terrestrial samples vs meteorites. Meteorit-
ics and Planetary Science, 38, A136.
Scherer, E. E., Whitehouse, M. J., and Münker, C., 2007. Zircon as
a monitor of crustal growth. Elements, 3, 19–24.
Segal, I., Halicz, L., and Platzner, I. T., 2003. Accurate isotope
ratio measurements of ytterbium by multiple collectionin-
ductively coupled plasma mass spectrometry applying erbium
and hafnium in an improved double external normalization pro-
cedure. Journal of Analytical Atomic Spectrometry, 18,
1217–1223.
Sguigna, A. P., Larabee, A. J., and Waddington, J. C., 1982. The
half-life of 176Lu by gamma-gamma coincidence measurement.
Canadian Journal of Physics, 60, 361–364.
Skora, S., Baumgartner, L. P., Mahlen, N. J., Johnson, C. M., Pilet, S.,
and Hellebrand, E., 2006. Diffusion-limited REE uptake by
eclogite garnets and its consequences for Lu–Hf and Sm–Nd
geochronology. Contributions to Mineralogy and Petrology,
152, 703–720.
Smit, M. A., Scherer, E. E., and Mezger, K., 2013. Lu-Hf and
Sm-Nd garnet geochronology: chronometric closure and impli-
cations for dating petrological processes. Earth and Planetary
Science Letters, 2013, 222–233.
Söderlund, U., Patchett, P. J., Vervoort, J. D., and Isachsen, C. E.,
2004. The 176Lu decay constant determined by Lu–Hf and
U–Pb isotope systematics of Precambrian mafic intrusions.
Earth and Planetary Science Letters, 219, 311–324.
Thoni, M., 2002. Sm-Nd isotope systematics in garnet from differ-
ent lithologies (Eastern Alps): age results, and an evaluation of
potential problems for garnet Sm-Nd chronometry. Chemical
Geology, 185, 255–281.
Thrane, K., Connelly, J. N., Bizzarro, M., Meyer, B. S., and The,
L.-S., 2010. Origin of excess 176Hf in meteorites. The Astrophys-
ical Journal, 717, 861–867.
Van Orman, J. A., Grove, T. L., Shimizu, N., and Layne, G. D.,
2002. Rare earth element diffusion in a natural pyrope single
crystal at 2.8 GPa. Contributions to Mineralogy and Petrology,
142, 416–424.
Vervoort, J. D., and Blichert-Toft, J., 1999. Evolution of the
depleted mantle: Hf isotope evidence from juvenile rocks
through time. Geochimica et Cosmochimica Acta, 63, 533–556.
Vervoort, J. D., and Patchett, P. J., 1996. Behavior of hafnium and
neodymium isotopes in the crust: constraints from Precambrian
crustally derived granites. Geochimica et Cosmochimica Acta,
60, 3717–3733.
Vervoort, J. D., Blichert-Toft, J., Patchett, P. J., and Albarède, F.,
1999. Relationships between Lu-Hf and Sm-Nd isotopic systems
in the global sedimentary system. Earth and Planetary Science
Letters, 168, 79–99.
Vervoort, J. D., Patchett, P. J., Söderlund, U., and Baker, M., 2004.
Isotopic composition of Yb and the determination of Lu concen-
trations and Lu/Hf ratios by isotope dilution using MC-ICPMS.
Geochemistry, Geophysics, Geosystems, 5, 1–15.
Vervoort, J., Ramsey, M. J., Aleinikoff, J. N., and Southworth, S.,
2013. Contrasting Lu-Hf and Sm-Nd ages in garnetiferous
gneisses of the Blue Ridge province, Virginia: Implications for
the timing and duration of the Grenville orogeny in the Central
Applaachians. Geol. Soc. Am. Abstr. w Prog., 45, 322–325.
Vitale Brovarone, A., and Herwartz, D., 2013. Timing of HP meta-
morphism in the Schistes Lustrés of Alpine Corsica: new Lu–Hf
garnet and lawsonite ages. Lithos, 172–173, 175–191.
Walder, A. J., and Freedman, P. A., 1992. Isotopic ratio measure-
ment using a double focusing magnetic sector mass analyser
with an inductively coupled plasma as an ion source. Journal
of Analytical Atomic Spectrometry, 7, 571–575.
Wells, M. L., Hoisch, T. D., Cruz-Uribe, A. M., and Vervoort, J. D.,
2012. Geodynamics of synconvergent extension and tectonic
mode switching: constraints from the Sevier-Laramide orogen.
Tectonics, 31, 1–20.
White, W. M., Albarede, F., and Telouk, P., 2000. High-precision
analysis of Pb isotope ratios by multi-collector ICP-MS. Chemi-
cal Geology, 167, 257–270.
Woodhead, J. D., Hergt, J. M., Shelley, M., Eggins, S., and Kemp,
R., 2004. Zircon Hf-isotope analysis with an excimer laser, depth
profiling, ablation of complex geometries, and concomitant age
estimation. Chemical Geology, 209, 121–135.
Xie, L. W., Zhang, Y. B., Zhang, H. H., Sun, J. F., and Wu, F. Y.,
2008. In situ simultaneous determination of trace elements,
U\Pb and Lu\Hf isotopes in zircon and baddeleyite. Chinese Sci-
ence Bulletin, 53, 1565–1573.
Yuan, H.-L., Gao, S., Dai, M.-N., Zong, C.-L., Günther, D., et al.,
2008. Simultaneous determinations of U–Pb age, Hf isotopes
and trace element compositions of zircon by excimer laser-
ablation quadrupole and multiple-collector ICP-MS. Chemical
Geology, 247, 100–118.
Zack, T., and Brumm, R., 1998. Ilmenite/liquid partition coefficients
of 26 trace elements determined through ilmenite/clinopyroxene
partitioning in garnet pyroxene. In Presented at 7th International
Kimberlite Conference, Cape Town, South Africa.
Zirakparvar, N., Vervoort, J., Lewis, R., and McClelland, W. C.,
2010. Insights into the thermal evolution of the Belt-Purcell
basin; evidence from Lu-Hf garnet geochronology. Canadian
Journal of Earth Sciences, 47, 161–179.
Zirakparvar, N. A., Baldwin, S. L., and Vervoort, J. D., 2011. Lu–Hf
garnet geochronology applied to plate boundary zones: insights
from the (U)HP terrane exhumed within the Woodlark Rift.
Earth and Planetary Science Letters, 309, 56–66.
LUMINESCENCE DATING
Geoffrey A. T. Duller
Department of Geography and Earth Sciences,
Aberystwyth University, Aberystwyth, Wales, UK
Definition
Luminescence: emission of light from a semiconductor or
insulator in response to some form of stimulation such as
heat or light. In the context of using luminescence for dat-
ing, the luminescence signal arises from the release of
energy stored from prior exposure to ionizing radiation
such as alpha, beta, or gamma radiation. Thus, lumines-
cence provides a mechanism for measuring the amount
of ionizing radiation a sample has been exposed to.
Thermoluminescence Emission of luminescence
(TL) in response to heating of the
sample.
Optically Stimulated Emission of luminescence
Luminescence (OSL) in response to exposing
the sample to light. In the
laboratory this light is normally
restricted to a narrow range of
wavelengths.
 LUMINESCENCE DATING
Introduction
Radioactivity is ubiquitous in the natural environment.
Luminescence dating exploits the presence of radioactive
isotopes of elements such as uranium (U), thorium (Th),
and potassium (K). Naturally occurring minerals such as
quartz and feldspars act as dosimeters, recording the
amount of radiation to which they have been exposed.
A common property of some naturally occurring min-
erals is that when they are exposed to emissions resulting
from radioactive decay (e.g., alpha, beta, and gamma radi-
ations), they are able to store within their crystal structure
a small proportion of the energy delivered by the radiation.
This stored energy increases as exposure to radioactive
decay continues through time. At some later date this
energy may be released, and in some minerals this energy
is released in the form of light. This light is termed
luminescence.
What makes this a useful phenomenon for dating, and
what event is being dated? The answer lies in the fact that
this energy stored in minerals can be reset by two pro-
cesses. The first is by heating the sample to high tempera-
tures (typically above 300  C, as would occur during
firing of pottery, or in a hearth). The second process is
exposure of the minerals to daylight, as may occur during
erosion, transport, and deposition of sediments. Either of
these processes will release preexisting energy stored
within a mineral and therefore set the “clock” to zero.
Thus, in luminescence dating, the event being dated is this
resetting, either by heat or by exposure to light.
Major applications utilizing the resetting of the signal
by heat include dating when a piece of pottery was pro-
duced or when stones in a hearth were last heated. The
application to burnt stones includes both stones that were
used to form a hearth, but also a large number of important
studies have dated flint tools that have inadvertently been
burnt. Luminescence dating is now most commonly used
to date the last exposure to sunlight of sediments; this is
assumed to correspond with the time of deposition of the
sediment. The method is ideally suited to aeolian sedi-
ments where there is ample opportunity for sediments to
be exposed to daylight such as coastal and desert dunes
(see “Luminescence, Coastal Sediments” and “Lumines-
cence, Desert Dunes”) and windblown dust either depos-
ited on land to form loess (see “Luminescence Dating,
Loess”) or in the oceans (see “Luminescence Dating,
Deep-Sea Marine and Lacustrine”). Provided that due care
is taken with measurements (as detailed below), the
method can also be applied to fluvial sediments (see
“Luminescence, Fluvial Sediments”), and in recent years
the method has been extended to look at glacially derived
materials (see “Luminescence, Glacial Sediments”),
though great care is needed in such applications to assess
whether the sedimentary grains were exposed to sufficient
daylight at deposition to reset the luminescence signal.
Measurements of the brightness of the luminescence
signal can be used to calculate the total amount of radia-
tion that the sample was exposed to during the period of
burial. If this is divided by the amount of radiation that
391
the sample receives from its surroundings each year, then
this will give the duration of time that the sample has been
receiving energy:
Total energy accumulated during burial
Age ¼
Energy delivered each year from radioactive decay
ð1Þ
The S.I. (Système International) unit of absorbed radiation
is the Gray, abbreviated to Gy. It is a measure of the
amount of energy absorbed by a sample, or its dose, and
has the units Joules per kilogram. Luminescence measure-
ments are used to calculate the total absorbed dose as mea-
sured in the laboratory. The name given to the quantity is
the equivalent dose (DE). The amount of energy absorbed
per year from radiation in the environment surrounding
the measured material (known as the dose rate) can either
be derived by directly measuring the amount of radioactiv-
ity or by chemically analyzing the surrounding material
and calculating the concentration of radioisotopes in it;
this has the units of Gray per year. Thus, the age equation
for luminescence can formally be expressed as
Equivalent dose ðGyÞ
Age ðyearsÞ ¼ ð2Þ
Dose rate ðGy per yearÞ
A range of different minerals emit luminescence, but the
majority of dating applications have used quartz, feld-
spars, or mixed phases of SiO2 in flint. A number of differ-
ent luminescence signals can be obtained from each of
these minerals (or rocks in the case of flint). The result is
an increasingly diverse range of luminescence dating
methods that are possible. They all share the same basic
principles outlined above, but the age range over which
they can be applied, and the details of their application
vary. To understand the reason for these differences and
some of the details of the method, it is necessary to under-
stand more about the physical basis of the luminescence
signals that are observed.
Physical basis
Luminescence occurs as result of energy stored within a
crystalline solid being released and being emitted in the
form of a photon of light. To understand this process,
Figure 1 shows a schematic representation of the band
structure of an insulator or semiconductor and various pro-
cesses that occur within and among the bands. In such
materials the valence band is fully occupied by electrons,
while no electrons reside in the conduction band. Further
information about the band structure of crystalline insula-
tors and semiconductors is given in see “Band Structure.”
There is a forbidden energy gap between the valence and
the conduction band and most electrons are unable to
reside in this gap. However, if chemical impurities exist
within the crystal or other structural defects are present,
then these may give rise to traps within the forbidden
gap (marked T1 and T2 on Figure 1a) where electrons
can be stored in an excited state.
392 LUMINESCENCE DATING

Luminescence Dating, Figure 1 Schematic bandgap diagram of the processes involved in ionization, storage, and eviction of
charge that underpin the production of luminescence signals in crystalline materials (Adapted from Aitken, 1985). Further details
about the sequence of processes illustrated here are given in the text.

Ionizing radiation interacting with this theoretical crys- Thermoluminescence (TL)

tal can transfer energy to an electron in the valence band,
and this process of ionization will excite the electron into
the conduction band (Figure 1a) and leave a hole in the
valence band. The electron is unable to remain in the con-
duction band and one of two events will follow. Either the
electron will immediately de-excite to the valence band,
releasing the energy absorbed from the radiation, or the
electron in rare cases may lose a small amount of energy
and become trapped at T1 or T2 within the forbidden gap
as shown in Figure 1a. Similarly, the hole created in the
valence band by the eviction of an electron can
diffuse through the crystal and may become trapped at
another class of defects, called recombination centers.
A proportion of these recombination centers are capable
of producing luminescence (“L” on Figure 1a). The net
effect is that a proportion of the energy deposited by
the radioactivity has become stored within the crystal
structure. As further ionization occurs, the number of elec-
trons stored in traps (e.g., T1 or T2), and hence the amount
of energy stored, will increase. If the trap is stable, then
these electrons may be stored for periods of thousands
to millions of years (Figure 1b). The stability of the
charge in the trap depends upon the amount of energy
required to excite charge out of the trap. This energy
(indicated by “E” in Figure 1b) varies from one trap
to another. Generally, the deeper the trap, the more
stable it is.
The electrons stored in the defects within the forbidden
gap can be released by providing sufficient energy (such
as light or heat) to excite them into the conduction band.
From there they must relax and may recombine with a hole
stored at a luminescence center “L” (Figure 1c). This
recombination may result in the emission of light – this
is the luminescence signal that is observed.
Historically (see “Luminescence Dating, History”), the
first method used to measure luminescence was by
heating the sample at a fixed rate (typically between
1  C/s and 5  C/s) from room temperature up to tempera-
tures varying between 450  C and 700  C, depending
upon the mineral. The emission is termed thermolumines-
cence (TL) and a graph showing the luminescence signal
emitted as a function of temperature (commonly referred
to as a glow curve) consists of a series of peaks (e.g., Fig-
ure 2). Each peak may be due to a single type of trap
within the mineral being measured, but more commonly
the signal is a composite of several traps. Although it is
not always possible uniquely to identify the source of
the electrons, it is normally true that the TL signal
observed at higher measurement temperatures originates
from traps that are deeper below the conduction band: this
follows because more energy is required to release elec-
trons from deeper traps than electrons released from traps
at lower temperatures, and so this only occurs at higher
measurement temperatures. A corollary of this is that
electrons in deeper traps are more stable than those in
shallower traps. For instance, Figure 2 shows a TL mea-
surement of quartz. The electrons in the trap giving rise
to the TL peak observed at
110  C are relatively unsta-
ble and at room temperature have a mean lifetime of

20 h, making them useless for dating. In contrast, mea-
surements suggest that the electrons giving rise to the TL
peak observed at
325  C are stable over many millions
of years. These two peaks could be thought of as being
derived from traps T1 and T2, respectively, in Figure 1.
The stability of the deeper trap has been confirmed by dat-
ing of sediments up to almost a million years old, and it is
this trap that is the focus of most methods used for dating
with quartz.
 LUMINESCENCE DATING
Luminescence Dating, Figure 2 A typical thermoluminescence
(TL) measurement, known as a glow curve. The intensity of the
TL signal is shown as a function of the sample temperature
as it is steadily increased from room temperature to 450  C at
a rate of 5  C/s.
Optically stimulated luminescence (OSL)
A second way of stimulating the emission of luminescence
is by exposing a sample to light. In ground breaking work,
Huntley et al. (1985) shone the beam from an argon-ion
laser emitting at 514 nm onto their samples of quartz and
feldspar. They then observed the luminescence emitted
at a wavelength of
400 nm. This optically stimulated
luminescence (OSL) signal occurs as soon as the sample
is illuminated, and the intensity of the OSL signal
decreases as electrons are removed from the traps as mea-
surement continues (Figure 3). Unlike TL, the OSL data
typically obtained gives little indication from which traps
the electrons have been released. Thus, prior to making an
OSL measurement, it is important to thermally pretreat the
sample so that a signal is obtained from the group of traps
of interest. This is achieved by heating the sample prior to
measurement so that the shallow traps, whose electrons
are unstable over the burial period (e.g., T1 in Figure 1),
are emptied, leaving only the electrons in deeper, stable
traps (e.g., T2 in Figure 1). This thermal treatment is called
a preheat.
The light used to stimulate the minerals is restricted to a
narrow range of wavelengths, and this light must be
prevented from reaching the sensitive light detector (the
photomultiplier tube) used to measure the luminescence
signal (see “Luminescence Dating, Instrumentation”).
This is because the light given off by the sample, the lumi-
nescence that is being measured, is emitted at a different
wavelength from the stimulation light. Blue light-emitting
diodes (LEDs) are commonly used for stimulation and
will generate an OSL signal from both quartz and feldspar.
An alternative method of stimulation is to use LEDs
that emit beyond the visible part of the spectrum, in the
infrared. These are the type of LEDs used for remote
controls in televisions, and for luminescence work, those
centered at 880 nm have been used extensively in
393
Luminescence Dating, Figure 3 A typical quartz optically
stimulated luminescence (OSL) decay curve. When the optical
stimulation source is first switched on, the OSL signal is large
(18,000 counts per second in this example) but then rapidly
decreases as the charge stored in the traps is removed. Note that
the vertical axis is on a log scale.
infrared-stimulated luminescence (IRSL) dating of feld-
spar. OSL signals are produced but are more commonly
referred to as infrared-stimulated luminescence (IRSL).
IRSL is only observed from feldspars; quartz grains do
not give an IRSL signal when the sample is measured at
room temperature unless the quartz grains contain feldspar
inclusions. The fact that pure quartz does not emit an IRSL
signal at room temperature can be exploited to provide a
method for assessing the purity of quartz separated for
luminescence measurements.
In naturally occurring minerals, a range of different
traps occur. In quartz it is thought that aluminum substitut-
ing for silicon within the SiO2 structure is an important
source of defects, though many other impurities are also
important (e.g., Ge) (see “Quartz Defects, Optically Stim-
ulated Luminescence and Thermoluminescence”). The
relative concentration of these different types of trap
varies from one sample of quartz to another. The result is
that the intensity and detailed nature of the luminescence
signal from quartz varies for different areas. Similarly, dif-
ferent minerals have different impurities (e.g., manganese
in calcite) and thus different luminescence behavior. In
feldspars the mechanisms by which luminescence is gen-
erated are thought to be more complex than in quartz, with
some recombination from traps to luminescence centers
not involving the conduction band (see “Feldspar,
Infrared-Stimulated Luminescence”).
The physical characteristics of the different traps con-
trol whether they are suitable for luminescence dating.
Only some of the traps that are known in naturally occur-
ring minerals are sensitive to light, while charge in others
can only be evicted by heating the sample. Those traps
which are sensitive to light are the ones that are most
394 LUMINESCENCE DATING

Luminescence Dating, Figure 4 (a) The additive dose method for determining equivalent dose (De). Red data points show
luminescence measurements made after different laboratory irradiations. (b) The regenerative dose method for determining
equivalent dose. The luminescence signal resulting from irradiation in nature (plotted as a square on the y-axis) is interpolated onto a
dose response curve that is generated by making measurements after different laboratory irradiations.

appropriate for dating sediments. Equally important is
whether the charge stored within the traps is stable over
the period of time that is being dated. A key consideration
here is the depth of the trap below the conduction band.
This can be measured as the energy required to excite an
electron from the trap to the conduction band (marked
as E on Figure 1b). In general terms the deeper the trap
(the greater the value of E), the more stable the trap.
Measuring equivalent dose
In order to calculate the radiation dose that a sample has
been exposed to (the equivalent dose, De), it is necessary
to calibrate the signal using an artificial radiation source
in the laboratory whose dose rate is known. The intensity
of the luminescence signal arising from the dose received
in nature can then be compared with the intensity of a
luminescence signal following exposure to known radia-
tion doses in the laboratory. There are two broad strategies
for such measurements, additive dose methods and regen-
eration methods (Figure 4).
In the additive dose method (Figure 4a), the intensity of
the luminescence signal arising from the dose absorbed in
nature is measured. Additional measurements of the lumi-
nescence signal are then made after adding known radia-
tion doses in the laboratory. The change in luminescence
intensity with laboratory radiation is then fitted and
extrapolated to the intersection with the x-axis to calculate
what dose must have been absorbed by the sample prior to
measurement of the natural – the De.
In the regeneration method (Figure 4b), after measure-
ment of the luminescence signal arising from exposure to
radiation in nature, the sample has its luminescence signal
reset (normally by exposure to an artificial light source),
and then different laboratory radiation doses are given and
the subsequent luminescence signal measured. The growth
of the luminescence signal with laboratory dose is fitted,
and the intensity of the signal resulting from exposure in
nature is interpolated onto that curve in order to calculate De.
The value of De calculated by extrapolation in the addi-
tive dose method (Figure 4a) can be very sensitive to the
choice of which mathematical model is used to fit the dose
response curve. Mathematically, the regeneration method
is simpler since the natural signal (shown as a square on
Figure 4b) is being interpolated between measured data
points, and not extrapolated.
Two broad approaches have been developed for making
the luminescence measurements necessary for the additive
dose or regenerative dose methods. When luminescence
was first developed, a number of different subsamples,
or aliquots, of each material would be prepared. Different
laboratory doses would be given to each group of aliquots.
Thus, multiple aliquots were required to calculate a single
equivalent dose (De). Both multiple aliquot additive dose
(MAAD) and multiple aliquot regenerative dose (MAR)
methods have been used in the past. Duller (1991) was
the first to propose practical methods where all of the mea-
surements necessary to determine De could be made using
a single aliquot. Since this innovation a series of important
breakthroughs have been made, developing a variety of
single-aliquot methods for feldspars (Duller, 1995) and
for quartz (Murray et al., 1997), and there has been a ten-
dency for the majority of measurements to be made using
single-aliquot methods.
A challenge when using single-aliquot methods is that
the luminescence sensitivity, defined as the amount of
light emitted per unit of absorbed radiation, may vary
when repeated measurements are made. This problem
became apparent when applying single-aliquot methods
to quartz (e.g., Stokes et al., 2000). Early attempts to cor-
rect for this change in sensitivity used the response of the
110 TL peak in quartz (Murray and Roberts, 1998), but it
became apparent that this signal did not completely mimic
the changes in sensitivity of the OSL signal. The single-
aliquot regenerative dose (SAR) method proposed by
Murray and Wintle (2000) for quartz explicitly measures
any change in luminescence sensitivity during repeated
measurements by monitoring the OSL response to a test
 LUMINESCENCE DATING 395

Luminescence Dating, Figure 5 Schematic of the single-aliquot regenerative dose (SAR) measurement protocol and how the
different luminescence measurements are combined in order to determine the equivalent dose (De). The method was first developed
by Murray and Wintle (2000) (Figure adapted from Duller (2008b)).
396 LUMINESCENCE DATING
dose and corrects for these changes. Figure 5 summarizes
the sequence of measurements that are undertaken. After
measurement of the luminescence signal arising from irra-
diation in nature (Ln) or measurement of the luminescence
signal arising from a laboratory dose (Lx) (shown at the
top of the panel), the aliquot is exposed to a fixed radiation
dose to test the luminescence sensitivity. This test dose is
kept constant for any given aliquot, and the subsequent
luminescence measurement (Tx) would be constant if
there were no change in sensitivity. If sensitivity change
does occur, then it can be seen as changes in Tx.
A simple way to correct for these changes in sensitivity
is to plot a dose response curve of the ratio of Lx divided
by Tx as a function of laboratory radiation dose (bottom
of Figure 5). The critical development in Murray and Win-
tle (2000) was developing a protocol which made it possi-
ble to make all of these measurements using OSL signals.
Advantages of single-aliquot measurements
The use of single-aliquot measurements makes it feasible
to make replicate measurements of the De of a sample.
It is common practice for 20–50 aliquots to be measured
for a single sample. This makes it possible to determine
whether the values of De for a sample are consistent with
one another, or whether they show variability. For samples
which were either heated uniformly or which were
exposed to sufficient daylight at deposition to completely
remove their signal, replicate De values will form a single
population (Figure 6a, b), and some measure of central
tendency (e.g., the mean or weighted mean) can be calcu-
lated as used in the age equation (Eq. 2). More complex
patterns are seen in samples where the amount of exposure
to daylight at deposition varied from one grain to another
or where mixing of the deposit, for example, by bioturba-
tion, has occurred. This is common for samples where the
exposure to daylight is limited, as may be the case for flu-
vial or glacial samples (Figure 6c, d). In this situation,
where a measurement includes grains that were not
bleached at deposition, the value of De obtained will be
the sum of the dose acquired since deposition, plus an
inherited signal whose size may vary from a few Gy to
many tens of Gy. The result is a skewed distribution where
the De value corresponding to deposition of the sediment
is the lowest value observed. A range of statistical
approaches have been proposed to deal with such situa-
tions (Galbraith and Roberts, 2012). The most commonly
used statistical model is the Minimum Age Model
(MAM). Complex patterns may also be seen in replicate
measurements of De if bioturbation has occurred and
mixed sediments of different ages (Rink et al., 2013), in
which case standard models to deal with this situation
have not yet been developed.
An additional advantage of single-aliquot methods of
equivalent dose determination is that the number of grains
in each aliquot can be varied (Duller, 2008a). Where sam-
ples consist of a complex mixture of grains with different
De values, then it may be advantageous to make
measurements on single mineral grains (see “Lumines-
cence Dating, Single-Grain Dose Distribution”). This is
not possible for silt-sized samples (<90 mm), but for
sand-sized materials, it is possible either to handpick the
grains or to use focused laser systems designed to make
luminescence measurements on single grains (Figure 7).
Such single grain measurements have proved essential
for dating some complex archaeological sites (Jacobs
et al., 2008), glacial sediments (Duller, 2006), and applica-
tions looking at rates of bioturbation in soils (Heimsath
et al., 2002). Single grain measurements of quartz consis-
tently show that only a small proportion of grains emit
luminescence of sufficient intensity to be useful. It is com-
mon for 90 % of the luminescence signal that is emitted to
originate in 5 % of the grains. In extreme cases less than
1 % of the grains may emit a signal of sufficient intensity
to be useable. When making measurements of feldspars, a
higher proportion of grains emit a luminescence signal,
but this approach is still in its infancy.
Measuring dose rate
The denominator in the luminescence age equation Eq. (2)
is the dose rate; this is a measure of the ionizing radiation
energy that a sample has been exposed to each year during
burial. This energy derives from two main sources (see
“Luminescence Dating, Dose Rates” and “Radiation Dose
Rate”). The first of these are naturally occurring radioiso-
topes which are found in the sediments making up the
samples collected for dating. Alpha, beta, and gamma
radiations are produced by a variety of radioisotopes, pri-
marily associated with potassium, uranium, and thorium,
and to a lesser degree rubidium. The second source of radi-
ation is cosmic rays. These are high-energy particles that
originate from high-energy sources in the universe such
as supernovae and which bombard the earth and interact
with atmospheric gas molecules to produce earth-surface
showers of cosmic rays. Cosmic rays are critical in the for-
mation of radiocarbon (see “Radiocarbon Dating”) and
form the basis for cosmogenic nuclide dating (see “Terres-
trial Cosmogenic Nuclide Dating”). In luminescence dat-
ing cosmic rays are significant because they are
sufficiently energetic to excite charge into the conduction
band and thus are part of the total radiation dose rate for a
given sample.
Cosmic rays are absorbed as they pass through sedi-
ments, and thus, the dose rate due to cosmic rays is
greatest near the surface and decreases in a predictable
manner as samples are buried. Cosmic rays are weakly
absorbed by the atmosphere, so cosmic dose rates are
greater for samples at high altitude than those near sea
level. There is also a weak latitudinal dependence because
of interactions between cosmic rays and the Earth’s mag-
netic field. The cosmic dose rate can be easily calculated
based on measurements of burial depth, altitude, longi-
tude, and latitude, using equations described by Prescott
and Hutton (1994). Typical dose rates due to cosmic rays
are 0.2–0.05 Gy/ka depending upon the depth of burial.
 LUMINESCENCE DATING 397

Luminescence Dating, Figure 6 Replicate measurements of equivalent dose (De) for two different samples. In each case the same
data set is shown both as a histogram and as a radial plot. (a, b) Replicate measurements of De for an aeolian sample form a
single population, and an average value can be used for calculating an age (shown by the vertical dashed line in (a) and the grey bar
in (b)). (c, d) Fluvial samples (or other incompletely bleached samples) may exhibit much greater variability in De. High values
(in this case above about 3 Gy) occur because some grains were not reset at deposition and thus inherited a signal. The value of De
appropriate for calculating an age from this distribution can be found using statistical models and is shown in (c) as a vertical
dashed line and by a grey bar in (d).

In most applications of luminescence dating, the
greatest contribution to the total dose rate arises from
the decay of radioisotopes in the immediate vicinity of
the sample. The main elements of relevance here are
potassium, uranium, and thorium. The radioactive isotope
of potassium (40 K) undergoes a single decay to either
40
Ca or 40Ar, emitting beta particles and gamma rays. Ura-
nium (238U and 235U) and thorium (232Th) have more
complex decay patterns. Each of these three radionuclides
is the parent isotope of a series of radioactive daughter iso-
topes (see “Radiation and Radioactivity”) and they form
three decay series. The different radioisotopes making up
these decay series emit a combination of alpha, beta, and
gamma radiations. These three types of radiation have
very different properties. Alpha particles consist of two
neutrons and two protons and are only able to penetrate
very short distances (
10–40 mm) in Earth materials. Beta
particles are electrons, and these are able to penetrate up to
2 mm. Gamma rays are part of the electromagnetic spec-
trum and are able to penetrate up to 30 cm in rocks and
sediments.
When calculating the dose rate, it is necessary to differ-
entiate between the dose from radionuclides that are found
inside the material being used for luminescence dating and
the dose from radionuclides found in the sediment sur-
rounding the material separated for dating. The former is
known as the internal dose, and the latter the external dose.
If luminescence measurements are made on grains of
quartz separated from sediment, the internal dose is nor-
mally very small in comparison with the external dose,
398 LUMINESCENCE DATING
Luminescence Dating, Figure 7 Upper: Focused laser used to
stimulate single grains. Lower: Sample holder (10 mm diameter)
with a grid of 100 holes (each hole 300 mm in diameter). The
focused laser strikes each hole in turn, measuring the OSL from
each grain of quartz (Photos: G.A.T. Duller).
and in many cases it is assumed to be zero. In contrast if
the luminescence signal from feldspars is used, these
grains may contain up to 14 % by weight of potassium,
and this yields a large internal dose which forms a critical
part of the total dose rate to the grains.
An additional consideration in calculating the dose rate
to a sample is the impact of the water content of a sample.
In the case of sediments, there will be interstices between
mineral grains, and depending upon the environmental
conditions at a specific site, these interstices may either
be wholly air-filled, wholly filled with water, or more
commonly a mixture of the two. Water between grains
absorbs a proportion of the radiation energy emitted by
radionuclides and thus alters the dose rate. The impact of
water upon alpha, beta, and gamma radiations is well
understood and is an integral part of the process of calcu-
lating a dose rate. A useful rule of thumb is that a 1 %
increase in the water content leads to approximately a
1 % increase in the luminescence age. It is therefore neces-
sary to estimate the average water content of the sample
during the period being dated. Estimating the water con-
tent of sediments during the entire period of burial can
be a challenge, and uncertainty in this value is commonly
one of the most important controls on the precision with
which it is possible to calculate luminescence ages (see
“Luminescence Dating, Uncertainties and Age Range”).
There are two primary approaches for measuring the
dose rate due to radionuclides: chemical methods and
emission counting methods. A variety of chemical
methods such as AAS, ICP-MS, and NAA are available.
It is normal practice to measure only the concentration
of K, U, and Th in samples and to assume that the U and
Th decay series are in secular equilibrium (see “Radiation
and Radioactivity”). Emission counting methods directly
measure alpha, beta, or gamma emissions from a sample.
Alpha particles and gamma rays are emitted with specific
energies which means that spectroscopic methods can be
used to identify which radioisotope individual alphas
and gammas originate from. Alpha spectroscopy is not
commonly used because of the lengthy sample preparation
procedure, but high-resolution gamma spectroscopy using
high-purity germanium detectors is widely used. Beta par-
ticles arising from a specific isotope are emitted at a range
of energies and thus spectroscopic methods are not possi-
ble. However, it is possible to measure the total beta dose
rate and to account for the range of expected ranges of beta
emission energy which occur with known proportion. All
of the chemical and emission counting methods described
above are applied to subsamples returned to the laboratory
for testing. These are from the location of the sample.
They will give an accurate assessment of the dose rate pro-
vided that they are representative of the materials around
the sample. The short range that alpha and beta particles
can travel means that material from the sample collected
for luminescence measurement will be appropriate. How-
ever, gamma rays may travel up to 30 cm in sediment. If
the characteristics of the material surrounding the sample
vary over this scale, then it is either necessary to take mul-
tiple subsamples in the field to characterize these varia-
tions or to make measurements of the gamma dose in the
field in situ. Field measurements can be made using a por-
table gamma spectrometer (Figure 8). These portable sys-
tems do not have the energy resolution that is achieved by
laboratory-based gamma spectrometry, and so it is not
possible to identify emissions from a large number of iso-
topes. However, it is possible to assess the U, Th, and
K concentrations.
A final issue relating to dose rate is whether the dose
that is measured today has remained constant during the
period being dated. A primary concern here is whether
the decay series of U or Th are in secular equilibrium or
not (see “Radiation and Radioactivity” and “Radiation
Dose Rate” for discussion of secular equilibrium). Where
 LUMINESCENCE DATING
Luminescence Dating, Figure 8 Using a portable gamma
spectrometer to measure the in situ gamma dose rate. The
detector is in the upper hole that remains after collecting the
sample for luminescence dating, and the results of the
measurement are being displayed and stored in the yellow and
grey unit at the bottom of the picture.
disequilibrium is detected in the decay series, it may
be possible to correct for this if the nature and timing of
the disequilibrium can be determined (Olley et al.,
1996). Fortunately disequilibrium is not a common prob-
lem provided that geochemically active environments are
avoided. Typical environments at greater risk of disequi-
librium include organic-rich sediments, samples with evi-
dence of cementation since deposition, and environments
at the interface between saline and fresh water.
Practicalities
Sampling
Sampling is a critical part of the application of lumines-
cence dating. Samples for sediment dating should be col-
lected in a way that ensures that there is no exposure to
daylight. A common approach is to clean a sediment face
and then hammer an opaque plastic or metal tube
(typically 5–7 cm diameter and 20–30 cm length) into
the section (Figure 9). Once in the laboratory, material at
each end of the tube which may have briefly been exposed
to daylight can be removed. This material can be used for
measurement of the dose rate. Material from the middle of
the tube can then be used for luminescence analysis and
prepared in the manner described below.
At some sites it is not possible to insert tubes, and in this
case an alternative method of sampling is to cover the
399
Luminescence Dating, Figure 9 Sampling sediments for
luminescence dating by hammering a metal tube into a section
(Photo: G.A.T. Duller).
sampling position with a large (
2 3 m) sheet of thick
black plastic to exclude daylight and then extract a sample
while working under the plastic sheet in dim red light con-
ditions (as used in the laboratory).
Sample selection is critical, and if possible samples
should be taken at least 30 cm from a major stratigraphic
boundary to ensure that the volume from which the
gamma dose originates is homogeneous. Where samples
have to be taken nearer than 30 cm from a boundary, then
measurements should be made of the distance from the
sample to the boundary, and a subsample of the adjacent
material should be collected so that the dose rate contribu-
tion from that material can be calculated.
Sample preparation
Many of the luminescence signals used for dating are sen-
sitive to light, and all are sensitive to heat. Care needs to be
taken during sample preparation to ensure that light expo-
sure is kept to a minimum and that samples are not unduly
heated. Dim red lights similar to those used in photo-
graphic laboratories are used to provide room lighting
both for sample preparation and luminescence measure-
ment. Most luminescence signals are less strongly affected
by light at the red end of the spectrum than at the blue, and
the eye retains sufficient sensitivity at these wavelengths
to allow safe working conditions.
400 LUMINESCENCE DATING
Sample preparation aims to isolate material suitable for
luminescence measurements (Wintle, 1997). The majority
of luminescence dating applications use either quartz or
feldspar as the target for luminescence measurements. If
present, it is necessary to remove any organics and any
carbonates using hydrogen peroxide (H2O2) and
hydrochloric acid (HCl). Both reactions are exothermic,
and care is needed to ensure that samples are not heated
above about 60  C. The grain size used for luminescence
measurements are commonly either fine grains
(4–11 mm diameter) or coarse grains (within the range of
90–250 mm). Fine grains are sufficiently small that alpha
particles will penetrate the grains completely, while for
coarse grains the external
30–40 mm of the grain surface
can be chemically removed using hydrofluoric acid, and
this removes the external alpha dose to the grains. Thus,
choosing one of these two grain sizes simplifies calcula-
tion of the external alpha dose rate. The use of grains
between these two sizes is more complex.
When using coarse grains, different mineral phases can
be obtained using density separation. Quartz is extracted
by isolating material with a density between 2.62 and
2.70 g.cm3. Some plagioclase feldspar persists through
this process but is removed when grains are etched in con-
centrated hydrofluoric acid to remove the external alpha
dose and then re-sieved. Potassium-rich feldspars can be
concentrated by isolating grains with a density between
2.58 and 2.53 g.cm3.
When using fine grains (4–11 mm), density separation
is ineffective. It is still possible to chemically isolate
quartz by etching grains in hydrofluorosilicic acid
(H2SiF6) for extended periods of time (typically
1–2 weeks). This will remove feldspars, but not quartz.
After preparation, samples are mounted on steel or alu-
minum discs about 10 mm in diameter suitable for mea-
surement in instruments built by a number of
manufacturers. Fine grains are deposited from suspension
in acetone to form a uniform layer covering the entire sur-
face of the disc. Coarse grains are fixed using silicone oil,
and this makes it possible to control the proportion of the
disc surface that is covered and hence the number of grains
on each disc.
Application of luminescence to dating
Heated materials
As luminescence dating has become more established, the
number of applications has increased dramatically. The
technique was originally developed to date the period of
time since pottery had been fired, and this remains an
important application (see “Luminescence, Pottery and
Bricks”). Barnett (2000) undertook an extensive set of
analyses of pottery sherds from 18 sites around the United
Kingdom that had been assigned a later prehistoric age on
the basis of their typology. A total of 135 samples were
measured. Of these, 24 had independent age estimates
based in part on radiocarbon dating of associated finds
and good agreement was found. A much larger suite of
samples was also measured, and here the age of the pottery
was poorly constrained. The study demonstrated the value
of luminescence dating in providing numerical ages for
individual pottery sherds. The results were especially
valuable where only small fragments of pottery were
found, and thus typological distinctions were difficult.
An additional application to heated materials has been to
dating when bricks were fired. This has been extensively
applied to medieval buildings across northwestern Europe
and is now being applied further afield. Bailiff et al. (2013)
applied OSL to date bricks from domed stupas in Sri
Lanka, providing some of the first numerical ages for
these important archaeological features.
Pottery and bricks are deliberately fired during their
manufacture, but other materials may be inadvertently
heated and thus dated using luminescence. This is espe-
cially useful at Paleolithic sites where fires were lit. Stone
tools, especially those manufactured from flint, cannot
normally be dated using luminescence methods, but at
some sites a small proportion of tools may have been inad-
vertently heated and this resets the luminescence signal,
making it possible to date the occupation (see “Lumines-
cence, Flints and Stones”). This method is proving valu-
able across a range of sites and has yielded very early
ages for Middle Paleolithic tanged tools at the site of Ifri
n’Ammar in Morocco, ranging from 83  6 ka to 171 
12 ka (Richter et al., 2010). Stones which were not tools
but were heated by fires may also be dated using lumines-
cence. These too have provided valuable chronological
information, for example, at the site of Blombos in South
Africa (Tribolo et al., 2006). Here, luminescence dating of
sediments and burnt lithics have provided a chronology
extending back to 140 ka for a wide range of Middle Stone
Age artifacts including engraved ochres, bone tools, and
Nassarius kraussianus shell beads that are thought to have
been used for personal ornamentation.
Unheated sediments
In the last three decades, the greatest focus for the applica-
tion of luminescence dating has been to unheated sedi-
ments, where the event being dated is the last exposure
to daylight. The method is ideally suited to dating the for-
mation of both coastal and desert dunes because there is
normally ample opportunity for the mineral grains to be
exposed to daylight prior to deposition. Bristow
et al. (2007) collected samples from a transect of boreholes
across a linear dune in the northern part of the Namib Sand
Sea in order to understand the processes involved in
forming the feature and the rates of these processes
(Figure 10). A ground penetrating radar survey was under-
taken along the same transect prior to OSL sampling and
provided stratigraphic evidence of the pattern of sediment
accumulation. This stratigraphic evidence was used to
choose the locations of the boreholes and the depths at
which samples were collected for OSL dating. The OSL
ages demonstrated that this very large dune (
500 m wide
and over 60 m high) had been completely reworked within
 LUMINESCENCE DATING
Luminescence Dating, Figure 10 Linear dune in the northern
Namib Sand Sea. A series of nine boreholes were drilled along a
500 m transect across the dune to collect samples for OSL
dating. Vehicles provide the scale (Photo: G.A.T. Duller).
the last 5,700 years and that the flank of the dune had
moved 300 m laterally during the last 2,500 years.
Luminescence provides insights into very recent geo-
morphological events and provides a chronological
method that can span both the historic record and the more
distant past. Roberts and Plater (2007) used quartz OSL
measurements on sands underlying the gravel ridges of
Dungeness, a coastal foreland in southern Britain, to pro-
vide a rate for the growth of this feature (Figure 11). Ages
for these underlying sands ranged from 430  20 to 5,120
 220 years, and the younger ages were consistent with
documentary historical evidence for growth of Dungeness
during the Middle Ages. A number of other impressive
data sets utilizing luminescence to provide rates for geo-
morphological processes have been produced, including
Madsen et al. (2010), who calculated 78 OSL ages for
quartz extracted from samples collected from the island
of Rømø in the Wadden Sea, Denmark, with beach ridges
ranging from 220  20 years ago to over 8,000 years ago.
The ability of luminescence dating in these environments
to date very recent deposition makes the method ideal
for this type of study (see also “Luminescence, Geomor-
phological Processes”).
Limits in the age range dated using luminescence
The youngest age that can be obtained using luminescence
dating is difficult to define (see “Luminescence Dating,
Uncertainties and Age Range”). The two major limitations
are the extent to which the signal was reset by exposure to
daylight at deposition, and the sensitivity of the equipment
used to measure the luminescence signal (Madsen and
Murray, 2009). Ballarini et al. (2003) measured quartz
OSL ages for a series of coastal dunes on the island of
Texel on the coast of the Netherlands. Maps of this area
stretching back 260 years provided excellent independent
401
age control, and the OSL ages were accurate across this
entire time period, with one of the youngest samples giv-
ing a date of AD 1996  2 when it was known to have
been deposited between 1996 and 2002 AD. For aeolian
and beach sediments where there is extended exposure
of grains to sunlight prior to deposition, it is normally pos-
sible to obtain accurate ages for samples deposited as
recently as the last couple of decades. In fluvial environ-
ments it may also be possible to achieve such young ages
(e.g., Rustomji and Pietsch, 2007), though because of the
potential for some grains not to have been exposed to suf-
ficient daylight to completely reset their signal, more com-
plex statistical analysis is normally required (e.g.,
application of the minimum age model). In environments
where the exposure to sunlight is more uncertain (e.g., gla-
cial sediments) or where the luminescence sensitivity of
the material is low, the limit for dating may be several hun-
dred to a thousand years.
The upper limit to the age that can be dated using lumi-
nescence is also difficult to define. The major controls on
this limit are the saturation of the luminescence signal, and
the stability of the signal. As quartz or feldspar is exposed
to ionizing radiation, the amount of charge trapped within
defects increases (Figure 4). At low doses this increase in
trapped charge, and hence in the brightness of the lumines-
cence signal measured in the laboratory, is approximately
linear. As exposure to radiation continues, traps within the
crystalline material fill, and it becomes increasingly unlikely
that charge will be trapped. The result is that the rate of
growth of the luminescence signal with radiation dose
decreases, until at some point no further increase in lumines-
cence is seen; at this point the luminescence is said to be sat-
urated. Growth of the luminescence signal can be described
by Eq. (3) where LD is the intensity of the luminescence sig-
nal at a dose D, LMax is the maximum intensity of the lumi-
nescence signal, and D0 describes the rate at which the
luminescence signal approaches saturation:
 
D
LD ¼ LMax 1  e D0 ð3Þ
As a sample approaches saturation, the uncertainties in the
De increase, and if the measured natural luminescence sig-
nal is within 15 % of the maximum luminescence signal
(LMax), then it is normally considered to be beyond the
range of luminescence dating methods. This same thresh-
old can also be defined as two times the value of D0. The
value of D0 varies from one sample to another, and from
one luminescence signal to another. For the OSL signal
from quartz, typical values of D0 are 50–100 Gy, meaning
that the maximum value of equivalent dose that can be
measured ranges from 100 to 200 Gy. The age range that
can be dated in this situation depends upon the dose rate.
For sites where the dose rate is low (e.g., 0.5 Gy/ka), it
may be possible to date up to 400 ka, whereas in areas
where the dose rate is high (e.g., 4 Gy/ka), then it may only
be possible to date as far as 50 ka.
The IRSL signal from feldspar typically has a higher D0
value than the OSL signal from quartz and so may be able
402 LUMINESCENCE DATING

Luminescence Dating, Figure 11 Dungeness foreland, United Kingdom. The rate of growth of this coastal foreland was determined
by applying OSL dating to quartz extracted from sands underlying the gravel ridges that form the surface expression of this
feature (Figure adapted from Roberts and Plater, 2007).

to date older samples. However, potassium-rich feldspars
have a significant internal dose from the potassium within
the grains, meaning that their total dose rate is higher than
that for quartz, so reducing the advantage that they would
otherwise have. A great deal of research has been under-
taken looking for other luminescence signals, both from
quartz and feldspar and from other minerals (e.g., calcite,
see “Luminescence, Biogenic Carbonates”) that may be
able to extend the age range over which luminescence
methods can be applied. It still remains to be seen whether
these new signals such as Thermally Transferred OSL
(TT-OSL) and Violet Stimulated Luminescence (VSL) from
quartz, or the TL signal from calcite, will have suitable
characteristics for dating.
Novel applications
Luminescence geochronology has seen rapid develop-
ment in the last few decades and this rate of progress is
continuing. Two exciting new applications of lumines-
cence that have been proposed in recent years are
(i) its use in thermochronology and (ii) its use for
 LUMINESCENCE DATING
dating the duration that rock surfaces have been exposed
at the Earth’s surface.
Thermochronology (see “Thermochronology, Landform
Evolution”) attempts to understand the thermal history of a
rock and primarily the rate at which it has cooled. Most
commonly this is of interest in understanding the rate at
which rocks have risen in the Earth’s crust. Luminescence
signals can be reset by holding them at high temperatures.
Rocks held at high temperatures will not retain charge in
defects and thus will not yield any luminescence signal.
However, as a rock approaches the Earth’s surface and
cools, it will become able to store charge. Thus, rocks that
have been rapidly uplifted will yield an apparent lumines-
cence age, and this can be used to yield thermochro-
nological information. Initial work (Herman et al., 2010)
used the OSL signal from quartz, and this has an effective
closure temperature of 30  C, far lower than other
thermochronometers, thus making new areas of study pos-
sible. More recent work has provided a sound theoretical
framework for this exciting area (Guralnik et al., 2013),
and it seems very likely that this will be a major growth area
in the application of luminescence in geochronology.
A second area that has seen recent rapid advances has
been the use of luminescence to date rock surfaces (see
“Luminescence, Rock Surfaces”). Here luminescence is
being used to date how long a particular rock surface has
been exposed to daylight. Some luminescence signals
are reset by exposure to daylight. Commonly these signals
are reset rapidly, within periods of seconds to minutes. As
daylight penetrates through rock, it is very rapidly attenu-
ated, meaning that while a particular luminescence signal
(e.g., the OSL signal from quartz) may be reduced to
0.1 % of its initial signal within a few minutes when
exposed to daylight at the Earth’s surface, the OSL signal
in samples found a few millimeters inside a rock may take
days, years, or millennia to be reset. By drilling a core
through the upper few centimeters of a rock and making
measurements of the luminescence signal remaining at
different depths in the rock, it is possible to build up a pic-
ture of how the signal has been reset, and this can be used
to determine the duration of time that the rock surface has
been exposed (Sohbati et al., 2012).
Further reading
A simple overview of luminescence dating is provided by
Duller (2008b) which is freely available at http://www.
english-heritage.org.uk/publications/luminescence-dating/.
Wintle (2008) gives an excellent overview of the develop-
ment of luminescence dating in the last half of the twenti-
eth century. Rhodes (2011) provides an excellent
overview of the application to dating sediments, while
Richter (2007) describes applications to dating burnt
flints. Jacobs and Roberts (2007) describe the application
of single grain OSL measurements in archaeology. Wintle
and Murray (2006) give an excellent review of the applica-
tion of the single-aliquot regenerative dose (SAR) method
applied to quartz.
403
Bibliography
Aitken, M. J., 1985. Thermoluminescence Dating. London:
Academic.
Bailiff, I. K., Lacey, H. R., Coningham, R. A. E., Gunawardhana, P.,
Adikari, G., Davis, C. E., Manuel, M. J., and Strickland, K. M.,
2013. Luminescence dating of brick stupas: an application to the
hinterland of Anuradhapura, Sri Lanka. Antiquity, 87(335),
189–201.
Ballarini, M., Wallinga, J., Murray, A.S., van Heteren, S., Oost, A.
P., Bos, A. J. J., and van Eijk, C. W. E., 2003. Optical dating of
young coastal dunes on a decadal time scale. Quaternary Science
Reviews 22, 1011–1017.
Barnett, S. M., 2000. Luminescence dating of pottery from later pre-
historic Britain. Archaeometry, 42, 431–457.
Bristow, C. S., Duller, G. A. T., and Lancaster, N., 2007. Age and
dynamics of linear dunes in the Namib Desert. Geology,
35, 555–558.
Duller, G. A. T., 1991. Equivalent dose determination using single
aliquots. Nuclear Tracks and Radiation Measurements,
18, 371–378.
Duller, G. A. T., 1995. Luminescence dating using single aliquots:
methods and applications. Radiation Measurements, 24,
217–226.
Duller, G. A. T., 2006. Single grain optical dating of glacigenic sed-
iments. Quaternary Geochronology, 1, 296–304.
Duller, G. A. T., 2008a. Single grain optical dating of Quaternary
sediments: why aliquot size matters in luminescence dating.
Boreas, 37, 589–612.
Duller, G. A. T., 2008b. Luminescence Dating: Guidelines on Using
Luminescence Dating in Archaeology. Swindon: English
Heritage.
Galbraith, R. F., and Roberts, R. G., 2012. Statistical aspects of
equivalent dose and error calculation and display in OSL dating:
an overview and some recommendations. Quaternary Geochro-
nology, 11, 1–27.
Guralnik, B., Jain, M., Herman, F., Paris, R. B., Harrison, T. M., Mur-
ray, A. S., Valla, P. G., and Rhodes, E. J., 2013. Effective closure
temperature in leaky and/or saturating thermochronometers. Earth
and Planetary Science Letters, 384, 209–218.
Heimsath, A. M., Chappell, J., Spooner, N. A., and Questiaux,
D. G., 2002. Creeping soil. Geology, 30, 111–114.
Herman, F., Rhodes, E. J., Braun, J., and Heiniger, L., 2010. Uni-
form erosion rates and relief amplitude during glacial cycles in
the Southern Alps of New Zealand, as revealed from
OSL-thermochronology. Earth and Planetary Science Letters,
297(1–2), 183–189.
Huntley, D. J., Godfrey-Smith, D. I., and Thewalt, M. L. W., 1985.
Optical dating of sediments. Nature, 313, 105–107.
Jacobs, Z., and Roberts, R. G., 2007. Advances in optically stimu-
lated luminescence (OSL) dating of individual grains of quartz
from archaeological deposits. Evolutionary Anthropology, 16,
210–223.
Jacobs, Z., Roberts, R. G., Galbraith, R. F., Deacon, H. J., Grun, R.,
Mackay, A. W., Mitchell, P., Vogelsang, R., and Wadley, L.,
2008. Ages for the middle stone age of Southern Africa: implica-
tions for human behavior and dispersal. Science, 322, 733–735.
Madsen, A. T., and Murray, A.S., 2009. Optically stimulated lumi-
nescence dating of young sediments: A review. Geomorphology
109(1–2), 3–16.
Madsen, A. T., Murray, A. S., Andersen, T. J., and Pejrup, M., 2010.
Luminescence dating of Holocene sedimentary deposits on
Rømø, a barrier island in the Wadden Sea, Denmark. Holocene,
20(8), 1247–1256.
Murray, A. S., and Roberts, R. G., 1998. Measurement of the equiv-
alent dose in quartz using a regenerative-dose single-aliquot pro-
tocol. Radiation Measurements, 29, 503–515.
404 LUMINESCENCE DATING OF ARCHAEOLOGICAL SEDIMENTS
Murray, A. S., and Wintle, A. G., 2000. Luminescence dating of
quartz using an improved single-aliquot regenerative-dose pro-
tocol. Radiation Measurements, 32(1), 57–73.
Murray, A. S., Roberts, R. G., and Wintle, A. G., 1997. Equivalent
dose measurement using a single aliquot of quartz. Radiation
Measurements, 27, 171–184.
Olley, J., Murray, A., and Roberts, R. G., 1996. The effects of
disequilibria in the uranium and thorium decay chains on burial
dose rates in fluvial sediments. Quaternary Geochronology
(QSR), 15, 751–760.
Prescott, J. R., and Hutton, J. T., 1994. Cosmic ray contributions to
dose rates for luminescence and ESR dating: large depths and
long-term time variations. Radiation Measurements, 23, 497–500.
Rhodes, E. J., 2011. Optically stimulated luminescence dating of
sediments over the past 200,000 years. Annual Review of Earth
and Planetary Sciences, 39, 461–488. R. Jeanloz and
K. H. Freeman.
Richter, D., 2007. Advantages and limitations of thermolumines-
cence dating of heated flint from paleolithic sites.
Geoarchaeology, 22(6), 671–683.
Richter, D., Moser, J., Nami, M., Eiwanger, J., and Mikdad, A., 2010.
New chronometric data from Ifri n’Ammar (Morocco) and the
chronostratigraphy of the Middle Palaeolithic in the Western
Maghreb. Journal of Human Evolution, 59(6), 672–679.
Rink, W. J., Dunbar, J. S., Tschinkel, W. R., Kwapich, C., Repp, A.,
Stanton, W., and Thulman, D. K., 2013. Subterranean transport
and deposition of quartz by ants in sandy sites relevant to age
overestimation in optical luminescence dating. Journal of
Archaeological Science, 40(4), 2217–2226.
Roberts, H. M., and Plater, A. J., 2007. Reconstruction of Holocene
foreland progradation using optically stimulated luminescence
(OSL) dating: an example from Dungeness, UK.
The Holocene, 17, 495–505.
Rustomji, P., and Pietsch, T., 2007. Alluvial sedimentation rates
from southeastern Australia indicate post-European settlement
landscape recovery. Geomorphology, 90(1–2), 73–90.
Sohbati, R., Murray, A. S., Chapot, M. S., Jain, M., and Pederson, J.,
2012. Optically stimulated luminescence (OSL) as a chronome-
ter for surface exposure dating. Journal of Geophysical
Research, 117, B09202.
Stokes, S., Colls, A. E. L., Fattahi, M., and Rich, J., 2000.
Investigations of the performance of quartz single aliquot D-E
determination procedures. Radiation Measurements, 32(5–6),
585–594.
Tribolo, C., Mercier, N., Selo, M., Valladas, H., Joron, J. L., Reyss,
J. L., Henshilwood, C. S., Sealy, J. C., and Yates, R., 2006. TL
dating of burnt lithics from Blombos Cave (South Africa): fur-
ther evidence for the antiquity of modern human behaviour.
Archaeometry, 48, 341–357.
Wallinga, J., Murray, A. S., and Wintle, A. G., 2000. The single-
aliquot regenerative-dose (SAR) protocol applied to coarse-
grain feldspar. Radiation Measurements, 32, 529–533.
Wintle, A. G., 1997. Luminescence dating: laboratory procedures
and protocols. Radiation Measurements, 27, 769–817.
Wintle, A. G., 2008. Fifty years of luminescence dating.
Archaeometry, 50(2), 276–312.
Wintle, A. G., and Murray, A. S., 2006. A review of quartz optically
stimulated luminescence characteristics and their relevance in
single-aliquot regeneration dating protocols. Radiation
Measurements, 41, 369–391.
Cross-references
Mass Spectrometry
Secondary Ion Mass Spectrometry (SIMS)
Uranium–Lead Dating
Zircon
LUMINESCENCE DATING OF ARCHAEOLOGICAL
SEDIMENTS
James Feathers
Department of Anthropology, University of Washington,
Seattle, WA, USA
Definition
“Archaeological sediments” can be defined in two ways.
First, it can refer to any sediment at archaeological
sites even if the depositional agent is entirely
geologic. A site is any locus where human activity, as
identified by remains of that activity (artifacts), took
place. The timing of that activity interests archaeologists.
Dating sediments which contain or bracket artifacts is
a proxy for dating the activity itself. Luminescence
is a powerful tool for this because the event addressed
is the deposition itself, assuming sunlight was sufficient
to reset the signal during deposition or, if not fully,
a well-bleached portion can be isolated. But lumines-
cence is not dating the activity (Richter et al., 2009). Arti-
facts can be redeposited by geologic forces making their
current location not where the activity occurred. To avoid
misleading information, bridging arguments are required
to link the event dated (last exposure to sunlight) to the
activity of interest.
Secondly, “archaeological sediments” can refer to sedi-
ments who owe their deposition entirely or in part to
human activity. Here the link between human activity
and the dating event is direct, and bridging arguments
are not required, as long as it can be shown that humans
were indeed the depositional agent. The latter can be
obscure if the human role is indirect, for example, when
humans denude the environment of vegetation and
become just a trigger for subsequent erosion.
Introduction
Luminescence dating has been applied to sediments in
both cases (see earlier reviews by Roberts (1997), Feathers
(2003), Jacobs and Roberts (2007)). In the former, the
rationale is that the human activity cannot be dated with
more precision in another way, either because the event
of interest is outside the dating range of other methods,
because the bridging arguments are less tenable than they
are for luminescence, or because other suitable dating
materials are not present. Luminescence becomes the
method of choice only because other options are poorer.
To be sure, the suitability of any method is not always
known in advance, so many archaeologists recommend
multiple methods on the premise that convergence in
results lends more credibility to results because different
bridging arguments are needed for each method. In the
case of human-caused deposition, however, luminescence
is the method of choice, because of its value as a direct dat-
ing method with no need for bridging arguments.
 LUMINESCENCE DATING OF ARCHAEOLOGICAL SEDIMENTS
Where luminescence dating of sediments is a proxy for
dating human activities, the contributions are largely sub-
stantive, with the only methodological (in terms of archae-
ology) issue related to bridging arguments. To argue that
sediment dates are good proxies, the stratigraphy must
have integrity with little postdepositional activity, and
luminescence single-grain dating is uniquely suited to
address this issue. On the other hand, where luminescence
is applied to anthropogenic sediments, the contributions
are methodological (e.g., how can earthen mounds be
dated?), as well as substantive.
Africa and the Middle East
Most scholars agree that anatomically modern humans
originated in sub-Saharan Africa and that Middle Stone
Age (MSA) lithics represent their handiwork. However,
a more contentious question is when did humans became
cognitively modern and how and when did they colonize
the rest of the world? Dating of complex stone industries
called the Howiesons Poort and the Still Bay, which
appear within the MSA in South Africa, was dated with
this question in mind. Prior to luminescence applications,
both industries remained poorly dated. Work by Jacobs
and colleagues, who applied innovative single-grain dat-
ing to several sites, culminated in an argument (Jacobs
et al., 2008) that Howiesons Poort dated 60–65 ka and
Still Bay around 71 ka with a significant gap between
them. Conversely, Tribolo et al. (2013), using single-grain
OSL along with TL dating of burned lithics at Diepkloof,
produced not only older dates but suggested the durations
of both industries overlap and are substantially longer, as
much as 50 ka. While this disagreement is bound to foster
additional research, both groups have already made an
important contribution. Prior dating was so coarse-grained
that the best explanatory framework was correlation with
environmental change. Now OSL provides dates of suffi-
cient precision to decouple direct changes in lithics from
direct changes in climate or other environmental variables.
A precise chronology of the MSA is now possible, forcing
archaeologists to build explanations that consider the
selective role of the environment in a more complex
fashion.
North Africa and the Middle East are also involved in
this issue. The question in North Africa is the age of
a complex lithic industry called Aterian, often cited as evi-
dence of modernity. Traditional dating placed the Aterian
no older than 40 ka, but luminescence has pushed back the
age to more than 100 ka (Jacobs et al., 2012). This has
muddied the issue, because industries attributed to modern
human behavior in North Africa are as old and even older
than those from South Africa, and both seem to have
developed independently. How modern humans got from
Africa to the Middle East centers on two routes: north
along the Nile into the Levant and south across the Red
Sea onto the Arabian Peninsula. Luminescence has con-
tributed to this debate in two ways: by dating sediments
relevant to climate change, because the routes were
405
hospitable to travel only at certain times (e.g., Rosenberg
et al., 2011), and by dating archaeological sites themselves
(e.g., Armitage et al., 2011).
Asia
How modern humans dispersed across Asia is a question
with few answers because the area is vast and the number
of sites investigated are few, but luminescence has been
applied here as well (e.g., Petraglia et al., 2007; Pei
et al., 2010; Haslam et al., 2011; Demeter et al., 2012).
In East Asia, thermally transferred OSL (TT-OSL) was
introduced to achieve older dates on quartz (Kim et al.,
2010). Perhaps the best-known luminescence dating in
Asia has been applied to sediments containing remains
of the small hominin, Homo floresiensis, in Indonesia.
Despite the lack of a fast-bleaching component, TL using
red emission helped determine a 18–38 ka duration for the
relict archaic hominin, well into the time of modern
humans (Morwood et al., 2004).
Australia
Two issues dominate the early prehistory of Australia:
When did humans first arrive and were they responsible
for the extinction of the Pleistocene megafauna (Roberts
et al., 2001)? Prior to the 1990s, radiocarbon had
established that humans were present in Australia by about
40 ka, but the application of luminescence dating pro-
duced dates of 60 ka or older (e.g., Roberts et al.,
1998b). The work is critically reviewed by O’Connell
and Allen (2004), who claim little evidence for arrival ear-
lier than 45 ka. Postdepositional mixing is the issue. Sin-
gle-grain dating measures an equivalent dose or “age”
for each grain and can identify mixtures of different-aged
grains, if other sources of variability, including heteroge-
neous dose rates, can be controlled (David et al., 2007).
An early application overturned previous TL dating of
more than 100 ka at Jinmium rock shelter in northern Aus-
tralia, by showing that a portion of grains were poorly
bleached (Roberts et al., 1998a). However, at one site,
Malakunanja, with OSL ages in the 50–60 ka range,
single-grain dating suggested no postdepositional distur-
bance (Roberts et al., 1998b). Understanding the demise
of the megafauna depends on whether dating the extinc-
tions coincides with dating human arrival, and lumines-
cence has been used to argue for and against the
correlation (Roberts et al., 2001; Price et al., 2011).
Europe
Modern humans spread into Europe 40–50 ka. Exactly
when they arrived and how they replaced resident Nean-
derthals continue to generate a large literature, and lumi-
nescence has played a key role. The Bohunician lithic
tradition in south-central Europe, dated by luminescence
to about 48 ka, appears to be the oldest evidence for
modernity (Hoffecker, 2009). Richter et al. (2009) found
the most reliable dates to be TL on lithics because of poor
association for radiocarbon and bad postdepositional
406 LUMINESCENCE DATING OF ARCHAEOLOGICAL SEDIMENTS
mixing for OSL. Among the oldest hominid sites in
Europe, the Atapuerca karstic complex in northern Spain
has been dated by luminescence. Berger et al. (2008) used
TL and IRSL on potassium feldspars to produce stratigra-
phically consistent dates up to 960  120 ka.
North and South America
Much of the modern human debate involves times beyond
the reach of radiocarbon, but not so early settlement of the
Americas. However, OSL is still being applied, not only as
a supplemental dating method but more importantly to
provide a direct measure of sediment mixing. The issue
is whether people arrived prior to the appearance of the
distinctive Clovis lithic tradition that spread across North
America around 13 ka. Luminescence has been applied
for several pre-Clovis claims (Feathers et al., 2006; Waters
et al., 2009; Lahaye et al., 2013). The Debra L. Friedkin
site in Texas (Waters et al., 2011) was dated by 49 closely
spaced OSL dates on quartz-dominated fine grains, which
remarkably are almost all in correct stratigraphic order.
The pre-Clovis artifacts are dated 15–16 ka, but an anom-
aly is that the Clovis layer is dated about 14 ka and the
succeeding Folsom layer at 13 ka, both older than their
normal age range. Lubinski et al. (2013) applied single-
grain and multigrain single aliquot IRSL dating at
a mammoth site in Washington. With the bones was found
a piece of flake debitage in colluvium dated 16 ka.
Although most of the site has stratigraphic integrity, one
sample showed an intrusion of 5 ka grains, raising the
possibility that the flake is also intruded.
Dating built structures and human environmental
impact using associated sediments
A continual problem in archaeology is obtaining dates for
prehistoric structures, whether simple rock arrangements
or monumental architecture. While often the artifacts asso-
ciated with these structures are reasonably well dated, the
structures themselves are not.
Earthen structures are sediments deposited by humans.
Any stratum within these structures can presumably be
dated to the time of deposition by OSL as long as the mode
of construction insured well-bleaching. Our laboratory
found fill to be poorly bleached in mounds in Louisiana,
but soils underneath the mounds could be dated (Bush
and Feathers, 2003). The rationale is that soil (or any sur-
face exposed for some length of time) will become par-
tially bleached by turbation processes that bring grains to
the surface. Single-grain analysis using minimum age
models provides age estimates for when the soil was cov-
ered over and grains could no longer reach the surface.
The same principle of dating buried surfaces has been
applied to rock structures, by dating sediment below the
rocks (e.g., Rink and Bartoll, 2005; Outram et al., 2010).
In dating soils beneath rocks in anthropogenic rock pat-
terns in Montana and Wyoming, the author demonstrated
the zeroing effect of turbation by collecting samples verti-
cally under the rocks. Five vertical segments were dated
separately, and in all samples, the youngest grains were
concentrated in the segment just below the buried surface
(Feathers, 2012).
For rock walls, Porat et al. (2012) recommended dating
sediments trapped in between rocks as these are less likely
to suffer postdepositional disturbance. Work has also been
done dating mortar (Feathers et al., 2008; Goedicke,
2011). Superimposed sediment over rock art has been
dated by OSL (e.g., Yoshida et al., 2003; Mercier et al.,
2012). Susino (2010) used OSL to date microlithic
debitage, isolated using an SEM, to provide a direct date
for knapping activity.
Dating prehistoric canals is challenging (Sanderson
et al., 2007; Huckleberry et al., 2012). The event archaeol-
ogists want to know is the construction or use of canals.
Fill deposited during use may be present, although
distinguishing that from post-abandonment fill is difficult.
Original construction sediments may also be preserved in
berms, but the berms may also contain sediments from
subsequent cleanouts. Poor bleaching may be the rule, as
well as bioturbation that blurs boundaries. OSL has also
been used to date other agricultural features including
dams and terraces (e.g., Aiuvalasit et al., 2010; Avni
et al., 2012; Davidovich et al., 2012; Beckers et al., 2013).
Luminescence methods have also helped understand the
impact of human activities on the environment, most work
with human-induced erosion (Fuchs and Lang, 2009). Sev-
eral studies have also used luminescence dating to recon-
struct environmental context for archaeological remains,
often to show how artifact distributions might be related
to site formation studies (e.g., Rhodes et al., 2010).
Issues in application of luminescence dating
Postdepositional mixing is a common problem in many
archaeological sediments (Bateman et al., 2007). Single-
grain dating is the solution, and despite difficulties in
understanding single-grain distributions, it is still prefera-
ble to multigrain aliquots where averaging masks the
problem. Work in understanding the extent to which sand
grains move by various agents is just the beginning but
vital (e.g., Tribolo et al., 2010). Several publications report
ages for Paleo-Indian sites in Brazil, not to demonstrate
pre-Clovis occupations but to show stratigraphic integrity.
Two showed little postdepositional sediment mixing
(Araujo et al., 2008; Feathers et al., 2010), including one
with a skeleton dating 12–16 ka. However, two others
did show mixing of sediment grains, although chronolog-
ical information could still be obtained (Araujo et al.,
2013; Bueno et al., 2013). Profiling, or making minimal
measurements on several small samples, is also
recommended to identify badly mixed samples (Burbidge
et al., 2007), something that can now be done in the field
by portable luminescence readers (Sanderson and
Murphy, 2010). Work is also being done to understand
how postdepositional mixing can affect luminescence
results by studying, for example, the movement of grains
by burrowing insects (e.g., Rink et al., 2013).
 LUMINESCENCE DATING OF ARCHAEOLOGICAL SEDIMENTS
Partial bleaching is an important issue in luminescence
dating of archaeological sediments. Detection and accu-
rate dating of partially bleached sediments remain an issue
that is not fully resolved (e.g., Murray et al., 2012).
Many archaeologists think that quartz is best for dating
and know little about dating with feldspars. The recent
research emphasis on dating feldspars belies the old confi-
dence in quartz. Quartz has two major problems. First, it
saturates fairly early, making it problematic for dating older
sediments. This might be an underlying factor in the debate
over the dating at Diepkloof, discussed earlier, because the
quartz there is near saturation. Second, quartz is relatively
insensitive to luminescence on about half the planet, at least
with commonly used UVemission. For older sediments and
in areas where quartz is insensitive, feldspar becomes
advantageous, despite the disadvantage of anomalous fad-
ing. Many strides in understanding the luminescence prop-
erties of and in dating feldspars have been made in the last 5
years (e.g., Jain and Ankjærgaard, 2011; Lamothe et al.,
2012), particularly in the use of high-temperature stimula-
tions. High-precision feldspar dating should be a major goal
for dating archaeological sediments. In areas where even
the feldspars are insensitive, more research into red TL is
needed (Westaway, 2009).
Conclusions
In the last decade, application of luminescence dating to
archaeological sediments has increased in quantity, variety,
and innovativeness. Large portions of the pre-40 ka archae-
ological record, particularly in Africa, are now much better
known. Luminescence dating has allowed archaeologists to
greatly expand their knowledge on the age of built struc-
tures. Luminescence has increased information on how
humans have impacted the environment and how the envi-
ronment has affected humans. It has become a major tool in
site formation studies and establishing an environmental
context. Single-grain dating has become a tool to investi-
gate stratigraphic integrity and bioturbation.
Bibliography
Aiuvalasit, M. J., Neely, J. A., and Bateman, M. D., 2010. New
radiometric dating of water management features at the prehis-
toric Purrón Dam complex, Tehuacn Valley, Puebla, México.
Journal of Archaeological Science, 37, 1207–1213.
Araujo, A. G. M., Feathers, J. K., Arroyo-Kalin, M., and Tizuka,
M. M., 2008. Lapa das boleiras rockshelter: stratigraphy and
formation processes at a paleoamerican site in Central Brazil.
Journal of Archaeological Science, 35, 3186–3202.
Araujo, A., Strauss, A., Feathers, J., Paisani, J., and Schrage, T.,
2013. Paleo-Indian openair sites in tropical settings: a case study
in formation processes, dating methods and paleoenvironmental
models. Geoarchaeology, 28, 195–220.
Armitage, S. J., Jasim, S. A., Marks, A. E., Parker, A. G., Usik, V. I.,
and Uerpmann, H.-P., 2011. The southern route “out of Africa”:
evidence from an early expansion of modern humans into
Arabia. Science, 331, 453–456.
Avni, Y., Porat, N., Plakht, J., and Avni, G., 2006. Geomorphic
changes leading to natural desertification versus anthropogenic
land conservation in an arid environment, the Negev Highlands,
Israel. Geomorphology, 82, 177–200.
407
Avni, Y., Porat, N., and Avni, G., 2012. Pre-farming environment
and OSL chronology in the Negev Highlands, Israel. Journal
of Arid Environments, 86, 12–27.
Bateman, M. D., Boulter, C. H., Carr, A. S., Frederick, C. D., Peter,
D., and Wilder, M., 2007. Preserving the palaeoenvironmental
record in drylands: bioturbation and its significance for lumines-
cence-derived chronologies. Sedimentary Geology, 195, 5–19.
Beckers, B., Shütt, B., Tsukamoto, S., and Frechen, M., 2013. Age
determination of Petra’s engineered landscape – optically stimu-
lated luminescence (OSL) and radiocarbon ages of runoff terrace
systems in the Eastern Highlands of Jordan. Journal of Archae-
ological Science, 40, 33–348.
Berger, G. W., Pérez-Gonzlez, A., Carbonell, E., Arsuaga, J. I.,
Bermúdez de Castro, J.-M., and Ku, T.-L., 2008. Luminescence
chronology of cave sediments at the Atapuerca paleoanthropo-
logical site, Spain. Journal of Human Evolution, 55, 300–311.
Bueno, L., Feathers, J., and De Blasis, P., 2013. The formation pro-
cess of a paleo-Indian open-air site in Central Brazil: integrating
lithic analysis, radiocarbon and luminescence dating. Journal of
Archaeological Science, 40, 190–203.
Burbidge, C. I., Sanderson, D. C. W., Housely, R. A., and Allworth
Jones, P., 2007. Survey of Palaeolithic sites by luminescence
profiling, a case study from eastern Europe. Quaternary Geo-
chronology, 2, 296–302.
Bush, D. A., and Feathers, J. K., 2003. Application of OSL single-
aliquot and single-grain dating to quartz from anthropogenic soil
profiles in the SE United States. Quaternary Science Reviews,
22, 1153–1159.
Cochrane, G. W. G., Doelman, T., and Wadley, L., 2013. Another
dating revolution for prehistoric archaeology? Journal of
Archaeological Method and Theory, 20, 42–60.
David, B., Roberts, R. G., Magee, J., Mialanes, J., Turney, C., Bird,
M., White, C., Fifield, K., and Tibby, J., 2007. Sediment mixing
at Nonda Rock: investigations of stratigraphic integrity at an
early archaeological site in northern Australia and implications
for the human colonization of the continent. Journal of Quater-
nary Science, 22, 449–479.
Davidovich, U., Porat, N., Gadot, Y., Avni, A., and Lipschits, O.,
2012. Archaeological investigations and OSL dating of terraces
at Ramat Rahel, Israel. Journal of Field Archaeology,
37, 192–208.
Demeter, F., Shackelford, L. L., Bacon, A.-M., Duringer, P.,
Westaway, K., Sayavongkhamdy, T., Braga, J., Sichanthongtip,
P., Khamdalavong, P., Ponch, J.-L., Wang, H., Lundstrom, C.,
Patole-Edoumba, E., and Karpoff, A.-M., 2012. Anatomically
modern human in southeast Asia (Laos) by 46 ka. Proceedings
of the National Academy of Sciences, 109, 14375–14380.
Feathers, J. K., 2003. Use of luminescence dating in archaeology.
Measurement Science and Technology, 14, 1493–1509.
Feathers, J. K., 2012. Luminescence dating of anthropogenic rock
structures in the northern Rockies and adjacent High Plains,
North America: a progress report. Quaternary Geochronology,
10, 399–405.
Feathers, J. K., Rhodes, E. J., Huot, S., and McAvoy, J. M., 2006.
Luminescence dating of sand deposits related to late Pleistocene
human occupation at the Cactus Hill site, Virginia, USA. Quater-
nary Geochronology, 1, 167–187.
Feathers, J. K., Johnson, J. A., and Kimbel, S. R., 2008. Lumines-
cence dating of monumental stone architecture at Chavín de
Huntar, Perú. Journal of Archaeological Method and Theory,
15, 266–296.
Feathers, J. K., Piló, L., Arroyo-Kalin, M., Kipnes, R., and Coblenz,
D., 2010. How old is Luzia? Luminescence dating and strati-
graphic integrity at Lapa Vermelha, Lagoa Santa, Brazil.
Geoarchaeology, 25, 395–436.
Fuchs, M., and Lang, A., 2009. Luminescence dating of hillslope
deposits – a review. Geomorphology, 109, 17–26.
408 LUMINESCENCE DATING OF ARCHAEOLOGICAL SEDIMENTS
Goedicke, C., 2011. Dating mortar by optically stimulated lumines-
cence: a feasibility study. Geochronometria, 38, 42–49.
Haslam, M., Roberts, R. G., Shipton, C., Pal, J. N., Fenwick, J. L.,
Ditchfield, P., Boivin, N., Dubey, A. K., Guta, M. C., and
Petraglia, M., 2011. Late Acheulean hominins at the Marine Iso-
tope Stage 6/5e transition in north-central India. Quaternary
Research, 75, 670–682.
Hoffecker, J. F., 2009. The spread of modern humans in Europe.
Proceedings of the National Academy of Sciences, 106,
16040–16045.
Huckleberry, G., Hayashida, F., and Johnson, J., 2012. New insights
into the evolution of an intervalley prehistoric irrigation canal
system, north central Peru. Geoarchaeology, 27, 492–520.
Jacobs, Z., and Roberts, R. G., 2007. Advances in optically stimu-
lated luminescence dating of individual grains of quartz from
archeological deposits. Evolutionary Anthropology, 16,
210–223.
Jacobs, Z., Roberts, R. G., Galbraith, R. F., Deacon, H. J., Grün, R.,
Mackay, A., Mitchell, P., Vogelsang, R., and Wadley, L., 2008.
Ages for the Middle Stone Age of southern Africa: implications
for human behavior and dispersal. Science, 322, 733–735.
Jacobs, Z., Roberts, R. G., Nespoulet, R., El Hajraoui, M. A., and
Debénath, A., 2012. Single-grain OSL chronologies for Middle
Palaeolithic deposits at El Mnasra and El Harhoura 2 Morocco:
implications for Late Pleistocene human environment interac-
tions along the Atlantic coast of northwest Africa. Journal of
Human Evolution, 2012(62), 377–394.
Jain, M., and Ankjærgaard, C., 2011. Towards a non-fading signal
in feldspar: insight into charge transport and tunneling from
time-resolved optically stimulated luminescence. Radiation
Measurements, 46, 292–309.
Kim, J.–. C., Duller, G. A. T., Roberts, H. M., Wintle, A. G., Lee,
Y.-I., and Yi, S.-B., 2010. Re-evaluation of the chronology of
the palaeolithic site at Jeongokri, Korea, using OSL and TT-
OSL signals from quartz. Quaternary Geochronology, 5, 365.
Lahaye, C., Hernandez, M., Boëda, E., Felice, G. D., Guidon, N.,
Hoeltz, S., Lourdeau, A., Pagli, M., Pessis, A.-M., Rasse, M.,
and Viana, S., 2013. Human occupation in South America by
20,000 BC: the Toca da Tira Peia site, Piauí, Brazil. Journal of
Archaeological Science, 40, 2840–2847.
Lamothe, M., Barré, M., Huot, S., and Quimet, S., 2012. Natural
luminescence and anomalous fading in feldspar. Radiation
Measurements, 12, 682–687.
Lubinski, P. M., Feathers, J., and Lillquist, K., 2014. Single-grain
luminescence dating of sediment surrounding a possible late
Pleistocene artifact from the Wenas Creek Mammoth site,
Pacific Northwest, USA. Geoarchaeology (in press) 29:16–32.
Mercier, N., Le Quelle, J.-L., Hachid, M., Agsous, S., and Grenet,
M., 2012. OSL dating of quaternary deposits associated with
the parietal art of the Tassili-n-Ajjer plateau (central Sahara).
Quaternary Geochronology, 10, 367–373.
Morwood, M. J., Soejono, R. P., Roberts, R. G., Sutikna, T., Turney,
C. S. M., Westaway, K. E., Rink, W. J., Zhao, J.-X., van den
Bergh, G. D., Awe Due, R., Hobbs, D. R., Moore, M. W., Bird,
M. I., and Fifields, L. K., 2004. Archaeology and age of a new
hominin from Flores in eastern Indonesia. Nature,
431, 1087–1091.
Murray, A. S., Thomsen, K. J., Masuda, N., Buylaert, J.-P., and Jain,
M., 2012. Identifying well-bleached quartz using the different
bleaching rates of quartz and feldspar luminescence signals.
Radiation Measurements, 47, 688–695.
O’Connell, J. F., and Allen, J., 2004. Dating the colonization of
Sahul (Pleistocene Australia-New Guinea): a review of recent
research. Journal of Archaeological Science, 31, 835–853.
Outram, Z., Batt, C. M., Rhodes, E. J., and Dockrill, S. J., 2010. The
integration of chronological and archaeological information to
date building construction: an example from Shetland, Scotland,
UK. Journal of Archaeological Science, 37, 2821–2830.
Pei, S.-W., Gao, X., Feng, X.-W., Chen, F.-Y., and Dennell, R.,
2010. Lithic assemblages from the Jingshuiwan Paleolithic site
of the early Late Pleistocene in the Three Gorges, China. Quater-
nary International, 211, 66–74.
Petraglia, M., Korisettar, R., Boivin, N., Clarkson, C., Ditchfield, P.,
Jones, S., Koshy, J., Mirazón Lahur, M., Oppenheimer, C., Pyle,
D., Roberts, R., Schwenninger, J.-L., Arnold, L., and White, K.,
2007. Middle Paleolithic assemblages from the Indian subconti-
nent before and after the Toba super-eruption. Science, 317,
114–116.
Porat, N., Duller, G. A. T., Roberts, H. M., Piasetksy, E., and
Finkelstein, I., 2012. OSL dating in multi-strata Tel: Megiddo
(Israel) as a case study. Quaternary Geochronology, 10,
359–366.
Price, G. J., Webb, G. E., Zhao, J.-X., Feng, Y.-X., Murray, A. S.,
Cooke, B. N., Hocknull, S. A., and Sobbe, I. H., 2011. Dating
megafaunal extinction on the Pleistocene Darling Downs, east-
ern Australia: the promise and pitfalls of dating as a test of
extinction hypotheses. Quaternary Science Reviews, 30,
809–914.
Rhodes, E. J., Bronk Ramsey, C., Outram, Z., Batt, C., Willis, L.,
Dockrill, S., and Bond, J., 2003. Bayesian methods applied to
the interpretation of multiple OSL dates: high precision sediment
ages from Old Scatness Broch excavations, Shetland Isles. Qua-
ternary Science Reviews, 22, 1231–1244.
Rhodes, E. J., Fanning, P. C., and Holdaway, S. J., 2010.
Developments in optically stimulated luminescence age control
for geoarchaeological sediments and hearths in western
New South Wales, Australia. Quaternary Geochronology, 5,
348–352.
Richter, D., Tostevin, G., Škrdla, P., and Davies, W., 2009. New
radiometric ages for the early upper Palaeolithic type locality
of Brno-Bohunice (Czech Republic): comparison of OSL, IRSL,
TL and 14C dating results. Journal of Archaeological Science,
36, 708–720.
Rink, W. J., and Bartoll, J., 2005. Dating the geometric Nasca lines
in the Peruvian desert. Antiquity, 79, 390–401.
Rink, W. J., Dunbar, J. S., Tschinkel, W. R., Kwapich, C., Repp, A.,
Stanton, W., and Thulman, D. K., 2013. Subterranean transport
and deposition of quartz by ants in sandy sites relevant to age
overestimation in optical luminescence dating. Journal of
Archaeological Science, 40, 2217–2226.
Roberts, R. G., 1997. Luminescence dating in archaeology: from
origins to optical. Radiation Measurements, 27, 819–892.
Roberts, R., Bird, M., Olley, J., Galbraith, R., Lawson, E., Laslett,
G., Yoshita, H., Jones, R., Fullagar, R., Jacobsen, G., and Hua,
Q., 1998a. Optical and radiocarbon dating at Jinmium rock shel-
ter in northern Australia. Nature, 393, 358–362.
Roberts, R. G., Yoshida, H., Galbraith, R., Laslett, G., Jones, R., and
Smith, M., 1998b. Single-aliquot and single-grain optical dating
confirm thermoluminescence age estimates at Malakunanja II
rock shelter in northern Australia. Ancient TL, 16, 19–24.
Roberts, R. G., Lannery, T. F., Ayliffe, L. K., Yoshida, H., Olley,
J. M., Prideaux, G. J., Laslett, G. M., Baynes, A., Smith,
M. A., Jones, R., and Smith, B. L., 2001. New ages for the last
Australian megafauna: continent-wide extinction about 46,000
years ago. Science, 292, 1888–1892.
Rosenberg, T. M., Preusser, F., Fleitmann, D., Schwalb, A.,
Penkman, K., Schmid, T. W., Al-Shanti, M. A., Kadi, K., and
Matter, A., 2011. Humid periods in southern Arabia: windows
of opportunity for modern human dispersal. Geology,
39, 1115–1118.
Sanderson, D. C. W., and Murphy, S., 2010. Using simple portable
OSL measurements and laboratory characterisation to help
 LUMINESCENCE DATING, DEEP-SEA MARINE AND LACUSTRINE
understand complex and heterogeneous sediment sequences for
luminescence dating. Quaternary Geochronology, 5, 299–305.
Sanderson, D. C. W., Bishop, P., Stark, M., Alexander, S., and
Penny, D., 2007. Luminescence dating of canal sediments from
Angkor Borei, Mekong Delta, southern Cambodia. Quaternary
Geochronology, 2, 322–329.
Susino, G. J., 2010. Optical dating and microwaste – archaeological
applications. Quaternary Geochronology, 5, 306–310.
Tribolo, C., Mercier, N., Rasse, M., Soriano, S., and Huysecom, E.,
2010. Kobo 1 and L-Abri aux Vaches (Mali, West Africa): two
case studies for the optical dating of bioturbated sediments. Qua-
ternary Geochronology, 5, 317–323.
Tribolo, C., Mercier, N., Douville, E., Joron, J.-L., Reyss, J.-L.,
Rufer, D., Cantin, N., Lefrais, Y., Miller, C. E., Porraz, G.,
Parkington, J., Rigaud, J.-P., and Texier, P.-J., 2013. OSL and
TL dating of the Middle Stone Age sequence at Diepkloof Rock
Shelter (South Africa): a clarification. Journal of Archaeological
Science, 40, 3401–3411.
Waters, M. R., Forman, S. L., Stafford, T. W., and Foss, J., 2009.
Geoarchaeological investigations at the Topper and Big Pine
Tree sites, Allendale County, South Carolina. Journal of Archae-
ological Science, 36, 1300–1311.
Waters, M. R., Forman, S. L., Jennings, T. A., Nordt, L. C., Driese,
S. G., Feinberg, J. M., Keene, J. L., Halligan, J., Lindquist, A.,
Pierson, J., Hallmark, C. T., Collins, M. B., and Wiederhold,
J. E., 2011. The Buttermilk Creek complex and the origins
of Clovis at the Debra L. Friedkin site, Texas. Science,
331, 1599–1603.
Westaway, K. E., 2009. The red, white and blue of quartz lumines-
cence: a comparison of De values derived for sediments from
Australia and Indonesia using thermoluminescence and optically
stimulated luminescence emission. Radiation Measurements,
44, 462–466.
Yoshida, H., Roberts, R. G., and Olley, J. M., 2003. Progress
towards single-grain optical dating of fossil mud-wasp nests
and associated rock art in northern Australia. Quaternary Sci-
ence Reviews, 22, 1273–1278.
Cross-references
Luminescence Dating
Luminescence Dating, Single-Grain Dose Distribution
Luminescence, Coastal Sediments
Luminescence, Colluvial Sediments
Luminescence, Fluvial Sediments
Luminescence, Geomorphological Processes
Luminescence, Soils
LUMINESCENCE DATING, DEEP-SEA MARINE AND
LACUSTRINE
Helen M. Roberts
Department of Geography and Earth Sciences,
Aberystwyth University, Aberystwyth, UK
Synonyms
Infrared-stimulated luminescence (IRSL); Optical dating;
Optically stimulated luminescence (OSL); Thermolumi-
nescence (TL)
409
Definition
Luminescence dating: A family of chronologic methods
typically applied to the commonly occurring minerals
quartz and feldspar, which exploits a time-dependent sig-
nal that builds up in mineral grains by exposure to natu-
rally occurring ionizing radiation (principally from
uranium, thorium, and potassium). The methods assess
the time elapsed since these mineral grains were last
exposed to sunlight or to heating. In the case of marine
and lacustrine sediments, the event being dated is the last
exposure to sunlight, i.e., typically the time of deposition
of the sediments.
Deep-sea, marine: Of or pertaining to the deeper parts of
the sea or ocean (as opposed to shallow waters and coasts).
Lacustrine: Of or pertaining to lakes.
Introduction
Despite the fact that marine sediments were among the
first sediments from which a luminescence signal was
observed (Wintle and Huntley, 1979), subsequently little
work has been done using luminescence to date marine
sediments. Similarly, surprisingly little work has been
done to apply luminescence procedures to date lacustrine
deposits, in spite of the rapid development, the widespread
uptake, and the improved accuracy and precision of lumi-
nescence dating when applied to a wide variety of other
depositional settings. The reasons behind this reluctance
to use luminescence techniques to provide ages for marine
and lacustrine sediments share some common links, and
the challenges faced in dating these water-lain sediments
are similar. It is therefore appropriate that these two depo-
sitional environments are considered together.
Historical overview of luminescence dating in
deep-sea marine and lacustrine environments
Deep-sea and marine sediments
The term “luminescence dating” is a broad concept which
has itself developed over time, and today the term actually
incorporates a family of methods which have been devel-
oped over the last ~35–45 years for dating in different set-
tings including sedimentary environments. These various
luminescence dating methods use different signals (some
of which are stimulated by heat, “thermoluminescence”
(TL), and some by light, “optically stimulated lumines-
cence” (OSL)) and different measurement protocols to
determine the final numerical age estimate. A more
detailed consideration of luminescence dating (see “Lumi-
nescence Dating”) and its history (see “Luminescence
Dating, History”) are given elsewhere in this encyclope-
dia. However, it is interesting to note here that the poten-
tial of TL signals from quartz and feldspars to derive
a numerical age for the time of deposition of marine sedi-
ments was first noted in the western literature using
marine sediments. Wintle and Huntley (1979) observed
an increasing TL signal with increasing depth for a core
from the Crozet Plateau in the Antarctic Ocean and recog-
nized that this TL signal was coming from inorganic
410 LUMINESCENCE DATING, DEEP-SEA MARINE AND LACUSTRINE
sediment. Furthermore, they noted that exposure to sun-
light could act as a mechanism for resetting the TL signal
prior to deposition of the sediments. Comparing the TL
ages generated against independent age control, Wintle
and Huntley (1979) concluded that the findings
were “sufficiently encouraging to suggest that the method
is basically sound.” Following this pioneering work, TL
was then rapidly applied to terrestrial sediments, particu-
larly loess (see entry for “Luminescence Dating, Loess”
in this encyclopedia), but it was to be more than 20 years
until further work was done in deep-sea or marine settings
using luminescence signals of any kind.
In the intervening 20 years, luminescence dating
underwent a series of radical transformations (see “Lumines-
cence Dating, History”), the most critical being the develop-
ment of optical stimulation to yield a signal appropriate for
dating sediments, and the development of sensitivity-
corrected methods for measuring the luminescence signal
from single aliquots (e.g., the single-aliquot regenerative
dose (SAR) measurement protocol of Murray and Wintle
(2000), which has subsequently been adapted and developed
for various luminescence signals and materials). Following
the introduction of the SAR protocol for quartz, and compar-
ison of the ages generated against independent age control
(e.g., Murray and Olley, 2002; Rittenour, 2008), confidence
grew in the OSL signal from quartz for dating and in the abil-
ity to look at complex equivalent dose (De) distributions;
attention could now be turned to more challenging environ-
ments, including the marine realm.
Stokes et al. (2003) were the first to apply a SAR OSL
measurement protocol to quartz prepared from deep-sea
marine sediments, studying cores taken from the Western
Arabian Sea. The fine-grained quartz ages generated were
in agreement with independent age control provided by
radiocarbon, tephrostratigraphy, and paleoecological data
(Stokes et al., 2003). The fine-grained (4–11 mm diameter)
quartz was believed to be eolian (i.e., wind-blown) in ori-
gin, and as such is likely to be well bleached prior to depo-
sition. Using fine grains for OSL dating precludes the
possibility of investigating the degree of bleaching
because of the large number (>105) of grains on each ali-
quot. In spite of this, it is interesting to note that Stokes
et al. (2003) still observed >4 % of aliquots with De values
>3 standard deviations from the mean estimate, and they
used this rather crude approach to screen and remove out-
lying De data; today a more sophisticated statistical
approach would be taken (e.g., the Minimum Age Model,
Galbraith et al. 1999; see “Luminescence Dating, Single-
Grain Dose Distribution”). Such an approach was taken
by Olley et al. (2004) who examined the degree of
bleaching of sediments from the eastern Indian Ocean
using single grains of fine-sand-sized (60–70 mm) quartz.
In spite of being eolian in origin and having relatively long
transport distances (and it would be thought, ample oppor-
tunity for bleaching), some samples were incompletely
bleached on deposition. However, following appropriate
statistical treatment, the single-grain quartz OSL ages gen-
erated were in agreement with independent age control
provided by radiocarbon, between ~1,500 and 50,000
years ago (Olley et al., 2004). If problems of incomplete
bleaching prior to deposition are found to be an issue in
these nonpolar marine settings where eolian transport
can dominate, it should perhaps come as little surprise that
problems of incomplete bleaching and age overestimation
have also been reported in polar marine settings (e.g.,
Jakobsson et al., 2003; Berger, 2006), where daylight
can be restricted and where transport pathways are more
complex and can occur in association with ice and turbid
water. Nevertheless, some of these OSL ages have been
found to be in agreement with the expected ages (e.g.,
Jakobsson et al., 2003). However, it was not possible to
develop criteria based on luminescence evidence alone
by which accurate and inaccurate luminescence ages
could be distinguished.
More recently, quartz grains of fine-silt (4–11 mm),
coarse-silt (~38–63 mm), and fine-sand (63–90 or
90–125 mm) size have been analyzed in nonpolar deep-
sea marine settings, and these yield SAR OSL ages which
are typically in agreement with independent age control or
expected ages, e.g., the Sea of Okhotsk (Sugisaki
et al., 2010a; Sugisaki et al., 2010b) and the south Bohai
Sea, China (Yi et al., 2012a; Yi et al., 2012b). A notable
exception is the work undertaken in the south Bohai Sea
where radiocarbon ages of 38–44 cal kyr BP have been
used to assign a particular sedimentary unit to Marine Oxy-
gen Isotope Stage (MIS)-3 (Liu et al., 2009), while OSL
ages suggest that the unit actually relates to MIS-5
(Yi et al., 2012b) and hence imply that the radiocarbon ages
may be at the limit of the method and thus unreliable. This
example highlights the potential value of luminescence dat-
ing in any depositional setting where it can be used to
extend the numerical chronology for a sedimentary
sequence beyond the upper (and indeed lower) limit of
radiocarbon dating. Where agreement can be established
between luminescence ages and any other independent
chronology in a time span clearly appropriate for both
methods (e.g., in Holocene sediments from the south Bohai
Sea; Yi et al., 2012b), then there is additional confidence
that as one chronometer approaches its limit, then the sec-
ond chronometer can be used to extend the chronology.
In carbonate-restricted zones such as high-
accumulation-rate deposits where there is a lack of bio-
genic material, and in deep areas of the oceans below the
carbonate compensation depth (CCD), there is even
greater incentive to use luminescence techniques to gener-
ate a numerical chronology. Given the relatively limited
number of luminescence dating studies conducted thus
far in marine settings, further studies are probably required
in order to build up confidence in the ability of lumines-
cence dating to generate accurate ages in settings where
no independent chronology exists. However, the initial
studies that have been conducted are extremely promising.
 LUMINESCENCE DATING, DEEP-SEA MARINE AND LACUSTRINE
Lacustrine sediments
Lakes preserve some of the best terrestrial records of cli-
mate and environmental change on Earth, containing
numerous proxy records which can be seen to vary over
time. Surprisingly little work has been done to apply lumi-
nescence techniques to date lacustrine sediments. Instead,
such sediments have hitherto primarily been dated using
radiocarbon techniques, and extrapolation is frequently
used to infer the age of sediments beyond ~40–50,000
years ago (e.g., Burnett et al., 2011). However, given the
potentially long time period over which lake sediments
can accumulate, and the comparatively short time range
over which radiocarbon can be applied, there is tremen-
dous scope for the use of luminescence techniques for dat-
ing lacustrine deposits.
The reluctance to apply luminescence techniques to
date lacustrine sediments is probably linked primarily to
concerns over the degree of bleaching of the sediments
prior to deposition. For this reason, where luminescence
techniques have been used to date lacustrine sediments,
it is primarily lake shoreline deposits (rather than lake
floor sediments) that have been dated thus far (e.g.,
Burrough et al., 2009). These concerns over incomplete
bleaching of the luminescence signals prior to deposition
were particularly pertinent in the early days of lumines-
cence dating when TL signals were used, because the TL
signal is known to bleach much more slowly in nature than
the OSL signal. Additionally, bleaching in the water col-
umn adds an extra layer of complication due to the turbid-
ity of the water and the attenuation of the spectrum with
increasing depth through the water column. Lake mor-
phology (i.e., shape) is therefore likely to be an important
factor in determining the success of a lacustrine lumines-
cence study; for example, an early study by Berger
et al. (1987) demonstrated that large lakes with broad,
low-gradient lake floors were likely to be most appropri-
ate. In spite of these concerns over bleaching prior to
deposition, particularly using TL signals, the successful
application of luminescence dating of lacustrine sediments
is demonstrated by Berger and Anderson (1994) for Squir-
rel Lake, Alaska, where TL and radiocarbon ages were
found to be in agreement for the upper 3 m of sediment,
which was within the range of radiocarbon dating; beyond
this, a further 11 m of sediment was dated using TL with
the maximum age being 125  31 ka (Berger and
Anderson, 1994). The IRSL signal from feldspars
bleaches more rapidly than the TL signal, and for this rea-
son, Berger and Doran (2001) favored the IRSL signal for
dating lacustrine sediments from Lake Hoare, Antarctica,
which gave IRSL ages that were younger than TL ages
for corresponding samples. Even though these IRSL ages
were younger than the TL ages generated, there is still
some question as to whether the IRSL signal was
completely bleached prior to deposition because examina-
tion of a modern analogue sample (which should give an
age which is essentially zero years) returned an age of
~600 years. The OSL signal from quartz bleaches even
411
more rapidly than that of feldspar and also offers the
advantage that it does not suffer from anomalous fading
(which can cause age underestimation in feldspars), and
hence today since the advent of SAR methods (Murray
and Wintle, 2000), quartz would probably be the mineral
of choice for dating lake sediments. An example of the
successful use of quartz OSL dating to determine the ages
of lacustrine sediments is provided by Long et al. (2011)
for Qingtu Lake, in northern China, where good agree-
ment was obtained between OSL ages from coarse-silt-
sized quartz and radiocarbon ages derived from shell frag-
ments and organic-rich bulk sediments.
Although the quartz OSL signal bleaches most rapidly
of all the commonly used signals from the family of lumi-
nescence dating techniques, this fast component OSL sig-
nal is also the fastest to saturate in nature, i.e., it has the
youngest upper limit for dating (~150 Gy (Chapot
et al., 2012), which is likely to correspond to
a maximum age for lacustrine sediments of ~130 ka).
For example, Lowick et al. (2010) and Lowick and
Preusser (2011) generate fine-grained quartz OSL ages
that are in agreement with independent biochronologic
evidence to ~140 Gy (which in this study corresponds to
an age of only ~70 ka), but beyond this the OSL ages are
increasingly underestimated. For this reason, Zander and
Hilgers (2013) explored the use of various different lumi-
nescence signals to date their Middle- to Upper-
Pleistocene sediments (~145–760 ka) from Lake
El’gygytgyn, Russia. In this study, the quartz OSL signals
were saturated and hence gave ages which significantly
underestimated the expected age based on magnetostra-
tigraphy, and the IRSL signals measured at 50  C from
polymineral fine grains also gave rise to age underestima-
tions throughout the study for similar reasons. The ther-
mally transferred OSL (“TT-OSL”) quartz ages were also
unsatisfactory as they were found to be of variable reliabil-
ity. However, ages generated using a post-IR IRSL290
measurement protocol (Thiel et al., 2011) which mini-
mizes anomalous fading were in agreement with the
expected ages back to ~465 ka and potentially even to
~760 ka (Zander and Hilgers, 2013).
Practical considerations, concerns, and benefits of
using luminescence techniques to date deep-sea
marine and lacustrine sediments
Several practical considerations and concerns have contrib-
uted to the slow uptake of luminescence techniques for dat-
ing deep-sea marine and lacustrine sediments. However,
there are several key benefits offered by luminescence dat-
ing which it can be argued now outweigh these concerns.
Practicalities
Deep-sea marine and lake floor sediments are typically
obtained as cores, often requiring deep drilling to obtain
them. This logistically complex and hence also relatively
expensive method of retrieval leads to further complica-
tions. Being taken in cores, the material is often relatively
412 LUMINESCENCE DATING, DEEP-SEA MARINE AND LACUSTRINE
limited in terms of the sample size available. The wide
variety of proxies available from lake cores also means
that the precious core material is required for numerous
different analyses, which again restricts the amount of
core material available for any one technique. In many
studies, cores are often split and logged and exposed to
daylight prior to sampling for luminescence dating. While
light-exposed cores can still be successfully sampled for
luminescence dating (by removing the outer, light-
exposed portions of the core which can be used for dose-
rate determination), it should be noted that X-ray scanning
prior to sampling for luminescence can add a radiation
dose to sample material, potentially leading to erroneous
age determinations (Davids et al., 2010).
Concerns
One of the principal concerns regarding the use of lumines-
cence dating in deep-sea marine and lacustrine sediments is
the accurate assessment of the water content over the post-
depositional history of the sediments; this is important
because water attenuates the radiation dose reaching the
sediment grains and hence influences the age determina-
tion. Uncertainty in the water content is typically the largest
single source of uncertainty in luminescence ages generated
for deep-sea marine and lake floor sediments. The water
content of such water-lain sediments can be extremely var-
iable through the sedimentary sequence and also changes
over time, with the lowermost sediments becoming increas-
ingly compacted as new sediments accumulate above, caus-
ing dewatering and hence a reduction in water content over
time. Compaction of the sediments can also occur on
retrieval of sedimentary cores, and care must therefore be
taken to assess the degree of compaction during the drilling
process to facilitate the reconstruction of water content after
sampling. It is also important to consider the changing
depth of water (and sediment) above the chosen sample
location within the sedimentary profile because it will influ-
ence the cosmic dose rate (and the dose rate from adjacent
sediments) reaching the sample over time.
Another key concern is the degree of bleaching of any
preexisting luminescence signal prior to deposition in
deep-sea marine and lacustrine settings; investigating this
is particularly difficult in these settings due to the typically
fine-grained nature of the sediments precluding the determi-
nation of De values for individual grains which would
otherwise be used to assess the degree of bleaching.
An alternative approach is to compare the De values
obtained using different luminescence signals from the same
sample and exploit the different bleaching rates of the lumi-
nescence signals as a means of identifying incompletely
bleached sedimentary units. For example, Buylaert
et al. (2013) proposed using the ratio of IR50 and pIRIR290
De values throughout a lacustrine sediment profile at Laguna
Potrok Aike, Argentina, to screen out incompletely bleached
samples; the ages which passed this screening criterion
(29 ages out of 47 analyzed) are chronostratigraphically
consistent and show good agreement with independent age
control.
A further concern is the geochemistry of deep-sea
marine and lacustrine settings, which can complicate the
assessment of the dose rate. In such cases, the assumptions
typically made regarding the presence of secular equilib-
rium in the uranium- and thorium-decay series chains
and thus a constant dose rate (Gy/ka) over time would
not hold true, and hence a more detailed assessment of
the dose rate over time would need to be conducted to
account for the unsupported and supported radionuclides
(e.g., Stokes et al., 2003).
Benefits
Given the concerns outlined above, and the relatively small
number of luminescence studies in deep-sea marine and
lacustrine settings thus far, it would seem prudent for new
studies to have some form of independent age control for
verification of the luminescence ages generated. However,
this will not be possible in all cases, and indeed the lack
of age control in these settings is a key motivation for apply-
ing luminescence dating here. Where other chronologic evi-
dence does exist, it can sometimes be complicated by
problems within that technique itself (e.g., radiocarbon res-
ervoir effects). Notwithstanding these issues, luminescence
dating of deep-sea marine and lacustrine sediments is
becoming an increasingly attractive prospect, particularly
given advances in equipment and methodology since the
year 2000, and also in the accuracy and precision of the
ages generated. Luminescence dating offers a number of
internal quality control checks that can be performed to
offer assurance of the suitability of the signals used for dat-
ing (e.g., Roberts, 2008), and unlike some other methods, it
is not beset by issues of calibration. Luminescence dating is
also an attractive method because the two minerals princi-
pally used for dating (quartz and feldspar) are very com-
mon, so it is rarely problematic to find sufficient material
for dating. Furthermore, luminescence dating provides
a numerical chronology and hence allows the assessment
of accumulation rates through long sedimentary sequences
without the need for extrapolation and the inherent assump-
tion of a constant accumulation rates over time periods
typically from the present day to 105 years.
Summary and conclusions
Relatively few studies of deep-sea marine or lacustrine sed-
iments have taken place using luminescence dating. The
most successful approach demonstrated thus far is the use
of quartz OSL signals for dating, measured use of a single-
aliquot sensitivity-corrected method such as SAR, although
newly developed post-IR IRSL methods developed to min-
imize anomalous fading in feldspars also show great prom-
ise. The luminescence ages generated tend to be in good
agreement with the independent age control. Given the dif-
ficulties of dating marine and lacustrine sediments using
many other methods (e.g., “Lacustrine Environments
(14C)”, “210Pb Dating,” “Amino Acid Racemization, Marine
 LUMINESCENCE DATING, DEEP-SEA MARINE AND LACUSTRINE
Sediments”, “Radiocarbon Dating of Marine Carbonates”,
“Carbonates, Lacustrine (U-Series),” “Carbonates, Marine
Carbonates (U-Series)”), and the particular problems that
can be associated with the use of radiocarbon in some set-
tings (Radiocarbon Dating), luminescence dating could
potentially make an important contribution to the study of
deep-sea marine and lacustrine sediments in the future.
Bibliography
Berger, G. and Doran, P., 2001. Luminescence-dating zeroing tests
in Lake Hoare, Taylor Valley, Antarctica. Journal of
Palaeolimnology, 25, 519–529.
Berger, G. W., 2006. Trans-arctic-ocean tests of fine-silt lumines-
cence sediment dating provide a basis for an additional
geochronometer for this region. Quaternary Science Reviews,
25, 2529–2551.
Berger, G. W., and Anderson, P. M., 1994. Thermoluminescence
dating of an Arctic lake core from Alaska. Quaternary Geochro-
nology (Quaternary Science Reviews), 13, 497–501.
Berger, G. W., Clague, J. J., and Huntley, D. J., 1987. Thermolumi-
nescence dating applied to glaciolacustrine sediments from cen-
tral British Columbia. Canadian Journal of Earth Sciences,
24, 425–434.
Burnett, A. P., Soreghan, M. J., Scholz, C. A., and Brown, E. T.,
2011. Tropical East African climate change and its relation to
global climate: a record from Lake Tanganyika, Tropical East
Africa, over the past 90+ kyr. Palaeogeography, Palaeocli-
matology, Palaeoecology, 303, 155–167.
Burrough, S. L., Thomas, D. S. G., and Bailey, R. M., 2009. Mega-
Lake in the Kalahari: a Late Pleistocene record of the Palaeolake
Makgadikgadi system. Quaternary Science Reviews, 28,
1392–1411.
Buylaert, J.-P., Murray, A. S., Gebhardt, A. C., Sohbati, R.,
Ohlendorf, C., Thiel, C., Wastegård, S., and Zolitschka, B.,
2013. Luminescence dating of the PASADO core 5022-1D from
Laguna Potrok Aike (Argentina) using IRSL signals from feld-
spar. Quaternary Science Reviews, 71, 70–80.
Chapot, M. S., Roberts, H. M., Duller, G. A. T., and Lai, Z. P., 2012.
A comparison of natural- and laboratory-generated dose response
curves for quartz optically stimulated luminescence signals from
Chinese Loess. Radiation Measurements, 47, 1045–1052.
Davids, F., Roberts, H. M., and Duller, G. A. T., 2010. Is X-ray core
scanning non-destructive? Assessing the implications for opti-
cally stimulated luminescence (OSL) dating of sediments. Jour-
nal of Quaternary Science, 25, 348–353.
Galbraith, R. F., Roberts, R. G., Laslett, G. M., Yoshida, H., and
Olley, J. M., 1999. Optical dating of single and multiple grains
of quartz from Jinmium rock shelter, northern Australia: part I,
experimental design and statistical models. Archaeometry,
41, 339–364.
Jakobsson, M. J., Backman, J., Murray, A., and Løvlie, R., 2003.
Optically stimulated luminescence dating supports central Arctic
Ocean cm-scale sedimentation rates. Geochemistry, Geophysics,
Geosystems, 4, 1016, doi:10.1029/2002GC000423.
Liu, J., Saito, Y., Wang, H., Zhou, L., and Yang, Z., 2009. Strati-
graphic development during the Late Pleistocene and Holocene
offshore of the Yellow River delta, Bohai Sea. Journal of Asian
Earth Sciences, 36, 318–331.
Long, H., Lai, Z.-P., Wang, N.-A., and Zhang, J.-R., 2011.
A combined luminescence and radiocarbon dating study of
Holocene lacustrine sediments from arid northern china. Quater-
nary Geochronology, 6, 1–9.
Lowick, S., and Preusser, F., 2011. Investigating age underestima-
tion in the high dose region of optically stimulated luminescence
using fine grain quartz. Quaternary Geochronology, 6, 33–41.
413
Lowick, S., Preusser, F., Pini, R., and Ravazzi, C., 2010. Underesti-
mation of fine grain quartz OSL dating towards the Eemian:
comparison with palynostratigraphy from Azzano Decimo,
northeastern Italy. Quaternary Geochronology, 5, 583–590.
Murray, A. S., and Olley, J. M., 2002. Precision and accuracy in the
optically stimulated luminescence dating of sedimentary quartz:
a status review. Geochronometria, 21, 1–16.
Murray, A. S., and Wintle, A. G., 2000. Luminescence dating of
quartz using an improved single-aliquot regenerative-dose pro-
tocol. Radiation Measurements, 32, 57–73.
Olley, J. M., De Deckker, P., Roberts, R. G., Fifield, L. K., Yoshida,
H., and Hancock, G., 2004. Optical dating of deep-sea sediments
using single grains of quartz: a comparison with radiocarbon.
Sedimentary Geology, 169, 175–189.
Rittenour, T. M., 2008. Luminescence dating of fluvial deposits:
applications to geomorphic, palaeoseismic and archaeological
research. Boreas, 37, 613–635.
Roberts, H. M., 2008. The development and application of lumines-
cence dating to loess deposits: a perspective on the past, present
and future. Boreas, 37, 483–507.
Stokes, S., Ingram, S., Aitken, M. J., Sirocko, F., Anderson, R., and
Leuschner, D., 2003. Alternative chronologies for Late Quater-
nary (Last Interglacial-Holocene) deep sea sediments via optical
dating of silt-sized quartz. Quaternary Science Reviews, 22,
925–941.
Sugisaki, S., Buylaert, J. P., Murray, A. S., Harada, N., Kimoto, K.,
Okazaki, Y., Sakamoto, T., Iijima, K., Tsukamoto, S., Miura, H.,
and Nogi, Y., 2010a. High resolution optically stimulated lumi-
nescence dating of a sediment core from the southwestern Sea
of Okhotsk. Geochemistry, Geophysics, Geosystems, 13,
Q0AA22, doi:10.1029/2011GC0004029.
Sugisaki, S., Buylaert, J. P., Murray, A. S., Tsukamoto, S., Nogi, Y.,
Miura, H., Sakai, S., Iijima, K., and Sakamoto, T., 2010b. High
resolution OSL dating back to MIS 5e in the central Sea of
Okhotsk. Quaternary Geochronology, 5, 293–298.
Thiel, C., Buylaert, J.P., Murray, A.S., Terhorst, B., Hofer, I.,
Tsukamoto, S., Frechen, M., 2011. Luminescence dating of the
Stratzing loess profile (Austria)-Testing the potential of an ele-
vated temperature post-IR IRSL protocol. Quaternary Interna-
tional, 234, 23–31.
Wintle, A. G., and Huntley, D. J., 1979. Thermoluminescence dat-
ing of a deep-sea sediment core. Nature, 279, 710–712.
Yi, L., Yu, H.-J., Ortiz, J. D., Xu, X.-Y., Chen, S. L., Ge, J.-Y., Hao,
Q.-Z., Yao, J., Shi, X.-F., and Peng, S.-Z., 2012a. Late Quater-
nary linkage of sedimentary records to three astronomical
rhythms and the Asian monsoon, inferred from a coastal bore-
hole in the south Bohai Sea, China. Palaeogeography,
Palaeoclimatology, Palaeoecology, 329–330, 101–117.
Yi, L., Lai, Z., Yu, H., Xu, X., Su, Q., Yao, J., Wang, X., and Shi, X.,
2012b. Chronologies of sedimentary changes in the south Bohai
Sea, China: constraints from luminescence and radiocarbon dat-
ing. Boreas, 42, 267–284.
Zander, A., and Hilgers, A., 2013. Potential and limits of OSL,
TT-OSL, IRSL and pIRIR290 dating methods applied on
a Middle Pleistocene sediment record of Lake El’gygytgyn,
Russia. Climate of the Past, 9, 719–733.
Cross-references
210
Pb Dating
Amino Acid Racemization, Marine Sediments
Carbonates, Lacustrine (U-Series)
Feldspars
Lacustrine Environments (14C)
Luminescence Dating
Luminescence Dating, History
Luminescence Dating, Single-Grain Dose Distribution
414 LUMINESCENCE DATING, DOSE RATES
Luminescence, Coastal Sediments
Quartz
Radiocarbon Dating
Sediments, Terrestrial (Paleomagnetism)
LUMINESCENCE DATING, DOSE RATES
Guillaume Guérin
Centre for Nuclear Technologies, Technical University of
Denmark, Roskilde, Denmark
Definition
Dose rate: the amount of energy absorbed by a mineral
that can be used as a natural dosimeter per unit time and
mass, resulting from naturally occurring radiation both
in the sample and its environment. Unit: Gy.a1 (more
commonly given in Gy.ka1).
Radioactivity: sources and types
The focus of dose rate measurements in luminescence dat-
ing is to best quantify the amounts of naturally occurring
radiation that a dated sample has been exposed to through
time. It is well understood that naturally occurring radia-
tion is generated by the decay of radioactive isotopes
stored in rocks and sedimentary bodies that make up the
Earth’s surface. The most common source of such radioac-
tive energy comes from potassium (K), uranium (U), and
thorium (Th). 40K is a radioisotope of potassium with
a half-life of 1.25 109 years; 232Th (half-life, 1.41 1010
years), 238U (half-life, 4.47 109 years), and 235U (half-life,
7.04 108 years) are the top members of distinct series of
radioelements. 40K decays in stable daughter products
and is a source of both beta and gamma radioactivity. In
comparison to the potassium decay, the radioactive series
of thorium and uranium (232Th, 238U, and 235U) are more
complex as they both comprise several alpha and beta
emitters (see, e.g., Aitken, 1985, for the decay schemes);
gamma rays are also emitted in the de-excitation of nuclei,
so these series contribute alpha, beta, and gamma dose
rates. In an ideal case of a closed system, all radioelements
in the series are at equilibrium, i.e., all the members
of each chain have the same nuclear activity, defined
by the concentration and half-life of the parent.
As a consequence, the nuclear activity of all the decay series
members from U and Th can be deduced from the concen-
tration, or activity, of the parent. In typical sediments, beta
and gamma dose rates contribute roughly two-third and
one-third, respectively, of the total dose rates to sand-sized
quartz grains; more detailed analysis of the repartition of
dose rates can be found in Ankjærgaard and Murray
(2007). In some cases (see below), alpha dose rates can also
contribute a significant fraction of dose rates.
Beta and gamma radiations have low ionizing powers,
i.e., the linear energy transfer is rather small compared to
other radiation types such as alpha particles. As
a consequence, both beta and gamma dose rates are com-
monly assumed to have the same efficiency in producing
luminescence in a given mineral grain. For example, the
amount of luminescence induced by a 1 Gy.ka1 beta dose
rate from 40K is assumed to be equivalent to the amount of
luminescence induced by a 1 Gy.ka1 gamma dose rate
from the 232Th series. Conversely, alpha particles are
heavy, highly ionizing particles which saturate along their
tracks the electron traps and recombination centers (the
energy stores of the mineral grain) in the mineral grains
at a much faster rate. As a consequence, their efficiency
in producing luminescence per unit dose is much lower
than that of gamma and beta radiations and has to be taken
into account for accurate determination of effective alpha
dose rates (see, e.g., Tribolo et al., 2001).
Another source dose rate to mineral grains is that of
cosmic radiation. The contribution from cosmic radiation
has been tabulated as a function of longitude, latitude, alti-
tude (the two latter factors being the most influent), and
burial depth of the dated samples, as the cosmic radiation
is attenuated in the ground (Prescott and Hutton, 1988).
With a focus on burial depth, the contributions from cos-
mic radiation are classically grouped into a “soft” and
a “hard” component. The “soft” component becomes
totally attenuated within several tens of centimeters,
whereas the “hard” component is capable of penetrating
much further into the ground (several meters). Cosmic
dose rates generally contribute a low percentage of the
total dose rates (<10 % in most cases).
Infinite matrix approximation
The most important concept behind the calculation of dose
rates in a sedimentary medium is that of the infinite matrix
theory (Fano, 1954; Roesch and Attix, 1968; Aitken,
1985). In a radioactive, homogeneous medium, whose
dimensions are greater than the range of ionizing radia-
tions, all the energy that is emitted per unit time and mass
in the medium is also absorbed in the same medium. In
other words, the infinite matrix dose rate in any volume
of homogeneous sediment is equal to the rate of energy
emitted per unit time and mass in that same volume.
Ultimately, from the knowledge of the radioelements’
nuclear properties (nuclear activity, average energy emit-
ted per disintegration), applying this assumption to
a known measurement of the concentration of the main
radioelements (K, U, and Th) within the volume of sedi-
ment allows one to convert these concentrations into infi-
nite matrix dose rates for the homogeneous medium
(Guérin et al., 2011).
Measurement
As a consequence, spectrometry techniques are the most
widely used techniques to determine dose rates: different
available techniques include high-resolution gamma
spectrometry, neutron activation analysis (NAA), and
 LUMINESCENCE DATING, DOSE RATES
inductively coupled plasma mass spectrometry (ICP-MS).
In all methods the concentrations of K, U, and Th are
determined for dry samples in the laboratory and resul-
tantly converted in infinite matrix dose rates. Another set
of techniques, called counting techniques, exploits the
proportionality between disintegration count rates and
dose rates: for example, beta counting techniques
(Ankjærgaard and Murray, 2007) and thick source alpha
counting (TSAC), or threshold techniques used to derive
gamma dose rates from portable field gamma spectrome-
ters (Guérin and Mercier, 2011). However, these tech-
niques generally provide less information than spectra
analysis, since the different contributions to the calculated
dose rates are not individually quantified.
Sample heterogeneity and implications for dose
rate calculations
The infinite matrix theory assumes homogeneity of the
sample environment from a radiation point of view. How-
ever, alpha, beta, and gamma radiations travel at different
distances within a given sample and around it, so this uni-
formity assumption has different meanings. Alpha parti-
cles have the ability to travel only short distances within
a sample, at typically less than ~20 mm. In comparison,
lighter beta particles travel up to 2–3 mm and gamma rays
travel greater distances, up to 40–50 cm. Departure from
homogeneity over these different scales has notable con-
sequences on the accuracy of the corresponding dose
rates. It is, therefore, of paramount importance to attempt
to avoid the effects of sample heterogeneity in all stages
of sample collection. Primarily this is achieved in the field
stage by collecting samples from a sedimentary unit that is
both uniform in grain size and composition within an area
of roughly 1 m3 and devoid of secondary processes such
as pedogenesis, bioturbation, or carbonate replacement,
which may act to change the rates of radiation transfer
within a given volume of sediment.
One of the most common, and generally unavoidable,
sources of dose rate heterogeneities is the presence of
water. Primarily, water acts to dilute the concentrations
of the radioelements in the sediment, hence requiring an
adjustment of the dose rates when determined from labo-
ratory measurements performed on dry samples (as in
the case of, e.g., laboratory spectrometry techniques). In
addition to this, electron stopping powers and photon
interaction cross sections are higher in water than in typi-
cal soil constituents. The ratio of these physical quantities
provides a correction factor to account for this effect
(Zimmermann, 1971). Aitken and Xie (1990) have aver-
aged these ratios on the beta and gamma spectra in typical
sediments, and more recently Nathan and Mauz (2008)
have proposed new working values based on Monte Carlo
simulations.
The general dosimetric concepts presented so far apply
to most minerals dated by luminescence. However, con-
stituent rock and/or mineral grains that make up
415
a sedimentary unit have their own distinct characteristics
in terms of dose rate determination. For example, flints
usually have homogeneous spatial distributions of radioel-
ements at the scale of both alpha and beta radiations
(Schmidt et al., 2013) and are thus simple dosimeters from
an alpha and beta dose rate calculation perspective.
However, such rocks are often embedded in sediments
of different characteristics, both in terms of chemical
composition and radioactivity. Given the range of gamma
rays (40–50 cm), gamma dose rates to rock samples thus
mainly depend on the characteristics of their environment.
As a consequence, direct measurements of gamma dose
rates in the field (using either field gamma spectrometry
techniques or artificial dosimeters) are often preferred to
determine gamma dose rates to such samples; an alterna-
tive consists of determining gamma dose rates from sedi-
ment samples taken close to the rock samples.
Many mineral forms such as quartz and potassium feld-
spar that are commonly utilized in luminescence dating
are more complicated in terms of dose rate determination.
This is a result of their chemical composition and grain
dimensions which ultimately break the key infinite matrix
assumption of sample homogeneity. With focus on alpha
and beta dose rate contributions, quartz grains contain
very few radioelements (Vandenberghe et al., 2008) and
thus represent cold spots in sedimentary matrices. In com-
parison, due to their high potassium content, feldspar
grains act as radioactive sources themselves, which gener-
ate an internal dose rate that must be accounted for in final
dose rate calculations. Internal dose rate to potassium feld-
spar grains can be calculated using their potassium content
(Huntley and Baril, 1997) together with dose rate conver-
sion factors (Guérin et al., 2011). The self-dose fraction of
the infinite matrix dose rate has been tabulated by Guérin
et al. (2012). The small but significant contribution from
87
Rb can be calculated according to Readhead (2002)
and Huntley and Hancock (2001). Finally, the diameter
of the mineral (quartz and feldspar) grains commonly
studied in luminescence (typically ~100–250 mm) is not
negligible compared to the range of beta particles. As
a consequence, the grains attenuate beta radiation from
the surrounding matrix. Grain-size-dependent attenuation
factors of external beta dose rates from the environment
can be found in Guérin et al. (2012).
Due to the small range of alpha particles, they get
strongly attenuated in sand-sized grains and their contri-
bution is negligible ~20 mm in the grains; furthermore,
homogeneity of the environment of the grains is very dif-
ficult to assess at this scale. It is thus common to bypass
potentially inaccurate dose rate determinations, to avoid
working with alpha dose rates. For that matter, coarse
grains are often treated with hexafluorosilicic acid (HF)
to remove the outer rim of the grains, which has been irra-
diated by alpha particles, and thus eliminate the alpha dose
rate contribution. For smaller grains, however, where
chemical etching is not possible, a proper assessment
416 LUMINESCENCE DATING, DOSE RATES
of alpha dose rates and of alpha particle efficiency in
luminescence production is crucial.
Current investigations
The infinite matrix assumption is a very powerful tool in
deriving dose rates from rocks and sedimentary units.
Although fundamentally set in principle, it does allow
for small departures from the concept of sample homoge-
neity through variations in water content and grain size.
In particular, the infinite matrix assumption has been of
significant use when applied to the dating of ceramics,
where quartz grains are included in a uniform clay matrix
at the scale of alpha and beta radiations. However, the cur-
rent focus of luminescence dating has shifted to the dating
of sediments where the requirements of the infinite matrix
assumption are rarely met. For example, the presence of
sand-sized feldspar is problematic in deriving infinite
matrix dose rates, due to their non-negligible size com-
pared to beta ranges which acts to absorb some amount
of radiation and hence reduce the relative levels of radia-
tion in other localized minerals.
Furthermore, the use of grain-size attenuation factors to
calculate beta dose rates to quartz grains is based on the
assumption that quartz grains represent nonradioactive
inclusions in an otherwise uniform, radioactive matrix
(as is the case, e.g., of sand-sized quartz grains surrounded
by clay in pottery). In a sand dune, where quartz grains are
the main constituent of the sediment, such a uniform
matrix surrounding the grains does not exist; small grains
will attenuate beta radiation less than coarse grains, but
this is a relative effect, and therefore, the use of classical
dose rate attenuation factors is not suited for absolute dose
rate calculations (see Guérin et al., 2012).
Because of these problems related to spatial variations
in dose rates, especially at the range of beta particles, some
research in the field is devoted to the development of
numerical and experimental methods to characterize and
quantify sources of radioactivity (see, e.g., Rufer and
Preusser, 2009). In particular, the spatial distribution of
dose rates, in relationship with the distribution of natural
dosimeters (e.g., quartz grains), is a field of special inter-
est. It is well known that some of the dispersion in single
grain OSL equivalent dose distributions arises from het-
erogeneities of radioactivity at the scale of beta radiations:
grains close to a beta source such as a potassium feldspar
grain will receive a higher dose rate during burial than
grains more distant from such sources. However, until
now only few models have been proposed to quantify dose
rate distributions at the single grain level. Mayya et al.
(2006) modelled such distributions arising from the pres-
ence of potassium feldspar grains in sand made of
200 mm quartz grains; they have shown that potassium
concentration in such environments is an important factor
because as the potassium content is decreased, the average
distance between source (potassium feldspar) and dosime-
ter (quartz) is increased. As a result, few grains located
close to the sources will receive high dose rates, whereas
most quartz grains will receive small beta dose rates,
which leads to dispersion in dose rate distributions to sin-
gle grains of quartz. However, more focused studies are
required in order to identify the different factors driving
the dispersion of dose rates. Such studies would then
allow direct comparisons between equivalent dose and
dose rate distributions at a single grain level and thus
allow distinguishing the different factors of dispersion in
equivalent doses, for example, insufficient bleaching at
the time of sediment deposition.
At a gamma dose rate scale (10 cm and more), impor-
tant spatial variations may occur, for example, at the inter-
face between sediment and bedrock or between sediment
and air when samples are taken close to the surface. The
presence of rocks in sediments may also influence gamma
dose rate to luminescence samples. For example, it is well
known that limestone is generally less radioactive than
other sediment types such as clay. In complex environ-
ments where calcareous lumps are found in fine sediments
richer in radioelements, Brennan et al. (1997) have thus
recommended using in situ gamma dose rate measure-
ments to obtain representative measurements.
In closure, one of the main problems in dose rate esti-
mation lies in the evolution of this physical quantity with
time: present-day dose rates may not be representative of
the sample history. Water circulation, for example, can
induce movement of radioelements in sediments. 238U
can be soluble in some environments as well as 226Ra
and this can cause disequilibrium in uranium series.
222
Rn has a short half-life (3.8 days) and is gaseous, so it
can easily escape from the ground when surface is
exposed to open air. High-resolution gamma spectrometry
and alpha spectrometry can reveal differences in the activ-
ity of different members of the decay chain; in such cases,
different scenarios for the origin of the disequilibrium can
be investigated (e.g., Guibert et al., 2009) and then dose
rate variations in time can be accounted for.
Bibliography
Aitken, M. J., 1985. Thermoluminescence Dating. London:
Academic.
Aitken, M. J., and Xie, J., 1990. Moisture correction for annual
gamma dose. Ancient TL, 8, 6–9.
Ankjærgaard, C., and Murray, A. S., 2007. Total beta and gamma
dose rates in trapped charge dating based on beta counting. Radi-
ation Measurements, 42, 352–359.
Brennan, B. J., Schwarcz, H. P., and Rink, J., 1997. Simulation of
the gamma radiation field in lumpy environments. Radiation
Measurements, 27, 299–305.
Fano, U., 1954. A note on the Bragg-Gray cavity principle for mea-
suring energy dissipation. Radiation Research, 1, 237–240.
Guérin, G., and Mercier, N., 2011. Determining gamma dose rates
by field gamma spectroscopy in sedimentary media: results of
Monte Carlo simulations. Radiation Measurements, 46,
190–195.
Guérin, G., Mercier, N., and Adamiec, G., 2011. Dose-rate conver-
sion factors: update. Ancient TL, 29(1), 5–8.
Guérin, G., Mercier, N., Nathan, R., Adamiec, G., and Lefrais, Y.,
2012. On the use of the infinite matrix assumption and associated
 LUMINESCENCE DATING, HISTORY
concepts: a critical review. Radiation Measurements, 47,
778–785.
Guibert, P., Lahaye, C., and Bechtel, F., 2009. The importance of U-
series disequilibrium of sediments in luminescence dating: a case
study at the Roc de Marsal Cave (Dordogne, France). Radiation
Measurements, 44, 223–231.
Huntley, D. J., and Baril, M. R., 1997. The K content of the
K-feldspars being measured in optical dating or in thermolumi-
nescence dating. Ancient TL, 15, 11–13.
Huntley, D. J., and Hancock, R. G. V., 2001. The Rb contents of the
K-feldspar grains being measured in optical dating. Ancient TL,
19, 43–46.
Mayya, Y. S., Morthekai, P., Murari, M. K., and Singhvi, A. K.,
2006. Towards quantifying beta microdosimetric effects in sin-
gle-grain quartz dose distribution. Radiation Measurements,
41, 1032–1039.
Nathan, R. P., and Mauz, B., 2008. On the dose-rate estimate of car-
bonate-rich sediments for trapped charge dating. Radiation Mea-
surements, 43, 14–25.
Prescott, J. R., and Hutton, J. T., 1988. Cosmic ray and gamma ray
dosimetry for TL and ESR. Nuclear Tracks and Radiation Mea-
surements, 14, 223–227.
Readhead, M. L., 2002. Absorbed dose fraction for 87Rb beta parti-
cles. Ancient TL, 20, 25–28.
Roesch, W. C., and Attix, F. H., 1968. Basic concepts of dosimetry.
In Attix, F. H., and Roesch, W. C. (eds.), Radiation Dosimetry,
volume 1: Fundamentals. New York/London: Academic.
Rufer, D., and Preusser, F., 2009. Potential of autoradiography to
detect spatially resolved radiation patterns in the context of
trapped charge dating. Geochronometria, 34, 1–13.
Schmidt, C., Rufer, D., Preusser, F., Krbetschek, M., and Hilgers,
A., 2013. Assessment of radionuclide distribution in silex by
autoradiography. Archaeometry, 55, 407–422.
Tribolo, C., Mercier, N., and Valladas, H., 2001. Alpha sensitivity
determination in quartzite using an OSL single aliquot proce-
dure. Ancient TL, 19, 47–50.
Vandenberghe, D., De Corte, F., Buylaert, J.-P., Kučera, J., and Van
den Haute, P., 2008. On the internal radioactivity in quartz. Radi-
ation Measurements, 43, 771–775.
Zimmerman, D. W., 1971. Thermoluminescence dating using fine
grains from pottery. Archaeometry, 13, 29–52.
LUMINESCENCE DATING, HISTORY
Ludwig Zöller1 and Günther A. Wagner2
1
Geographisches Institut, Universität Bayreuth, Bayreuth,
Germany
2
Geographisches Institut der Universität Heidelberg,
Heidelberg, Germany
Definition
Luminescence dating is based on the perception in solid
state physics that energy from the absorption of ionizing
radiation can be stored in the crystal lattice of insulators
as radiation damage. Thermal or optical stimulation of
dosed mineral grains results in release of radiation-
induced luminescence. The amount of radiation damage
and, thus, the acquired luminescence signal is at least
within certain limits, proportional to the time during which
the insulating minerals were exposed to ionizing radiation.
417
Assuming a constant rate of natural ionizing radiation the
age is calculated as the ratio of the accumulated dose over
the dose-rate. The present contribution deals with the his-
tory of luminescence dating from the first observations of
the luminescence phenomenon via the first proposal to use
it for dating purposes, and the first dating application to
the present wide field of applications in archaeology, Qua-
ternary geology, geomorphology and geoarchaeology.
Early observations of the phenomenon
The term “Luminescenz” was first used in 1888 by the Ger-
man physicist Eilhardt Wiedemann at the University of
Erlangen for “all those phenomena of light which are not
solely conditioned by the rise in temperature,” i.e., the condi-
tion of incandescence or “hot light” in contrast to lumines-
cence or “cold light” (cited after Harvey, 1957). He
differentiated various kinds of luminescence according to
the method of excitation or of stimulation, such as
photoluminescence, thermoluminescence, tribolumines-
cence and chemiluminescence. When irradiating various
minerals with “Entladungsstrahlung” (cathode rays)
Wiedemann and his colleague Gerhart Schmidt (Wiedemann
and Schmidt, 1895) observed, after gentle heating, a light-
emission and thus first described the phenomenon of “ther-
moluminescence” (TL), which is a thermally stimulated
radioluminescence induced by ionizing radiation. Experi-
ments with the phenomenon of luminescence led in the same
year to the epochal discoveries of X-rays by Conrad Roent-
gen at Würzburg and of the natural radioactivity by Henri
Becquerel at Paris only 1 year later.
Actually, a luminescence phenomenon had been
already scientifically described more than 200 years
before by Robert Boyle. He reported on 28 October
1663 to the Royal Society “About a diamond that Shines
in the Dark” and adding experimental details, particularly
“I also brought it to some kind of Glimmering Light, by
taking into bed with me, and holding it to a good while
upon a warm part of my Naked Body” (cited after Aitken,
1985). A few years later Johann Sigismund Elsholtz
(1676) at Berlin described a similar property for the min-
eral fluorspar. There are also much earlier reports on lumi-
nescent gems and stones which glimmer in the dark after
being exposed to light, probably a kind of photolumi-
nescence (Harvey, 1957).
Evolution of TL dating: from first suggestions
to first results
It was Farrington Daniels and coworkers at the University
of Wisconsin in 1953, while measuring exposure of min-
erals to nuclear radiation, who first proposed to use ther-
moluminescence also for geological and archaeological
age determination (Daniels et al., 1953). Ceramics and
burned bricks, mainly from the Roman period, were the first
archaeological materials on which the suggested method
was successfully tested by Norbert Grögler and colleagues
at the University of Bern (Grögler et al., 1958, 1960),
followed by Kennedy and Knopff (1960) at the University
418 LUMINESCENCE DATING, HISTORY
of California. Starting from this point, luminescence dating
gradually evolved into a powerful tool for heated archaeo-
logical materials. This was mainly due to the efforts by
Martin Aitken and his team at the University of Oxford dur-
ing the 1960s (Aitken et al. 1964, 1968b, cited after Aitken
1985), but gradually also other laboratories became
involved. The essential procedures of pottery TL-dating
were developed: the quartz inclusion technique (Fleming,
1966) and the fine-grain technique (Zimmerman, 1967).
An important by-product of these methodological develop-
ments was the TL authenticity-testing of ceramic objects
(Fleming et al., 1970). In addition to baked clay, TL dating
has been also applied to heated stones, in particular to flint
artefacts (Göksu et al., 1974).
Dating of sediments
With the recognition that not only heating, but also optical
bleaching resets the TL signal (Wintle and Huntley, 1979),
the method became also applicable to the dating of sedi-
ments. The observation made on quartz that intense green
illumination, instead of heating, also stimulates lumines-
cence (Huntley et al., 1985) opened the door for utilizing
the phenomenon of Optically Stimulated Luminescence
(OSL) for dating. Soon afterwards the luminescence stim-
ulation by infrared illumination of feldspar was discovered
(Hütt et al., 1988).
TL dating of sediments
The observation that the natural TL intensities from Qua-
ternary sediments increase with geological age stimulated
researchers from the former USSR to propose TL as a tool
for the chronology of sediments. Pioneer research was
done by V. N. Shelkoplyas and G. V. Morozov at Kyiv
(Dreimanis et al., 1978). The lack of English publications
by the Kyiv group makes it difficult to fully understand
sample preparation and measurement protocols. The
apparently most important early work is noted in
Morozow (1968) and Shelkoplyas and Morozow (1981)
even if a first note by Shelkoplyas and Morozow (1965)
dates a few years earlier. Favourable material for the Kyiv
group was loess from which they first tried to establish
a relative TL chronology by a kind of regeneration of the
natural signal using X-rays or g-irradiation. Later, numer-
ical ages were calculated assuming that almost all relevant
natural growth of the TL signal was due to natural alpha
irradiation but this idea was revised by Shelkoplyas and
Morozow (1981) accounting for an alpha efficiency of
ca. 0.1. A fully reliable TL dosimetry satisfying modern
requirements (e.g., Aitken, 1985; Guerin et al., 2011)
was not delivered in the early work at Kyiv. But also the
estimation of the equivalent dose ED remained question-
able. Thus, Wintle and Huntley (1982) doubted the reli-
ability of these TL ages.
Despite this, several merits of the early work at Kyiv
and some technical details deserve mention. The heating
rate was only 1 K/s up to 300  C or 350  C. Photomul-
tipliers of the Russian type FEU-19 and FEU-29 with
maximum sensitivity at ca. 380 nm were used for detec-
tion. The peak of the natural TL was found at ca. 230  C
hotplate temperature. Despite the lack of rigorous physical
and chemical mineral separation the Kyiv group dubiously
stated that the TL signal was obtained from quartz grains.
Experiments using coarse (1 mm) as well as small ground
“quartz” grains (0.01 mm) showed much smaller TL from
fine grains. Although resetting of the TL signal by UV
light was also observed, Morozow (1968) concluded that
triboluminescence due to weathering and transport was
the main factor to reset the inherited TL signal. Wintle
and Huntley (1979, 1980) first demonstrated that sunlight
exposure bleaches the TL signal of sediments which, how-
ever, does not rule out the effect of triboluminescence.
Clearly, the merits of the early work done in Kyiv are to
have shown the capability of TL from sediments to estab-
lish a relative and stratigraphically consistent chronology
from Quaternary sediments. For details of further develop-
ment of TL sediment dating we refer to Wintle and
Huntley (1982).
The work by Wintle and Huntley (1979, 1980) boosted
the development of TL dating for sediments. Since day-
light bleaching was identified as the main resetting mech-
anism further work concentrated on TL dating of wind-
lain sediments, in particular loess and dune sands. But also
fluvial and coastal sediments were tested. Special atten-
tion was directed to determine the unbleached residual of
the TL at deposition to circumvent age overestimates.
The originally applied “R-Г” respectively “R-ß”-method
(Wintle and Huntley, 1980) was complemented by addi-
tive and regenerative “total bleach” methods for aeolian
sediments of middle and lower latitudes for which
a resetting of the TL at deposition to an unbleachable
residual was assumed. Further “partial bleach” methods
to account for partial resetting at deposition were proposed
by Mejdahl (1985). Meanwhile refined techniques for TL
dosimetry allowing for calculation of more accurate and
reliable effective dose-rates (e.g., Mejdahl, 1979; Aitken,
1985) contributed further steps towards meaningful TL
ages of sediments. The “feldspar inclusion technique”
(Mejdahl 1983) was extended to dating of sediments
and, thus, advantage was taken from dating of different
mineral phases with different TL characteristics.
The euphoria at the early 1980s was dampened begin-
ning in 1985 when significant TL age underestimates from
loess older than 30–100 ka (1,000 years ago) became
obvious from many publications. The poorly understood
phenomenon of “anomalous fading” from feldspars was
often made responsible for this, but some authors claimed
that reliable ages up to 300 ka or even 800 ka could be
obtained (e.g., Singhvi et al., 1989; Berger et al., 1992).
To some extent, non-standardized laboratory protocols
contributed to the uncertainty. Zöller et al. (1988, 1991)
demonstrated that the reliable TL age range from loess
could be extended to a minimum of about 100 ka. Besides
the “effective mean lifetime” t of an electron trap (c.f.
Wagner, 1998), “dose-dependent sensitivity change”
(Wintle, 1985) was suggested to explain observed age
 LUMINESCENCE DATING, HISTORY
underestimates from older loess. Finally, efforts were
made to additionally investigate the characteristics and
stability of luminescence centres using a CCD camera-
based luminescence spectrometer (Rieser et al., 1994).
On the other hand, the risk of TL age overestimates
from sediments due to incomplete optical resetting (often
referred to as incomplete bleaching) at deposition gave
increasing distinction to the discussion. This resulted in
the development of OSL dating (Huntley et al., 1985; Hütt
et al., 1988; Aitken 1998). Finally, the TL dating of sedi-
ments was almost completely abandoned since the mid-
1990s. Some laboratories, however, have continued with
TL dating of loess and have published reasonable results
even for loess >200 ka (Kuziak et al., 2007).
Recent developments of TL dating
TL dating of burned flint as direct dating of ancient human
activity has been an important application since 1974
(Göksu et al., 1974). Recent developments, including
microdosimetry problems of some flint tools, are
described in Richter (this volume). Successful TL dating
results for volcanic eruptions using quartz, feldspars or
volcanic glass are reviewed by Fattahi and Stokes
(2003). The more recent availability of band-pass filter
glasses for detecting luminescence emissions in a narrow
wave-length band has resulted in the intensified use of
orange-red emissions (ca. 620 nm) from quartz. Despite
the disadvantage of a much lower signal/noise ration the
great advantage of the 620 nm emission from quartz is
its high saturation dose which exceeds the saturation dose
of UV-A (315–400 nm wavelength) or blue emissions by
more than an order of magnitude. Meaningful TL ages
>1 Ma using the orange-red emission from quartz appear
obtainable (Fattahi and Stokes, 2000). As volcanic quartz
is not a significant component in most mafic rocks, Zöller
and Blanchard (2009) used heated crustal xenoliths and
maar tephra for TL dating, but different techniques yielded
diverging ages so far, which points to pending problems.
TL dating of baked loess from a Lower Gravettian hearth
in Austria, however, resulted in congruent ages obtained
from blue TL emission of polymineral fine grains and
orange-red emissions from quartz separates in agreement
with results from independent dating methods (Zöller
et al., 2013). Aeolian quartz grains baked by lava flows
were successfully dated to ca. 170 ka using orange-red
TL emissions (Suchodoletz et al., 2012).
Surface dating
In addition to sediment dating, it is also possible to deter-
mine the age when the surface of a stony object, which had
been exposed to daylight before, became ultimately
shielded from light. This allows dating various archaeo-
logical and geological events, such as the construction as
well as the destruction of stone buildings, the burial of
stone artefacts and the sedimentary deposition of boulders.
Liritzis (1994) proposed to use TL to date the construction
of a megalithic limestone building. But the applied
419
methodology was hampered by the fact that in calcite –
even after extended exposure to sunlight – retains
a significant residual TL signal (Liritzis and Galloway,
1999). As in sediment dating, this problem of
unbleachable luminescence signals is overcome by the
use of OSL instead of TL. It seems, therefore, consequent
to employ OSL dating also to stone surfaces. Early
attempts to determine the burial age of quartzite pebbles
by this approach were reported by Huntley and Richards
(1997). After scraping off the uppermost part from the
bleached surface of lithic artefacts, IRSL dating was
applied to the fine-grained feldspar fraction, whereby
uncertainties with insufficiently bleached grains were
encountered (Morgenstein et al., 2003). When utilizing
a high spatial resolution OSL-detection technique (HR-
OSL) of minerals that are left in their original petrologic
context, i.e., without any mineral separation, Greilich
et al. successfully dated a stonewall of the medieval castle
of Lindenfels, Germany, and the pre-Columbian Nasca
lines (geoglyphs) around Palpa in southern Peru (Greilich
et al., 2002, 2005).
Perspectives of TL dating
The appropriate method for age determination of sediments
is presently optical dating, in particular for young sedi-
ments. The upper age limit of optical dating is, despite more
recent attempts using post-IR-IRSL (pIR-IRSL) or Thermal
Transfer-OSL (TT-OSL) dating, hampered by either low
saturation doses or by age underestimates. As for older sed-
iments the unbleached TL residual becomes increasingly
negligible, TL dating protocols deserve further attention.
The more promising application of TL dating will, how-
ever, be concentrated on heated materials such as artefacts
or rocks heated by, e.g., volcanic eruptions.
Recent developments of OSL and IRSL dating
As IRSL and, in particular, OSL minimizes the problem of
age overestimates due to unbleached residuals at deposi-
tion, the rapid development of OSL dating enabled the
assessment of reliable chronologies of sediments from
Late Glacial and Holocene and of archaeosediments. The
Single Aliquot-Regeneration (SAR) protocol (Murray
and Wintle, 2000) became soon established as a routine
protocol. High resolution deciphering of landscape devel-
opment under the influence of mankind not only supplied
chronostratigraphies of geoarchaeological sites but also
initiated further development and verification of
geomorphologic and geoarchaeologic concepts (e.g.,
Lang and Hönscheid, 1999; Lang et al., 1999; Fuchs
et al., 2004).
The very rapid bleaching of quartz-OSL and the appli-
cation of statistical methods to discriminate between well
and poorly bleached sub-samples measured by the SAR
protocol opened up the ability to obtain meaningful dating
results from sediments or geomorphologic events not dat-
able by luminescence before, such as periglacial slope
deposits (Hülle et al., 2009) or tsunamites (Brill et al.,
420 LUMINESCENCE DATING, HISTORY
2012). The rather low saturation dose of quartz-OSL,
however, limits the upper dating range to less than or
around 100 ka, depending on the natural radioactivity of
the sediment. The much higher saturation dose of feld-
spar-IRSL may allow for an upper dating limit of
200–300 ka or even more, but the limiting factor was
found in the phenomenon of “anomalous fading,” an
athermal long term instability of the luminescence from
feldspars first described by Wintle (1973). Procedures to
correct for anomalous fading were proposed by Huntley
and Lamothe (2001) and Auclair et al. (2003) but they
are expected to yield reliable results only in the linear part
of the dose–response curve. More recently, the “post IR-
IRSL” (pIR-IR) protocol was developed and tested
(Buylaert et al., 2009; Thiel et al., 2011; Buylaert et al.,
2012; Murray et al., 2013). It was found that the fading
of the pIR-IR signal is significantly lower allowing rea-
sonable age estimates of a few hundred ka. To extend the
dating limit of quartz OSL by an order of magnitude
a “Thermal Transfer-OSL” (TT-OSL) protocol using recu-
perated OSL was suggested (Wang et al., 2006, 2007;
Porat et al., 2009) but results from testing the protocol still
appear ambiguous (e.g., Zander and Hilgers, 2013).
Perspectives of OSL and IRSL dating
On the one hand, the new methodological developments
of OSL dating as well as technical improvements of the
luminescence readers enable dating of sediments depos-
ited as young as a few years or decades. Thus, lumines-
cence dating is ready for bridging the gap between
historical geology (dating of past events) and process geo-
morphology (estimation of recent process rates). On the
other hand, the extension of the dating range beyond
100 ka towards 1 Ma remains a challenge, but first prom-
ising steps are breaking new ground.
OSL dating of rock surfaces (i.e., last exposure to day-
light) is expected to gain increasing relevance for archae-
ology and geomorphology, in particular since the latest
developments of commercial luminescence readers pro-
vide the option of spatially resolved luminescence
measurements.
Infra-red radiofluorescence dating
Although the infrared radiofluorescence (IR-RF) tech-
nique follows the same basic principles as TL and OSL
dating, its dose determination differs from them. The
underlying phenomenon is unequivocally a fluorescence
effect (Krbetschek et al., 2000), and thus the original
name was later changed from “radioluminescence”
to “radiofluorescence” (Erfurt and Krbetschek, 2003).
IR refers to the near infrared emission (865 nm), which
is characteristic for potassium feldspars, such as micro-
cline and orthoclase. The physical background and radia-
tion-dose dependence of this phenomenon have been
intensively investigated, along with the methodological
aspects of dating Quaternary sediments (Degering and
Krbetschek, 2007). Of particular interest for the data
presented is the long-term stability of the IR-RF signal.
It is possible to obtain IR-RF ages in agreement with inde-
pendent age controls back to ~700 ka, such as demon-
strated for the type-site of Homo heidelbergensis at
Mauer, Germany (Wagner et al., 2010).
Summary
Within the last ca. 50 years the development of lumines-
cence dating has been fast-paced. Originally, thermal stim-
ulation (TL) was applied to date ceramics and heated
artefacts. Later, the recognition of solar resetting of the
luminescence signal broadened the application of lumi-
nescence dating strongly in a sustainable matter. For the
first time, the sedimentation age of Quaternary sediments
could be dated directly, first by TL and then later by opti-
cal stimulation (OSL). Beyond archaeology, luminescence
dating has developed to become an indispensable scien-
tific tool in Quaternary geology, geomorphology, and
geoarchaeology world-wide. Challenges for ongoing
research appear to focus on extension of the dating range
towards the more recent as well towards older dating
limits.
Further reading
Further reading is recommended: Duller (2008), Preusser
et al. (2008), Roberts (2008), and Wintle (2008). Periodi-
cals informing about latest progress in luminescence dat-
ing are mainly Ancient TL, Archaeometry, Quaternary
Geochronology and Radiation Measurements. Proceed-
ings of the triennial international conference on Lumines-
cence and ESR Dating (LED) are published in special
volumes of Quaternary Geochronology and Radiation
Measurements.
Bibliography
Aitken, M. J., 1985. Thermoluminescence dating. London:
Academic.
Aitken, M. J., 1998. An Introduction to Optical Dating. Oxford:
Oxford University Press.
Auclair, M., Lamothe, M., and Hout, S., 2003. Measurement of
anomalous fading for feldspar IRSL using SAR. Radiation
Measurements, 37, 487–492.
Berger, G. W., Pillans, B. J., and Palmer, A. S., 1992. Dating loess
up to 800 ka by thermoluminescence. Geology, 20, 403–406.
Boyle, R., 1664. Experiments and Considerations Upon Colours
with Observations on a Diamond that Shines in the Dark.
London: Henry Herringham.
Brill, D., Klasen, N., Brückner, H., Jankaew, K., Scheffers, A.,
Kelletat, D., and Scheffers, S., 2012. OSL dating of tsunami
deposits from Phra Thong Island, Thailand. Quaternary
Geochronology, 10, 224–229.
Buylaert, J. P., Murray, A. S., Thomsen, K. J., and Jain, M., 2009.
Testing the potential of an elevated temperature IRSL signal
from K-feldspar. Radiation Measurements, 44, 560–565.
Buylaert, J.-P., Jain, M., Murray, A. S., Thomsen, K. J., Thiel, C.,
and Sohbati, R., 2012. A robust feldspar luminescence dating
method for middle and late Pleistocene sediments. Boreas, 41,
435–451.
Daniels, F., Boyd, C. A., and Saunders, D. F., 1953. Thermolumi-
nescence as a research tool. Science, 117, 343–349.
 LUMINESCENCE DATING, HISTORY
Degering, D., and Krbetschek, M. R., 2007. Dating of interglacial
sediments by luminescence methods. In Sirokko, F., Claussen,
M., Sanchez Goni, M. F., and Litt, T. (eds.), The Climate of Past
Interglacials. Amsterdam: Elsevier, pp. 157–172.
Dreimanis, A., Hütt, G., Raukas, A., and Whippey, P. W., 1978. Dat-
ing methods of Pleistocene deposits and their problems: 1.
Thermoluminescence dating. Geoscience Canada, 5, 55–60.
Duller, G. A. T., 2008. Single-grain optical dating of quaternary sed-
iments: why aliquot size matters in luminescence dating. Boreas,
37, 589–612.
Elsholtz, J. S., 1676. De phosphoris quatuor: observatio. Berlin
(Georg Schultze).
Erfurt, G., and Krbetschek, M. R., 2003. Studies on the physics of
the infrared radioluminescence of potassic feldspar and on the
methodology of its application to sediment dating. Radiation
Measurements, 37, 505–510.
Fattahi, M., and Stokes, S., 2000. Extending the time range of
luminescence dating using red TL (RTL) from volcanic quartz.
Radiation Measurements, 32, 479–485.
Fattahi, M., and Stokes, S., 2003. Dating volcanic and related sedi-
ments by luminescence methods: a review. Earth-Science
Reviews, 62, 229–264.
Fleming, S. J., 1966. Study of thermoluminescence of crystalline
extracts from pottery. Archaeometry, 9, 170–173.
Fleming, S. J., Moss, H. M., and Joseph, A., 1970. Thermolumines-
cence authenticity testing of some six dynasties figures.
Archaeometry, 12, 57–68.
Fuchs, M., Lang, A., and Wagner, G. A., 2004. The history of Holo-
cene soil erosion in the Phlious Basin, NE Peloponnese, Greece,
based on optical dating. The Holocene, 14, 334–345.
Göksu, H. Y., Fremlin, J. H., Irwin, H. T., and Fryxell, R., 1974. Age
determination of burned flint by a thermoluminescence method.
Science, 183, 651–654.
Greilich, S., Glasmacher, U. A., and Wagner, G. A., 2002. Spatially
resolved detection of luminescence: a unique tool for
archaeochronometry. Naturwissenschaften, 89, 371–375.
Greilich, S., Glasmacher, U. A., and Wagner, G. A., 2005. Optical
dating of granitic stone surfaces. Archaeometry, 47, 645–665.
Grögler, N., Houtermans. F. G., and Stauffer, H., 1958. Radiation
damage as a research tool for geology and prehistory. Convegno
sulle dotazioni con metodi nucleari. In 5aRass Internazion Elettr
Nucl, Sezione Nucl, Roma, pp. 5–15.
Grögler, N., Houtermans, F. G., and Stauffer, H., 1960. Über die
Datierung von Keramik und Ziegel durch Thermolumineszenz.
Helvetica Physica Acta, 33, 595–596.
Guerin, G., Mercier, N., and Adamiecz, G., 2011. Dose-rate conver-
sion factors: update. Ancient TL, 19, 5–8.
Harvey, K. N., 1957. A History of Luminescence – From the Earliest
Times Until1900. Philadelphia, PA: The American Philosophical
Society.
Hülle, D., Hilgers, A., Kühn, P., and Radtke, U., 2009. The potential
of optically stimulated luminescence for dating periglacial slope
deposits – a case study from the Taunus area, Germany. Geomor-
phology, 109, 66–78.
Huntley, D. J., and Lamothe, M., 2001. Ubiquity of anomalous fad-
ing in K-feldspars and the measurement and correction for it in
optical dating. Canadian Journal of Earth Sciences, 38,
1093–1106.
Huntley, D. J., and Richards, M., 1997. The age of the Diring
Quriakh archaeological site. Ancient TL, 15, 48–51.
Huntley, D. J., Godfrey-Smith, D. I., and Thewalt, M. L. W., 1985.
Optical dating of sediments. Nature, 313, 105–107.
Hütt, G., Jaek, I., and Tchonka, J., 1988. Optical dating: K-feldspars
optical response stimulation spectra. Quaternary Science
Reviews, 7, 381–385.
Kennedy, G. C., and Knopff, L., 1960. Dating by thermolumines-
cence. Archaeology, 13, 147–148.
421
Krbetschek, M. R., Trautmann, T., Dietrich, A., and Stolz, W., 2000.
Radioluminescence dating: methodological aspects. Radiation
Measurements, 32, 493–498.
Kuziak, J., Łanczont, M., and Wilgat, M., 2007. TL ages of loesses
from the last two glacials in SE Poland. Geochronometria, 27,
33–40.
Lang, A., and Hönscheid, S., 1999. Age and source of colluvial sed-
iments at Vaihingen-Enz, Germany. Catena, 38, 89–207.
Lang, A., Kadereit, A., Behrends, R. H., and Wagner, G. A., 1999.
Optical dating of anthropogenic sediments at the archaeological
excavation site Herrenbrunnenbuckel, Bretten-Bauerbach, Ger-
many. Archaeometry, 41, 397–411.
Liritzis, I., 1994. A new dating method by thermoluminescence of
carved megalithic stone building. Comptes Rendus Academie
des Sciences, Série II, 319, 603–610.
Liritzis, I., and Galloway, R. B., 1999. Dating implications from
solar bleaching of thermoluminescence of ancient marble. Jour-
nal of Radioanalytical and Nuclear Chemistry, 241, 361–368.
Mejdahl, V., 1979. Thermoluminescence dating: beta-dose attenua-
tion in quartz grains. Archaeometry, 21, 61–72.
Mejdahl, V., 1983. Feldspar inclusion dating of ceramics and burnt
stones. PACT, 9, 351–364.
Mejdahl, V., 1985. Thermoluminescence dating of partially
bleached sediments. Nuclear Tracks and Radiation Measure-
ments, 10, 711–715.
Morgenstein, M. E., Luo, S., Ku, T. L., and Feathers, J., 2003. Ura-
nium series and luminescence dating of volcanic lithic artefacts.
Archaeometry, 45, 503–518.
Morozow, G. V., 1968. Oprodelenie otnositelnogo vosrasta
antropogenovikh otloshenii Ukraini termoljuminestsentnim
metodom. Kiev: Akademija Nauk Ukrainskoi SSR, Instit.
Geolog. Nauk Komm. poizuch. chetvert. perioda (in Russian).
Murray, A. S., and Wintle, A. G., 2000. Luminescence dating of
quartz using an improved single-aliquot regenerative-dose pro-
tocol. Radiation Measurements, 32, 57–73.
Murray, A., Schmidt, E. D., Stevens, T., Buylaert, J. -P., Marković,
S. B., Tsukamoto, S., and Frechen, M., 2013. Dating middle
Pleistocene loess from Stari Slankamen (Vojvodina, Serbia) –
limitations imposed by the saturation behaviour of an elevated
temperature IRSL signal. Catenahttp://dx.doi.org/10.1016/j.
catena.2013.06.029.
Porat, N., Duller, G.A.T., Roberts, H.M., and Wintle, A.G. 2009.
A simplified SAR protocol for TT-OSL. Radiation Measure-
ments, 44, 538-542.
Preusser, F., Degering, D., Fuchs, M., Hilgers, A., Kadereit, A.,
Klasen, N., Krbetschek, M., Richter, D., and Spencer, J. Q. G.,
2008. Luminescence dating: basics, methods and applications.
Eiszeitalter und Gegenwart Quaternary Science Journal, 57,
95–149.
Rieser, U., Krbetschek, M. R., and Stolz, W., 1994. CCD-camera
based high sensitivity TL/OSL-spectrometer. Radiation Mea-
surements, 23, 523–528.
Roberts, H. M., 2008. The development and application of lumines-
cence dating to loess deposits: a perspective on the past, present
and future. Boreas, 37, 483–507.
Shelkoplyas, V. N., and Morozow, G. V., 1965. Oprodelenie
otnositelnogo vosrasta chetvertichnikh otlozhenii srednego
Pridneproviia termoljuminestsentnom metodom. In Osnovie
Problemi Izucheniia Chertvertichnogo Perioda. Moscow:
Nauka (in Russian).
Shelkoplyas, V. N., and Morozow, G. V., 1981. Primenenie
Termoljuminestsentnogo Metoda Dlja Izuchenija
Antropogenovych Otlozhenij. Kiev: Akademija Nauk
Ukrainskoi SSR.
Singhvi, A. K., Bronger, A., Sauer, W., and Pant, R. K., 1989. Ther-
moluminescence dating of Loess–paleosol sequences in the
Carpathian basin (East-Central Europe): a suggestion for
422 LUMINESCENCE DATING, INSTRUMENTATION
a revised chronology. Chemical Geology: Isotope Geoscience
Section, 73, 307–317.
Suchodoletz, H. V., Blanchard, H., Hilgers, A., Radtke, U., Fuchs,
M., Dietze, M., and Zöller, L., 2012. TL and ESR dating of mid-
dle Pleistocene lava flows on Lanzarote island, Canary Islands
(Spain). Quaternary Geochronology, 9, 54–64.
Thiel, E., Terhorst, B., Jaburov, I., Buylaert, J.-P., Murray, A. S.,
Fladerer, F. A., Damm, B., Frechen, M., and Ottner, F., 2011.
Sedimentation and erosion processes in Middle to Late Pleisto-
cene sequences exposed in the brickyard of Langenlois/Lower
Austria. Geomorphology, 135, 295–307.
Wagner, G. A., 1998. Age Determination of Young Rocks and Arti-
facts. Berlin/Heidelberg: Springer.
Wagner, G. A., Krbetschek, M., Degering, D., Bahain, J. J., Shao,
Q., Falguères, C., Voinchet, P., Dolo, J. M., Garcia, T., and
Rightmire, G. P., 2010. Radiometric dating of the type-site for
Homo heidelbergensis at Mauer, Germany. Proceedings of the
National Academy of Sciences of the United States of America,
107, 19726–19730. http://www.pnas.org/content/early/2010/
10/27/1012722107-aff-5#aff-5.
Wang, X. L., Wintle, A. G., and Lu, Y. C., 2006. Thermally trans-
ferred luminescence in fine-grained quartz from Chinese loess:
basic observations. Radiation Measurements, 41, 649–658.
Wang, X. L., Wintle, A. G., and Lu, Y. C., 2007. Testing a single-
aliquot protocol for recuperated OSL dating. Radiation Mea-
surements, 42, 380–391.
Wiedemann, E., 1888. Ueber fluorescenz und phosphorescenz.
Annalen der Physik, 34, 446–449.
Wiedemann, E., and Schmidt, G. C., 1895. Ueber Luminescenz.
Annalen der Physik und Chemie, 54, 604–625.
Wintle, A. G., 1973. Anomalous fading of thermoluminescence in
mineral samples. Nature, 245, 143–144.
Wintle, A. G., 1985. Stability of TL signal in fine grains from loess.
Nuclear Tracks, 10, 725–730.
Wintle, A. G., 2008. Luminescence dating: where it has been and
where it is going. Boreas, 37, 471–482.
Wintle, A. G., and Huntley, D. J., 1979. Thermoluminescence dat-
ing of a deep-sea core. Nature, 279, 710–712.
Wintle, A. G., and Huntley, D. J., 1980. Thermoluminescence dat-
ing of ocean sediments. Canadian Journal of Earth Sciences,
17, 348–360.
Wintle, A. G., and Huntley, D. J., 1982. Thermoluminescence dat-
ing of sediments. Quaternary Science Reviews, 1, 31–53.
Zander, A., and Hilgers, A., 2013. Potential and limits of OSL, TT-
OSL, IRSL and pIRIR290 dating methods applied on a Middle
Pleistocene sediment record of Lake El’gygytgyn, Russia.
Climate of the Past, 9, 719–733.
Zimmerman, D. W., 1967. Thermoluminescence from fine grains
from ancient pottery. Archaeometry, 10, 26–28.
Zöller, L., and Blanchard, H., 2009. The partial heat – longest pla-
teau technique: testing TL dating of Middle and Upper Quater-
nary volcanic eruptions in the Eifel Area, Germany.
Eiszeitalter und Gegenwart Quaternary Science Journal,
58, 88–106.
Zöller, L., Stremme, H. E., and Wagner, G. A., 1988. Thermolu-
mineszenz-Datierung an Löß-Paläoboden-Sequenzen von
Nieder-, Mittel- und Oberrhein. Chemical Geology: Isotope
Geoscience Section, 73, 39–62.
Zöller, L., Conard, N. J., and Hahn, J., 1991. Thermoluminescence
dating of middle Palaeolithic open air sites in the middle rhine
valley/Germany. Naturwissenschaften, 78, 408–410.
Zöller, L., Richter, D., Blanchard, H., Einwögerer, T., Händel, M.,
and Neugebauer-Maresch, C. (2013). Our oldest children – age
constraints for the Krems-Wachtberg site obtained from various
thermoluminescence dating approaches. Quaternary Interna-
tional. http://dx.doi.org/10.1016/j.quaint.2013.05.003.
Cross-references
Alpha Spectroscopy
Chert
Electron Spin Resonance (ESR) Dating, General Principles
Feldspar, Infrared-Stimulated Luminescence
Feldspars
Historical Development of Dating Methods
Luminescence Dating
Luminescence Dating, Loess
Luminescence, Flints and Stones
Luminescence, Fluvial Sediments
Luminescence, Geomorphological Processes
Luminescence, Pottery and Bricks
Luminescence, Volcanic Rocks
Neutron Activation Analysis
Quartz
Quartz Defects, Optically Stimulated Luminescence and Thermo-
luminescence
Radiation Defect
LUMINESCENCE DATING, INSTRUMENTATION
Kristina Jørkov Thomsen
Centre for Nuclear Technologies, Technical University
of Denmark, Roskilde, Denmark
Definition
The basic instrumentation required to obtain a dose based
on optically stimulated luminescence (OSL) signals con-
sists of four main components: (i) a luminescence detec-
tion system, (ii) a stimulation light source, (iii) a heating
system, and (iv) an irradiation facility. To obtain
a thermoluminescence (TL) dose, a stimulation light
source is not required.
Introduction
Luminescence dating is an absolute chronological tech-
nique widely used in archaeological and geological dating
but also with applications in other areas of retrospective
dosimetry such as environmental, accidental, medical,
and forensic dosimetry. In dating applications an age is
determined by measuring the dose absorbed by the target
mineral, primarily quartz and feldspars, and dividing it
by an independently determined dose rate in the media.
This dose rate is mainly derived from the flux of ionizing
radiation emitted from naturally occurring isotopes (from
the thorium and uranium series as well as potassium-40)
and from cosmic rays and is determined using methods
such as gamma spectrometry, beta counting, alpha
counting, inductively coupled plasma-mass spectrometry
(ICP-MS), and neutron activation analysis. These mea-
surement techniques and the associated instrumentation
are described elsewhere. Here, the main emphasis is
placed on the instrumentation required to determine the
dose absorbed by the target mineral using luminescence
techniques.
 LUMINESCENCE DATING, INSTRUMENTATION
Luminescence detection system
Many natural phosphors are only weakly luminescent
making good light collection important. This requires
a sensitive photon detector with a low noise (background)
level to obtain a good signal-to-noise ratio. Luminescence
is emitted in all directions, but the use of reflective metal
sample holders helps to maximize the light emitted toward
the detector. The distance between the photon detector and
the sample should be as short as possible, since only
a small increase in this distance will lead to
a considerable loss in light collection efficiency. Alterna-
tively, suitable optics can be used to improve the light
detection.
The light detection system consists of two components:
(i) a photon detector and (ii) detection filters.
Photon detector
The most commonly used photon detector in lumines-
cence dating is the photomultiplier tube (PMT), which
converts photons into cascades of electrons. The quantum
efficiency of PMTs is typically up to
30 % (number of
electron pulses per photon crossing the detection area)
depending on the wavelength of the incident light. The
bialkali EMI 9235QB PMT is widely used; this has
a quantum efficiency of 25–30 % in the range from
200
to
450 nm making it suitable for detection of UV/blue
luminescence from both quartz and feldspar (emitting
mainly in the range 350–450 nm, e.g., Huntley et al.,
1991). PMTs with an extended sensitivity in the red region
(e.g., R943-02 Hamamatsu GaAs tube) are also used for
red TL measurements of quartz and feldspar (e.g., Fattahi
and Stokes, 2000; Jain et al., 2005; Vandenberghe et al.,
2009) or for IR-RF dating of feldspars (e.g., Krbetschek
et al., 2000; Buylaert et al., 2012). Such PMTs need to
be cooled to
10  C to reduce dark count.
PMTs do not provide any information on spatial varia-
tion of the emitted luminescence. Various imaging
systems have been developed (see summary in Bøtter-
Jensen et al., 2003) including the use of charged coupled
devices (CCD). Recently, the use of EMCCD cameras
has been suggested (Clark-Balzan and Schwenninger,
2012; Richter et al., 2013).
Detection filters
The main function of the detection filters is to define the
spectral detection window, but they must also shield the
photon detector from scattered stimulation light. The
intensity of the emitted luminescence is <1015 times
the intensity of the stimulation light, so for most applica-
tions it is very important to ensure that the spectral stimu-
lation and detection windows are well separated. UV
emissions (e.g., quartz, with stimulation with blue
470 nm light) are commonly detected through the broad
band Hoya U340 glass filter, which has a peak transmis-
sion at 340 nm (full width at half maximum (FWHM)
80 nm). Blue emissions (e.g., feldspar, with stimulation
with IR 870 nm light) are typically measured using
423
a combination of BG3/Schott 7–59 and BG 39 glass fil-
ters, which has a peak transmission at
380 nm (FWHM

60 nm). This filter combination is also widely used
for TL measurements, usually with the addition of
heat-absorbing filters to reduce the contribution from
blackbody radiation.
Stimulation light sources
In OSL, optical stimulation results in the eviction of
trapped charge; subsequent recombination may lead to
the emission of a luminescence signal. The probability of
eviction and hence the luminescence signal intensity
depends on the photoionization cross section of the target
trap(s). This cross section is dependent on both the wave-
length and the intensity (effective stimulation power) of
the optical stimulation. In general, a brighter luminescence
signal is obtained as the wavelength is decreased and the
stimulation power is increased. Optical stimulation is usu-
ally achieved using a constant light intensity (CW mode),
but the intensity can also be modulated to give a better
visual insight into the number of traps contributing to the
OSL signal (e.g., linearly modulated OSL, LM-OSL;
Bulur, 1996). In both CW and LM modes, the lumines-
cence is recorded during stimulation. In pulsed OSL
(POSL, e.g., McKeever et al., 1996; Denby et al., 2006),
the stimulation light is pulsed and the OSL is measured
between pulses. In luminescence dating applications,
POSL is mainly used to separate quartz and feldspar
OSL signals instrumentally (e.g., Thomsen et al., 2008)
and in combination with time-resolved detection to obtain
insight into luminescence recombination processes (e.g.,
Jain and Ankjærgaard, 2011).
Optical stimulation is often achieved using inexpensive
and compact arrays of individual light-emitting diodes
(LEDs, e.g., Bøtter-Jensen et al., 2010), which can be eas-
ily intensity modulated and can have a stable light output.
A disadvantage of LED stimulation is the relatively broad
emission (half width
20–30 nm) with significant tails.
Thus, additional cutoff filters are often required to obtain
sufficient spectral separation of stimulation and detection
windows (Bøtter-Jensen et al., 1999).
Nevertheless, optical stimulation can be achieved using
any optical stimulation source, e.g., filtered lamp systems
(Bøtter-Jensen and Duller, 1992), lasers (Huntley et al.,
1985; Duller et al., 1999; Richter et al., 2013), and laser
diodes (Bøtter-Jensen et al., 2000; Richter et al., 2013).
One advantage of using a laser-based stimulation source
is the monochromatic or very narrow emission band,
which reduces the need for filtration to separate the detec-
tion and stimulation spectral windows. These stimulation
sources further have the advantages of being focusable
to narrow spots and high power densities, which greatly
reduces the readout time. However, laser sources are often
relatively expensive, are sensitive to temperature, and can
have long-term stability issues.
Feldspar measurements are generally obtained using
infrared (800–900 nm) LED stimulation at
200 milliwatts
424 LUMINESCENCE DATING, INSTRUMENTATION
per square cm (mW/cm2). For quartz measurements optical
stimulation using LEDs is usually obtained using blue
(470 nm) and green (520 nm) stimulations at power densi-
ties typically varying between 50 and 100 mW/cm2.
Jain (2009) suggested the use of a violet laser diode
(405 eV) to extend the age range using quartz by probing
deeper traps not accessible by blue light stimulation. In sin-
gle-grain dating applications, solid state lasers emitting at
532 and 830 nm are used as stimulation light sources
(Duller et al., 2003). The power densities at the sample
for these light sources are estimated to be >2 orders of mag-
nitude higher than those achievable with conventional
LEDs.
Heating system
In TL dating it is important to be able to heat the samples
reproducibly using adjustable (although usually constant)
heating rates. The ability to vary the heating rates is partic-
ularly important for understanding TL processes.
Although OSL is an optical method, samples usually
require some thermal pretreatment to be able to compare
the luminescence resulting from the natural and
regenerated doses (e.g., Ankjærgaard et al., 2006). In
quartz and feldspar OSL dating, for instance, it is custom-
ary to preheat the samples in the range of 160–320  C.
Quartz OSL signals are also generally measured at an ele-
vated but constant temperature (125  C) to improve the
signal-to-noise ratio. Most luminescence dating systems
can heat up to 700  C using heating rates ranging between
0.01 and 10 K/s. From a purely instrumental point of view,
it is beneficial to perform heatings above 200  C in a N2
atmosphere to prevent oxidization of the heating element.
In TL measurements to high temperatures, it has the fur-
ther advantage of reducing spurious signals.
Irradiation sources
In luminescence dating the measured natural lumines-
cence signal is converted into an absorbed dose by com-
parison with regenerated signals induced in the
laboratory by administering known doses to the sample.
These signals can be regenerated using for instance
gamma, beta, or alpha emitters. For practical reasons it
is beneficial to be able to perform in situ irradiations.
Due to radiation protection issues, beta emitters, particu-
larly 90Sr/90Y, are usually preferred as in situ reference
sources. The maximum 90Sr/90Y electron energy is
2.27 MeV and the mean electron energy is about
600 keV. Two of the major disadvantages of using a beta
source are that the dose rate is fixed and that it may not
provide a spatially uniform dose across the sample area
(e.g., Spooner and Allsop, 2000; although this disadvan-
tage is not so important with a new type of 90Sr/90Y
sources). In contrast, X-ray sources have electronically
adjustable dose rates and have been shown to have
a high degree of spatial uniformity and are easily
shielded. However, X-ray sources are not widely used
in luminescence dating, because the dose rate is highly
sample dependent, probably because of high local ioni-
zation effects (Thomsen et al., 2006; Jain et al., 2007).
It has been suggested that by hardening the X-ray spec-
trum, one can get rid of this sample dependency, but this
suggestion has not yet been thoroughly tested. Alpha
sources (primarily 241Am) are only used in fine-grained
(4–11 mm) polymineral dating and only if one wishes to
determine the a-value (alpha to beta efficiency) for the
samples in question.
Bibliography
Ankjærgaard, C., Murray, A. S., and Denby, P. M., 2006. Thermal
pre-treatment in the OSL dating of quartz: is it necessary?
Radiation Protection Dosimetry, 119, 470–473.
Bøtter-Jensen, L., and Duller, G. A. T., 1992. A new system for
measuring OSL from quartz samples. Nuclear Tracks Radiation
Measurements, 20, 549–553.
Bøtter-Jensen, L., Duller, G. A. T., Murray, A. S., and Banerjee, D.,
1999. Blue light emitting diodes for optical stimulation of quartz
in retrospective dosimetry and dating. Radiation Protection
Dosimetry, 84, 335–340.
Bøtter-Jensen, L., Bulur, E., Duler, G. A. T., and Murray, A. S.,
2000. Advances in luminescence instrument systems. Radiation
Measurements, 32, 523–528.
Bøtter-Jensen, L., McKeever, S. W. S., and Wintle, A. G., 2003.
Optically Stimulated Luminescence Dosimetry. Amsterdam:
Elsevier.
Bøtter-Jensen, L., Thomsen, K. J., and Jain, M., 2010. Review of
optically stimulated luminescence (OSL) instrumental develop-
ments for retrospective dosimetry. Radiation Measurements,
45, 253–257.
Bulur, E., 1996. An alternative technique for optically stimulated
luminescence (OSL) experiment. Radiation Measurements,
26, 701–709.
Buylaert, J. P., Jain, M., Murray, A. S., Thomsen, K. J., and Lapp, T.,
2012. IR-RF dating of sand-sized K-feldspar extracts: a test of
accuracy. Radiation Measurements, 47, 759–765.
Clark-Balzan, L., and Schwenninger, J., 2012. First steps toward
spatially resolved OSL dating with electron multiplying
charge-coupled devices (EMCCDs): system design and image
analysis. Radiation Measurements, 47, 797–802.
Denby, P. M., Bøtter-Jensen, L., Murray, A. S., Thomsen, K. J., and
Moska, P., 2006. Application of pulsed OSL to the separation of
the luminescence components from a mixed quartz/feldspar
sample. Radiation Measurements, 41, 774–779.
Duller, G. A. T., Bøtter-Jensen, L., Murray, A. S., and Truscott,
A. J., 1999. Single grain laser luminescence (SGLL) measure-
ment using a novel automated reader. Nuclear Instruments and
Methods B, 155, 506–514.
Duller, G. A. T., Bøtter-Jensen, L., and Murray, A. S., 2003.
Combining infrared- and green-laser stimulation sources in sin-
gle-grain luminescence measurements of feldspar and quartz.
Radiation Measurements, 37, 543–550.
Fattahi, M., and Stokes, S., 2000. Extending the time range of lumi-
nescence dating using red TL (RTL) from volcanic quartz. Radi-
ation Measurements, 32, 479–485.
Huntley, D. J., Godfrey-Smith, D. I., and Thewalt, M. L. W., 1985.
Optically dating of sediments. Nature, 313, 105–107.
Huntley, D. J., Godfrey-Smith, D. I., and Haskell, E. H., 1991.
Light-induced emission spectra from some quartz and feldspars.
Nuclear Tracks and Radiation Measurements, 18, 127–131.
Jain, M., 2009. Extending the dose range: probing deep traps in
quartz with 3.06 eV photons. Radiation Measurements, 44,
445–452.
 LUMINESCENCE DATING, LOESS
Jain, M., and Ankjærgaard, C., 2011. Towards a non-fading signal
in feldspar: insight into charge transport and tunnelling from
time-resolved optically stimulated luminescence. Radiation
Measurements, 46, 292–309.
Jain, M., Bøtter-Jensen, L., Murray, A. S., Denby, P. M.,
Tsukamoto, S., and Gibling, M. R., 2005. Revisiting TL: dose
measurement beyond the OSL using SAR. Ancient TL, 23, 9–24.
Jain, M., Bøtter-Jensen, L., and Thomsen, K. J., 2007. High local
ionization density effects in X-ray excitations deduced from
optical stimulation of trapped charge in Al2O3:C. Journal of
Physics: Condensed Matter, 19, 116201–116215.
Krbetschek, M. R., Trautmann, T., Dietrich, A., and Stolz, W., 2000.
Radioluminescence dating of sediments: methodological
aspects. Radiation Measurements, 32, 493–498.
McKeever, S. W. S., Markey, B. G., and Axelrod, M. S., 1996.
Pulsed optically stimulated luminescence dosimetry using
a-Al2O3:C. Radiation Protection Dosimetry, 65, 267–272.
Richter, D., Richter, A., and Dornich, K., 2013 (Accepted).
Lexsyg – a new instrument for luminescence research.
Geochronometria, 40, 220–228.
Spooner, N. A., and Allsop, A., 2000. The spatial variation of dose-
rate from 90Sr/90Y beta sources for use in luminescence dating.
Radiation Measurements, 32, 49–55.
Thomsen, K. J., Bøtter-Jensen, L., Denby, P. M., and Murray, A. S.,
2006. Luminescence response to irradiation using mini X-ray
generators. Nuclear Instruments and Methods in Physics
Research Section B, 252, 267–275.
Thomsen, K. J., Jain, M., Murray, A. S., Denby, P. M., Roy, N., and
Bøtter-Jensen, L., 2008. Minimizing feldspar OSL contamina-
tion in quartz UV-OSL using pulsed blue stimulation. Radiation
Measurements, 43, 752–757.
Vandenberghe, D. A. G., Jain, M., and Murray, A. S., 2009. Equiv-
alent dose determination using a quartz isothermal TL signal.
Radiation Measurements, 44, 439–444.
Cross-references
Luminescence Dating
LUMINESCENCE DATING, LOESS
Helen M. Roberts
Department of Geography and Earth Sciences,
Aberystwyth University, Aberystwyth, UK
Synonyms
Infrared stimulated luminescence (IRSL); Optical dating;
Optically stimulated luminescence (OSL); Thermolumi-
nescence (TL)
Definition
Luminescence dating. A family of chronologic methods
typically applied to the commonly occurring minerals
quartz and feldspar, which exploits a time-dependent
signal that builds up in mineral grains by exposure to nat-
urally occurring ionizing radiation (principally from ura-
nium, thorium, and potassium). The methods assess the
time elapsed since these mineral grains were last exposed
to sunlight or to heating. In the case of sediments,
425
the event being dated is the last exposure to sunlight,
i.e., typically the time of deposition of the sediment.
Loess. A fine-grained terrestrial sedimentary deposit
derived from wind-blown dust.
Introduction
Loess deposits preserve one of the most detailed and
lengthy terrestrial records of Quaternary climate
change on Earth; the key to unlocking these records is
the development of a chronology. Early chronologies for
loess-paleosol sequences were based largely on correla-
tive techniques, comparing variations in paleoenvir-
onmental proxy data, such as particle size, loess/paleosol
thicknesses, and magnetic susceptibility records, and
where available using other tie points such as paleomag-
netic signatures and tephra layers to link various records.
A common assumption in such correlative techniques is
a continuous and near-constant sediment accumulation
rate over time, with no major hiatus. However, recent
high-resolution luminescence dating studies (e.g., Antoine
et al., 2001; Frechen et al., 2001; Roberts et al., 2001,
2003; Lai and Wintle, 2006) demonstrate that sediment
accumulation rates can vary considerably both spatially
and temporally and that hiatus can occur (e.g., Lu
et al., 2006; Stevens et al., 2006; Thiel et al., 2011a) show-
ing that this assumption is not always valid. This has
implications for comparisons of terrestrial, ice-core,
and marine oxygen isotope records (e.g., Zhou and
Shackleton, 1999) and emphasizes the importance of
establishing independent numerical chronologies where
possible. Luminescence dating has played an increasing
role in providing such chronologies for loess sequences,
from the earliest applications of the technique to terrestrial
sediments through to the most recent proposals for new
and/or novel luminescence dating methods and signals.
Much of the early work to derive numerical ages from
the upper portions of loess sequences used radiocarbon
dating, often applied to paleosol units weathered into the
loess. In contrast, luminescence dating offers the opportu-
nity to date the deposition event directly because it records
the last exposure to sunlight of the grains of quartz or feld-
spar that make up the sedimentary deposit. The family of
luminescence dating techniques also extends beyond the
limits of radiocarbon dating in most sedimentary environ-
ments and does not rely on the discovery of appropriate in
situ material (e.g., organic matter or snail shells in the case
of dating using radiocarbon or amino acid racemization)
which may pre- or postdate the deposition event and/or
which may be rare in wind-blown sediments. These rea-
sons, coupled with improvements in the accuracy and pre-
cision of luminescence dating over the last 15 years, mean
that increasingly, luminescence dating is becoming the
technique of choice for determining a numerical chronol-
ogy for eolian deposits including loess.
An excellent general review of luminescence dating is
given by Rhodes (2011) and see also the volume “Lumi-
nescence Dating.” Specific reviews of the application of
426 LUMINESCENCE DATING, LOESS
luminescence dating to loess are given by Wintle (1990)
and Zöller and Wagner (1990) for thermoluminescence
(TL), while Singhvi et al. (2001) consider the dating of
loess using both TL and optically stimulated luminescence
(OSL), and Roberts (2008) charts the developments of
luminescence dating as applied to loess from the earliest
days to the most recent developments (including OSL) at
the time of publication. There have, however, been some
significant advances since the publication of this last
review, several of which are mentioned within this
encyclopedia entry.
Historical overview of luminescence dating
of loess
Loess was the first terrestrial clastic sediment to which
luminescence dating was systematically applied, and loess
deposits have remained important both for the develop-
ment and the application of luminescence dating tech-
niques since the earliest pioneering work was conducted
(Wintle, 1981, 1982; Wintle and Brunnacker, 1982). The
fine-grained nature of loess tends to imply long-transport
distances prior to deposition, ideal for the effective
zeroing of the luminescence signal, which is a fundamen-
tal assumption and requirement of luminescence dating.
The dominance of silt-sized particles in loess means that
it is the 4–11 mm diameter “fine-grain” fraction, which is
typically prepared for dating. Most of the published lumi-
nescence dating studies of loess, particularly the earlier
studies, are based on the “polymineral fine-grain fraction,”
which is thus termed because it is a mixture of both quartz
and feldspar minerals of 4–11 mm diameter. More
recently, with the establishment of methods to monitor
and correct for sensitivity changes during the measure-
ment procedure, and given the faster optical bleaching of
the quartz OSL signal (Godfrey-Smith et al., 1988) com-
pared to other luminescence signals, some workers have
used extended fluorosilicic acid treatments of several
days’ duration to isolate pure quartz of 4–11 mm grain size
or to isolate quartz of coarse silt size (35–50 mm) from
loess which is more representative of the grain size distri-
bution. Occasionally, relatively coarse grains (>63 mm)
from loess deposits, usually of quartz, have been exam-
ined using luminescence techniques; coarse grains cannot
be transported over long distances, and such deposits are
typically associated with the transition between deposits
of sand (source-proximal deposits) and loess (more distal
deposits). Use of the 4–11 mm fine-grain fraction or the
hydrofluoric acid-etched >63 mm coarse-grained fraction
circumvents the need for complex dosimetry assessments
of the alpha and beta dose to the sample that are otherwise
required when working with intermediate grain sizes (i.e.,
11–63 mm).
The earliest luminescence works published on loess
were conducted using thermoluminescence (TL) of
polymineral fine grains. Here, the TL is dominated by
the signal from feldspars, although the quartz grains
present do also contribute to the TL signal; the relative
contributions of these two mineral groups are likely to
vary between samples due to changes in mineralogy, and
it is not always clear in TL measurements which mineral
is being analyzed. Another concern identified in TL dating
is the long daylight exposure times (several hours)
required to reset the TL signal to a low level at deposition
and the fact that an unbleachable residual TL signal
remains after this time. This is of particular significance
for young material, e.g., of Holocene age, as it could lead
to age overestimation; Berger (1987) showed that modern
silt could yield an apparent TL age of ~1 ka, and TL ages
determined in a study of Japanese loess by Watanuki and
Tsukamoto (2001) overestimated the tephra-derived age
of the early Holocene and last glacial sediments by more
than 10 and 5 ka, respectively.
Thermoluminescence remained the only signal used for
luminescence dating of sediments until the development
of optically stimulated luminescence signals for dating.
Initially, costly Ar-ion lasers (514 nm) were used to stim-
ulate an optical signal from quartz and feldspar (Huntley
et al., 1985), and then shortly afterward much less expen-
sive light sources were used to give an infrared stimulated
luminescence (IRSL) signal (880 nm) from feldspar (Hütt
et al., 1988). Later still, halogen lamp systems and then
green and then blue light-emitting diodes (LEDs) were
used for optical stimulation and made the study of OSL
signals from quartz more practical on financial grounds
than the argon-ion laser that was initially used. Use of an
OSL signal for dating sediments has the advantage over
TL signals since OSL signals are bleached more rapidly
and typically to a lower residual level than the TL signal.
For these various reasons, the IRSL signal grew in popu-
larity through the 1990s for dating loess, typically applied
to polymineral fine grains. During this time, it was not
uncommon to see studies generating and comparing both
TL and IRSL ages from loess and also comparing the ages
generated for these signals using different methods of
measurement (e.g., regenerative dose and additive dose,
multiple aliquot methods) too. At this time, there was
no clear consensus on the luminescence signal or measure-
ment method which gave rise to the most accurate and pre-
cise luminescence ages for loess or for other sediments.
One problem which potentially affected both TL and
IRSL signals from feldspars is the phenomenon of
“anomalous fading” (Wintle, 1973), whereby the loss
of unstable charge over time leads to underestimation
of the true age of deposition if it is not (or cannot be)
corrected. Some authors suggest that anomalous fading
of luminescence signals from feldspar is a ubiquitous
phenomenon (e.g., Huntley and Lamothe, 2001; Little
et al., 2002), while others find no significant evidence
for anomalous fading in their studies (e.g., Frechen and
Dodonov, 1998; Frechen et al., 2001), and some other
authors using feldspars do not consider it at all in their
studies. There is no consensus on the most appropriate
 LUMINESCENCE DATING, LOESS
method for assessing anomalous fading nor for
correcting it if it is detected.
Dating using quartz avoids the problems and uncer-
tainties associated with feldspars because the OSL signal
from quartz does not suffer from anomalous fading. How-
ever, only since the introduction of the single-aliquot
regenerative-dose (SAR) measurement procedure of Mur-
ray and Wintle (2000) for dating quartz has the quartz
component of loess been systematically examined. The
SAR procedure brought about a step change in the accu-
racy and precision of luminescence dating of quartz and
successfully addressed the key impediment to the earlier
use of quartz for dating loess, namely, monitoring and
correcting for sensitivity change which occurs either dur-
ing the measurement procedure as a result of thermal pre-
treatments or during burial of the sample. Since this
advance, there have been numerous successful applica-
tions of luminescence dating using SAR methods applied
to quartz from loess, but the upper age limit of this OSL
signal is relatively restricted in loess compared to other
sedimentary deposits due to the high dose rate of these
fine-grained deposits. An upper limit for quartz OSL ages
from loess of ~50–70 ka has been reported by several
authors (e.g., Zhou and Shackleton, 2001; Buylaert
et al., 2007, 2008; Lai, 2010), although older ages have
been reported by others. Feldspars therefore remained
potentially attractive for luminescence dating because
they saturate at higher values of equivalent dose (De) than
quartz and hence potentially extend the upper age limit of
dating; however, feldspar methods were still troubled by
the specter of anomalous fading.
A major breakthrough came in 2008 when Thomsen
et al. reported that the IRSL signal observed at temperature
of 225  C following a prior IR stimulation at 50  C was
much more stable than the conventional IRSL signal that
had been previously commonly used, measured at 50  -
C. Use of this and other elevated temperature “post-IR
IRSL” signals from feldspars potentially offered a means
of minimizing or even overcoming the phenomenon of
anomalous fading and quickly gave rise to methods which
were applied to and tested on loess deposits (e.g., Thiel
et al., 2011b; Li and Li, 2011; Roberts, 2012), extending
the upper age limit far beyond that of quartz OSL to
~200–300 ka in loess (e.g., Buylaert et al., 2012; Li and
Li, 2012). Work to test and validate these post-IR IRSL sig-
nals continues, but the various methods proposed show
great promise (e.g., see review by Buylaert et al., 2012),
and the work of Thomsen et al. (2008) potentially marks
the beginning of a revolution in feldspar dating similar to
that which occurred in quartz dating following the proposal
of the SAR measurement protocol by Murray and
Wintle (2000).
Applications of luminescence dating of loess
Loess deposits have played an important role in the recon-
struction of long-term climatic and paleoenvironmental
427
change and in correlations between the terrestrial and
marine record. These correlations have often been done
in the past by wiggle-matching of proxy records, but
increasingly where possible, these records are being
linked using independently derived absolute or numerical
age control in order to explore whether such records are
synchronous between sites and to facilitate the potential
investigation of leads and lags within paleoclimate and
paleoenvironmental systems. Although the uncertainty
associated with an individual luminescence age can be rel-
atively large (typically ~5–10 %), where numerous ages
are available, these uncertainties can be reduced using
Bayesian age-depth models which exploit the information
regarding the relative sequence of ages down-section (e.g.,
Muhs et al., 2013). The ability of luminescence techniques
to generate numerical ages relating directly to the deposi-
tion event, and the fact that the methods are applied
directly to mineral grains which can be found continu-
ously throughout the loess-paleosol sequence, means that
there have been very many applications of luminescence
dating of loess relating to paleoenvironmental or paleocli-
matic reconstruction. For example, paleoprecipitation
(e.g., Rao et al., 2013) and paleotemperature (e.g., Jia
et al., 2013) records have been reconstructed on a numer-
ical timescale, and changing pressure gradients have been
inferred (e.g., Muhs et al., 2013), as have wind-strength
fluctuations, and wind directions, and changes in the mon-
soon intensity (e.g., Maher and Hu, 2006; Stevens
et al., 2008). The influence of soil-forming durations on
the development of mineral magnetic signatures has also
been investigated, using luminescence ages to constrain
maximum soil-forming durations by dating the loess units
that bracket various paleosols (e.g., Maher et al., 2003).
One of the applications which has grown in importance
in recent years is the use of luminescence dating applied to
loess sequences to assess the magnitude and changes in
dust mass accumulation rates (MARs) over time, mea-
sured in g/m2/a. Through this work, the traditional view
of a constant and linear sediment accumulation rate over
time, with no major hiatus occurring in the sedimentary
record, has been challenged. Loess deposits were also tra-
ditionally viewed as a passive record of past climate
change, but increasingly the wind-blown dust that the
loess deposits are comprised of is being viewed as a poten-
tial agent of climate change capable of regional modifica-
tion of climate and is now being incorporated into some
climate models (e.g., Mahowald et al., 2006). Lumines-
cence dating has helped to provide some of the first empir-
ical evidence for this potential for regional modification of
climate by dust (e.g., Roberts et al., 2003), and critically
luminescence-derived chronologies underpin several of
the high-resolution terrestrial MAR determinations which
are providing key inputs to the dust component of climate
models (e.g., Albani et al., 2014).
428 LUMINESCENCE DATING, LOESS
Current challenges in luminescence dating of loess
The two key challenges posed in the application of lumi-
nescence dating to loess both concern the upper limit of
dating using luminescence signals. The first key challenge
is recognizing and defining the reliable upper age limit for
each of the signals within the family of luminescence dat-
ing techniques. For example, OSL signals have been
observed to grow in the laboratory far beyond what has
been described as a prudent upper limit for dating (i.e.,
beyond 85 % of the maximum achievable luminescence
signal when fitted with a single-saturating exponential
function). The reliability of OSL ages beyond this range,
particularly when obtained using an additional linear com-
ponent or a second saturating exponential function, is not
clearly established, and the origin of this additional com-
ponent or second saturating exponential function is not
understood. Recent work by Chapot et al. (2012) proposed
the comparison of laboratory-generated dose-response
curves and a “natural dose-response curve” generated
using loess samples of known age taken from the
Luochuan section of the Chinese Loess Plateau. This work
represents the first clear demonstration that OSL ages can
increasingly underestimate the true age of the deposits
long before the luminescence signal is saturated, and finds
the upper limit of reliable dating at this site to be consistent
with (although derived in a totally different way to) the
upper limits described by others and discussed earlier in
this entry. This novel approach of comparing the response
in the laboratory and in nature can also potentially be used
to investigate the upper limit of dating using other lumines-
cence signals and measurement protocols and is only made
possible due to the long record and stratigraphic framework
offered by the loess-paleosol sequence studied.
The second key challenge is to seek new methods and
signals which might potentially extend the upper limit
(s) of luminescence dating, particularly given the long
temporal records preserved in many loess deposits. The
situation is exaggerated for loess sediments in particular
because the environmental dose rate (Gy/ka) is high com-
pared to many other sedimentary deposits, due in part to its
fine-grained size and hence large surface area-to-volume
ratio. This high dose rate (commonly ~3.5–5 Gy/ka, i.e.,
~2–3 times greater than dose rates typically observed in
other deposits) means that the minerals within loess
deposits are not capable of extending as far back in time
(ka) as they can in other sediments even though the total
amount of trapped charge (Gy) can be the same. Several
methods of extending the age range of luminescence dat-
ing have been proposed and are still being explored,
including thermally transferred optically stimulated lumi-
nescence (TT-OSL; Wang et al., 2006), which was first
developed and applied to loess, and use of the violet stim-
ulated luminescence (VSL) signal which may potentially
extend the dating range of quartz to span the entire Quater-
nary period (Ankjærgaard et al., 2013). However, while
these new methods are potentially very exciting, they are
in the early stages of development, and much remains to
be done to demonstrate their ability to generate reliable
numerical ages.
Summary and conclusions
Loess deposits have played, and continue to play, a critical
role in the development of luminescence dating tech-
niques. In turn, chronologies derived using luminescence
dating techniques are playing an increasingly significant
role in deciphering the long and quasi-continuous record
of paleoenvironmental and paleoclimatic change pre-
served within these temporally and areally extensive
deposits of wind-blown dust.
Bibliography
Albani, S., Mahowald, N. M., Winckler, G., Anderson, R. F.,
Bradtmiller, L. I., Delmonte, B., Francois, R., Goman, M.,
Heavens, N. G., Hesse, P. P., Hovan, S. A., Kohfeld, K. E., Lu,
H., Maggi, V., Mason, J. A., Mayewski, P. A., McGee, D., Miao,
X., Otto-Bliesner, B. L., Perry, A. T., Pourmand, A., Roberts,
H. M., Rosenbloom, N., Stevens, T., and Sun, J. (2014) Twelve
thousand years of dust: the Holocene global dust cycle
constrained by natural archives. Climate of the Past Discus-
sions,10, 4277–4363.
Ankjærgaard, C., Jain, M., and Wallinga, J., 2013. Towards dating
Quaternary sediments using the quartz Violet Stimulated Lumi-
nescence (VSL) signal. Quaternary Geochronology, 18,
99–109.
Antoine, P., Rousseau, D.-D., Zöller, L., Lang, A., Munaut, A.-V.,
Hatté, C., and Fontugne, M., 2001. High-resolution record of
the last interglacial-glacial cycle in the Nussloch loess-palaeosol
sequences, Upper Rhine Area, Germany. Quaternary Interna-
tional, 76(77), 211–229.
Berger, G. W., 1987. Thermoluminescence dating of the Pleistocene
Old Crow tephra and adjacent loess, near Fairbanks, Alaska.
Canadian Journal of Earth Sciences, 24, 1975–1984.
Buylaert, J. P., Vandenberghe, D., Murray, A. S., Huot, S., De Corte,
F., and Van den Haute, P., 2007. Luminescence dating of old
(>70 ka) Chinese loess: a comparison of single-aliquot OSL
and IRSL techniques. Quaternary Geochronology, 2, 9–14.
Buylaert, J. P., Murray, A. S., Vandenberghe, D., Vriend, M., De
Corte, F., and van den Haute, P., 2008. Optical dating of Chinese
loess using sand-sized quartz: establishing a time frame for Late
Pleistocene climate changes in the western part of the Chinese
Loess Plateau. Quaternary Geochronology, 3, 99–113.
Buylaert, J.-P., Jain, M., Murray, A. S., Thomsen, K. J., Thiel, C.,
and Sohbati, R., 2012. A robust feldspar luminescence dating
method for Middle and Late Pleistocene sediments. Boreas,
41, 435–451.
Chapot, M. S., Roberts, H. M., Duller, G. A. T., and Lai, Z. P., 2012.
A comparison of natural- and laboratory-generated dose
response curves for quartz optically stimulated luminescence
signals from Chinese Loess. Radiation Measurements, 47,
1045–1052.
Frechen, M., and Dodonov, A. E., 1998. Loess chronology of the
Middle and Upper Pleistocene in Tadjikistan. Geologische
Rundschau, 87, 2–20.
Frechen, M., van Vliet-Lanoë, B., and van den Haute, P., 2001. The
Upper Pleistocene loess record at Harmignies/Belgium – high res-
olution terrestrial archive of climate forcing. Palaeogeography,
Palaeoclimatology, Palaeoecology, 173, 175–195.
 LUMINESCENCE DATING, LOESS
Godfrey-Smith, D. I., Huntley, D. J., and Chen, W.-H., 1988.
Optical dating studies of quartz and feldspar sediment extracts.
Journal of Luminescence, 14, 373–380.
Huntley, D. J., and Lamothe, M., 2001. Ubiquity of anomalous fad-
ing in K-feldspars and the measurement and correction for it in
optical dating. Canadian Journal of Earth Sciences, 38,
1093–1106.
Huntley, D. J., Godfrey-Smith, D. I., and Thewalt, M. L. W., 1985.
Optical dating of sediments. Nature, 313, 105–107.
Hütt, G., Jaek, I., and Tchonka, J., 1988. Optical dating: K-feldspars
optical response stimulation spectra. Quaternary Science
Reviews, 7, 381–385.
Jia, G., Rao, Z., Zhang, J., Li, Z., and Chen, F., 2013. Tetraether bio-
marker records from a loess-paleosol sequence in the western
Chinese Loess Plateau. Frontiers in Microbiology, 4, 199.
Lai, Z.-P., 2010. Chronology and the upper dating limit for loess
samples from Luochuan section in the Chinese Loess Plateau
using quartz SAR protocol. Journal of Asian Sciences, 37,
176–185.
Lai, Z. P., and Wintle, A. G., 2006. Locating the boundary between
the Pleistocene and the Holocene in Chinese loess using lumi-
nescence. The Holocene, 16, 893–899.
Li, B., and Li, S.-H., 2011. Luminescence dating of K-feldspar from
sediments: a protocol without anomalous fading correction.
Quaternary Geochronology, 6, 468–479.
Li, B., and Li, S.-H., 2012. Luminescence dating of Chinese loess
beyond 130 ka using the non-fading signal from K-feldspar.
Quaternary Geochronology, 10, 24–31.
Little, E. C., Lian, O. B., Velichko, A. A., Morozova, T. D.,
Nechaev, V. P., Dlussky, K. G., and Rutter, N. W., 2002. Quater-
nary stratigraphy and optical dating of loess from the east Euro-
pean Plain (Russia). Quaternary Science Reviews, 21,
1745–1762.
Lu, H. Y., Stevens, T., Yi, S. W., and Sun, X. F., 2006. An erosional
hiatus in Chinese loess sequences revealed by closely spaced
optical dating. Chinese Science Bulletin, 51, 2253–2259.
Maher, B. A., and Hu, M. Y., 2006. A high-resolution record of
Holocene rainfall variations from the western Chinese Loess Pla-
teau: antiphase behaviour of the African/Indian and East Asian
summer monsoons. The Holocene, 16, 309–319.
Maher, B. A., Hu, M. Y., Roberts, H. M., and Wintle, A. G., 2003.
Holocene loess accumulation and soil development at the west-
ern edge of the Chinese Loess Plateau: implications for magnetic
proxies of palaeorainfall. Quaternary Science Reviews,
22, 445–451.
Mahowald, N. M., Yoshioka, M., Collins, W. D., Conley, A. J., Fill-
more, D. W., and Coleman, D. B., 2006. Climate response and
radiative forcing from mineral aerosols during the glacial maxi-
mum, pre-industrial, current and doubled-carbon dioxide cli-
mates. Geophysical Research Letters, 33, L20705.
Muhs, D. R., Bettis, E. A., III, Roberts, H. M., Harlan, S. S., Paces,
J. B., and Reynolds, R. L., 2013. Chronology and provenance of
last-glacial (Peoria) loess in western Iowa and paleoclimatic
implications. Quaternary Research, 80, 468–481.
Murray, A. S., and Wintle, A. G., 2000. Luminescence dating of
quartz using an improved single-aliquot regenerative-dose pro-
tocol. Radiation Measurements, 32, 57–73.
Rao, Z., Chen,F., Cheng, H., Liu, W., Wang, G., Lai, Z., and
Bloemendal, J., 2013. High-resolution summer precipitation var-
iations in the western Chinese Loess Plateau during the last gla-
cial. Scientific Reports 3: Article number 2785
Rhodes, E. J., 2011. Optically stimulated luminescence dating of
sediments over the past 200,000 years. Annual Review of Earth
and Planetary Sciences, 39, 461–488.
429
Roberts, H. M., 2008. The development and application of lumines-
cence dating to loess deposits: a perspective on the past, present
and future. Boreas, 37, 483–507.
Roberts, H. M., 2012. Testing Post-IR IRSL protocols for
minimizing fading in feldspars, using Alaskan loess with inde-
pendent chronological control. Radiation Measurements, 47,
716–724.
Roberts, H. M., Wintle, A. G., Maher, B. A., and Hu, M. Y., 2001.
Holocene sediment accumulation rates in the western Loess Pla-
teau, China, and a 2,500 year record of agricultural activity,
revealed by OSL dating. The Holocene, 11, 477–483.
Roberts, H. M., Muhs, D. R., Wintle, A. G., Duller, G. A. T., and
Bettis, E. A., 2003. Unprecedented last-glacial mass accumula-
tion rates determined by dating of loess from western Nebraska.
Quaternary Research, 59, 411–419.
Singhvi, A. K., Bluszcz, A., Bateman, M. D., and Someshwar,
R. M., 2001. Luminescence dating of loess-palaeosol sequences
and coversands: methodological aspects and palaeoclimatic
implications. Earth-Science Reviews, 54, 193–211.
Stevens, T., Armitage, S. J., Lu, H. Y., and Thomas, D. S. G., 2006.
Sedimentation and diagenesis of Chinese loess: implications for
the preservation of continuous, high-resolution climate records.
Geology, 34, 849–852.
Stevens, T., Lu, H., Thomas, D. S. G., and Armitage, S. J., 2008.
Optical dating of abrupt shifts in the late Pleistocene East Asian
monsoon. Geology, 36, 415–418.
Thiel, C., Buylaert, J.-P., Murray, A. S., Terhorst, B., Tsukamoto, S.,
Frechen, M., and Sprafke, T., 2011a. Investigating the
chronostratigraphy of prominent palaeosols in Lower Austria
using post-IR IRSL dating. Quaternary Science Journal
(E&G), 60, 137–152.
Thiel, C., Buylaert, J.-P., Murray, A. S., Terhorst, B., Hofer, I.,
Tsukamoto, S., and Frechen, M., 2011b. Luminescence dating
of the Stratzing loess profile (Austria) – testing the potential of
an elevated temperature post-IR IRSL protocol. Quaternary
International, 234, 23–31.
Thomsen, K. J., Murray, A. S., Jain, M., and Bøtter-Jensen, L.,
2008. Laboratory fading rates of various luminescence signals
from feldspar-rich sediment extracts. Radiation Measurements,
43, 1474–1486.
Wang, X. L., Lu, Y. C., and Wintle, A. G., 2006. Recuperated OSL
dating of fine-grained quartz in Chinese loess. Quaternary Geo-
chronology, 1, 89–100.
Watanuki, T., and Tsukamoto, S., 2001. A comparison of GLSL,
IRSL and TL dating methods using loess deposits from Japan
and China. Quaternary Science Reviews, 20, 847–851.
Wintle, A. G., 1973. Anomalous fading of thermoluminescence in
mineral samples. Nature, 245, 143–144.
Wintle, A. G., 1981. Thermoluminescence dating of Late Devensian
loesses in southern England. Nature, 289, 479–480.
Wintle, A. G., 1982. Thermoluminescence properties of fine-grain
minerals in loess. Soil Science, 134, 164–170.
Wintle, A. G., 1990. A review of current research on TL dating of
loess. Quaternary Science Reviews, 9, 385–397.
Wintle, A. G., and Brunnacker, K., 1982. Ages of volcanic tuff in
Rheinhessen obtained by thermoluminescence dating of loess.
Naturwissenschaften, 69, 181–183.
Zhou, L. P., and Shackleton, N. J., 1999. Misleading positions of geo-
magnetic reversal boundaries in Eurasian loess and implications
for correlation between continental and marine sedimentary
sequences. Earth and Planetary Science Letters, 168L, 117–130.
Zhou, L. P., and Shackleton, N. J., 2001. Photon-stimulated lumi-
nescence of quartz from loess and effects of sensitivity change
on palaeodose determination. Quaternary Science Reviews,
20, 853–857.
430 LUMINESCENCE DATING, METEORITES
Zöller, L., and Wagner, G. A., 1990. Thermoluminescence dating of
loess-Recent developments. Quaternary International, 7–8,
119–128.
Cross-references
Feldspars
Luminescence Dating
Luminescence Dating, History
Luminescence Dating, Uncertainties and Age Range
Luminescence, Desert Dunes
Magnetostratigraphic Dating
Paleosol
Quartz
Radiation Dose Rate
Radiocarbon Dating of Terrestrial Carbonates
Sediments, Terrestrial (Paleomagnetism)
LUMINESCENCE DATING, METEORITES
Derek W. G. Sears
NASA Ames Research Center, Mountain View, CA, USA
Definition
Thermoluminescence is the light produced as a result of
heating. In crystalline silicate rocks, there are a variety of
possible energy levels at which electrons can reside, such
as the valence band and the conduction band. Ionizing radi-
ation passing through the crystals can transfer to the elec-
trons sufficient energy to promote them from the valence
band to the conduction band where they freely move
through the crystal lattice. If they locate “traps,” they can
remain at an elevated energy level. Traps are typically lat-
tice defects or impurity ions, an important example being
Mn2+ substituting for Ca2+ in a silicate lattice. Calcium is
an abundant element, and Ca-bearing minerals are common
in meteorites, most notably as the mineral plagioclase, a Ca
aluminosilicate. The traps are typically an electron volt or
so below the conduction band, and this energy has to be
supplied to release them. One means of supplying the
required energy is by heating the crystal. Once released,
the excited electrons can find their way to the ground state
by way of a “luminescence center” when their excitation
energy is released as a photon, normally in the visible range.
The number of excited electrons determines the intensity of
light emitted, and the temperature at which it is emitted, as
the crystal is heated, is determined by the depth of the trap
below the conduction band. Deeper traps require higher
heating temperatures. Being a solid-state property, there is
a distribution of traps about some mean value, and the
release of electrons from traps is governed by
a probability function, e.g., the Arrhenius equation, so TL
peaks are fairly broad. In addition, there are usually many
kinds of trap in a typical crystal lattice, so there are multiple
overlapping peaks in a typical experiment. The data are
recorded as light emitted as a function of heating tempera-
ture, and the resulting plot is called the “glow curve.”
Meteorites have a wide variety of composition and min-
eralogy, but plagioclase is present in most of them and
gives rise to their thermoluminescence. There are a few
exceptions, in which plagioclase is absent and the thermo-
luminescence is produced by pyroxene, an Mg silicate.
Pyroxene has at least two types of trap, one being Mn2+
substituting for Mg2+, but pyroxene luminescence is
“quenched” (destroyed) in the presence of Fe2+ which pro-
vides pathways for the electrons to reach the ground state
without a visible transition. A recent review of the applica-
tion of thermoluminescence to the studies of extraterres-
trial materials is given by Sears et al. (2013).
The thermoluminescence mechanism enables several
potential means of dating a meteorite. An early idea was
that the traps were dislocations caused by radiation dam-
age so that the number of traps built up with time and thus
provided a “clock.” The basis of the idea was an observed
correlation between induced TL and potassium-argon age
(Komovsky, 1961). Induced TL (sometimes referred to as
“artificial TL” in the early literature) is the TL signal
obtained by draining the meteorite of its natural TL and
exposing the meteorite to a standard test dose of radiation
in the laboratory. The idea of induced TL as a clock has not
withstood the test of time, however, since exposure to
radiation doses and fluencies greater than the meteorite
experienced in nature do not change the induced TL
response (Sears, 1980). Instead, it seems that meteorites
with low induced TL and too young potassium-argon ages
have experienced severe shock-heating events that melted
the plagioclase and caused the loss of Ar. Thus, while
induced TL failed as a proposed chronometer, it succeeded
as a new means of evaluating the shock-heating history of
meteorites.
Thermoluminescence dating applications that have
been proposed and developed in the literature and that
have some degree of success are as follows:
• Terrestrial age determination by TL fading. Terrestrial
age determination for meteorites, the time that has
lapsed since their fall from space, has taken on particu-
lar interest with the discovery of large numbers of mete-
orites in the prairies, deserts, and Antarctic ice fields.
A meteorite in space is exposed to large doses of radia-
tion from internal radioactivities and cosmic rays, cos-
mic rays being the more important in the short term.
The level of TL is determined by a competition between
decay at ambient temperatures and buildup depending
on the radiation dose rate. Once on Earth, the meteorite
is shielded from the high dose of cosmic rays experi-
enced in space and stored at a higher temperature on
Earth than in space, so the TL level adjusts to the new
environment by slowly fading to its new lower equilib-
rium level. Thus, one has the potential to determine how
long the meteorite has been on Earth from the extent of
fading. There are two issues to be resolved. One is that
meteorites enter the atmosphere with a wide variety of
natural TL levels, depending on the temperature of the
object in space and the degree of “shielding” available
 LUMINESCENCE DATING, SHELL-RICH DEPOSITS
to the sample, that is, how deeply it was buried within
a larger body while in space. There is no way around
this problem, and shielding is a problem shared with
the isotopic methods for determining terrestrial age
such as those using the decay of radiocarbon. The sec-
ond issue is the effect of different storage temperatures
on Earth, but these can be handled fairly well from
knowledge of climate temperatures and theoretical
modeling of the decay process. Several major studies
of this process have been reported (Sears and Durrani,
1980; Melcher, 1981; Benoit et al., 1991; Sears et al.,
2011), and it was one reason for routine natural TL mea-
surements being incorporated in the US Antarctic
Meteorite Program (Hasan et al., 1987).
• Terrestrial age by buildup of TL in the meteorite just
under the fusion crust. This is sometimes referred to,
somewhat misleadingly, as “fusion crust dating.” As
a meteorite passes through the atmosphere, its surface
is heated to the extent that a conspicuous melt crust is
formed. The high heat capacities of silicate rocks and
the ability of the melt to flow away from the meteorite
mean that the heat of atmospheric entry does not pene-
trate far into the meteorite. However, for a few millime-
ters under the fusion crust, the meteorite will have been
heated enough to drain the natural TL. Once on Earth,
environmental radiations will cause the TL to build up
again, and there is the potential to determine terrestrial
ages using the same methods as used for dating pottery.
The level of TL is measured and converted into an
absorbed dose using laboratory calibration experi-
ments. Then, the environmental radiation dose rate is
determined by placing dosimeters in the field. Finally,
absorbed dose is divided by dose rate to determine the
age. Only a few studies have been made of this tech-
nique, but the results seem promising (Miono and
Nakanishi, 1994; Akridge et al., 2000).
• Cosmic ray exposure age by natural TL buildup. There
has always been a sense that the thermoluminescence
phenomenon should yield a means of determining cos-
mic ray exposure age, literally the time that has lapsed
since the meteorite became exposed to cosmic rays.
Consistent with this idea is the observation that the nat-
ural TL levels observed throughout large pieces of
meteorite vary, reflecting the buildup and decay of pri-
mary and secondary radiations during cosmic ray expo-
sure. However, it is only recently that quantitative
efforts have been made (Biswas et al., 2011) in this
direction, and the technique is in its infancy.
Bibliography
Akridge, J. M. C., Benoit, P. H., and Sears, D. W. G., 2000. Terres-
trial age measurements using natural thermoluminescence of
a drained zone under the fusion crust of Antarctic ordinary chon-
drites. Meteoritics and Planetary Science, 35, 869–874.
Benoit, P. H., Sears, D. W. G., and McKeever, S. W. S., 1991. The
natural thermoluminescence of meteorites – II. Meteorite orbits
and orbital evolution. Icarus, 94, 311–325.
431
Biswas, R. H., Morthekai, P., Gartia, R. K., Chawla, S., and Singhvi,
A. K., 2011. Thermoluminescence of the meteorite interior:
a possible tool for the estimation of cosmic ray exposure ages.
Earth and Planetary Science Letters, 304, 36–44.
Hasan, F. A., Haq, M., and Sears, D. W. G., 1987. Natural thermo-
luminescence levels in meteorites, I: 23 meteorites of known
Al-26 content. In Proceedings of the 17th Lunar and Planetary
Science Conference, Part 2 Journal of Geophysical Research,
Vol. 92, pp. E703–E709.
Komovsky, G. F., 1961. Thermoluminescence of stony meteorites.
Meteoritika, 21, 64–69.
Melcher, C. M., 1981. Thermoluminescence of meteorites and their
terrestrial ages. Geochimica et Cosmochimica Acta, 45,
615–626.
Miono, S., and Nakanishi, A., 1994. Terrestrial ages of the Antarctic
meteorites measured by thermoluminescence of the fusion crust,
II. In Proceedings of the NIPR Symposium on Antarctic Meteor-
ites, Vol. 7, pp. 225–229.
Sears, D. W., 1980. Thermoluminescence of meteorites; relation-
ships with their K-Ar age and their shock and reheating history.
Icarus, 44, 190–206.
Sears, D. W., and Durrani, S. A., 1980. Thermoluminescence and
the terrestrial age of meteorites: some recent results. Earth and
Planetary Science Letters, 46, 159–166.
Sears, D. W. G., Yozzo, J., and Ragland, C., 2011. The natural ther-
moluminescence of Antarctic meteorites and their terrestrial ages
and orbits: a 2010 update. Meteoritics & Planetary Science,
46(1), 79–91.
Sears, D. W. G., Ninagawa, K., and Singhvi, A. K., 2013. Lumines-
cence studies of extraterrestrial materials: insights into their
recent radiation and thermal histories and into their metamorphic
history. Chemie der Erde, 73, 1–37.
Cross-references
Luminescence Dating
LUMINESCENCE DATING, SHELL-RICH DEPOSITS
Robert Hendricks and Alex Hodson
School Geography and Earth Sciences, McMaster
University, Hamilton, ON, Canada
Synonyms
OSL of anthropological shell middens; OSL of shell
mounds; OSL of geological layers rich in shells
Definition
Shell-Rich Deposits. Any stratum composed primarily of
shells which consist of either aragonite or calcite biomin-
eral. The remaining materials maybe a complex composi-
tion of siliciclastic or carbonate grains with or without
carbonate cement.
Introduction to shell-rich deposits and OSL
Although optically stimulated luminescence dating (OSL)
and/or infrared stimulated luminescence dating (IRSL) is
typically used in environments where the predominant
clasts or grains consist of sand-size particles of quartz and
feldspars, shell-rich deposits containing these grains can
432 LUMINESCENCE DATING, SHELL-RICH DEPOSITS
also be dated using this method. These shell-rich deposits
can be either of marine or lacustrine origin, or artificial con-
centrations created by humans. This allows OSL to be used
to answer chronologic questions pertaining to geologic pro-
cesses and features and well as anthropological sites and
events. Many studies traditionally obtain chronologic infor-
mation from shell deposits using radiocarbon (14C),
230
Th/234U, and amino acid racemization.
Geological shell deposits are formed when coastal pro-
cesses concentrate the shells by either winnowing fines or
depositing larger shells within a particular portion of the
near-shore zone. Examples of these environments include
shell lags created as the result of sea-level transgressions
and regressions as well as chenier plains formed from tidal
action (Rosendahl et al., 2007; Masselink et al., 2011).
Determining the age of these deposits can provide impor-
tant information about the history of coastal processes in
a region.
Anthropological concentrations of shells are generally
termed shell middens and are common in coastal environ-
ments. Shell middens are best defined by Waselkov (1987)
as a human created cultural deposit in which the main
material constructing the deposit is shells and shell frag-
ments. This loose definition can include deposits of shells
discarded by humans in a haphazard manner (i.e., a trash
pile) or shells used for a specific purpose, such as the con-
struction of a building or monument. The age of a shell
midden can be used to determine the age of artifacts of cul-
tural significance found within the midden. Shell middens
are formed from processes completely different from
coastal deposits, but differentiating them from natural
shell accumulations can be difficult (Rosendahl et al.,
2007). For further discussion on the cultural significance
and identification of shell middens, readers are referred
to Waselkov (1987) and Rosendahl et al. (2007).
While OSL has been used with success in the shallow
marine and coastal environments (see Jacobs (2008) for
a detailed review), the deposits that are targeted are gener-
ally composed of sand. Most studies that endeavor to
determine the age of a shell-rich horizon use OSL on sand
layers above or below the shell layers (Stone and Cupper,
2003; Parker and Goudie, 2007; Bateman et al., 2008).
However, OSL on quartz sand grains within shell layers
has been used with success (Lopez, 2007; Thompson
et al., 2007; Burdette et al., 2009; Pluckhahn et al.,
2014). There has also been a published report of using
IRSL to determine the age of feldspar grains from within
a shell-rich deposit (Mauz 1999). The dating of sediments
with abundant calcite cements commonly referred to
“beach rock” (Frechen et al. 2001; Tatumi et al. 2003;
Frechen et al. 2004) is similar to the dating of shell-rich
layers and shares the same challenges with regard to deter-
mination of the long-term dose rate.
This article will highlight as a case study in the use of
OSL in shell-rich deposits in several anthropological shell
middens at the Crystal River Site on the Crystal River in
Dixie County, Florida USA. Based on artifactual content
and radiocarbon dating, this site experienced human
occupation during the Woodland Period from 1000 BC
to 1000 AD and includes a number of burial and ceremo-
nial mounds (Thompson and Pluckhahn, 2010). The
mounds at the Crystal River Site provide an opportunity
to compare results of OSL and radiocarbon dating.
Advantages and disadvantages of using OSL
relative to other geochronologic methods
The main advantage that OSL has over other methods
commonly used to determine the age of shells and shell-
rich horizons, such as 14C, (230Th/234U), and amino acid
racemization (AAR), is that it determines the age of
a depositional process, not a biological event. This allows
for the age of the shell deposit to be determined regardless
of the age of the individual shells that compose the
deposit. Studies of modern marine shell deposits (Flessa
et al., 1993; Wehmiller et al., 1995; Carroll et al., 2003)
have shown that the average age of the shells composing
these deposits are far from modern. Flessa et al. (1993)
discovered that the average shell in their modern study
area was approximately 450 years old, with shells over
3.5 ka present. These results are not surprising when con-
sidering that these types of deposits represent time-
averaged assemblages (Kowalewski, 1996). This attribute
can prove to be a significant barrier to determining the true
age of the depositional process that brought all the shells
together. These studies show that any depositional process
occurring on a thousand year or smaller time scale might
not be resolved by dating of the shells themselves due to
the range of shell ages (Flessa et al., 1993).
Radiocarbon
Radiocarbon dating of marine shell requires a carbon res-
ervoir correction in order to account for the age of the inor-
ganic carbon taken up by the shell during growth (see
section on Cross-references). These corrections can be
quite large and require information about the conditions
where the shell formed. The use of OSL eliminates the
uncertainties associated with the estimation of the reser-
voir effects of a particular location. Studies such as
Bateman et al. (2008) have used the combination of OSL
and 14C to highlight the need for further investigation of
the reservoir effects of local sites. Dating sand found
within a shell deposit can provide independent age control
to sites that have had extensive research into the areas
local carbon reservoir. Stone and Cupper (2003) also
exemplify the usefulness of using OSL when contamina-
tion of radiocarbon samples is suspected.
230
Th/234U
The 230Th/234U dating method uses the ratio of parent
(234U)-to-daughter (230Th) isotopes within the uranium
decay chain to determine the age of a particular object
(see Cross-references). Mollusks, particularly oyster, are
a major component in many shell middens (Waselkov,
1987). Unfortunately oysters often exhibit post-burial
open-system behavior where parent and daughter isotopes
 LUMINESCENCE DATING, SHELL-RICH DEPOSITS
are added or removed through digenetic processes (Kaufman
et al., 1971; Blackwell and Schwarcz, 1995; McLaren and
Rowe, 1996; Edwards et al., 2003; Calsteren and Thomas,
2006). This makes oysters difficult to date with U-series.
OSL does not rely on isotopic ratios to determine the age
of a deposit and can be used on shell deposits that may expe-
rience this open-system behavior. However, the open-system
behavior can lead to problems with dose-rate determination,
because uranium is mobilized through this process.
Amino acid racemization (AAR)
AAR dating uses the racemization, or change, of
amino acid peptides that occur over time after the death
of an organism to determine the time since death (see sec-
tion on Cross-references). The rate of the racemization
process is dependent on the temperature, ph, and species
of the sample. Many of these factors have to be estimated
when attempting to determine the age of a sample
(Rutter and Blackwell, 1995; Wehmiller and Miller,
2000; Wehmiller, 2012). Due to the many factors associ-
ated with the determination of an AAR shell age, this
technique often lacks the high resolution needed in many
Holocene and Pleistocene geologic and archaeological
sites. OSL can provide the independent age confirmation
in shell deposits in which one may wish to attempt to
use AAR.
Heterogeneity and uranium uptake/loss
OSL requires that the annual radioactive dose rate of the
material be known to calculate the age of a sample. In
sandy deposits, this is relatively straightforward as sedi-
ment is generally homogeneous in composition and radio-
active dose within distances that affect the samples. In
shell-rich deposits, the larger shells can lead to the forma-
tion of heterogeneous radioactive doses to the sample
complicating the calculation of the annual dose rate. Fur-
thermore the open-system behavior of oyster shells with
respect to uranium can have a significant impact on dose
rates through time, leading to larger errors in the age esti-
mation of older shell-rich horizons (Zander et al., 2007;
Burdette et al., 2009). While this open-system behavior
has been studied (Prescott and Hutton, 1995; Olley et al.,
1996; Zander et al., 2007), more research specific to
OSL dose rates in shell-rich deposits is needed.
Obtaining and isolating quartz
There are some specific problems with sample collection
and preparation. Unlike sand deposits, the large shells or
dense shell concentrations can make it difficult to collect
a sample using a sample tube or core. To overcome this
problem, a sample taken by hand at night maybe neces-
sary. Samples of a shell-rich deposit may also need to con-
tain more material than a bulk sample from a sand-rich
environment so that enough quartz can be collected. After
the sample is prepared, it generally yields only a very
small amount of pure quartz. Quartz sand will adhere to
shells, and a gentle agitation maybe required to separate
433
quartz from the larger shells (brushing the sand off works
well). The sieved fraction needs some special care because
its high calcium carbonate fraction causes a vigorous reac-
tion at the beginning of the acid dissolution step and may
require several repeat dissolutions to isolate the non-
carbonate fraction.
Case study of Crystal River
Crystal River and Roberts Island are neighboring archaeo-
logical sites. Of the many mounds present, three mounds
were studied. One composed of about 98 % whole oyster
shells (Roberts Mound) and two constructed of sand with
few shell fragments (Mounds A and H). Radiocarbon
dates obtained from Roberts Mound as well as other arti-
facts at the site provide general age control and make the
site an excellent opportunity to use OSL on a shell-rich
deposit and compare it to the results of sand-rich deposits
(Pluckhahn et al., 2014).
Vertical push-core samples were taken from the sandy
Mounds A and H, while a sample collected by hand at
night was collected at Roberts Mound. A Harwell gamma
spectrometer was used in vertical holes to determine the
annual gamma dose rates in Roberts Mound to try to avoid
any homogeneity in dose rate that might have been pre-
sent. Less than about 2 % of the total materials in the
mound were sediment, the remainder being air space and
oyster shell. In Roberts Mound, the beta dose rate
(reconstructed from uranium, thorium, and potassium
concentrations) of the sediment and the beta dose rate con-
tributed by the shells were calculated separately. The shell
beta dose rate proved clearly more appropriate, as using
the sediment itself to calculate this leads to major underes-
timates of the age based on radiocarbon dating. Moreover,
the results of these calculations showed that the shells con-
tributed less than 10 % of the total annual dose compared
to the dose sourced from the sediment and cosmic radia-
tion sources (Pluckhahn et al., 2014). It was also found
that the oyster shell beta plus gamma dose rate comprised
only 50–60 micrograys per year, which was only about
20–25 % of the total dose rate to the quartz grains (the
remainder coming from cosmic rays). To a large extent,
this ameliorates any affects associated with changing dose
rate in the shells due to uranium uptake or loss during
open-system behavior at Roberts Mound. However, this
is only the case for sites where shell makes up such
a large proportion of the buried material.
Results of the OSL analysis of Roberts Mound correlate
well with radiocarbon ages from the mound. At the Crystal
River site, OSL ages of Mounds A and H showed some
significant age inversions, but a minimum age model in
other samples yielded agreement with radiocarbon ages.
This latter result is best explained by incomplete bleaching
of the sand during the construction of Mounds A and H
(Pluckhahn et al., 2014), which were ceremonial in nature.
434 LUMINESCENCE DATING, SHELL-RICH DEPOSITS
Summary and conclusions
OSL is not dependent upon dating biological processes,
thus having obvious advantages in some cases over radio-
carbon, 230Th/234U, and AAR in the dating of shell-rich
deposits. It can provide direct determination of
a depositional event and enclosing of archaeological mate-
rials, eliminating some of the possibilities of age mixing
and allowing for better resolution within Holocene
deposits. The main challenges of using OSL within
shell-rich horizons include open-system behavior affect-
ing dose rates, and in sandy ceremonial mounds, there is
a considerable possibility of incomplete zeroing as sand
was poured onto the mound surface during construction.
Bibliography
Bateman, M. D., Carr, A. S., Murray-Wallace, C. V., Roberts, D. L.,
and Holmes, P. J., 2008. A dating intercomparison study on Late
Stone Age coastal midden deposits, South Africa.
Geoarchaeology, 23(6), 715–741.
Blackwell, B., and Schwarcz, H. P., 1995. The uranium series dis-
equilibrium dating methods. In Rutter, N. W., and Catto, N. R.
(eds.), Dating Methods for Quaternary Deposits. St. John’s
Newfoundland: GEOtext2, Memorial University of Newfound-
land. St. John's, Newfoundland Canada.
Burdette, K. E., Jack Rink, W., Harley Means, G., and Portell,
R. W., 2009. Optical dating of the Anastasia Formation, north-
eastern Florida, USA. Southeastern Geology, 46(4), 173–185.
Carroll, M., Kowalewski, M., Simoes, M. G., and Goodfriend,
G. A., 2003. Quantitative estimates of time-averaging in
terebratulid brachiopod shell accumulations from a modern trop-
ical shelf. Paleobiology, 29(3), 381–402.
Edwards, R. L., Gallup, C. D., and Cheng, H., 2003. Uranium-series
dating of marine and lacustrine carbonates. Reviews in Mineral-
ogy and Geochemistry, 52(1), 363–405.
Flessa, K. W., Cutler, A. H., and Meldahl, K. H., 1993. Time and
taphonomy: quantitative estimates of time-averaging and strati-
graphic disorder in a shallow marine habitat. Paleobiology,
19(2), 266–286.
Frechen, M., Dermann, B., Boenigk, W., and Ronen, A., 2001.
Luminescence chronology of aeolianites from the section at
Givat Olga-coastal plain of Israel. Quaternary Science Reviews,
20, 805–809.
Frechen, M., Neber, A., Tsatskin, A., Boenigk, W., and Ronen, A.,
2004. Chronology of Pleistocene sedimentary cycles in the Carmel
Coastal Plain of Israel. Quaternary International, 121, 41–52.
Pluckhahn, T.J., Hodson, A.D., Rink, W.J., Thompson, V.D., Hen-
dricks, R.R., Doran, G., Farr, G., Cherkinsky, A., Norman, S.
P., 2014. Radiocarbon and luminescence age determinations on
mounds at Crystal River and Roberts Island, Florida, USA.
Jacobs, Z., 2008. Luminescence chronologies for coastal and
marine sediments. Boreas, 37, 508–535.
Kaufman, A., Broecker, W. S., Ku, T.-L., and Thurber, D. L., 1971.
The status of U-series methods of mollusk dating. Geochemica
et Cosmochimica Acta, 35, 115–1183.
Kowalewski, M., 1996. Time-averaging, overcompleteness, and the
geological record. The Journal of Geology, 104(3), 317–326.
Lopez-Cadavid, G., 2007. The Late Quaternary Evolution of the
Apalachicola Barrier Island Complex, Northeast Gulf of Mex-
ico, as Determined from Optical Dating. Ph.D. Dissertation,
Hamilton, Canada, McMaster University.
McLaren, S.J., and Rowe, P.J., 1996, The reliability of uranium-
series mollusc dates from the western Mediterranean basin: Qua-
ternary Science Reviews (Quaternary Geochronology), vol. 15,
pp. 709-717.
Masselink, G., Hughes, M. G., and Knight, J., 2011. Introduction to
Coastal Processes and Geomorphology, 2nd edn. London:
Hodder Education.
Mauz, B., 1999. Late Pleistocene records of littoral processes at the
Tyrrhenian Coast (Central Italy): depositional environments and
luminescence chronology. Quaternary Science Reviews, 18,
1173–1184.
Olley, J. M., Murray, A., and Roberts, R. G., 1996. The effects of
disequilibria in the uranium and thorium decay chains on burial
dose rates in fluvial sediments. Quaternary Science Reviews,
15, 751–760.
Parker, A. G., and Goudie, A. S., 2007. Development of the Bronze
Age landscape in the southeastern Arabian Gulf: new evidence
from a buried shell midden in the eastern extremity of the Rub’
al-Khali desert, Emirate of Ras al-Khaimah, U.A.E. Arabian
Archaeology and Epigraphy, 18, 132–138.
Pluckhahn, T. J., Thompson, V. D., and Weisman, B. R., 2010.
Toward a new view of history and process at Crystal River
(8CI1). Southeastern Archaeology, 29(1), 164–181.
Prescott, J. R., and Hutton, J. T., 1995. Environmental dose rates
and radioactive disequilibrium from some Australian lumines-
cence dating sites. Quaternary Science Reviews, 14, 439–448.
Rink, W. J., Dunbar, J., and Burdette, K. E., 2012. The Wakulla
Springs Lodge Site 8WA329 2008 excavations and new OSL
dating evidence. Florida Anthropologist, 65, 85–105.
Rosendahl, D., Ulm, S., and Weisler, M. I., 2007. Using foraminif-
era to distinguish between natural and cultural shell deposits in
coastal eastern Australia. Journal of Archaeological Sciences,
34, 1584–1593.
Rutter, N. W., and Blackwell, B., 1995. Amino acid racemization
dating. In Rutter, N. W., and Catto, N. R. (eds.), Dating Methods
for Quaternary Deposits. St. John’s, Newfoundland: Memorial
University of Newfoundland. GEOtext2.
Staubwasser, M., Henderson, G. M., Berkman, P. A., and Hall,
B. L., 2004. Ba, Ra, Th, and U in marine mollusk shells and
the potential of 226Ra/Ba dating of Holocene marine carbonate
shells. Geochimica et Cosmochimica Acta, 68(1), 89–100.
Stone, T., and Cupper, M. L., 2003. Last glacial maximum ages for
robust humans at Kow Swamp, southern Australia. Journal of
Human Evolution, 45, 99–111.
Tatumi, S. H., Kowata, E. A., Gozzi, G., Kassab, L. R. P., Suguio, K.,
Barreto, A. M. F., and Bezerra, F. H. R., 2003. Optical dating
results of beachrock, eolic dunes and sediments applied to sea-level
changes study. Journal of Luminescence, 102–103, 562–565.
Thompson, V. D., and Pluckhahn, T. J., 2010. History, complex
hunter-gatherers, and the mounds and monuments of Crystal
River, Florida, USA: a geophysical perspective. The Journal of
Island and Coastal Archaeology, 5, 33–51.
Thompson, J., Rink, W. J., and Lopez, G. I., 2007. Optically stimu-
lated luminescence age of the Old Cedar midden, St. Joseph Pen-
insula State Park, Florida. Quaternary Geochronology, 2,
350–355.
van Calsteren, P., and Thomas, L., 2006. Uranium-series dating
applications in natural environmental science. Earth-Science
Reviews, 75, 155–175.
Waselkov, G. A., 1987. Shellfish gathering and shell midden
archaeology. Advances in Archaeological Method and Theory,
10, 93–210.
Wehmiller, J. F., 2013. United States Quaternary coastal sequences
and molluscan racemization geochronology – what have they
meant for each other over the past 45 years. Quaternary Geo-
chronology, 16, 3–20.
Wehmiller, J. F., and Miller, G. H., 2000. In Noller, J. S., Sower,
J. M., and Lettis, W. R. (eds.), Quaternary Geochronology
Methods and Applications. Washington, DC: American Geo-
physical Union. Aminostratigraphic dating methods in quater-
nary geochronology.
 LUMINESCENCE DATING, SINGLE-GRAIN DOSE DISTRIBUTION
Wehmiller, J. F., York, L. L., and Bart, M. L., 1995. Amino acid
racemization geochronology of reworked Quaternary mollusks
on U.S. Atlantic coast beaches: implications for chronostra-
tigraphy, taphonomy, and coastal sediment transport. Marine
Geology, 124, 303–337.
Wehmiller, J. F., Harris, W. B., Boutin, B. S., and Farrell, K. M.,
2012. Calibration of amino acid racemization (AAR) kinetics
in United States mid-Atlantic Coastal Plain Quaternary mollusks
using 87Sr/86Sr analysis: evaluation of kinetic models and esti-
mation of regional Late Pleistocene temperature history. Quater-
nary Geochronology, 7, 21–36.
Zander, A., Degering, D., Preusser, F., Kasper, H. U., and
Brunckner, H., 2007. Optically stimulated luminescence dating
of sublittoral and intertidal sediments from Dubai, UAE: radio-
active disequilibria in the uranium decay series. Quaternary
Geochronology, 2, 123–128.
Cross-references
Amino Acid Racemization Dating
Carbonates, Marine Carbonates (U-Series)
Luminescence Dating
Luminescence Dating of Archaeological Sediments
Luminescence, Coastal Sediments
Radiocarbon Dating of Marine Carbonates
LUMINESCENCE DATING, SINGLE-GRAIN DOSE
DISTRIBUTION
Richard G. Roberts and Zenobia Jacobs
Centre for Archaeological Science, School of Earth and
Environmental Sciences, University of Wollongong,
Wollongong, NSW, Australia
Synonyms
Single-grain De distribution; Single-grain equivalent dose
distribution
Definition
A single-grain dose distribution refers to the spread in
measured dose values (and their associated uncertainties)
for individual mineral grains in luminescence dating;
usually these are independent dose estimates for tens to
hundreds of single grains from the same sample.
Introduction
In luminescence dating, an estimate is made of the equiv-
alent dose for a single grain or for a group of mineral
grains, the latter commonly referred to as an “aliquot.”
The equivalent dose (expressed in grays, Gy) corresponds
to the radiation energy absorbed by a mineral grain, and
stored in the form of trapped electrons, since the grain
was last exposed to sunlight (“bleached”) or was last
heated to a high temperature. But this only holds true for
an aliquot if each and every grain on the aliquot had been
bleached or heated to the same, sufficient extent. For this
reason, a single grain is the smallest meaningful unit of
analysis in luminescence dating, because each grain in a
435
sample may, in principle, have had a different bleaching,
heating, or postdepositional history.
An equivalent dose estimate for a single grain, or for a
single aliquot composed of multiple grains, is obtained
experimentally, so it has an associated uncertainty. The
uncertainty is usually expressed as the standard error, at
the 1-sigma (1s) or 68 % confidence interval. The esti-
mate of uncertainty is as important as the estimate of the
equivalent dose itself, and this is especially true for single
grains, because the size of the uncertainty can vary greatly
from grain to grain. A set of bivariate observations
(of equivalent dose and standard error) are, therefore,
commonly obtained for each sample, and the distribution
of these observed values should be examined before calcu-
lating the burial age of the sample or some other event of
interest.
As a general note of caution, it is important to ensure
that the quoted standard error encapsulates all of the mea-
surement uncertainties associated with estimation of the
equivalent dose. For single grains, the standard error
includes uncertainties associated with photon counting
statistics, instrumental reproducibility, mathematical
fitting of the dose–response curve to the measured lumi-
nescence data, and corrections for any spatial heterogene-
ity in the laboratory beta sources. Equivalent dose values
might appear more scattered than is actually the case if
the standard errors are underestimated, whereas dose dis-
tributions may look misleadingly homogeneous if the
standard errors are overestimated.
Graphical displays of equivalent dose
distributions
Researchers often wish to plot the distribution of equiva-
lent dose estimates, and there are several means of doing
so. Galbraith and Roberts (2012) review some of the most
popular approaches to displaying equivalent dose values.
Frequency distributions
A commonly used form of data display is the histogram.
This requires equivalent dose values to be sorted into
“bins” of some fixed size and then the number of observa-
tions summed for each bin to generate a frequency distri-
bution. In Figure 1, the bottom left-hand plot is a
histogram of the equivalent dose values for 120 aliquots
composed of sand-sized quartz grains from a fluvial sam-
ple. Most values cluster between 20 and 30 Gy, with a few
smaller values and some much larger ones. The distribu-
tion appears to be asymmetric and positively skewed.
Despite their apparent simplicity, however, histograms
present various problems of implementation and interpre-
tation. The bin size, for example, is an arbitrary choice, so
the shape of the histogram can be affected by the choice of
bin size. Furthermore, equivalent dose histograms include
information about the distribution of the “true” doses (i.e.,
the doses that would have been observed if we could do so
without measurement error) and the distribution of the
standard errors, among other things.
436 LUMINESCENCE DATING, SINGLE-GRAIN DOSE DISTRIBUTION

Standard error (Gy) Relative standard error (%)

10 25

5 15

5
0
0 40 80 120 10 20 40 60 120

Number of aliquots Number of aliquots

50 30
40
30 20
20
10
10
0 0
0 40 80 120 10 20 40 60 120
Equivalent dose (Gy) Equivalent dose (Gy)
Luminescence Dating, Single-Grain Dose Distribution, Figure 1 Histograms of equivalent dose values and scatter plots of their
standard errors for 120 single aliquots of a fluvial sample. The plots in the left-hand column use a linear scale for the equivalent dose
and show the absolute standard errors in Gy. The right-hand panels use a logarithmic scale for the equivalent dose (i.e., equal
divisions of log dose) and show the relative standard errors in % (From Galbraith and Roberts (2012)).

In Figure 1, the chosen bin size is 5 Gy. This is
larger than the standard errors for most of the equivalent
dose estimates, which are displayed in the top left-hand
plot. The combination of these two plots is useful, as it
allows the reader to see the distribution of equivalent
doses and the size of their standard errors, which is
not something that can be judged from the histogram
alone. The standard errors of this sample show an inter-
esting feature that is typical of many luminescence
samples: the size of the uncertainty increases with the
size of the equivalent dose. Because of this, larger equiv-
alent dose values will tend to scatter more in a histogram
than smaller values and thereby create a positive skew,
regardless of the bleaching and burial history of the
grains.
To compensate for this effect, it is useful to take the nat-
ural logarithm of each equivalent dose value and to
express the standard error as the relative standard error
(i.e., the standard error in Gy divided by the equivalent
dose in Gy). Logarithmic transformations are common-
place in statistics and exploit the fact that the relative stan-
dard error of an estimate is approximately equal to the
absolute standard error of its natural logarithm.
The two right-hand plots in Figure 1 show the same
data as in the left-hand column but with the relative stan-
dard errors plotted in the top right-hand panel and the
equivalent doses presented on a logarithmic scale. The rel-
ative standard errors do not increase with the size of the
equivalent dose, and the equivalent dose distribution is
much closer to normal (Gaussian) in shape. As the latter
may be interpreted differently from the bottom left-hand
histogram, it is worthwhile displaying data in different
ways before drawing conclusions about the likely
bleaching and burial history of a sample from its equiva-
lent dose distribution.
Another form of the histogram, often referred to as a
“probability density function,” is sometimes used in lumi-
nescence dating. Probability density functions have the
appealing appearance of “smoothed” histograms, but they
suffer from many statistical shortcomings, and their use is
strongly discouraged (Galbraith, 1998, 2010; Galbraith
and Roberts, 2012).
Radial plots
The preferred form of data display for equivalent dose
values is the “radial plot” (Galbraith, 1988; Galbraith
et al., 1999; Galbraith and Roberts, 2012). This plot is par-
ticularly useful to display single-grain (and single-aliquot)
dose distributions because both the equivalent dose value
and its standard error are shown for each and every grain
(or aliquot), and this allows a number of other features of
the distribution to be discerned at the same time. Figure 2
is a radial plot of the same data as shown in Figure 1. Each
of the open circles denotes the equivalent dose value and
relative standard error for a single aliquot. The equivalent
dose can be read by drawing a line from the zero point on
the “standardized estimate” axis (on the left-hand side),
through the data point of interest, to intersect the radial
axis on the right-hand side. The point of intersection is
the equivalent dose (De). The relative standard error on
this estimate is read by drawing a vertical line from the
same data point to intersect the horizontal axis at the bot-
tom. The relative standard error is shown in % (i.e., as
the standard error divided by the equivalent dose, multi-
plied by 100), and the “precision” is the reciprocal of the
relative standard error.
 LUMINESCENCE DATING, SINGLE-GRAIN DOSE DISTRIBUTION 437

120 density of data points consistent with equivalent dose

values of 20–30 Gy, so the “frequency” aspect of the his-
De (Gy) togram is inherent in the radial plot also. Second, the plot
shows clearly the existence of three very high equivalent
60 dose values and one very low value, each measured with
standardized reasonable precision (i.e., a relative standard error of
estimate
less than 10 %).
40 The use of log values may not be appropriate for young
2
0
samples that contain grains with equivalent dose values
−2 close to zero or negative dose estimates within error of
30 zero. In such instances, the unlogged or modified
log-transformed versions of the radial plot can be used
(Arnold et al., 2009; Galbraith, 2010; Galbraith and
20 Roberts, 2012). Example radial plots for samples depos-
ited in a variety of contexts are provided by Jacobs and
Roberts (2007) and Galbraith and Roberts (2012), the lat-
relative standard error ter summarizing the many appealing features of radial
20% 10% 7.5% 10 plots as follows:
0 5 10 15 • Equivalent dose values can be viewed simultaneously
precision with their standard errors.
Luminescence Dating, Single-Grain Dose Distribution, • Data are sorted automatically so that the equivalent
Figure 2 A radial plot of the same data as displayed in Figure 1. dose values can be distinguished easily in terms of their
The equivalent dose values are plotted on a log scale, and the relative precisions.
relative standard errors (in %) are indicated on the horizontal • Any overdispersion can be identified at a glance.
axis (From Galbraith and Roberts (2012)).
• Patterns in the data, such as the clustering of equivalent
dose values, the presence of outliers, or the existence of
A benefit of displaying the data in this form is that the multiple, discrete dose components (see below), can be
equivalent dose values measured with the highest preci- recognized before calculating the sample age.
sion (i.e., the smallest relative standard errors) fall furthest
to the right, whereas those measured with least precision
lie furthest to the left. In addition to this self-sorting of Age models for equivalent dose distributions
values by precision, the use of the standardized estimate After a visual inspection of the equivalent dose distribu-
allows other aspects of the distribution to be conveniently tion, it is common for researchers to then estimate the
assessed at a glance. The standardized estimate represents equivalent dose for the population of grains (or aliquots)
the unit standard deviation applicable to all of the data related most closely to the event of interest. This may be
points. For each point, it is calculated as the equivalent the burial time of the most fully bleached grains in the
dose minus a chosen reference value (such as the average sample or the last time that the grains were heated to a high
equivalent dose for the entire distribution), divided by the temperature. In sediment dating, the value often sought is
relative standard error for that point. Accordingly, any the equivalent dose corresponding to the most recent
band of width  2 units projecting from the standardized bleaching event. But if a sample had been disturbed after
estimate axis will capture 95 % of the points if they are sta- burial and become contaminated by younger or older intru-
tistically consistent at 2s. sive grains, then it is useful to be able to identify and
In Figure 2, the equivalent doses are plotted on a log exclude these grains before calculating the sample age.
scale (as in the right-hand panels of Figure 1), so the stan- To allow for these and other possibilities requires a
dardized estimates and relative standard errors are based range of models based on sound statistical principles. Sev-
on the log equivalent dose values. Clearly, no single band eral parametric models, developed originally for fission
extending  2 units from the standardized estimate axis track dating (Galbraith and Green, 1990; Galbraith and
can capture 95 % of the equivalent dose values, so there Laslett, 1993), have been adapted for luminescence data
is some additional spread, which can be expressed in terms and are now in widespread use. Galbraith and Roberts
of the “overdispersion.” The latter term is used to describe (2012) review these different “age models,” although it
the scatter among the equivalent dose values over and should be borne in mind that they are rarely applied to ages
above that due to the measurement error associated with per se in luminescence dating. Instead, these models are
each observation (Galbraith et al., 2005; Galbraith and applied to equivalent dose values, because the information
Roberts, 2012). The data in Figure 2 are overdispersed required to determine single-grain ages – namely, the
by 32.0  2.3 %, as calculated from the “central age environmental dose rate specific to each grain – is not gen-
model” (see below). Two other noteworthy features of this erally known. At present, the same (sample mean) dose
distribution are visually apparent. First, there is a high rate is usually employed for all grains, while research
438 LUMINESCENCE DATING, SINGLE-GRAIN DOSE DISTRIBUTION
continues into methods of measuring and modeling grain-
specific dose rates.
A critical element of any decision-making process
involving statistical models is sample context. This should
dictate which, if any, of the models may be the most appro-
priate for the sample of interest. On the basis of its deposi-
tional context, for example, one could ask if a sample is
likely to have been fully or incompletely bleached at the
time of deposition. Similarly, one could consider if there
is any independent evidence that the stratigraphic integrity
of the site has been compromised, perhaps resulting in
postdepositional disturbance of the sample. In short, sta-
tistical models should not be applied without first taking
into account the archaeological or geologic context of a
sample, stratigraphic considerations, and other relevant
information, such as independent age control (Galbraith
et al., 2005; Galbraith and Roberts, 2012).
Common and central age models
These two models represent the most straightforward case
in luminescence dating: one in which the “average” equiv-
alent dose is the quantity of interest. This may be a useful
value to estimate for samples composed of grains that have
been well bleached before deposition, exposed to the same
environmental dose rate after burial, and remained unaf-
fected by any processes of postdepositional sediment
mixing. The equations for the common and central age
models are given in Galbraith et al. (1999) and Galbraith
and Roberts (2012), together with worked examples in
the former. The two models are identical insofar as they
both weight each equivalent dose by the inverse square
of its standard error, which is a standard statistical method
of averaging values with differing precisions. A major dif-
ference, however, is that the central age model also calcu-
lates – and explicitly includes in the standard error – the
extent of any overdispersion among the equivalent dose
values. If there is zero overdispersion (i.e., all of the spread
among the equivalent doses can be accounted for by the
individual measurement uncertainties alone), then the cen-
tral age model reduces to the common age model. But
even for samples of quartz grains that are known or are
thought to have been fully bleached at deposition,
overdispersion values of 10–20 % are commonplace
(Galbraith et al., 2005; Jacobs and Roberts, 2007; Arnold
and Roberts, 2009), so the central age model is usually the
more appropriate of the pair.
As illustrated by comparing the two lower plots in
Figure 1, log equivalent doses typically produce a much
closer approximation to a normal (Gaussian) distribution
than do their unlogged counterparts. This is due to the size
of the absolute standard error increasing in concert with
the equivalent dose (see top left-hand panel), a common
feature of many luminescence samples. For this reason,
the most widely used form of the central age model utilizes
log equivalent dose values and the relative standard errors;
the resulting model estimate of equivalent dose approxi-
mates the geometric mean. If the measured equivalent
doses are zero or negative, as can occur with some young
samples and with modern samples, then log transforma-
tions are not applicable, and the unlogged version of the
central age model could be used instead (Arnold
et al., 2009).
Minimum age models
The assumption of complete bleaching before deposition
and lack of disturbance thereafter will not be true for many
samples. For example, sediment grains transported under-
water may not be fully bleached, and some deposits may
be subject to bioturbation or human disturbance. In such
cases, the “average” equivalent dose will be of little inter-
est. For samples that are thought to contain some grains
that were fully bleached before deposition and others that
were only partially bleached, the minimum age model can
provide a useful means of estimating the equivalent dose
associated with the subset of well-bleached grains. Details
of this model are given in Galbraith et al. (1999) and Gal-
braith and Roberts (2012). Alternative approaches that are
philosophically similar to the minimum age model are
rarely based on well-established statistical principles, so
the statistical properties of the resulting equivalent dose
and uncertainty estimates are ambiguous, at best.
The most widely used version of the minimum age
model assumes that the log equivalent doses form a trun-
cated normal distribution, with the lower truncation point
corresponding to the average log dose of the subset of
fully bleached grains. This assumed distribution is one
of several that could be made, but it is convenient to
assume a normal distribution in the absence of indepen-
dent support for another type. The minimum age model
was originally developed with four adjustable parameters,
but it has been shown that data sets with small numbers of
equivalent dose values or less dispersed distributions are
often fitted better using a three-parameter model, in which
the lower truncation point equals the mean of a normal dis-
tribution. As with the central age model, the unlogged ver-
sion of the minimum age model may be more appropriate
for samples with zero or negative measured equivalent
doses (Arnold et al., 2009).
When implementing the minimum age model, allow-
ance needs to be made for the extent of inherent
overdispersion among the equivalent dose values of the
subset of fully bleached grains. That is, well-bleached
samples commonly exhibit some amount of
overdispersion in their equivalent dose values, so an esti-
mate of this overdispersion should be added to the relative
standard error of each equivalent dose before running the
model. This estimate could be obtained from a well-
bleached sample of the same mineral that is similar in
age and derived from the same source. In the absence of
independent data, one could conduct a sensitivity test of
the model using overdispersion values of 10 % and
20 %, for example.
Minimum age models have been tested in numerical
simulations and in comparisons against independently
 LUMINESCENCE DATING, SINGLE-GRAIN DOSE DISTRIBUTION
dated Holocene samples from a variety of geomorphic set-
tings (Olley et al., 2004; Arnold et al., 2009). The latter
studies utilized single grains of quartz, but the minimum
age model can also be applied to single aliquots with one
additional caveat: the estimate of the minimum dose may
still be larger than the equivalent dose of the most recently
bleached grains unless at least one aliquot consists entirely
of fully bleached grains. For aliquots composed of rela-
tively few fully bleached grains, the minimum dose may
be much too large, and single-grain analysis may be
required to obtain an accurate estimate of depositional age.
Finite mixture model
For some samples, neither the mean nor minimum esti-
mates of equivalent dose may correspond to the event of
interest. This may be so for deposits affected by bioturba-
tion or for sites disturbed by human activities, both of
which may result in mixing of grains between units of dif-
ferent ages. In such circumstances, the finite mixture
model may be able to discern the existence of multiple,
discrete components in a single-grain equivalent dose dis-
tribution (Roberts et al., 2000; Galbraith and Roberts,
2012).
In the finite mixture model, the log equivalent doses for
single grains are assumed to represent a mixture of dis-
crete populations, each of which is normally distributed
with the same relative overdispersion. The extent of
overdispersion is one of the parameters in the model, as
is the number of discrete populations. In practice, the
model is run by initially fixing the number of components
at 2 and then inserting some reasonable overdispersion
values (such as 10 %, 15 %, and 20 %). The number of
fitted components is then increased to 3, using the same
overdispersion values as previously, and so on. The model
uses maximum likelihood to estimate the mean equivalent
dose and standard error for each fitted component and the
proportion of grains in each component. When the number
of components is set at 1, the finite mixture model is math-
ematically identical to the central age model.
The goodness-of-fit of the model to the data can be
assessed using the Bayes Information Criterion (BIC),
which takes into account the maximum likelihood esti-
mates and the number of fitted components; the smallest
BIC indicates the minimum number of components
needed to explain the dose distribution. It is important to
also check that the model estimates are sensible and to
compare them with the distribution of equivalent doses
on a radial plot, as no model should be used as a “black
box.” Straightforward explanations and worked examples
of the finite mixture model fitting procedure are given in
David et al. (2007) and Jacobs et al. (2008).
For 2- and 3-component mixtures of single grains that
were bleached and then given known laboratory doses,
the model could successfully recover the correct number
of discrete components, their relative proportions, and
the corresponding equivalent dose values, provided the
overdispersion parameter was small, known, or similar
439
for each component (Roberts et al., 2000; Jacobs
et al., 2006). But the finite mixture model may be unable
to identify discrete components in more complex equiva-
lent dose distributions, such as those associated with het-
erogeneously bleached grains that have been mingled
subsequently. The model should also not be applied to
the equivalent dose distributions of multigrain aliquots,
even for those consisting of very few grains. If aliquots
contain luminescent grains from more than one parent
population, then “phantom” components with intermedi-
ate doses can easily be generated – that is, equivalent dose
components that are not present in any of the single-grain
parent populations (Arnold and Roberts, 2009).
Summary
The graphical display of single-grain or single-aliquot
equivalent dose values can be accomplished in various
ways, of which the radial plot provides an effective means
of assessing all of the salient information at a glance. After
an initial visual inspection of the distribution, one or more
statistical “age models” can be used to estimate the equiv-
alent dose for the population of grains (or aliquots) related
most closely to the event of interest. Such models should
be supported by well-established statistical theory, but
the choice of model depends fundamentally on the scien-
tific context of each sample and on the purpose of the
investigation (Galbraith and Roberts, 2012).
Bibliography
Arnold, L. J., and Roberts, R. G., 2009. Stochastic modelling of
multi-grain equivalent dose (De) distributions: implications for
OSL dating of sediment mixtures. Quaternary Geochronology,
4, 204–230.
Arnold, L. J., Roberts, R. G., Galbraith, R. F., and DeLong, S. B.,
2009. A revised burial dose estimation procedure for optical dat-
ing of young and modern-age sediments. Quaternary Geochro-
nology, 4, 306–325.
David, B., Roberts, R. G., Magee, J., Mialanes, J., Turney, C., Bird,
M., White, C., Fifield, L. K., and Tibby, J., 2007. Sediment
mixing at Nonda Rock: investigations of stratigraphic integrity
at an early archaeological site in northern Australia, and implica-
tions for the human colonisation of the continent. Journal of
Quaternary Science, 22, 449–479.
Galbraith, R. F., 1988. Graphical display of estimates having differ-
ing standard errors. Technometrics, 30, 271–281.
Galbraith, R. F., 1998. The trouble with probability density plots of
fission track ages. Radiation Measurements, 29, 125–131.
Galbraith, R., 2010. On plotting OSL equivalent doses. Ancient TL,
28, 1–9.
Galbraith, R. F., and Green, P. F., 1990. Estimating the component
ages in a finite mixture. Nuclear Tracks and Radiation Measure-
ments, 17, 197–206.
Galbraith, R. F., and Laslett, G. M., 1993. Statistical models for
mixed fission track ages. Nuclear Tracks and Radiation Mea-
surements, 21, 459–470.
Galbraith, R. F., and Roberts, R. G., 2012. Statistical aspects of
equivalent dose and error calculation and display in OSL dating:
an overview and some recommendations. Quaternary Geochro-
nology, 11, 1–27.
Galbraith, R. F., Roberts, R. G., Laslett, G. M., Yoshida, H., and
Olley, J. M., 1999. Optical dating of single and multiple grains
of quartz from Jinmium rock shelter, northern Australia: Part I,
440 LUMINESCENCE DATING, UNCERTAINTIES AND AGE RANGE
experimental design and statistical models. Archaeometry,
41, 339–364.
Galbraith, R. F., Roberts, R. G., and Yoshida, H., 2005. Error varia-
tion in OSL palaeodose estimates from single aliquots of quartz:
a factorial experiment. Radiation Measurements, 39, 289–307.
Jacobs, Z., and Roberts, R. G., 2007. Advances in optically stimu-
lated luminescence dating of individual grains of quartz from
archeological deposits. Evolutionary Anthropology, 16,
210–223.
Jacobs, Z., Duller, G. A. T., Wintle, A. G., and Henshilwood, C. S.,
2006. Extending the chronology of deposits at Blombos Cave,
South Africa, back to 140 ka using optical dating of single and
multiple grains of quartz. Journal of Human Evolution,
51, 255–273.
Jacobs, Z., Wintle, A. G., Duller, G. A. T., Roberts, R. G., and
Wadley, L., 2008. New ages for the post-Howiesons Poort, late
and final Middle Stone Age at Sibudu, South Africa. Journal
of Archaeological Science, 35, 1790–1807.
Olley, J. M., Pietsch, T., and Roberts, R. G., 2004. Optical dating of
Holocene sediments from a variety of geomorphic settings using
single grains of quartz. Geomorphology, 60, 337–358.
Roberts, R. G., Galbraith, R. F., Yoshida, H., Laslett, G. M., and
Olley, J. M., 2000. Distinguishing dose populations in sediment
mixtures: a test of single-grain optical dating procedures using
mixtures of laboratory-dosed quartz. Radiation Measurements,
32, 459–465.
Cross-references
Luminescence Dating
Luminescence Dating, Uncertainties and Age Range
LUMINESCENCE DATING, UNCERTAINTIES AND
AGE RANGE
Jakob Wallinga1 and Alastair C. Cunningham2
1 sands for fine-grained dating). In the widely used single-
Soil Geography and Landscape group, Wageningen
University, Wageningen, The Netherlands
2 dose determination (Murray and Wintle, 2000, 2003;
Centre for Archaeological Science, School of Earth and
Environmental Sciences, University of Wollongong,
Wollongong, Australia
Definition
Luminescence ages have an uncertainty of at least 4–5 %,
mainly due to systematic errors in both dose rate
(conversion factors) and equivalent dose (source calibra-
tion) estimation. In most cases, the uncertainty will be
higher, due to random errors (e.g., spread in equivalent
doses) or uncertainty in assumptions (e.g., water content
fluctuations, burial history). Dating is possible for a wide
age range of a few decades to about half a million years,
although uncertainties are usually relatively large toward
the extremes of this range.
Uncertainties
As with any method, results of luminescence dating con-
tain errors or uncertainties. Adequate assessment of errors
is important, for instance, to correctly assess rates of pro-
cesses or leads and lags in natural or anthropogenic
systems, or contemporaneity of different sites (e.g.,
Guerin et al., 2013). It is common practice in the lumines-
cence community to present 1-sigma uncertainties, imply-
ing that there is a 68 % likelihood that the true answer is
within the presented range. This of course only holds if
all sources of uncertainty are adequately considered.
Error propagation in luminescence dating is not
straightforward. Uncertainties in both dose rate and
palaeodose estimation should be taken into account, as
both contribute equally to the uncertainty in the final age
estimate. Moreover, the errors should not only comprise
the measurement uncertainties (e.g., counting statistics)
but also the uncertainties in instrument calibration and
parameters used in calculations. Finally, difficult-to-
quantify errors may arise from assumptions made for
equivalent dose or dose rate calculation. The most-
important sources of uncertainty are discussed below,
differentiating between random and systematic errors.
The first arise mainly from counting statistics and may
be reduced by additional measurements (i.e., longer dura-
tion, more samples, larger samples). The latter are usually
related to instrument calibration, conversion factors used,
or assumptions made and cannot easily be reduced.
Equivalent dose uncertainty
Statistical aspects of equivalent dose estimation have
recently been discussed in detail by Galbraith and Roberts
(2012). Here we provide a short summary of the most rel-
evant information:
Individual aliquots
Luminescence measurements for luminescence dating are
carried out on subsamples; each of these aliquots consists
of a single grain or a group of grains (up to many thou-
aliquot regenerative dose (SAR) procedure for equivalent
Wallinga et al., 2000), repeated luminescence measure-
ments are made on the same sample, to compare the natu-
ral luminescence signal to artificially induced signals and
to monitor and correct for luminescence sensitivity
changes during this procedure. For each of the lumines-
cence measurements, the true luminescence signal is esti-
mated based on the measured value. The uncertainty in the
estimate is a function of the brightness of the signal and
that of the associated background signal that is subtracted
to obtain a net signal (Poisson statistics assumed). The
uncertainty in the equivalent dose estimate of the aliquot
should take the uncertainties in each of these individual
estimates into account, as well as the uncertainties in
fitting a dose–response curve through the data (see Duller,
2007; Galbraith and Roberts, 2012), and the additional
error arising from instrumental uncertainty (e.g., Duller,
2007).
Sample
In the next step, data obtained on the aliquots is combined
to obtain the best estimate for the equivalent dose of the
 LUMINESCENCE DATING, UNCERTAINTIES AND AGE RANGE
sample. For an ideal dataset showing a normal distribution
around the mean and identical error terms, this can be done
using the mean and standard error on the mean. Here the
standard error (standard deviation divided by the square
root of the number of measured aliquots) is used, as we
are interested in how well we know the mean value. For
estimates with widely differing uncertainties, it may be
beneficial to use a weighted mean, where the best known
estimates are given more importance than the lesser
known points. However, use of the weighted mean
assumes that there is a true burial dose to which all
grains/aliquots belong, and the departure of the measured
data from this value is due to the known uncertainties of
measurement (i.e., the error term). In the real world this
is not the case – even if the measurement uncertainties
were infinitely small, the equivalent doses would not be
identical.
The Central Age Model (CAM) (Galbraith et al., 1999)
provides a better estimate of the burial dose. The CAM
presumes instead that the single-aliquot equivalent doses
are drawn from a log-normal distribution; the parameters
of the distribution are then the “central age” (¼burial dose)
and the “overdispersion” (the relative standard deviation
of the underlying log-normal distribution) and are
determined by iteration or optimization. There is a
straightforward estimate of the uncertainty on both param-
eters implemented in the model.
For datasets affected by heterogeneous bleaching, the
true age is not represented by the mean or weighted mean,
but is more likely to be at the lower end of the equivalent
dose distribution. The correct estimate may be found using
methods that distinguish dispersion in the data due to het-
erogeneous bleaching and dispersion from other sources.
The Minimum Age Model (Galbraith et al., 1999) is most
widely applied, although an alternative approach has been
proposed by Thomsen et al. (2007); a study on young
floodplain deposits from Australia yielded similar results
for both methods (Sim et al., 2014). A crucial aspect in
interpreting equivalent dose distributions is to quantify
the spread due to sources other than heterogeneous
bleaching. Such spread firstly comes from counting statis-
tics, but this error can be adequately quantified and taken
into account; it does not contribute to the overdispersion
(Galbraith and Roberts, 2012). There are two categories
of overdispersion: (1) Intrinsic sources are the effect of
experimental limitations and can be quantified using dose
recovery experiments on uniformly gamma-dosed sam-
ples (Thomsen et al., 2007). (2) Extrinsic sources reflect
true variation in burial doses, e.g., due to beta dose rate
heterogeneity (e.g., Cunningham et al., 2012). Both
sources should be quantified to accurately interpret equiv-
alent dose distributions.
Arnold and Roberts (2009) report that 20 %
overdispersion is typical at single-grain level for well-
bleached samples, and Cunningham et al. (2011a) discuss
how this value changes with aliquot size. However, both
intrinsic and extrinsic sources of scatter are likely to vary
between samples (Arnold and Roberts, 2009) and with
441
absorbed dose (Thomsen et al., 2007, 2012). For this rea-
son, it is advisable to include an uncertainty in the
assumed overdispersion when adopting the Minimum
Age Model. Such an approach has been proposed by
Cunningham and Wallinga (2012); their bootstrapping
approach provides reasonable uncertainty estimates for
heterogeneously bleached samples, by linking uncertainty
to the number of aliquots consistent with the minimum
dose/age.
Calibration
The random errors in equivalent dose based on measure-
ment uncertainties and spread in distribution may in prin-
ciple be reduced by enlarging the dataset or improving the
measurements. For the final error estimate, these sources
of error should be combined with the systematic uncer-
tainty in calibration of the radioactive source used for
equivalent dose determination. Bos et al. (2006) have
described a detailed calibration procedure and argued that
the uncertainty in calibration is at least 2 %.
Other errors
The premise of luminescence dating methods is that the
dose–response of the natural luminescence signal can be
reconstructed in the laboratory, allowing the burial dose
to be inferred. Protocols like the SAR accomplish this by
taking shortcuts: using preheats to simulate the effect of
time on the OSL traps and high laboratory dose rates to
simulate the low dose rates in nature. Subtle imperfections
in this methodology can induce error in the equivalent
dose. For example, the OSL signal may be sensitive to
the rate at which the radiation dose is given, contributing
to the poorer accuracy observed for older samples
(Bailey, 2004); the signal is also sensitive to the choice
of preheat used (Wintle and Murray, 2006). These sources
of error are usually ignored, likely because it is very diffi-
cult to quantify the uncertainties associated with them. As
a consequence, the errors are not reflected in the quoted
uncertainty, but rather they define the practical limits of
OSL dating. Anomalous fading of the feldspar signal also
falls into this category of error; corrections for anomalous
fading are sometimes made, but uncertainties associated
with such corrections may be large and nevertheless are
rarely addressed explicitly (e.g., Kars et al., 2012).
Total equivalent dose uncertainty
For ideal samples, the total uncertainty in equivalent dose
is dominated by the uncertainty in the calibration of the
beta source and can be as small as 2 %. Measurement of
limited number of subsamples and/or scattered equivalent
dose distributions may lead to larger errors.
Dose rate uncertainty
Concentration measurements
As with the equivalent dose estimate, there are multiple
sources of error in the dose rate estimation. The sources
of error depend on the method used for dose rate determi-
nation. Usually radionuclide concentrations are estimated
442 LUMINESCENCE DATING, UNCERTAINTIES AND AGE RANGE
either by measuring activity concentrations (e.g., gamma
spectrometry) or by estimating radionuclide concentra-
tions based on measured chemical concentrations (e.g.,
ICP, neutron activation analysis). The quoted errors
should include both the random (counting statistics) and
systematic (instrument calibration) errors. Minimal feasi-
ble relative errors are about 3 % (Murray and Olley, 2002).
Conversion factors
In the second step, infinite matrix dose rates are calculated
from the radionuclide (activity) concentrations using con-
version factors (Guerin et al., 2011). Unfortunately, nei-
ther Guerin et al. (2011) nor previous publications on
this topic (e.g., Adamiec and Aitken, 1998) discuss uncer-
tainties in the conversion factors. Murray and Olley
(2002) estimated an uncertainty of 3 % on the conversion
factors, a value that is still reasonable for the updated
conversion factors (Guerin, pers. com.).
Attenuation factors
The third step involves the conversion of infinite matrix
dose rate to dose rates in the grains of interest. This dose
depends on the grain size due to partial shielding of the
inside of the grains to beta irradiation. This effect, com-
bined with effects of etching the surface of quartz grains
during sample preparation, is taken into account in the
beta attenuation factor (Brennan, 2003). The uncertainty
in beta attenuation factor is usually assumed to be around
2 % (Murray and Olley, 2002). Then, the shielding effect
of water should be considered (Aitken, 1998), where
1 % by weight of water decreases the dose rate by about
1 %. Any uncertainty on the time-integrated average water
content should be reflected in uncertainty in the water
attenuation factor and hence the dose rate. Uncertainties
in dose rate due to water content errors can be minor
(<3 %) for sandy deposits, especially when the water con-
tent history is well established, e.g., because it is clear that
the sampled material has always been above or below the
groundwater table. However, for clay-rich sediments
which can double in weight due to water uptake, uncer-
tainties in dose rate due to poorly known water content
history can be huge (>100 %). This is particularly relevant
to lake-bed sediment, where water content can also change
over time due to sediment compaction (e.g., Buylaert
et al., 2013)
Cosmic rays and alpha particles
Finally, the contribution of cosmic rays and internal alpha
particles to the total dose rate should be considered. The
cosmic dose contribution is well defined based on geo-
graphic location, height, and burial given by the present-
day burial depth (Prescott and Hutton, 1994; Madsen
et al., 2005). However, in many cases, the burial history
is not well established, leading to a relative uncertainty
of about 10 % in the cosmic dose contribution. In most
cases, the cosmic dose rate is a small portion of the total
dose rate, and uncertainties will have a minor impact on
overall uncertainties (<2 %). However, in exceptional
cases of shallow deposits with low radionuclide concen-
trations (e.g., quartz-rich deposits), the cosmic dose may
contribute up to 50 % of the total dose rate (e.g., Rink
and Lopez, 2010). The internal alpha dose is usually esti-
mated based on literature information (e.g., Vandenberghe
et al., 2008). Although the relative uncertainty in internal
alpha dose is large (up to 50 %), the effect on total dose
rate is usually limited (<2 %) due to the small relative con-
tribution to the total dose rate.
Other sources
In addition to the sources of error discussed above, addi-
tional uncertainties may arise for specific cases. When
feldspar minerals are used, the dose rate should include
the contribution from internal potassium and rubidium.
Usually, concentration of these radionuclides is not known
for the measured grains, and a standard value and uncer-
tainty is assumed (Huntley and Baril, 1997; Huntley and
Hancock, 2001). For fine-grained material, or
non-etched grains, the external alpha contribution should
be considered and require adequate assessment of the
alpha dose and alpha efficiency as well as their associated
uncertainties. Also, disequilibrium in decay chains may
add to the uncertainty in the final dose rate (e.g., Olley
et al., 1997).
Total dose rate uncertainty
For ideal samples, the uncertainty in dose rate is domi-
nated by uncertainties in radionuclide (activity) concentra-
tions (3 %) and conversion factors (3 %). The combined
effect of these is around 4 %, which is considered the min-
imum reasonably attainable uncertainty on dose rate. In
many cases, uncertainty will be greater due to uncer-
tainties in burial history and water content. Uncertainties
mostly vary between 5 % and 10 % for suitable and less-
suitable circumstances, respectively.
The calculation of dose rates and their uncertainties
described above make two key assumptions. The first is
the “infinite matrix” (IM) assumption, in which the radia-
tion emitted from the sample is equal to that coming in
from the surroundings. The second is the “uniform matrix”
assumption, in which all parts of the sample are identical.
When these assumptions are broken, a correction must be
made, implying an increase in the total uncertainty. For
example, an object or sedimentary bed within 20 cm of
the sample may break the IM assumption (e.g., Wallinga
and Bos, 2010) as will be the case for deposits that
remained at shallow depth for part of their burial history
(e.g., Madsen et al., 2005); an irregular sample compo-
nent, such as shells or pebbles, may break the UM assump-
tion (Cunningham et al., 2011b).
Age uncertainty
Simple error propagation is usually applied to obtain the
uncertainty on the age estimate from uncertainties in
equivalent dose and dose rate. Based on the minimal fea-
sible uncertainties presented above, an error of 4–5 % of
the age could be obtained (e.g., Ballarini et al., 2003;
 LUMINESCENCE DATING, UNCERTAINTIES AND AGE RANGE
Roberts and Plater, 2007). Smaller uncertainties are not
realistic, larger uncertainties are the norm due to limited
and/or scattered equivalent dose datasets, and uncer-
tainties in burial history and time-averaged water content.
For larger datasets, uncertainties may be reduced by
combining results from a single event layer or by incorpo-
rating prior information (e.g., the stratigraphic order) into
a Bayesian framework (Rhodes et al., 2003; Cunningham
and Wallinga, 2012). Care should be taken to differentiate
between random and systematic errors. When analyzing
the internal consistency of a dataset, only random errors
need to be included. However, when luminescence ages
are combined or compared with other age information,
the systematic errors should also be considered. This issue
is discussed in detail by Rhodes et al. (2003).
Age range
Luminescence dating is a family of methods, using differ-
ent luminescence signals of natural minerals. Quartz and
feldspar minerals are mainly used, and this discussion will
concentrate on these minerals.
Lower dating limit
Two aspects are important in discussing the lower age
limit for luminescence dating:
1. Resetting of the luminescence signal prior to deposi-
tion and burial. Incomplete resetting will result in an
overestimation of the burial age. Hence, it is advisable
to use a luminescence signal that is rapidly reset upon
light exposure. This is especially important for young
samples, as the remnant age is more likely to be large
compared to the burial age of the sample. As the fast-
component OSL signal of quartz is most rapidly reset,
quartz should be the mineral of choice and appropriate
methods should be used to maximize the contribution
of the fast component to the net signal used for analysis
(Cunningham and Wallinga, 2010). Madsen and Mur-
ray (2009) review the dating of young sediments and
show successful application to deposits less than
10 years old.
2. Luminescence sensitivity. The luminescence sensitiv-
ity of the quartz or feldspar minerals determines the
minimum detection limit of the signal. Usually this is
not an important constraint, as results can be improved
by measuring more aliquots and averaging results.
Ages of around 10 years can be obtained, although
the uncertainty tends to be large (c.50 %). In practice,
the lower age limit for precise ages is a few decades
(see Madsen and Murray, 2009).
Upper dating limit
The upper age limit of luminescence dating is determined
by the saturation of the luminescence signal used. Lumi-
nescence dose–response curves can usually be satisfacto-
rily approximated by a saturating exponential Eq. (1):
443
 
D
I ðDÞ ¼ I S 1  exp D0 ð1Þ
where I is the luminescence signal intensity (after back-
ground subtraction), IS is the saturation level, D is the
absorbed radiation dose, and D0 is indicative of the onset
of saturation, i.e., it characterizes the shape of the curve.
For high doses, the luminescence signal approximates the
saturation level and hardly increases with dose and cannot
be used to reliably infer the burial dose. Wintle and Murray
(2006) suggested that reliable burial dose estimation is lim-
ited to the area below 2*D0, and this (arbitrary) criterion is
generally applied. In some cases, dose–response curves
show an additional saturating exponential component or
linear component with continued growth at very high doses.
However, the suitability of these components for dating has
not been established (Murray et al., 2008), and using this
region for dating is therefore not advisable.
Saturation doses for the fast component of the quartz
OSL signal vary between samples, with average 2*D0
values around 150 Gy. The resulting age range depends
on the dose rate for the sample, but is usually no more than
150,000 years. However, due to sample-to-sample differ-
ences in both saturation levels and environmental dose
rates, the upper age limit for quartz OSL dating can widely
vary and may extend up to 400,000 years (Wallinga
et al., 2007).
Extending the luminescence age range is possible by
using other luminescence signals, either from quartz or
from feldspar. A range of methods have been proposed
that use luminescence signals from quartz that saturate at
higher doses. Although promising results have been
obtained using thermally transferred OSL (reviewed by
Duller and Wintle, 2012), and violet stimulated lumines-
cence (VSL, Ankjaergaard et al., 2013), these methods
are still in development and should be further tested before
routine application is feasible.
Feldspar IRSL signals saturate at higher doses than the
fast-component OSL signal of quartz, but the application
of feldspar dating has long been held back by athermal insta-
bility of feldspar luminescence signals (“anomalous fading,”
e.g., Wallinga et al., 2007). Recently, these problems have
been largely overcome using elevated temperature post-
infrared IRSL signal (Thomsen et al., 2008). Post-IR IRSL
methods have recently been tested and applied and are
now routinely determining ages up to approximately
500,000 years (e.g., Thiel et al., 2011; Buylaert et al., 2012;
Kars et al., 2012). Excellent agreement between post-
infrared IRSL dating and coupled ESR/230Th/234U dating
of tooth enamel at Mala Balanica cave in Serbia (Rink
et al., 2013) further supports the use of this method near
500,000 years. Following these encouraging findings, we
expect post-infrared IRSL dating methods to be further
developed in the near future and be widely used for dating
Quaternary sediments. However, the method is less suitable
for younger deposits as the post-infrared IRSL signal is not
readily bleachable (Kars et al., 2014), which could poten-
tially result in age overestimation.
444 LUMINESCENCE DATING, UNCERTAINTIES AND AGE RANGE

Bibliography Galbraith, R. F., Roberts, R. G., Laslett, G. M., Yoshida, H., and
Olley, J. M., 1999. Optical dating of single and multiple grains
Adamiec, G., and Aitken, M. J., 1998. Dose-rate conversion factors:
of quartz from Jinmium rock shelter, northern Australia: part I,
update. Ancient TL, 16, 37–50.
experimental design and statistical models. Archaeometry, 41,
Aitken, M. J., 1998. An Introduction to Optical Dating. London:
339–364.
Oxford University Press, p. 267.
Guerin, G., Mercier, N., and Adamiec, G., 2011. Dose rate
Ankjaergaard, C., Jain, M., and Wallinga, J., 2013. Towards dating
conversion factors: update. Ancient TL, 29, 5–8.
quaternary sediments using the quartz violet stimulated lumines-
Guerin, G., Murray, A. S., Jain, M., Thomsen, K. J., and Mercier,
cence (VSL) signal. Quaternary Geochronology, 18, 99–109.
N., 2013. How confident are we in the chronology of the transi-
Arnold, L. J., and Roberts, R. G., 2009. Stochastic modelling of
tion between Howieson’s Poort and Still Bay? Journal of Human
multi-grain equivalent dose (De) distributions: implications for
Evolution, 64, 314–317.
OSL dating of sediment mixtures. Quaternary Geochronology,
Huntley, D. J., and Baril, M. R., 1997. The K content of the
4, 204–230.
K-feldspars being measured in optical dating or in thermolumi-
Bailey, R. M., 2004. Paper I – simulation of dose absorption in
nescence dating. Ancient TL, 15, 11–13.
quartz over geological timescales and its implications for the pre-
Huntley, D. J., and Hancock, R. G. V., 2001. The Rb contents of the
cision and accuracy of optical dating. Radiation Measurements,
K-feldspar grains being measured in optical dating. Ancient TL,
38, 299–310.
19, 43–46.
Ballarini, M., Wallinga, J., Murray, A. S., Van Heteren, S., Oost,
Kars, R. H., Busschers, F. S., and Wallinga, J., 2012. Validating post
A. P., Bos, A. J. J., and Van Eijk, C. W. E., 2003. Optical dating
IR-IRSL dating on K-feldspars through comparison with quartz
of young coastal dunes on a decadal time scale. Quaternary Sci-
OSL ages. Quaternary Geochronology, 12, 74–86.
ence Reviews, 22, 1011–1017.
Kars, R. H., Reimann, T., Ankjaergaard, A., and Wallinga, J. 2014.
Bos, A. J. J., Wallinga, J., Johns, C., Abellon, R. D., Brouwer, J. C.,
Bleaching of the post-IR IRSL signal: new insights for feldspar
Schaart, D. R., and Murray, A. S., 2006. Accurate calibration of
luminescence dating. Boreas, 43, 780–791.
a laboratory beta particle dose rate for dating purposes. Radia-
Madsen, A. T., and Murray, A. S., 2009. Optically stimulated lumi-
tion Measurements, 41, 1020–1025.
nescence dating of young sediments; A review. Geomorphology,
Brennan, B. J., 2003. Beta doses to spherical grains. Radiation Mea-
109, 3–16.
surements, 37, 299–303.
Madsen, A. T., Murray, A. S., Andersen, T. J., Pejrup, M., and
Buylaert, J. P., Jain, M., Murray, A. S., Thomsen, K. J., Thiel, C.,
Breuning-Madsen, H., 2005. Optically stimulated luminescence
and Sohbati, R., 2012. A robust feldspar luminescence dating
dating of young estuarine sediments: a comparison with 210Pb
method for middle and late Pleistocene sediments. Boreas, 41,
and 137Cs dating. Marine Geology, 214, 251–268.
435–451.
Murray, A. S., and Olley, J. M., 2002. Precision and accuracy in the
Buylaert, J. P., Murray, A. S., Gebhardt, A. C., Sohbati, R.,
optically stimulated luminescence dating of sedimentary quartz:
Ohlendorf, C., Thiel, C., Wastegard, S., and Zolitschka, B.,
a status review. Geochronometria, 21, 314–317.
2013. The PASADO science team 1. Quaternary Science
Murray, A. S., and Wintle, A. G., 2000. Luminescence dating of
Reviews, 71, 70–80.
quartz using an improved single-aliquot regenerative-dose pro-
Cunningham, A. C., and Wallinga, J., 2010. Selection of integration
tocol. Radiation Measurements, 32, 57–73.
time-intervals for quartz OSL decay curves. Quaternary Geo-
Murray, A. S., and Wintle, A. G., 2003. The single aliquot regener-
chronology, 5, 657–666.
ative dose protocol: potential for improvements in reliability.
Cunningham, A. C., and Wallinga, J., 2012. Realizing the potential
Radiation Measurements, 37, 377–381.
of fluvial archives using robust OSL chronologies. Quaternary
Murray, A., Buylaert, J.-P., Henriksen, M., Svendsen, J.-I., and
Geochronology, 12, 98–106.
Mangerud, J., 2008. Testing the reliability of quartz OSL ages
Cunningham, A. C., Wallinga, J., and Minderhoud, P., 2011a.
beyond the Eemian. Radiation Measurements, 43, 776–780.
Expectations of scatter in equivalent-dose distributions when
Olley, J. M., Roberts, R. G., and Murray, A. S., 1997. Disequilibria
using multi-grain aliquots for OSL dating. Geochronometria,
in the uranium decay series in sedimentary deposits at Allen’s
38, 424–431.
cave, nullarbor plain, Australia: implications for dose rate deter-
Cunningham, A. C., Bakker, M. A. J., Van Heteren, S., Van der
minations. Radiation Measurements, 27, 433–443.
Valk, B., Van der Spek, A. J. F., Schaart, D. R., and Wallinga,
Prescott, J. R., and Hutton, J. T., 1994. Cosmic ray contributions to
J., 2011b. Extracting storm-surge data from coastal dunes for
dose rates for luminescence and ESR dating: large depths and
improved assessment of flood risk. Geology, 39, 1063–1066.
long-term time variations. Radiation Measurements, 23, 497–500.
Cunningham, A. C., DeVries, D. J., and Schaart, D. R., 2012.
Rhodes, E. J., Bronk-Ramsey, C., Outram, Z., Batt, C., Willis, L.,
Experimental and computational simulation of beta-dose hetero-
et al., 2003. Bayesian methods applied to the interpretation of
geneity in sediment. Radiation Measurements, 47, 1060–1067.
multiple OSL dates: high precision sediment age estimates from
Duller, G. A. T., 2007. Assessing the error on equivalent dose esti-
Old Scatness Broch excavations, Shetland Isles. Quaternary
mates derived from single aliquot regenerative dose measure-
Science Reviews, 22, 1231–1244.
ments. Ancient TL, 25, 15–24.
Rink, W. J., and Lopez, G. I., 2010. OSL-based lateral progradation
Duller, G. A. T., 2008. Single-grain optical dating of quaternary sed-
and aeolian sediment accumulation rates for the Apalachicola
iments: why aliquot size matters in luminescence dating. Boreas,
Barrier Island complex, North Gulf of Mexico, Florida. Geomor-
37, 589–612.
phology, 123, 330–342.
Duller, G. A. T., and Wintle, A. G., 2012. A review of the
Rink, W. J., Mercier, N., Mihailović, D., Morley, M. W., Thompson,
thermally transferred optically stimulated luminescence signal
J. W., and Roksandic, M., 2013. New radiometric ages for the
from quartz for dating sediments. Quaternary Geochronology,
BH-1 hominin from Balanica (Serbia): implications for understand-
7, 6–20.
ing the role of the Balkans in middle Pleistocene human evolution.
Galbraith, R. F., and Roberts, R. G., 2012. Statistical aspects of
PLoS ONE, 8(2), e54608, doi:10.1371/journal.pone.0054608.
equivalent dose and error calculation and display in OSL dating:
Roberts, H. M., and Plater, A. J., 2007. Reconstruction of Holocene
an overview and some recommendations. Quaternary Geochro-
foreland progradation using optically stimulated luminescence
nology, 11, 1–27.
 LUMINESCENCE, BIOGENIC CARBONATES
(OSL) dating: an example from Dungeness, UK. The Holocene,
17, 495–505.
Sim, A. K., Thomsen, K. J., Murray, A. S., Jacobsen, G., Drysdale,
R., and Erskine, W., 2014. Dating recent floodplain sediments in
the Hawkesbury-Nepean river system, eastern Australia using
single-grain quartz OSL. Boreas, 43, 1–21.
Thiel, C., Buylaert, J.-P., Murray, A., Terhorst, B., Hofer, I.,
Tsukamoto, S., and Frechen, M., 2011. Luminescence dating of
the Stratzing loess profile (Austria) – testing the potential of an
elevated temperature post-IR IRSL protocol. Quaternary Inter-
national, 234, 23–31.
Thomsen, K. J., Murray, A. S., Bøtter-Jensen, L., and Kinahan, J.,
2007. Determination of burial dose in incompletely bleached flu-
vial samples using single grains of quartz. Radiation Measure-
ments, 42, 370–379.
Thomsen, K. J., Murray, A. S., Jain, M., and Bøtter-Jensen, L.,
2008. Laboratory fading rates of various luminescence signals
from feldspar-rich sediment extracts. Radiation Measurements,
43, 1474–1486.
Thomsen, K. J., Murray, A. S., and Jain, M., 2012. The dose depen-
dency of the over-dispersion of quartz OSL single grain dose dis-
tributions. Radiation Measurements, 47, 732–739.
Vandenberghe, D., De Corte, F., Buylaert, J. P., Kučera, J., and
Van den Haute, P., 2008. On the internal radioactivity in quartz.
Radiation Measurements, 43, 771–775.
Wallinga, J., and Bos, I. J., 2010. Optical dating of clastic lake-fill
sediments – a feasibility study in the Holocene Rhine delta
(western Netherlands). Quaternary Geochronology, 5, 602–610.
Wallinga, J., Murray, A. S., and Wintle, A. G., 2000. The single-
aliquot regenerative-dose (SAR) protocol applied to coarse-
grain feldspar. Radiation Measurements, 32, 529–533.
Wallinga, J., Bos, A. J. J., Dorenbos, P., Murray, A. S., and
Schokker, J., 2007. A test case for anomalous fading correction
in IRSL dating. Quaternary Geochronology, 2, 216–221.
Wintle, A. G., and Murray, A. S., 2006. A review of quartz optically
stimulated luminescence characteristics and their relevance in
single aliquot regeneration dating protocols. Radiation Measure-
ments, 41, 369–391.
Cross-references
Luminescence Dating
LUMINESCENCE, BIOGENIC CARBONATES
Geoffrey A. T. Duller
Department of Geography and Earth Sciences,
Aberystwyth University, Aberystwyth, Wales, UK
Definition
Biogenic carbonates. Carbonates, especially calcite and
aragonite, precipitated as a result of biological processes.
Luminescence dating. The use of the luminescence signal
emitted from a mineral to date its formation or some other
event.
Introduction
The emission of luminescence from quartz and feldspar
has been widely exploited as the basis for dating
445
sediments, pottery, and burnt stones (see “Luminescence
Dating”). Calcite (CaCO3) and its polymorph aragonite
yield thermoluminescence (TL) signals which grow with
exposure to ionizing radiation in the same way as seen
for quartz and feldspar. A number of studies have been
undertaken to explore whether geochronological informa-
tion can be obtained from this TL signal from calcite, and
these have focused on trying to date the formation of the
mineral. Initial efforts in the 1970s and 1980s (Wintle,
1978; Debenham and Aitken, 1984) studied stalagmitic
calcite, but more recently a small number of researchers
have looked at the potential of biogenic carbonates, pro-
duced by bivalves and molluscs. The TL signal from cal-
cite does not bleach rapidly in light, and dating
applications have dated the crystallization event rather
than the last exposure to daylight. Nonetheless, it is pru-
dent to minimize light exposure during sampling, prepara-
tion, and measurement.
Marine shells
Shells of the Pectinidae family were shown to give
TL signals. Ninagawa et al. (1992) used this TL signal
to determine ages of samples with some independent
age control and found that they were able to get ages
that were near those expected at several hundred
thousand years old but that shells expected to be older
than a million years gave underestimates. A major
challenge with such work was the presence of organic
components within the shell which were intimately
intermixed with the calcite structure. To avoid combus-
tion of the organic components within the shells, samples
were only heated to a maximum of 240  C, but even
when using a low heating rate (1.7  C/s), this meant that
the measured TL signal may not have had sufficient ther-
mal stability over the long geological periods being stud-
ied. Studying a wide range of shellfish, Carmichael
et al. (1994) found that the mineralized component
varied from 40 % to 80 % by mass, with the remainder
being organic.
Terrestrial molluscs
Two other types of biogenic carbonates have received
attention in recent years: slug plates and snail opercula
(Duller et al., 2009; Stirling et al., 2012). Slugs of the
Limacidae family secrete a small plate of calcite, about
2–3 mm in diameter and 0.2 mm thick, enclosed by their
mantle. Snails commonly use an operculum to protect
themselves and prevent excessive loss of moisture.
In many species this operculum is highly organic, but it
is highly mineralized in the species Bithynia tentaculata
(Figure 1). These two target materials have been shown
to yield intense TL signals that grow with radiation to
doses over 6,000 Gy. The methods are still experimental
but potentially could be used to date the whole of the
Quaternary period (2.58 Ma).
446 LUMINESCENCE, COASTAL SEDIMENTS
Luminescence, Biogenic Carbonates, Figure 1 Operculum of
Bithynia tentaculata. The operculum is 4 mm across. The upper
image is taken using a normal microscope, while the lower image
shows the TL emitted during measurement following a
laboratory irradiation. In this lower image, blue represents low
intensity TL emission, while green, yellow, and red indicate
increasing intensity (Photos: G.A.T. Duller).
Conclusions
Biogenic carbonates, especially calcite, have very great
potential for providing geochronological information over
the entire Quaternary period. In comparison with other
materials used for luminescence dating, they have the
advantage that the event being dated is crystallization,
and thus, there are no uncertainties about whether the
signal was reset at the time of deposition as may occur
when dating sediments or analyzing heated materials.
A number of challenges remain to be resolved before the
full potential of biogenic carbonates can be realized,
including improving the accuracy with which the dose rate
can be determined and most critically comparing the
results of these new methods with robust independent
chronometers.
Bibliography
Carmichael, L. A., Sanderson, D. C. W., and Ni Riain, S., 1994.
Thermoluminescence measurement of calcite shells. Radiation
Measurements, 23, 455–463.
Debenham, N. C., and Aitken, M. J., 1984. Thermo-luminescence
dating of stalagmitic calcite. Archaeometry, 26, 155–170.
Duller, G. A. T., Penkman, K. E. H., and Wintle, A. G., 2009.
Assessing the potential for using biogenic calcites as dosemeters
for luminescence dating. Radiation Measurements, 44,
429–433.
Ninagawa, K., Adachi, K., Uchimura, N., Yomamoto, I., Wada, T.,
Yamashita, Y., Takashima, I., Sekimoto, K., and Hasegawa, H.,
1992. Thermoluminescence dating of calcite shells in the
Pectinidae family. Quaternary Science Reviews, 11, 121–126.
Stirling, R. J., Duller, G. A. T., and Roberts, H. M., 2012.
Developing a single-aliquot protocol for measuring equivalent
dose in biogenic carbonates. Radiation Measurements, 47,
725–731.
Wintle, A. G., 1978. A thermoluminescence dating study of some
Quaternary calcite: potential and problems. Canadian Journal
of Earth Sciences, 15, 1977–1986.
Cross-references
Luminescence Dating
LUMINESCENCE, COASTAL SEDIMENTS
Barbara Mauz
School of Environmental Sciences, University of
Liverpool, Liverpool, UK
Definition
Luminescence. Light emitted by an irradiated semicon-
ductor during exposure to light or heat. The light intensity
emitted by the semiconductor is a function of the absorbed
dose.
Luminescence dating technique. Dosimetric technique
which determines the amount of absorbed dose in
a dosimeter (semiconductor) such as quartz and feldspar
and the rate of dose to which the dosimeter was exposed
before the moment of sampling.
Coastal sediments. Deposits in the zone near where land
and sea meet.
Introduction
Luminescence dating technique relies on the absorption of
radiation dose in siliciclastic minerals such as quartz and
feldspar. When the grains of these minerals are exposed
to daylight for a sufficient length of time during sediment
transport and deposition, the luminescence clock of the
dosimeter is reset to zero. Besides sufficient resetting,
the accuracy of a luminescence age of a coastal deposit
depends upon sediment composition, the nature of wetting
and drying, and the degree of postdepositional geochemi-
cal alterations. In general, the coastal environment is
governed by the sea level. In the past, shelf areas have
 LUMINESCENCE, COASTAL SEDIMENTS
experienced migrating shorelines and associated deposits
are today submerged unless the rate of uplift of land is
faster than the rate of sea-level change or the sea level
was, at the time, higher than today. During glacial periods,
the sea level was low (e.g., ~120 m b.s.l. during MIS 2)
and was situated at or below the shelf edge, causing sub-
aerial processes to dominate the modern coastal area. Dur-
ing interglacial periods, the sea level was high, but, with
the exception of the last interglacial, rarely as high as it
is today. Thus, coastal sediments deposited at times of
sea-level lowstand (e.g., MIS 2) are rarely uncovered.
Likewise, coastal sediments deposited at times of maxi-
mum transgression (e.g., MIS 6/5.5) can be hard to find
in localities where high rate of sea-level rise exceeds the
sedimentation rate at the shoreline. Given the limited dose
that luminescence dosimeters can accumulate and the
preservation potential of coastal deposits, luminescence
ages of coastal deposits are generally <300 ka, and within
this range, we should mainly find ages of times when the
sea level was at its maximum height and when it was
falling.
Bleaching
Parameters controlling the degree of bleaching in coastal
environments are length of daylight exposure at a given
water depth, water turbulence, flow regime, and
suspended sediment concentration (SSC). In 4 m turbid
water, sunlight shorter than 500 nm and longer than
700 nm is almost completely attenuated (Berger and
Luternauer, 1987), and at 0.05 g l1 suspended sediment
concentration (SSC), feldspar grains are virtually
unbleached after 20 h exposure to light (Ditlefsen,
1992). In an estuarine environment, light levels are
reduced by three orders of magnitude in the upper 80 cm
of the water column (Richardson, 2001) due to the pres-
ence of SSC. But given that accurate ages were obtained
from tidal environments, it was concluded that reworking
of sediments through subsequent tidal cycles is crucial for
sufficiently resetting the OSL clock (Madsen et al., 2007;
Mauz et al., 2010). Table 1 shows that, when combined,
the influences of transport and deposition processes, the
interaction of tide- and wave-induced sediment transport,
and the efficacy of subaerial exposure during low water
dictate a high level of spatial and temporal variability in
the degree of bleaching. It is good practice to study the
suitability of a coastal environment for luminescence dat-
ing by examining the residual dose in modern samples
(e.g., Brill et al., 2012). However, residual dose data are
typically affected by large uncertainties which might
impact on the assessment of samples with low doses.
Postdepositional alteration
In general, any reorganization of the sediment, albeit
small, is undesired in luminescence dating because it
affects the mineral grains in terms of spatial distribution
(Cunningham et al., 2012) and chemical pristineness in
447
a largely unknown manner. Reorganizational processes
that may affect the accuracy of the dating result are
weathering, lithification, and bioturbation, where the first
two are a consequence of wetting and drying and the latter
depends on the biological activity in the coastal zone.
Wetting and drying as a result of tidal cyclicity was studied
by Roberts and Plater (2005) who compiled a readily
applicable water-content model based on particle-size dis-
tribution, pore-space saturation, and inundation period.
Wetting and drying as a result of climate seasonality was
studied by Nathan and Mauz (2008) who addressed pre-
cipitation of carbonate in the pore space when the deposit
falls dry. Mauz et al. (2009) report a 2-32 % reduction of
dose rate for samples with significant carbonate content.
Another challenge is represented by mollusc shells
(Zander et al., 2007) and evaporitic sediment components
(e.g., gypsum). These adsorb dissolved 234U from coastal
water that typically deviates from the open ocean with
regard to 234U/238U activity (see Ivanovich and Harmon
(1992) for details on U-activity in ocean waters and uptake
process). In this way, these components change the dose
rate during burial and create large equivalent-dose over-
dispersion (sb; Galbraith and Roberts, 2012). Rather
unstudied so far is unsupported uranium in organic-rich
tidal sediments (Bailiff and Tooley, 2000) and in Fe and
Mn oxyhydroxide coatings on the surface of mineral
grains. Madsen et al. (2005) assume that it is negligible
because the contribution of uranium to the total dose rate
of sand-sized dosimeters is rather small. Jeong et al.
(2007) show how chemical processes like illuviation and
leaching occurring under tropical climate conditions result
in mixing of mineral grains, change of dose rate, and, ulti-
mately, inaccurate ages.
Approaches to dating
Jacobs (2008) reviews 196 studies on coastal and marine
deposits originating from Europe, North Africa, South
Africa, Australia, New Zealand, and North America.
Few attempts were made for Scandinavia, west coast of
Africa, East Africa, Arabia and India, southeast Asia and
northeast Asia, South America, and the Arctic and Antarc-
tic coasts (Jacobs, 2008) reflecting the dominant applica-
tion of the dating technique to soft sediment coasts
outside arid and (formerly) glaciated environments.
Microtidal sediments
As shown by Jacobs (2008), the majority of studies used
soft sediment samples derived from mid-latitudinal
beaches, deltas, barriers and foredunes, marine terraces,
and beach ridges. These sediments are deposited in shal-
low, relatively clear water, often composed of well-sorted
and well-bleached siliciclastic sand. They can be dated
with high accuracy and precision (e.g., Thomas et al.,
2008) even if the sand is young (e.g., Sawakuchi et al.,
2008; Costas et al., 2012; Shen and Mauz, 2012).
448 LUMINESCENCE, COASTAL SEDIMENTS

Luminescence, Coastal Sediments, Table 1 Conceptual frame of tidal and wave processes and the link to daylight exposure of
sediment grains (Mauz et al., 2010)

Process Tide Wave Daylight exposure

Intertidal exposure N/A N/A Unrestricted, with the exception of

pools. Bleaching limited to surface
few millimeters, depth being
dependent upon grain-size
distribution (reduced sorting,
reduced penetration)
Early flood tide, Negligible waxing Significant wave action causing Effectiveness determined by location
sediment sediment resuspension. High- relative to wave breakpoint –
reworking by frequency fluctuation in turbidity higher efficiency in swash zone,
waves in (higher during breaking, lower lower in surf zone, lowest in zone
advancing surf during shoaling and post- of wave shoaling
zone breaking)
Flood tide, sediment Significant tidal currents orientated Negligible Effectiveness reduced with depth
transport within onshore (generally fastest at mid- through the water column,
water column and tide). Finer sediments carried in determined by turbidity (suspended
along bed suspension and intermittent sediment concentration).
suspension, coarser fraction Bleaching efficiency significantly
transported as traction load limited for traction load
Late flood tide/slack Waning-negligible Some wave action in shallow water Effectiveness determined by water
water, particle (depth < wave base) depth (i.e., location/altitude on the
settling linked to tidal flat relative to high water) and
grain size and residual turbidity. Bleaching likely
flocculation to be more effective for fines
retained in water column, but the
sedimentation potential for these
grains is limited
Ebb tide, potential for Waxing-significant tidal currents Negligible Effectiveness reduced with depth
bed sediment orientated offshore. Extent of ebb- through the water column,
resuspension and current reworking linked to determined by turbidity (suspended
reworking location/altitude on the tidal flat – sediment concentration).
limited on upper part and improved Bleaching efficiency significantly
on lower part where ebb currents limited for traction load
have sufficient time to gather speed
and entrain bed sediment. Finer
sediments carried in suspension
and intermittent suspension,
coarser fraction transported as
traction load
Shallow-water wave Negligible Significant wave action causing Effectiveness determined by location
reworking as ebb sediment resuspension. High- relative to wave breakpoint –
tide shoreline frequency fluctuation in turbidity higher efficiency in swash zone,
leaves the tidal flat (higher during breaking, lower lower in surf zone, lowest in zone
during shoaling and post- of wave shoaling
breaking)

Indurated (hard) coastal sediments (e.g., beachrock) are Tidal sediments

usually well-bleached but may present problems with
regard to dose rate. Most studies assume constant dose rate
over time in an infinite matrix (e.g., Jacobs and Roberts,
2009; Thiel et al., 2012). This is a fair approach for
“old” samples where the uncertainty associated with the
estimation of high doses exceeds the inaccuracy associ-
ated with the constant dose-rate assumption. For Holocene
beachrock samples, however, Thomas (2009) obtained
quartz OSL ages that seem to be underestimated, probably
due to an overestimated dose rate.
By outlining the conceptual frame and analyzing samples
derived from a well-established stratigraphic framework
(North Sea; 53 N), Mauz et al. (2010) find accurate as
well as inaccurate quartz OSL ages and conclude that
well-bleached samples may be obtained from all tidal
sub-environments. For the same area, Fruergaard et al.
(2011) present similar results which include few age over-
estimations derived from tidal channel samples. For quartz
fine-silt samples originating from a west Korean mudflat
(34 N), Kim et al. (2012) obtained ages in stratigraphic
 LUMINESCENCE, COASTAL SEDIMENTS
order between ~19 ka and ~1 ka. Also, Mauz et al. (2010)
show that quartz fine silt from mudflats yields
overestimated ages possibly due to high turbidity of the
coastal water alongside little opportunities for tidal
reworking. Even more challenging are coastal sediments
from arctic coasts because transport distances are short,
the coastal water is very turbid, and reworking opportuni-
ties are limited to the brief polar summer. Despite these
conditions, Alexanderson and Murray (2012) found negli-
gible residual doses in surface samples collected from up
to 10 m water depth.
Gravel beach
Conglomeratic deposits of gravel beaches are a challenge
for luminescence dating. Not only are sand-sized mineral
grains rare, the sediment matrix and geometry are inhomo-
geneous and standard assumptions (e.g., infinite matrix,
homogeneous distribution of radio sources) do not apply.
Mellett et al. (2012) avoided these problems by using the
sandy matrix of deposits that contained exclusively lime-
stone gravel of low radioactive element content. Jeong
et al. (2007) show that gravel deposited in a tropical envi-
ronment yields inaccurate OSL ages. Simms et al. (2011)
used the underside of cobbles with which the clast rests
on the beach surface. This approach assumes that the clast
was rotated by wave action the last time before the relative
sea level fell and was not moved by any environmental
activity afterwards. The results from this study are in
agreement with the local relative sea-level curve and
hence encouraging.
High-energy coastal deposits
High-energy events on coasts are storm surges and tsu-
namis, eventually lying down washover fan deposits or
sand layers of various thicknesses which are distinctively
different to the background sediments. Brill et al. (2012)
estimated the residual dose in tsunami sand deposited dur-
ing the Asian earthquake in 2004 to ~40 years, and Cun-
ningham et al. (2011) found negligible quartz residual
doses in storm-surge deposits. Sawakuchi et al. (2012)
use OSL sensitivity (i.e., OSL signal intensity per unit
dose) as a tracer to reconstruct frequency of storm impact
on a coastal barrier on the basis of known sediment path-
ways on the coast. Analyzing washover fan deposits
resulting from hurricane strikes, Davids et al. (2010)
obtained consistent ages for feldspar and quartz separates
indicating sufficient bleaching during such a high-energy
event.
Summary
Most coastal deposits are sufficiently bleached and can
be dated with high accuracy and precision owing to wave
and tidal reworking and homogeneous sediment texture.
Exceptions may be deposits laid down under high energy
and in turbid water. Inaccuracies, if any, are mostly
induced by dose-rate issues such as heterogeneous
449
radiation field in gravel deposits, unknown initial
U-activity, and changing dose rate in weathered or
lithified deposits.
Bibliography
Alexanderson, H., and Murray, A. S., 2012. Luminescence signals
from modern sediments in a glaciated bay, NW Svalbard. Qua-
ternary Geochronology, 10, 250–256.
Bailiff, I. K., and Tooley, M. J., 2000. Luminescence dating of fine-
grain Holocene sediments from a coastal setting. In Shennan, I.,
and Andrews, J. (eds.), Holocene Land-Ocean Interaction and
Environmental Change Around the North Sea. London: Geolog-
ical Society Special Publication, Vol. 166, pp. 55–67.
Berger, G. W., and Luternauer, J. J., 1987. Preliminary field work
for thermoluminescence studies at the Fraser River delta, British
Columbia. Geological Survey of Canada Paper, 87(1A),
901–904.
Brill, D., Klasen, N., Bruckner, H., Jankaew, K., Scheffers, A.,
Kelletat, D., and Scheffers, S., 2012. OSL dating of tsunami
deposits from Phra Thong Island, Thailand. Quaternary Geo-
chronology, 10, 224–229.
Costas, I., Reimann, T., Tsukamoto, S., Ludwig, J., Lindhorst, S.,
Frechen, M., Hass, H. C., and Betzler, C., 2012. Comparison
of OSL ages from young dune sediments with a high-resolution
independent age model. Quaternary Geochronology, 10, 16–23.
Cunningham, A. C., DeVries, D. J., and Schaart, D. R., 2012.
Experimental and computational simulation of beta-dose hetero-
geneity in sediment. Radiation Measurements, 47, 1060–1067.
Cunningham, A. C., Bakker, M. A. J., van Heteren, S., van der Valk,
B., van der Spek, A. J. F., Schaart, D. R., and Wallinga, J., 2011.
Extracting storm-surge data from coastal dunes for improved
assessment of flood risk. Geology, 39, 1063–1066.
Davids, F., Duller, G. A. T., and Roberts, H. M., 2010. Testing the
use of feldspars for optical dating of hurricane overwash
deposits. Quaternary Geochronology, 5, 125–130.
Ditlefsen, C., 1992. Bleaching of K-feldspars in turbid water sus-
pensions: a comparison of photo and thermoluminescence sig-
nals. Quaternary Science Reviews, 11, 33–38.
Fruergaard, M., Andersen, T. J., Nielsen, L. H., Madsen, A. T.,
Johannessen, P. N., Murray, A. S., Kirkegaad, L., and Pejrup,
M., 2011. Punctuated sediment record resulting from channel
migration in a shallow sand-dominated micro-tidal lagoon,
Northern Wadden Sea, Denmark. Marine Geology, 280, 91–104.
Galbraith, R. F., and Roberts, R. G., 2012. Statistical aspects of
equivalent dose and error calculation and display in OSL dating:
an overview and some recommendations. Quaternary Geochro-
nology, 11, 1–27.
Ivanovich, M., and Harmon, R. S., 1992. Uranium-series Disequi-
librium: Applications to Earth, Marine, and Environmental Sci-
ences. Oxford: Clarendon. 909p.
Jacobs, Z., 2008. Luminescence chronologies for coastal and
marine sediments. Boreas, 37, 508–535.
Jacobs, Z., and Roberts, D. L., 2009. Last interglacial age for aeo-
lian and marine deposits and the Nahoon fossil human footprints,
southeast coast of South Africa. Quaternary Geochronology,
4, 160–169.
Jeong, G. Y., Cheong, C.-S., and Choi, J.-H., 2007. The effect of
weathering on optically stimulated luminescence dating. Quater-
nary Geochronology, 2, 117–122.
Kim, J. C., Eum, C. H., Yi, S., Kim, J. Y., Hong, S. S., and Lee, J.-Y.,
2012. Optically stimulated luminescence dating of coastal sedi-
ments from southwestern Korea. Quaternary Geochronology,
10, 218–223.
Madsen, A.T., Murray, A.S., Andersen, T.J., Pejrup, M. and
Breuning-Madsen, H., 2005. Optically stimulated luminescence
450 LUMINESCENCE, COLLUVIAL SEDIMENTS
dating of young estuarine sediments: a comparison with 210Pb
and 137Cs dating. Marine Geology214, 251-268
Madsen, A. T., Murray, A. S., Andersn, T. J., and Pejrup, M., 2007.
Optical dating of young tidal sediments in the Danish Wadden
Sea. Quaternary Geochronology, 2, 89–94.
Mauz, B., and Bungenstock, F., 2007. How to reconstruct trends of
late Holocene relative sea level: a new approach using tidal flat
clastic sediments and optical dating. Marine Geology,
237, 225–237.
Mauz, B., Baeteman, C., Bungenstock, F., and Plater, A. J., 2010.
Optical dating of tidal sediments: potentials and limits inferred
from the North Sea coast. Quaternary Geochronology,
5, 667–678.
Mauz, B., Elmejdoub, N., Nathan, R., and Jedoui, Y., 2009. Last
interglacial coastal environments in the Mediterranean-Sahara
transition zone. Palaeogeography, Palaeoclimatology,
Palaeoecology, 279, 137–146.
Mellett, C. L., Mauz, B., Hodgson, D. H., Plater, A. J., and Lang, A.,
2012. Optical dating of drowned landscapes: a case study from
the English Channel. Quaternary Geochronology, 10, 201–208.
Nathan, R., and Mauz, B., 2008. On the dose rate estimate of
carbonate-rich sediments for trapped charge dating. Radiation
Measurements, 43, 14–25.
Richardson, C. A., 2001. Residual luminescence signals in modern
coastal sediments. Quaternary Science Reviews, 20, 887–892.
Roberts, H. M., and Plater, A. J., 2005. Optically Stimulated Lumines-
cence (OSL) dating of sands underlying the gravel beach ridges of
Dungeness and Camber, Southeast England, UK. English Heri-
tage Research Department Report Series, Vol. 27, p. 84.
Sawakuchi, A. O., Guedes, C. C. F., DeWitt, R., Giannini, P. C. F.,
Blair, M. W., Nascimento, D. R., Jr., and Faleiros, F. M., 2012.
Quartz OSL sensitivity as a proxy for storm activity on the south-
ern Brazilian coast during the Late Holocene. Quaternary
Geochronology, 13, 92–102.
Sawakuchi, A. O., Kalchgruber, R., Giannini, P. C. F., Nascimento,
D. R., Jr., Guedes, C. C. F., and Umisedo, N. K., 2008. The
development of blowouts and foredunes in the Ilha Comprida
barrier (Southeastern Brazil): the influence of Late Holocene cli-
mate changes on coastal sedimentation. Quaternary Science
Reviews, 27, 2076–2090.
Shen, Z., and Mauz, B., 2012. Optical dating of young deltaic
deposits in the Mississippi Delta with a single aliquot method.
Quaternary Geochronology, 10, 110–116.
Simms, A. R., DeWitt, R., Kouremenos, P., and Drewry, A. M.,
2011. A new approach to reconstructing sea levels in Antarctica
using optically stimulated luminescence of cobble surfaces.
Quaternary Geochronology, 6, 50–60.
Thiel, C., Buylaert, J.-P., Murray, A. S., Elmejdoub, N., and Jedoui,
Y., 2012. A comparison of TT-OSL and post-IR IRSL dating of
coastal deposits on Cap Bon peninsula, north-eastern Tunisia.
Quaternary Geochronology, 10, 209–217.
Thomas, P. J., 2009. Luminescence dating of beachrock in the
southeast coast of India – potential for Holocene shoreline
reconstruction. Journal of Coastal Research, 25, 1–7.
Thomas, P. J., Murray, A. S., Granja, H. M., and Jain, M., 2008.
Optical dating of late quaternary coastal deposits in Northwest-
ern Portugal. Journal of Coastal Research, 24, 134–144.
Zander, A., Degering, D., Preusser, F., Kasper, H. U., and Brückner,
H., 2007. Optically stimulated luminescence dating of sublittoral
and intertidal sediments from Dubai, UAE: radioactive
disequilibria in the uranium decay series. Quaternary Geochro-
nology, 2, 123–128.
Cross-references
Luminescence Dating
Luminescence Dating, Dose Rates
LUMINESCENCE, COLLUVIAL SEDIMENTS
Andreas Lang
School of Environmental Sciences, University of
Liverpool, Liverpool, UK
Definition
Colluvium is a sediment eroded from and transported
along hillslopes by water, gravity, or human action and
usually forms a wedge-shaped sediment body at the
footslope.
Hillslope deposits are important sedimentary archives
for geomorphological, paleoenvironmental, and increas-
ingly also archaeological research. Besides others, they
are utilized for reconstructing landscape response to cli-
mate change (e.g., Hanson et al., 2004), anthropogenic-
induced soil erosion (e.g., Lang, 2003), and fault activities
(e.g., Fattahi et al., 2006). Various definitions for the term
colluvium are in use in different scientific communities. In
general, the term describes sediments that are eroded from
and transported along hillslopes by running water or grav-
ity and that usually form wedge-shaped deposits on the
footslope. Typically, colluvial sediments are formed
through processes on slopes that are spatially diffusive
and are distinguished from alluvial sediments, which are
transported across slopes in spatially confined channels
to form alluvial fans and cones. The recent research focus
on colluvial systems is linked to the limited transport dis-
tance within such systems and their reduced complexity
that often allows cause-consequence analyses of sediment
formation.
Luminescence dating determines the time when sedi-
ments were formed if sufficient exposure to daylight
(bleaching) immediately before burial occurred. The
potential of using luminescence for dating colluvium has
already been explored by Forman et al. (1991) using ther-
moluminescence (TL). Wintle et al. (1993) were the first
to report successful infrared-stimulated luminescence
(IRSL) dating of colluvial sediments; later also optically
stimulated luminescence (OSL) has been utilized for dat-
ing colluvium (e.g., Fuchs et al., 2004). For luminescence
dating of colluvial sediments, limitations exist that are
related to (i) limited bleaching due to limited exposure of
sediment grains to daylight during the short and often
rapid transport of slope deposits and (ii) low-
luminescence sensitivity of quartz in many lithologically
immature deposits:
1. Several processes are involved in eroding and
transporting sediment along slopes, like sheet, rill and
gully erosion, mass movements (e.g., debris flows,
sliding, falling), and soil creep (e.g., solifluction).
While these processes act on hillslope material to move
it downslope, the sediment itself may move as a rigid
block as in the case of sliding. Thus, in respect of expo-
sure to daylight, only a small fraction of sediment
grains involved in the process (e.g., only the outside
 LUMINESCENCE, COLLUVIAL SEDIMENTS
layer of sediment grains on a rigid block) may be
exposed to daylight. If material is transported in sus-
pension, the intensity and spectral composition of the
daylight is subdued by the suspended load. Also, many
high-magnitude slope processes erode the material that
they are moving across and incorporate unbleached
material from the subsurface in the colluvial deposit
they leave behind.
Catchment morphology and erosion process con-
strain the sediment pathway, sediment travel distance,
and sediment sink. Because travel distance is usually
rather short, the probability of efficient bleaching is
reduced and events may occur fully during unfavorable
daylight conditions (e.g., transport entirely at night).
In addition, a large variety of sediment source materials
exist, leading to a large variety in sediment characteris-
tics, e.g., grain-size distribution, mineral composition,
or organic content. Some of these sediments tend to
be transported as single mineral grains, like well-sorted
dry sand, while others are prone to coagulation, like
clay-dominated sediments. The latter often hampers
sufficient zeroing of the luminescence signal, because
the inner grains of an aggregate are shielded from day-
light exposure (Lang, 1994). The same may result from
mineral coatings, where, e.g., iron, manganese, or car-
bonate coatings hamper the daylight to penetrate into
the mineral grains.
2. As in other sedimentary environments, the use of
quartz OSL in colluvial deposits can be hampered by
poor luminescence sensitivity. Often materials incor-
porated in colluvium are freshly eroded from bedrock
and enter the erosion-transport-deposition cycle for
the first time. Quartz from such environments usually
displays rather low-luminescence sensitivity as that
seems to increase only after repeated cycles of sedi-
ment recycling (Pietsch et al., 2008).
Despite the shortcomings, there are many examples
where luminescence dating of colluvium has successfully
been carried out (for reviews, see Lang, 2003; Fuchs and
Lang, 2009). Two factors influence this: (a) hillslope sed-
iments are often eroded, transported, and accumulated not
in a single event but temporary storage on the hillslope
occurs, before a further mobilization takes place (e.g.,
Lang and Hönscheidt, 1999). This increases the probabil-
ity of exposure to daylight and thus bleaching. (b) Biotur-
bation and mechanical processes in the soil as well as
cultivation allow that mineral grains are frequently
exposed to daylight before the sediment grains are eroded.
In many cases exposure to daylight continues after deposi-
tion until the sediment grains are covered through further
sedimentation. According to Berger and Mahaney
(1990), this process is more important for sediment
bleaching than the colluvial transport itself. Nevertheless,
as in other sedimentary environments, detecting and
excluding insufficient bleaching is a necessary prerequi-
site for successful luminescence dating of colluvial sedi-
ments. This is demonstrated by some investigations of
451
modern colluvial sediments, where residual luminescence
signals were identified, which would result in age overes-
timations of hundreds (Yanchou et al., 2002) or even thou-
sands of years (Porat et al., 2001). The large variety of
sedimentary process and regional and climatic settings,
however, make generic statements difficult: single-grain
dose distributions by Vandenberghe et al. (2009) derived
from fault scarp colluvium are unimodal, rather narrow,
and are only slightly skewed, indicative of sufficient
bleaching which is also confirmed by comparing the
resulting OSL ages with independent age information.
Porat et al. (2009) use OSL to date colluvial sediments
from Eilat, southern Israel. Based on single-grain OSL
measurements of quartz, they are able to identify bleached
grains and manage to establish a chronology spanning the
interval 1,300 years to 500 years ago. As in other sedimen-
tary environments, and where possible, the analyses of
dose distributions of single grains and small aliquots seem
crucial also in colluvial deposits.
Conclusions
Due to rather restricted bleaching possibilities, colluvial
sediments are challenging for OSL dating. Issuing generic
statements on the suitability is difficult as colluvial sedi-
ments exist in a great variety in very different environ-
ments. It has been demonstrated that OSL dating is
possible in many cases; but success is often difficult to
judge and success at one location does not guarantee suc-
cess elsewhere. To establish a trustworthy and robust OSL
chronology, internal (e.g., stratigraphic integrity) and
external (e.g., accordance with independent information)
consistency is a requirement.
Bibliography
Berger, G. W., and Mahaney, W. C., 1990. Test of thermolumines-
cence dating of buried soils from Mt. Kenya, Kenya. Sedimen-
tary Geology, 66, 45–56.
Fattahi, M., Walker, R., Hollingsworth, J., Bahroudi, A., Nazari, H.,
Talebian, M., Armitage, S., and Stokes, S., 2006. Holocene
slip-rate on the Sabzevar thrust fault, NE Iran, determined using
optically stimulated luminescence (OSL). Earth and Planetary
Science Letters, 245, 673–684.
Forman, S. L., Nelson, A. R., and McCalpin, J. P., 1991. Thermolu-
minescence dating of fault scarp-derived colluvium: deciphering
the time of paleoearthquakes on the Weber Segment of the
Wasatch fault zone, North Central Utah. Journal of Geophysical
Research, 96, 595–605.
Fuchs, M., and Lang, A., 2009. Luminescence dating of hillslope
deposits – a review. Geomorphology, 109, 17–26.
Fuchs, M., Lang, A., and Wagner, G. A., 2004. The history of
Holocene soil erosion in the Phlious Basin, NE-Peloponnese,
Greece, provided by optical dating. The Holocene, 14,
334–345.
Hanson, P. R., Mason, J. A., and Goble, R. J., 2004. Episodic Late
Quaternary slopewash deposition as recorded in colluvial
aprons, Southeastern Wyoming. Quaternary Science Reviews,
23, 1835–1846.
Lang, A., 1994. Infra-red stimulated luminescence dating of Holo-
cene reworked silty sediments. Quaternary Science Reviews,
13, 525–528.
452 LUMINESCENCE, DESERT DUNES
Lang, A., 2003. Phases of soil erosion-derived colluviation in the
Loess hills of Southern Germany. Catena, 51, 209–221.
Lang, A., and Hönscheidt, S., 1999. Age and source of colluvial
sediments at Vaihingen-Enz, Germany. Catena, 38, 89–107.
Pietsch, T. J., Olley, J. M., and Nanson, G. C., 2008. Fluvial trans-
port as a natural luminescence sensitiser of quartz. Quaternary
Geochronology, 3, 365–376.
Porat, N., Zilberman, E., Amit, R., and Enzel, Y., 2001. Residual
ages of modern sediments in an hyperarid region, Israel. Quater-
nary Science Reviews, 20, 795–798.
Porat, N., Duller, G. A. T., Amit, R., Zilberman, E., and Enzel, Y.,
2009. Recent faulting in southern Arava, Dead Sea Transform:
evidence from single grain luminescence dating. Quaternary
International, 199, 34–44.
Vandenberghe, D., Vanneste, K., Verbeeck, K., Paulissen, E.,
Buylaert, J. P., De Corte, F., and Vandenhaut, P., 2009. Late
Weichselian and Holocene earthquake events along the Geleen
fault in NE Belgium: OSL age constraints. Quaternary Interna-
tional, 199, 56–74.
Wintle, A. G., Li, S. H., and Botha, G. A., 1993. Luminescence
dating of colluvial deposits from Natal, South Africa. Suid-
Afrikaanse Tydskrif vir Wetenskap, 89, 77–82.
Yanchou, L., Prescott, J. R., Hua, Z., Jie, C., and Lanying, W., 2002.
Optical dating of colluvial deposits from Xiyangfang, China, and
the relation to palaeo-earthquake events. Quaternary Science
Reviews, 21, 1087–1097.
Cross-references
Feldspar, Infrared-Stimulated Luminescence
Luminescence Dating
Luminescence Dating of Archaeological Sediments
Luminescence Dating, Dose Rates
Luminescence Dating, Single-Grain Dose Distribution
Luminescence Dating, Uncertainties and Age Range
Luminescence, Earthquake and Tectonic Activity
Luminescence, Fluvial Sediments
Luminescence, Geomorphological Processes
Luminescence, Soils
Quartz Defects, Optically Stimulated Luminescence and Thermo-
luminescence
Radiation Dose Rate
LUMINESCENCE, DESERT DUNES
Matt Telfer
School of Geography, Earth and Environmental Science,
University of Plymouth, Plymouth, UK
Synonyms
Dune chronostratigraphy; OSL dating
Definition
Luminescence dating. Geochronological method relying
on the dosimetrical properties of common minerals to
accumulate a trapped charge proportional to burial age
and the dose rate.
Desert dunes. Discrete aeolian landforms, typically domi-
nated by sand, found in dryland regions of the world.
Introduction
In many respects, desert dunes are the ideal medium for
the application of luminescence dating. They are typically
dominated by sand-sized quartz, occur in an environmen-
tal setting typified by intense sunlight and relatively rapid
aeolian sedimentation (leading to well-defined
“bleaching” events), and are usually characterized by
a lack of other dateable material. Crucially, their dyna-
mism overlaps very closely with the typical range of time-
scales to which luminescence dating is applicable (i.e.,
10  5 105 years). It is, thus, unsurprising that it was
desert dunes that saw the first geoscience application of
luminescence dating, with Singhvi et al.’s (1982) land-
mark geochronological study of the Indian Thar desert.
The earliest studies used heat as the stimulation source
(thermoluminescence, or TL), and indeed TL continued
to be applied to dune sequences routinely for the next
20 years or so, but Huntley et al.’s (1985) use of light
(Optically Stimulated Luminescence, or OSL) has become
by far the most commonly used method today. It was not,
however, merely the ability to date dunes directly for the
first time that drove a rapid expansion in interest in lumi-
nescence dating of desert dunes; the research agenda was
driven primarily because it had long been recognized that
desert dunes are among the most dynamic and environ-
mentally sensitive of earth’s terrestrial systems.
The growth of luminescence dating of
desert dunes
Despite Pye and Tsoar’s (1990) cautious assertion that
“many difficulties remain to be resolved before lumines-
cence dating of sand can be regarded as entirely reliable,”
a key feature of today’s research was established early on;
it was not generally the active dunes of the world’s
hyperarid regions that attracted the most attention. It was
the fully or partially stabilized dunes of the desert margins.
The dunes of Australia (e.g., Gardner et al., 1987), South-
ern Africa (e.g., Buch et al., 1992), and the Western USA
(e.g., Stokes and Gaylord, 1993) attracted particular atten-
tion. These studies investigated a wide range of dune mor-
phologies. Many studies focused on individual sites or
localized areas, but some, notably Nanson et al. (1992)
and Stokes et al. (1997), were more ambitious in scope,
aiming to provide regional-scale syntheses of late Quater-
nary aridity. In general, desert dunes proved highly favor-
able for the method, typically offering well-defined
growth of the OSL signal and excellent repeatability
(best defined by overdispersion sensu (Galbraith
et al., 1999)). Figure 1 demonstrates a typical shine-down
curve, growth curve, and dose distribution for a favorable
desert dune sample. Thus, the majority of studies have
focused less on the technical suitability of dune sands for
luminescence dating and more on the interpretation of
the results of the dating work.
Some studies (e.g., Lomax et al., 2007), however, have
encountered poor repeatability of individual aliquots (i.e.,
high overdispersion) from dune sands. Such effects are
 LUMINESCENCE, DESERT DUNES 453

Luminescence, Desert Dunes, Figure 1 Many studies have found desert dunes present an ideal medium for luminescence dating,
with samples demonstrating: (a) bright, fast-component-dominated quartz OSL signals; (b) well-defined growth curves with low
recuperation and good recycling characteristics; (c) and (d) good repeatability of equivalent dose measurements between aliquots,
shown here as both as ranked and radial plots.

usually attributed to heterogenous dose environments,
inadequate bleaching at time of deposition, or postdepo-
sitional mixing of sediment. The latter is often considered
the most significant problem in massively bedded desert
dunes and has been investigated in detail (Bateman
et al., 2007). The low moisture levels and typically
homogenous sediment are also usually assumed to lead
to straightforward estimation of the dose rates, but as in
other environmental settings dated with luminescence,
dosimetrical issues have perhaps sometimes been
overlooked. In dunes which have periodically accumu-
lated over long periods of time, the cosmic contribution
to dose rate may be better estimated by an iterative proce-
dure to model the varying overburden (Munyikwa, 2000).
Commonly assumed moisture contents for dunes may also
have frequently been overestimated (Telfer and Hesse,
2013).
Desert dune sands have also been used in methodologi-
cal development of luminescence dating, although perhaps
less frequently than might be expected, perhaps due to the
difficulty of corroborating ages derived from new tech-
niques with alternative dating methods. Nonetheless, Li
and Li (2011) have applied post-IR IRSL (Infra-Red Stim-
ulated Luminescence) to Chinese dune sands, with a view
to extending the applicable age range of the method, and
Fujioka et al. (2009) paired OSL and cosmogenic dating
techniques to date the inception of an Australian dune field.
Field methodologies have perhaps developed more rapidly
in the study of desert dunes, with geophysical survey and
augering methods used to achieve full-depth survey and
sampling of dunes up to around 20 m depth. In a very few
instances (e.g., Preusser et al., 2002), heavy duty coring
equipment has been used to extract dateable sand cores in
excess of 150 m length, but challenges remain with the
recovery rate of such loose sediment.
Applications of luminescence dating
of desert dunes
Although the methodological challenges of luminescence
dating of desert dunes are less than some other environ-
ments (e.g., fluvial or glacial), the interpretation of ages
obtained from dunes remains highly complex. The major-
ity of studies of desert dunes around the turn of the millen-
nium remained focused on paleoenvironmental studies,
often with a view to paleoclimatic interpretation of the
data; a significant minority of research, however, used
the ability to provide chronometric control for dune devel-
opment to improve understanding of geomorphology.
It must be borne in mind that most dune systems will have
involved repeated cycles of deposition and erosion, and
an OSL age refers only to the most recent of these burial
events, with important possible implications for the
contextual interpretation of the age.
454 LUMINESCENCE, DESERT DUNES
A wide range of dune types has been studied, with differ-
ent dune types attracting different types of study on account
of their differing geomorphological characteristics. For
instance, the transition from active barchan dunes to stabi-
lized parabolic dunes on the Canadian prairies over
a period of ~200 years has been described by Wolfe and
Hugenholtz (2009). Linear dunes are typically considered
much more long-lived landforms, as their mode of forma-
tion does not lead to such frequent complete reworking of
the dune body, and hence these are often seen as offering
the greatest potential for long-term records of past aeolian
environments. Star dunes, which include the largest
dunes on earth, have proven challenging for dating
studies, with full-depth sampling of large star dunes not cur-
rently achieved, but they have occasionally formed part of
investigations into complex dune patterns (Beveridge
et al., 2006).
Paleoenvironmental applications
The prevailing paradigm of the 1990s was broadly that
increased frequency of observed ages within a currently
stabilized dune equated to past periods of aridity, although
there were already conceptual models which suggested
that dryland landscape response to external forcing stimuli
may be more complex than this (Kocurek, 1998). Some
long-standing hypotheses were challenged, such as the
assumption that the widespread arc of currently stabilized
dunes in Southern Africa resulted from drier conditions
during the Last Glacial Maximum (LGM) (O’Connor
and Thomas, 1999). Dunes offer a rare opportunity to
establish paleowind directions, which has been exploited
at both continental (e.g., Nanson et al., 1995) and regional
scale (e.g., Holmes et al., 2008). Luminescence dating also
allowed the direct dating of sites of high archaeological
significance, such as the human remains from the lunette
dune at Lake Mungo, Australia.
More recently, however, a number of studies have
begun to challenge the assumption that frequency of
observed luminescence-derived dune ages is a direct
proxy for past aridity. There are a number of reasons for
this, and perhaps the most important is the spatial and tem-
poral variability evident in the preservation of dune sands.
As most dunes rework themselves during periods of
reactivation, the sedimentary record from dunes is typi-
cally discontinuous. Some studies showed that adjacent
dunes did not respond identically to what must have been
identical climatic forcing (Telfer and Thomas, 2007).
Stone and Thomas (2008) sampled a dune at 0.5 m vertical
intervals – a better resolution than had previously been
attempted – and noted that if these data were subsampled
to 1 m resolution (i.e., jack-knifing), there was worrying
disparity between the conclusions drawn from some of
the older ages. Such studies implied that the stochastics
of net preservation in dune systems were complex and
led Telfer et al. (2010) to investigate such interpretations
via modeling, which suggested that current sample sizes
were in most cases inadequate to interpret anything other
than gross trends in the net accumulation rate in dunes.
Concurrent with increased realization of the implica-
tions of spatial variability in the preservation of dune
records came increased questioning as to what the record
of OSL-derived chronostratigraphies actually meant from
a paleoenvironmental, or even paleoclimatic, perspective.
Traditionally, an increase in observed frequency of dated
samples was interpreted as increased aridity (e.g., Stokes
et al., 1997). More recent work has identified a very wide
range of possible drivers of net dune accumulation rates,
and the complexity of the factors affecting dune accumu-
lation is shown in Figure 2 and must be considered site
specific. The interpretations of Roskin et al. (2011) are
a good example of current best practice, drawing in
a wide range of additional data before drawing inferences
on the driving factors in the region under investigation.
Geomorphological applications
There have been fewer chronological studies of dunes
with geomorphology as the primary aim, but those that
have addressed processes and rates of landform formation
have often been very successful. Luminescence-
constrained chronologies offer a longer-term perspective
than observational studies can allow, and although some
dunes do currently move on timescales that can be use-
fully monitored in the field, or using remotely sensed data,
it is also clear than many dunes have formed on timescales
greater than 103 years.
In some instances, luminescence dates have been used
to constrain the sequence of events in complex aeolian
environments, thus helping in the understanding of com-
plex interactions and feedbacks operating between the
bedform and the moving atmospheric boundary layer
(Lancaster et al., 2002). Luminescence dating has also
been used at the landform scale to help elucidate long-term
landform development, particularly in combination with
ground-penetrating radar (GPR) to reveal internal stratig-
raphy of dunes (Bristow et al., 2005; Bristow
et al., 2007). Telfer (2011) demonstrated that with suffi-
cient sampling resolution, chronostratigraphies alone
could inform long-term landform evolution studies, by
demonstrating the complex extension of a Kalahari linear
dune. Attempts have also been made to use the lumines-
cence characteristics of dune sands to determine whether
the dunes were formed in subaerial or subaqueous envi-
ronments (Burrough et al., 2012).
Summary and conclusions
Desert dunes are typically some of the least challenging
environments in terms of getting successful luminescence
dates. The challenge comes with the interpretation of these
dates. It is not possible to take a limited number of samples
in a nonsystematic manner from a stabilized dune field and
extract a regional climatic signal of aridity. There is an
urgent need for improved understanding of the long-term
 LUMINESCENCE, DESERT DUNES 455

Luminescence, Desert Dunes, Figure 2 Conceptual model of the drivers of net dune accumulation under changing climatic
regimes, demonstrating the complexity of the linkage between externals drivers and landform response. The presence of bedform/
boundary layer interaction as a result of changes in aeolian accumulation is perhaps the most significant aspect introducing
nonlinearities into the system. This schematic is not intended to be exhaustive, and numerous other linkages and feedback could be
proposed.

process geomorphology of dunes, and currently, the best
way of using dunes as a geomorphological proxy of past
environments is to consider them alongside other
paleoenvironmental archives and to recognize the com-
plexity of the climate change-landscape response relation-
ship in dunes. That said, luminescence dating of dunes still
offers an essential tool for understanding the nature and
rates of aeolian processes at scales beyond those which
can be observed directly. This will prove vital in the future
for improved understanding of dryland landscape
response to changing external stimuli, a pressing issue in
the light of twenty-first-century climatic change.
Bibliography
Bagnold, R. A., 1941. The Physics of Blown Sand and Desert
Dunes. London: Chapman and Hall.
Bateman, M. D., Boulter, C. H., Carr, A. S., Frederick, C. D., Peter,
D., and Wilder, M., 2007. Detecting post-depositional sediment
disturbance in sandy deposits using optical luminescence. Qua-
ternary Geochronology, 2, 57–64.
Beveridge, C., Kocurek, G., Ewing, R. C., Lancaster, N., Morthekai,
P., Singhvi, A. K., and Mahan, S. A., 2006. Development of spa-
tially diverse and complex dune-field patterns: Gran Desierto
Dune Field, Sonora, Mexico. Sedimentology, 53, 1391–1409.
Bristow, C. S., Lancaster, N., and Duller, G. A. T., 2005. Combining
ground penetrating radar surveys and optical dating to determine
dune migration in Namibia. Journal of the Geological Society,
162, 315–321.
Bristow, C. S., Duller, G. A. T., and Lancaster, N., 2007. Age and
dynamics of linear dunes in the Namib Desert. Geology, 35,
555–558.
Buch, M. W., Rose, D., and Zoller, L., 1992. A TL-calibrated
pedostratigraphy of the western lunette dunes of Etosha
Pan/northern Namibia: palaeoenvironmental implications for
the last 140 ka. Palaeoecology of Africa, 25, 129–147.
Burrough, S. L., Thomas, D. S. G., Bailey, R. M., and Davies, L.,
2012. From landform to process: morphology and formation of
lake-bed barchan dunes, Makgadikgadi, Botswana. Geomor-
phology, 161, 1–14.
Fujioka, T., Chappell, J., Fifield, L. K., and Rhodes, E. J., 2009.
Australian desert dune fields initiated with Pliocene-Pleistocene
global climatic shift. Geology, 37, 51–54.
Galbraith, R. F., Roberts, R. G., Laslett, G. M., Yoshida, H., and
Olley, J. M., 1999. Optical dating of single and multiple grains
of quartz from Jinmium rock shelter, northern Australia, part
1, Experimental design and statistical models. Archaeometry,
41, 339–364.
Gardner, G. J., Mortlock, A. J., Price, D. M., Readhead, M. L., and
Wasson, R. J., 1987. Thermoluminescence and radiocarbon dat-
ing of Australian desert dunes. Australian Journal of Earth Sci-
ences, 34, 343–357.
Holmes, P. J., Bateman, M. D., Thomas, D. S. G., Telfer, M. W.,
Barker, C. H., and Lawson, M. P., 2008. A Holocene-late Pleis-
tocene aeolian record from lunette dunes of the western Free
State panfield, South Africa. The Holocene, 18, 1193–1205.
Huntley, D. J., Godfrey-Smith, D. I., and Thewalt, M. L. W., 1985.
Optical dating of sediments. Nature, 313, 105–107.
Kocurek, G., 1998. Aeolian system response to external forcing
factors – a sequence stratigraphic view of the Saharan region.
456 LUMINESCENCE, EARTHQUAKE AND TECTONIC ACTIVITY
In Alsharhan, A. S., Glennie, K. W., Whittle, G. L., and Kendall,
C. G. S. (eds.), Quaternary Deserts and Climatic Change.
Rotterdam: Balkema.
Lancaster, N., 1979. Quaternary environments in the arid zone of
southern Africa. I. Environmental Studies, Occasional Paper
22, p. 77.
Lancaster, N., Kocurek, G., Singhvi, A., Pandey, V., Deynoux, M.,
Ghienne, J. F., and Lo, K., 2002. Late Pleistocene and Holocene
dune activity and wind regimes in the western Sahara Desert of
Mauritania. Geology, 30, 991–994.
Li, B., and Li, S. H., 2011. Luminescence dating of K-feldspar from
sediments: a protocol without anomalous fading correction.
Quaternary Geochronology, 6, 468–479.
Lomax, J., Hilgers, A., Twidale, C. R., Bourne, J. A., and Radtke,
U., 2007. Treatment of broad palaeodose distributions in OSL
dating of dune sands from the western Murray Basin, South Aus-
tralia. Quaternary Geochronology, 2, 51–56.
Munyikwa, K., 2000. Cosmic ray contribution to environmental
dose rates with varying overburden thickness. Ancient TL, 18,
27–34.
Nanson, G. C., Price, D. M., and Short, S. A., 1992. Wetting and
drying of Australia over the past 300 Ka. Geology, 20, 791–794.
Nanson, G. C., Chen, X. Y., and Price, D. M., 1995. Aeolian and flu-
vial evidence of changing climate and wind patterns during the
past 100 Ka in the Western Simpson Desert, Australia.
Palaeogeography Palaeoclimatology Palaeoecology, 113,
87–102.
O’Connor, P. W., and Thomas, D. S. G., 1999. The timing and envi-
ronmental significance of late quaternary linear dune develop-
ment in western Zambia. Quaternary Research, 52, 44–55.
Preusser, F., Radies, D., and Matter, A., 2002. A 160,000-year
record of dune development and atmospheric circulation in
southern Arabia. Science, 296, 2018–2020.
Pye, K., and Tsoar, H., 1990. Aeolian Sand and Sand Dunes. Lon-
don: Unwin Hyman.
Roskin, J., Tsoar, H., Porat, N., and Blumberg, D. G., 2011.
Palaeoclimate interpretations of Late Pleistocene vegetated lin-
ear dune mobilization episodes: evidence from the northwestern
Negev dunefield, Israel. Quaternary Science Reviews, 30,
3364–3380.
Singhvi, A. K., Sharma, Y. P., and Agrawal, D. P., 1982. Thermo-
luminescence dating of sand dunes in Rajasthan, India. Nature,
295, 313–315.
Stokes, S., and Gaylord, D. R., 1993. Optical dating of Holocene
dune sands in the Ferris Dune Field, Wyoming. Quaternary
Research, 39, 274–281.
Stokes, S., Thomas, D. S. G., and Washington, R., 1997. Multiple
episodes of aridity in southern Africa since the last interglacial
period. Nature, 388, 154–158.
Stone, A. E. C., and Thomas, D. S. G., 2008. Linear dune accumulation
chronologies from the southwest Kalahari, Namibia: challenges of
reconstructing late quaternary palaeoenvironments from aeolian
landforms. Quaternary Science Reviews, 27, 1667–1681.
Telfer, M. W., 2011. Growth by extension, and reworking, of a -
south-western Kalahari linear dune. Earth Surface Processes
and Landforms, 36, 1125–1135.
Telfer, M. W., and Hesse, P. P., 2013. Palaeoenvironmental recon-
structions from linear dunefields: recent progress, current chal-
lenges and future directions. Quaternary Science Reviews, 78,
1–21.
Telfer, M. W., and Thomas, D. S. G., 2007. Late quaternary linear
dune accumulation and chronostratigraphy of the southwestern
Kalahari: implications for aeolian palaeoclimatic reconstructions
and predictions of future dynamics. Quaternary Science
Reviews, 26, 2617–2630.
Telfer, M. W., Bailey, R. M., Burrough, S. L., Stone, A. E. S.,
Thomas, D. S. G., and Wiggs, G. S. F., 2010. Understanding
linear dune chronologies: insights from a simple accumulation
model. Geomorphology, 120, 195–208.
Wolfe, S. A., and Hugenholtz, C. H., 2009. Barchan dunes stabi-
lized under recent climate warming on the northern Great Plains.
Geology, 37, 1039–1042.
Cross-references
Luminescence Dating
Luminescence Dating of Archaeological Sediments
Luminescence Dating, Dose Rates
Luminescence Dating, History
Luminescence Dating, Instrumentation
Luminescence Dating, Loess
Luminescence Dating, Single-Grain Dose Distribution
Luminescence Dating, Uncertainties and Age Range
Luminescence, Soils
Quartz Defects, Optically Stimulated Luminescence and Thermo-
luminescence
LUMINESCENCE, EARTHQUAKE AND TECTONIC
ACTIVITY
Morteza Fattahi
Institute of Geophysics, University of Tehran, Tehran, Iran
Definition
Paleoearthquake. An old, often prehistoric, earthquake.
Paleoseismology. The study of paleoearthquakes, espe-
cially their geological expression, location, timing, and
size.
Recurrence interval. The repeat time of major
earthquakes.
Slip rate. The average velocity with which one side of the
fault moves with respect to the other.
Earthquake catalogue. A comprehensive earthquake
database.
Introduction
On average, 15 earthquakes with a magnitude of 7 or
greater occur on earth every year. Because of their devas-
tating potential, much effort has been devoted to finding
a reliable method for predicting the location and time of
large earthquakes. So far these efforts have been unsuc-
cessful. However, if the recurrence intervals for earth-
quakes of various magnitudes in the region of interest
were known, it is possible to estimate the probability of
an event in the future. Therefore, determining the average
rate of recurrence of strong earthquakes on a fault is a key
element in long-term hazard assessment.
Seismic hazard assessment
Seismic hazard assessment (see Kijko, 2011) is based
mainly on analysis of seismicity and knowledge of past
earthquake activity in a region. Past earthquake records
come in three categories: instrumental, historical, and
 LUMINESCENCE, EARTHQUAKE AND TECTONIC ACTIVITY
prehistoric. Instrumental data (post 1890s) cover a too
short period, and historical records are also short com-
pared to the repeat times of large earthquakes on a fault.
In most parts of the world, even in parts of old countries
such as China, Japan, Iran, and the Mediterranean regions,
historical information reaches back a maximum of a few
thousands of years. However, the recurrence interval of
large earthquakes on many important faults is several
thousands of years (e.g., 7 ka in Xiyangfang, China
(Yanchou et al., 2002); 3 ka in NE Iran (Fattahi et al.,
2006)). Therefore, analysis of paleoearthquakes is
required to complete the catalogue of earthquakes.
Fortunately, geologic sedimentary strata and geomor-
phic features preserve evidence of tectonic deformation
and landscape changes caused by large earthquakes at or
near the earth’s surface. Geology and geomorphology
identify and evaluate the location and extent of past fault
displacements near surface folding and features related
to coseismic deformation. Luminescence dating methods,
among other geochronology techniques, determine the
timing of these types of deformation. Paleoseismology
uses these records to understand the faulting process
(McCalpin, 2009). The majority of paleoseismic studies
have been undertaken to provide input for seismic hazard
assessments (e.g., Gurpinar, 2005). Paleoseismic data
have been used for producing probabilistic seismic hazard
maps (see Kijko, 2011) which include information on fault
location, geometry, dimensions, slip rates, and recurrence
intervals (see Kijko, 2011).
Recurrence interval, slip rate, and active fault
There are two ways to estimate recurrence intervals. One
method is direct dating of evidence of paleoearthquakes.
The second method is to divide the average displacement
per event by the long-term slip rate of the fault. Long-term
average slip rate is calculated by dividing the cumulative
displacement (net slip required to produce the observed
deformation across the fault) by the total time over which
this displacement occurred (the age of the deformation).
The deformation can be horizontal deformation of
a geomorphic or man-made feature such as river or road
or the vertical deformation of dated strata in the subsurface
or geomorphic surfaces overlying the blind thrust (mono-
clines, anticlines). The vertical slip rate, the net slip rate,
and the dip of the blind fault are trigonometrically related
to each other. Therefore, if vertical slip rate and dip are
known, the net slip rate can be calculated.
Slip rate is a key to understanding the relative impor-
tance of active faults in an area and the hazard those active
faults present to local residents and developments. There
are different definitions for “active fault” based on geo-
logic or engineering purposes in different countries. How-
ever, a fault is commonly considered to be active if it has
moved at least once, up to a certain time in the past, such
as the last 10 ka, 35 ka, 50 ka, or 1.8 million years ago
(e.g., Lee et al., 2006; USGS website http://earthquake.
usgs.gov/learn/glossary/term=activefault). An active fault
457
exhibits offsets and can therefore be identified by recog-
nizing and dating tectonically deformed surface or mate-
rial. Active faults, faulted land forms, and coseismic
surface ruptures can be identified by different methods,
including satellite images, aerial photographs (e.g.,
Lin et al., 2006; Fattahi et al., 2007), various geophysical
techniques, airborne light detection and ranging (LIDAR),
airborne laser swath, mapping, and field surveys
(see Kijko, 2011).
Evidence for prehistoric earthquakes
Earlier classifications of paleoseismic evidence are modi-
fied to 16 categories and 3 levels as primary or secondary,
on-fault or off-fault, and stratigraphic or geomorphic
(McCalpin, 2009, Table 1.2). Primary evidence is formed
by tectonic deformation and reflects seismic surface
faulting or folding, whereas secondary evidence is pro-
duced by earthquake shaking or is a response to strong
ground shaking. These evidences are either above a fault
trace or away from a fault trace and are coseismic or
postseismic features, which are preserved as stratigraphic
or geomorphic evidence. This distinction often determines
how to approach paleoseismic field investigations. The
geomorphic method usually measures and dates the amount
of landform deformation during paleoearthquakes. Fault
scarps, fissures, folds, colluvial aprons, tilted surfaces,
uplifted/subsided shorelines, and tectonic alluvial terraces
are typical examples of primary geomorphic evidence.
Landslides, sand blows, and rockfalls are examples of sec-
ondary paleoseismic geomorphic evidence.
The stratigraphic method usually measures and dates
deformed strata in exposures (McCalpin, 2009). Examples
of stratigraphic paleoseismic evidences can be divided
into primary and secondary evidence. Primary evidence
include displaced or folded strata, unconformities, scarp-
derived colluvial wedges, tsunami deposits, zones of
sheared sediment, fissures, and uplifted, subsided, or tilted
surfaces located some distance from the fault. Secondary
evidence includes, for instance, sand dikes and other liq-
uefaction features, filled craters, soft-sediment deforma-
tion structures, and turbidity current deposits.
Evidence of paleoearthquakes and old surface ruptures
are often minimal directly on the earth’s surface. There-
fore, trenches are excavated to expose faults and young
strata that record the evidence. In outcrops or trenches,
corresponding stratigraphic evidence can be found.
A trench log is used for defining and mapping the
sequence of deformation, sedimentation, and weathering
exposed at a site and for recording evidence of surface-
rupturing earthquakes. These records (see Toda, 2011,
Figure 4) can be preserved as direct fault exposure, such
as upward fault termination, fissure fill, scarp-derived col-
luvial wedge, wrapping, folding, minor fracturing, and
inferred offset, each of which is defined as an “event indi-
cator” (see Toda, 2011). Luminescence dosimeter min-
erals (quartz and feldspar) inside these event indicators
are used to find the age of paleoearthquakes.
458 LUMINESCENCE, EARTHQUAKE AND TECTONIC ACTIVITY
Luminescence dating of prehistoric
earthquakes and faulting
Luminescence dating of earthquake-related deposits offers
great potential for dating evidence of past earthquakes and
faulting for a number of reasons (see Fattahi, 2009). Day-
light or friction temperature can reset the luminescence
“clock” of sediments before burial due to earthquakes.
Therefore, luminescence dating techniques can be applied
for age determination of any of paleoearthquake evidence
if the clock of quartz or feldspar inside that evidence is reset
during, or just prior to, paleoearthquakes. They constrain
timing of paleoearthquakes directly or indirectly by calcu-
lating ages of deformed stratigraphic units and underlying
or overlying undeformed strata.
The most direct techniques for dating active faults and
paleoearthquakes are dating fault scarp, fault-scarp collu-
vium, fault gouge, or liquefaction features such as injec-
tion dike and sand blows (deposition from the surface
rupture of an injection dike). Although these techniques
directly date the formation of coseismic features, age
uncertainties can be relatively large.
Indirect dating methods involve bracketing the age of
the paleoearthquake by luminescence dating of structures
or deposits that predate and postdate the paleoseismic evi-
dence. Maximum ages are obtained from sediment depos-
ited before the paleoearthquake. Minimum ages are
obtained on material younger than the earthquake. The
accuracy and precision of luminescence dating may be
high, but the ages themselves may not provide close con-
straints on the age of faulting.
Fault-scarp and fault-collapse wedges have been identi-
fied as the most important records of the paleoearthquakes
(Allen, 1986). Fault-collapse wedges are a kind of collu-
vial deposit of sediments formed as wedge-shaped
deposits on the foot slope. Colluvium is a term that
describes sediments that are eroded from and transported
along hill slopes by gravity. Reverse and normal faulting
chronologies can be reconstructed from luminescence
ages on fault-scarp colluvial wedge sediments (e.g.,
Fattahi et al., 2006; Porat et al., 2008). Colluvial wedges
and fault planes are identified from exposures and
trenches, while materials from buried soils within the
faulted sediments are dated to obtain a fault-rupture chro-
nology. A suitable exposure or trench exhibits multiple
faulting events with clear displacements and consequent
colluviation. For reverse faulting, the approach assumes
that an overhanging fault scarp is produced coseismically
and immediately following the earthquake, hanging-wall
collapse to form an angle-of-repose debris slope. On the
assumption that the production process of colluvial wedge
is sufficiently slow for the sediments to have been exposed
to sunlight long enough to reset the luminescence clock,
direct dating of discrete fault-rupture events can be
achieved by OSL dating of fault-scarp colluvium. For nor-
mal faulting, a vertical throw of some meters is not
uncommon for the surface expression of faulting. Immedi-
ately following the earthquake, there would be a contour
step, the low side of which would then become progres-
sively filled with sediment. Normal faulting chronologies
can be reconstructed from luminescence ages on colluvial
wedge, slope-wash sediments, sag-pond deposit, and the
soil horizons that underlie and overlie the wedge. The
age of the event horizon decreases away from the fault
due to the time taken for soil to pour laterally, accreting
scarp-derived colluvium. The age of the event horizon
might also be extrapolated from the age of several subsam-
ples of the buried soil horizon. Luminescence surface-
exposure dating recently introduced by Sohbati et al.
(2012) theoretically has the potential to date fault scarp
directly.
The usual approach to dating individual
paleoearthquakes on strike-slip faults is to bracket the
earthquake by dating the youngest datable stratum
deformed by the quake and the oldest datable stratum that
overlies the earthquake horizon (McCalpin, 2009). How-
ever, for strike-slip fault, direct luminescence dating of
a fissure fill can provide the age of the faulting event
(e.g., Fattahi et al., 2010).
Fault gouge is formed by brittle fracturing of rocks as
a result of slip along the faults during an earthquake.
Luminescence dating of gouge to date neotectonic move-
ments and past seismic events is based on the assumption
that during the faulting event, the luminescence intensity
of the gouge material is “zeroed” by either frictional
heating or mechanical deformation or a combination of
these. Even with agreement between TL and radiocarbon
ages and despite consistent TL and IRSL data, there has
been debate about the resetting of the luminescence signal
in fault gouge. Some researchers propose a grain-size
dependence with finest grains almost certainly reset, while
others revealed no evidence for resetting of fault gouge
(see Spencer et al., 2012 and references there in). It is
suggested that luminescence emission from fault gouge
samples has the potential to determine the age distribution
of distinct deformation microstructures (Spencer et al.,
2012). Such age determination could help constrain some
of the proposed micromechanical models for shear locali-
zation in fault gouge, for example, identifying the relative
effect of stress-driven mass transfer processes.
Injection dikes correspond to episodic pulses of an
increasing hydraulic pressure generated mostly by seismic
loading during earthquakes. These injection dikes (formed
by liquefaction during earthquakes) were used for recon-
struction of paleo-epicenter locations (McCalpin, 1996
and references therein) and have been under investigation
to be dated by OSL in an attempt to directly date seismic
events (e.g., Thomas et al., 2007).
Sand blows are lenticular sand bodies that were erupted
onto paleo-ground surfaces and were fed by sand-filled
dikes that were sourced from subjacent liquefied sand
beds. Age constraint on the timing of liquefaction and
seismic activity may be obtained by OSL dating of sand
blows formed by the surface rupture of liquefaction fea-
tures and injection dikes (e.g., Thomas et al., 2007).
 LUMINESCENCE, EARTHQUAKE AND TECTONIC ACTIVITY
Dating tsunami events is important in estimating the
recurrence intervals of catastrophic events and hazard risk
assessment. OSL measures the times elapsed since sedi-
mentation, so direct dating of tsunamigenic sediments by
luminescence is possible if signal resetting is complete.
Usually, aeolian and fluvial resetting of the luminescence
signal takes place during sediment transport. However,
the process of sediment transport by tsunami is often rapid:
maximum tens of minutes for each half wave with total sed-
imentation lasting several hours. The flow may be too
muddy to expose all sediments to light and the tsunami
may occur at night. Consequently, only limited resetting
occurs during transport. The assumption for zeroed lumi-
nescence signals in tsunami-deposited sediments is that
the source sediments are well bleached prior to tsunami
entrainment. The source material is sediment from the
coastal and near-shore zone that is constantly reworked by
waves and tides. Intertidal sediments from various locations
and beach-face sediments deposited by a combination of
wind and wave action have been shown to have a near-zero
residual luminescence signal. OSL dating of deposits of the
2004 Indian Ocean tsunami and paleotsunami sediments of
known age proved the general applicability of the method
for tsunamigenic sand sheets (see Brill et al., 2012;
Prendergast et al., 2012 and references therein).
Luminescence dating of tectonic movements
Terraces, alluvial fans, fluvial, and dry lake deposits are
datable by OSL and can be important records of tectonic
activity. Fault activity or fold deformation can uplift or
subside land and change channel patterns near known
structures (e.g., Thamó-Bozsó et al., 2010). Land uplift
can block a river and make a lake or change the direction
of rivers. River and marine terraces can be dated by OSL
to determine the rate of uplift (e.g., Choi et al., 2003).
Thick basin-fill sediments can be dated to determine sub-
sidence rates (e.g., Chen et al., 2003). Late Quaternary
landforms which are displaced by a fault can be used to
determine the slip rate and average recurrence interval
(e.g., Fattahi et al., 2007). OSL can provide dates of
deformed and undeformed fluvial and aeolian sequences
which can be used to find the timing of neotectonic activ-
ity (e.g., Thamó-Bozsó et al., 2010). IRSL dating of the
inland fans/terminal fans (as the potential geomorphic
markers) can be used to determine the shift of neotectonic
activities in a region (Pati et al., 2012).
Luminescence dating has been used to assess tectonic
activity by studying the fluvial geomorphic response of
rivers to deformation (e.g., Rittenour 2008 and references
therein). These include:
• Determining the timing of the switch from a highly sin-
uous channel to a straightened channel as a response to
periodic clustering of seismic events and decreased gra-
dient due to fault displacement
• Fold migration during the Holocene era due to fault
length extension
459
• Examining the faulting and uplift history along a river
and comparing the influence of tectonic uplift and cli-
mate for the formation of strath terraces
• Rates of active faulting during the late Quaternary
fluvial terrace formation at a river.
Conclusion
Luminescence dating is rapidly evolving into a very
powerful tool for determining the depositional age of earth-
quake-related sediments. It can do this because of its versa-
tility in directly sampling targeted features and deposits
using the most common minerals on the earth: quartz and
feldspar. Following the successful application of TL dating
for paleoseismic reconstruction (reviewed by Fattahi,
2009), a new approach using OSL was employed to date
paleoearthquakes and related sediments that have been
displaced and exposed to daylight by any physical phenom-
ena associated with an earthquake (e.g., ground motion,
ground failure, surface ruptures, liquefaction, and tsunami).
These dates are also essential for estimating the activity of
earthquake faults, understanding neotectonic processes
and rates, determining fault slip rates, and establishing
recurrence intervals of fault activities. These are only the
most prominent uses.
Luminescence dating results can be used not only to
resolve paleoseismological problems on their own but also
to prove if a fault is active for engineering purposes. Ages
of paleoearthquake features plus all possible background
information (geomorphology, structural geology, sedimen-
tology, various geophysical records, results from other
investigations, etc.) make paleoseismological interpretation
of the results possible, which provide useful new insights
and information for earthquake hazard risk assessment.
Bibliography
Allen, C. R., 1986. Seismological and paleoseismological tech-
niques of research in active tectonics. In Wallace, R. E. (ed.),
Active Tectonics: Studies in Geophysics. Washington, DC:
National Academy Press, pp. 148–154.
Brill, D., Klasen, N., Brückner, H., Jankaew, K., Scheffers, A.,
Kelletat, D., and Scheffers, S., 2012. OSL dating of tsunami
deposits from Phra Thong Island, Thailand. Quaternary Geo-
chronology, 10, 224–229.
Chen, Y. W., Chen, Y. G., Murray, A. S., Liu, T. K., and Lai, T. C.,
2003. Luminescence dating of neotectonic activity on the south-
western coastal plain, Taiwan. Quaternary Science Reviews,
22, 1223–1229.
Choi, J. H., Murray, A. S., Cheong, C. S., Hong, D. G., and Chang,
H. W., 2003. The resolution of stratigraphic inconsistency in the
luminescence ages of marine terrace sediments from Korea.
Quaternary Science Reviews, 22, 1201–1206.
Fattahi, M., 2009. Dating past earthquakes and related sediments by
thermoluminescence methods: a review. Quaternary Interna-
tional, 199, 104–146.
Fattahi, M., Walker, R., Hollingsworth, J., Bahroudi, A., Nazari, H.,
Talebain, M., Armitage, S., and Stokes, S., 2006. Holocene slip-
rate on the Sabzevar thrust fault, NE Iran, determined using
optically stimulated luminescence (OSL). Earth and Planetary
Science Letters, 245, 673–684.
460 LUMINESCENCE, FLINTS AND STONES
Fattahi, M., Walker, R. T., Khatib, M. M., Dolati, A., and Bahroudi,
A., 2007. Slip-rate estimate and past earthquakes on the Doruneh
fault, eastern Iran. Geophysical Journal International, 168,
691–709.
Fattahi, M., Nazari, H., Bateman, M. D., Meyer, B., and Sebrier, M.,
2010. Refining the OSL age of the last earthquake on the
Dheshir fault, Central Iran. Quaternary Geochronology, 5,
286–292.
Gurpinar, A., 2005. The importance of paleoseismology in seismic
hazard studies for critical facilities. Tectonophysics, 408(1–4),
23–28.
Kijko, A., 2011. Seismic hazard. In Gupta, H. K. (ed.), Earth and
Environmental Science Encyclopedia of Solid Earth Geophysics.
Dordrecht: Springer, pp. 1107–1120.
Lee, H., Im, C. B., Shim, T. M., Choi, H. S., Noh, M., and Jeong,
J. H., 2006. Technical backgrounds of active fault definitions
used for nuclear facility sitting: a review. Geophysical Research
Abstracts, 8, 09228.
Lin, A., Guo, J., Kano, K., and Awata, Y., 2006. Average slip rate
and recurrence interval of large-magnitude earthquakes on the
western segment of the strike-slip Kunlun Fault, northern Tibet.
Bulletin of the Seismological Society of America, 96,
1597–1611.
McCalpin, J. P., 1996. Paleoseismology. New York: Academic.
McCalpin, J. P., 2009. Paleoseismology, 2nd edn. Burlington: Aca-
demic Press. International Geophysics Series, Vol. 95, 613 pp.
Pati, P., Parkash, B., Awasthi, A. K., and Jakhmola, R. P., 2012. Spa-
tial and temporal distribution of inland fans/terminal fans
between the Ghaghara and Kosi rivers indicate eastward shift
of neotectonic activities along the Himalayan front. A study from
parts of the upper and middle Gangetic plains. India Earth-
Science Reviews, 115, 201–216.
Porat, N., Duller, G. A. T., Amit, R., Zilberman, E., and Enzel, Y.,
2008. Recent faulting in the southern Arava, Dead Sea Trans-
form: evidence from single grain luminescence dating. Quater-
nary International, 199, 34–44.
Prendergast, A. L., Cupper, M. L., Jankaew, K., and Sawai, Y.,
2012. Indian Ocean tsunami recurrence from optical dating of
tsunami sand sheets in Thailand. Marine Geology, 295–298,
20–27.
Rittenour, T. M., 2008. Luminescence dating of fluvial deposits:
applications to geomorphic, paleoseismic and archaeological
research. Boreas, 37, 613–635.
Sohbati, R., Murray, A. S., Chapot, M. S., Jain, M., and Pederson, J.,
2012. Optically stimulated luminescence (OSL) as a chronome-
ter for surface exposure dating. Journal of Geophysical
Research, 117, 1–7.
Spencer, J. Q. G., Hadizadeh, J., Gratier, J.-P., and Doan, M. L.,
2012. Dating deep? Luminescence studies of fault gouge from
the San Andreas Fault zone 2.6 km beneath Earth's surface.
Quaternary Geochronology, 10, 208–284.
Thamó-Bozsó, E., Magyari, Á., Musitz, B., and Nagy, A., 2010.
OSL ages and origin of Late Quaternary sediments in the North
Transdanubian Hills (Hungary): timing of neotectonic activity.
Quaternary International, 222, 209–220.
Thomas, P. J., Reddy, D. V., Kumar, D., Nagabhushanam, P.,
Sukhija, B. S., and Sahoo, R. N., 2007. Optical dating of lique-
faction features to constrain prehistoric earthquakes in Upper
Assam, NE India - some preliminary results. Quaternary
Geochronology, 2, 278–283.
Toda, S., 2011. Paleoseismology. In Gupta, H. K. (ed.), Earth and
Environmental Science, Encyclopedia of Solid Earth Geophys-
ics. Dordrecht: Springer, pp. 964–976.
Yanchou, L., Prescott, J. R., Zhao, H., Chen, J., and Wei, L. Y.,
2002. Optical dating of colluvial deposits from Xiyangfang
China and the relation to paleoearthquake events. Quaternary
Science Reviews, 21, 1083–1097.
Cross-references
Luminescence Dating
Luminescence, Coastal Sediments
Luminescence, Colluvial Sediments
Luminescence, Fluvial Sediments
Luminescence, Geomorphological Processes
Tsunamigenic Sediments
LUMINESCENCE, FLINTS AND STONES
Daniel Richter
Department of Human Evolution, Max Planck Institute for
Evolutionary Anthropology, Leipzig, Germany
Geomorphologie, University of Bayreuth, Bayreuth,
Germany
Introduction
Most applications of luminescence dating of flints and
stones are related to archaeology and the timing of prehis-
toric sites. Here, prehistoric human activities are directly
dated in most cases, using the last heating of flint, chert,
limestone, quartzite, or quartz. This is in contrast to surface
exposure dating of such materials, where the event dated –
the last exposure or burial – is often not related to the event
of the human occupation or activities. Consideration of the
relation between the event dated and the research question
asked (i.e., sedimentology, human activity, etc.) is crucial.
This is well discussed in Dean (1978), who describes
a “typology of events.” Luminescence dating is especially
useful in archaeological context at the limits or beyond
the range of radiocarbon dating of organic material or when
the latter is lacking (see “Luminescence Dating”). The time
span of application ranges from a few to several hundred
ka. However, recent developments employing orange-red
TL (Richter, 2011) suggest applicability close to 1 Ma
under favorable circumstances.
Thermoluminescence (see “Quartz Defects, Optically
Stimulated Luminescence and Thermoluminescence”) is
often employed for determining the heat exposure age,
but in principle, optically stimulated luminescence (OSL,
see “Quartz Defects, Optically Stimulated Luminescence
and Thermoluminescence”) can be employed as well.
However, OSL is not generally applicable to materials
classed as flint, obsidian, etc., and therefore, only few
studies investigate such theoretical venues (Poolton
et al., 1995; Polymeris et al., 2010; Schmidt and Kreutzer,
2013). Rock material without significant amorphous
phases, i.e., containing significant amounts of a-quartz
(see “Quartz”) such as unheated quartz pebbles, is also
suitable for surface exposure dating (see “Luminescence,
Rock Surfaces”). Here, the time of the last exposure to
light is dated (e.g., Kouremenos, 2008; Sohbati
et al., 2012) which is not necessarily a prehistoric human
activity, such as turning rocks over (e.g., Greilich and
Wagner, 2009).
 LUMINESCENCE, FLINTS AND STONES 461

Luminescence of heated flint and stone
The principles of dosimetric dating (see “Luminescence
Dating”) and associated physical phenomena (see “Radia-
tion and Radioactivity”) apply. The radiation dose, as
determined by luminescence methods, is caused by ioniz-
ing radiation from within the sample as well as its sur-
roundings. The ratio of radiation dose (palaeodose, P) to
dose rate (Ḋ) provides the age. Archaeological flint and
stones (luminescence properties) as well as their context
(dosimetry) exhibit some specific aspects which are differ-
ent from other applications of such methods, such as ESR
or optical dating (Richter, 2007).
Samples and sample selection
Samples are primarily selected on the basis of suitability
for dating. In the case of heated material, macroscopic fea-
tures like crenation, reddening, craquelation, potlids, etc.,
are strong, but not conclusive (e.g., Richter, 2007) indica-
tors of heat alteration of lithic material. In most contexts,
natural fires can be excluded as cause of the heating
(Richter, 2007; Alperson-Afil et al., 2007); however, the
time difference between anthropogenic and natural firing
is likely not significant in most circumstances (Richter
and Krbetschek, 2014a). The sufficiency of the prehistoric
zeroing for dating application can be tested for by using
the material’s luminescence properties, such as the
presence of a heating plateau and an equivalent dose pla-
teau (Aitken, 1985). Sample size requirements depend
on the technique/protocol used to evaluate the palaeodose
(see below). In any case, the outer 2 mm of samples are
sawn off in order to exclude those parts which have
received a radiation dose from a-/b-radiation originating
from outside the sample. Thus, the number of parameters
in age calculation is reduced and precision is increased.
However, this requires minimum sample thicknesses of
5–6 mm, and thus several grams of weight, mostly depen-
dent on sample shape. For standard techniques based on
multiple aliquots, about 1 g of material has to remain after
the removal of the outer 2 mm, but much less is required
for single-aliquot procedures (Richter, 2011).
For optical dating of surfaces of rocks, the outer 2 mm sur-
face area is not to be removed because of limitations in the
bleaching depth of the light causing the zeroing of the signal.
Palaeodose determination
The thermoluminescence (TL) signal of most heated flint
and stone samples exhibit (significant for flint) supralinear
growth of the TL signal with dose/age (Figure 1). This
behavior is also seen in inclusion pottery dating (see “Lumi-
nescence, Pottery and Bricks”; Aitken, 1985). In the stan-
dard application of the technique, several groups of
multiple aliquots (subsamples) receive a suite of irradia-
tions from a calibrated ionizing source (see “Radiation

Luminescence, Flints and Stones, Figure 1 Three different methods for estimating palaeodoses (P1, P2, P3) (a) additive TL signals in
blue, where NTL is the natural TL signal and NTL+D is the natural TL plus the TL induced by additional artificial irradiation; (b) solid
lines show the theoretical signal growth with dose (d) of the additive (light blue) and regeneration dose growth curves (light red), both
exhibiting supralinear behavior. Dotted lines show fitted data (plotted as peak heights) with regression analysis respectively. (c)
Regeneration TL signals in red, where RTL+D is the TL from artificial irradiation after having heated the sample. The sum of the
regression (dotted blue) of the additive growth curve (dotted blue line providing the equivalent dose, DE) and the regeneration growth
curve (dotted red line providing the supralinearity correction, DI) are giving the total absorbed dose (palaeodose, P1) in a multiple-
aliquot additive regeneration (MAAR) protocol. Shifting of either growth curve (slide) along the x-axis (dose) provides P as well (P2),
while interpolation of the natural signal on the regenerated dose growth curve determines P in a single-aliquot regeneration (SAR)
protocol (P3).
462 LUMINESCENCE, FLINTS AND STONES
and Radioactivity”). While the dose producing the natural
signal (palaeodose, P, or equivalent dose, DE) is not known,
increasing the luminescence signal by several artificial irra-
diation steps of known dose (dose points) provides the sen-
sitivity (slope of luminescence versus dose) of each
individual sample to ionizing radiation. Regression analysis
of the measured luminescence signals that are given artifi-
cial doses (additive dose growth curve in Figure 1) provides
an equivalent dose (DE) in the multiple-aliquot additive
dose (MAAD or MAD) protocol. The supralinear growth
of the signal at low doses is considered by a second regres-
sion (supralinearity correction, DI) of luminescence signals
from aliquots which were zeroed in the laboratory prior to
irradiation (regeneration dose growth curve). In this
multiple-aliquot-additive-regeneration (MAAR) protocol,
the sum of DE and DI provides the total absorbed dose
(palaeodose, P) the sample has received since its zeroing
in antiquity (Aitken, 1985). Extrapolation procedures for
large palaeodoses when the natural luminescence signal is
at the onset of saturation, however, are prone to error and
not precise. In such cases, the palaeodose can be determined
by sliding one of the two dose growth curves along the dose
axis until they match (Figure 1) and the absolute value of
the slide provides the palaeodose (e.g. Mercier
et al., 2004). In single-aliquot regeneration (SAR) tech-
niques, the natural luminescence signal is just interpolated
on a regenerated dose growth curve (Figure 1) (Richter
and Krbetschek, 2014b). So far this technique appears to
be restricted to heated flint using the orange-red thermolu-
minescence (Richter and Krbetschek, 2006; Richter and
Temming, 2006) and has not been applied regularly. How-
ever, it allows the dating of rather small samples (Richter,
2011), which is also possible in OSL techniques (see
“Luminescence Dating”) which are commonly based on
a SAR protocol, as well as for dating heated materials or
in surface exposure dating.
Dosimetry
The dosimetry (see “Luminescence Dating, Dose Rates”)
in luminescence dating of flints and stones can be grouped
into internal and external dose rates. The former stems
from the content in radionuclides of the sample
(commonly U, Th, K) and is determined by geochemical
analysis (e.g., see “Neutron Activation Analysis”).
A common value for the sensitivity of the sample to
a-radiation (a- or b-value) is often assumed for quartz peb-
bles (in addition to the assumption of a small internal dose
rate) but should be individually determined for at least the
majority of samples especially in heated rock TL dating,
because of large variability even within supposed single
sources of the origin of the material (especially flint).
Recent analysis has shown that significant variation is
exclusively correlated to macroscopic visible differences
in flint (Schmidt et al., 2013), which is the most com-
monly employed material. These visible differences
include veins, inclusions, etc., which have to be removed
in any case, because they obviously belong to different
mineral phases. The luminescence properties of different
phases are likely to be rather different, and such composite
signals cannot, a priori, provide accurate results. There-
fore, with the removal of such phases, and other geochem-
ically instable parts like patination, the assumption of
insignificant variability is valid, and the internal dose rate
can be considered as stable over the entire time range of
the method.
The external dose rate originates from outside the sam-
ple, i.e., radionuclides contained in the surrounding matrix
(sediment, éboulis, gravel, blocks of stone, cave walls,
lithics, bones, etc.) and a cosmic contribution. Surpris-
ingly in most cases the assumption of constancy of the
external dose rate (see “Luminescence Dating, Dose
Rates”) appears to be valid but should always be verified
by laboratory gamma spectrometry to the extent possible.
In heated rock dating, the contribution from external
a-/b-radiation does not have to be considered because of
the removal of the outer 2 mm. While these could be
modeled, the complex shapes of especially artifacts would
presumably result in low-accuracy age determinations.
The, usually small, cosmic dose rate depends on depth of
burial, altitude, and latitude and is considered as stable.
The external g-dose rate often dominates the total dose
rate, and ages are thus very dependent on the accurate
determination of this parameter (see Richter, 2007; Table 1
for a hypothetical example). Because samples are almost
always removed from context prior to dosimetric analysis,
the external g-dose rate cannot be measured at the exact
location of the samples. If sediments are heterogeneous
(e.g., in caves), analogous locations of geometries similar
to the sample positions are required for in situ measure-
ments by portable g-spectrometry or dosimeters in 30 cm
deep holes (4-p geometry). Alternatively, each excavated
surface (2-p geometry) can be measured with portable
instruments (Guérin and Mercier, 2012). A discussion on
the implications of the external dosimetry and methods
used can be found in Richter (2007); and Richter
et al. (2014). Because the external g-dose rates cannot be
determined at the find spot, it is advisable to approach
the age of a lithic assemblage by measuring multiple sam-
ples and employ an average external dosimetry from sev-
eral readings (Richter, 2007). This often results in large
uncertainties associated with this parameter and conse-
quently large uncertainties in individual ages. Such
derived average external dose rates do not necessarily cor-
respond to the dose rate the sample actually was exposed
to during burial, and the resulting individual TL ages
may be too young or too old. This results in large spreads
of TL ages for a given context, and the individual sample
ages have to be interpreted with caution. However, this
can be overcome by dating multiple samples from
a context, yielding an average TL age derived from all
samples. This provides a good representation of the con-
text age (e.g., Richter and Krbetschek, 2014a). Obviously,
averaging requires strong indications of the validity of the
assumption that the heating of samples took place roughly
at the same time. Alternatively, the external g-dose rate
 LUMINESCENCE, FLINTS AND STONES
can be modeled, and the dosimetry estimated for single
samples (Guibert et al., 1998; Guérin, 2012), provided that
the heterogeneity of the sediments is extremely well
documented. In general the confidence in TL-dating
results of heated flint depends on the ratio of internal to
external dose rate, where samples with large internal
(stable) dose rates are considered more reliable than others
(Richter, 2007).
Application of luminescence dating of
flint or stone
Luminescence dating of heated flint and stone is playing
a major role in establishing chronostratigraphies, e.g., for
the expansion of humans out of Africa or the settlement
of Europe. TL dating of heated flint established the quasi
contemporaneity of early modern humans at Skhul
(Mercier et al., 1993) and Qafzeh (Valladas et al., 1988)
at >100 ka and with Neanderthals (Valladas et al., 1987)
at about 60 ka in the Levant. These studies showed for
the first time a potential time period for these species to
meet; however, actual evidence of such is still missing.
The antiquity of elaborate figurative art of ca. 40 ka was
shown by standard TL protocols for heated flint by Richter
et al. (2000) for the site of Geißenklösterle cave in South-
ern Germany. This site yields some of the most intriguing
examples of the Upper Paleolithic art, with figurines made
from ivory depicting animals as well as humans.
TL-dating results agreed only with some of the available
radiocarbon data, the latter of which showed a variability
considered as inconsistent with site formation. However,
it was not until the improvement of pretreatment methods
of radiocarbon dating of bone until excellent agreement
was shown between 14C and TL (Higham et al., 2012).
While these samples had their natural TL well within the
linear dose range of heated flint, artifacts from Menez-
Dregan I (France) could only be dated by the slide method,
due to their antiquity of ca. 200 ka (Mercier et al., 2004).
Heated flint artifacts from the Middle Paleolithic lake site
of Neumark-Nord 2 rarely exceed weights of 5 g,
inhibiting the application of standard MAAR regression
procedures. The application of a SAR protocol resulted
in an Eemian (early last interglacial) age for the last
heating, which is in excellent agreement with stratigraphy,
palynology, sediment OSL, and magnetic dating (Sier
et al., 2011; Richter and Krbetschek, 2014b).
Less frequently used materials sometimes provide spe-
cific challenges, especially for dosimetry. The general
assumption of homogeneous distribution of radionuclides
is not valid in quartzite, and internal microdosimetric
modeling is required to determine reliable ages (Tribolo
et al., 2006). An age of 74  5 ka was obtained for five
burnt lithics of the Still Bay layers at Blombos Cave
(South Africa). This result is in good agreement with both
ESR dates on teeth and OSL dates on sediment (Tribolo
et al., 2006), demonstrating the antiquity of “modern
behavior” in Africa. Heated limestone was dated by Roque
et al. (2001) for the Upper Paleolithic site of Combe
463
Saunière (France). The problem of spurious TL, usually
interpreted as the result of a decarbonation process, was
avoided by measuring the TL from the calcite crystals in
a carbon dioxide atmosphere. TL dates obtained were
found in good agreement with radiocarbon dates.
The last heating of lithic material can be dated by opti-
cally stimulated luminescence (OSL) for some materials
as well. OSL dating of heated rock sandstone revealed
consistent ages around 8 ka for clusters of hearths in the
Saharan Fazzan Basin (Libya), which will allow more
detailed reconstruction of the Neolithic use of the land-
scape of this area (Armitage and King, 2013). Quartz does
not need to be heated for dating purposes, but the event
dated is then the last exposure to light, routinely used to
determine deposition ages of sediments (see “Quartz
Defects, Optically Stimulated Luminescence and Thermo-
luminescence”). Excellent agreement was obtained for
OSL dating of the outer millimeter of the buried surface
of a rockfall boulder with the OSL dating of quartz grains
from the underlying sediment, as well as with AMS radio-
carbon estimates of a cottonwood leaf found immediately
between the clast and the underlying sediment (Chapot
et al., 2012), implying a minimum age of 900 years for
the creation of the rock art. Application of OSL is by no
means limited to archaeological context, and satisfying
agreement was obtained for unheated Holocene beach
cobbles, from which the buried undersides were dated
(Kouremenos, 2008).
TL dating of heated flint artifacts has also served to
detect site disturbances. A previously unnoticed severe
mixing of a Paleolithic site was indicated by the large var-
iability in TL-dating results of heated flint from an assem-
blage believed to represent a short time span. The age
differences of two samples from the same layer with ages
of 38 and 5 ka can only be explained by sedimentary dis-
turbance (Debenham, 1994). TL dating can further pro-
vide indications for the correct attribution of mixed lithic
assemblages to technocomplexes. In the case of the Paleo-
lithic site of Chez-Pinaud Jonzac, the TL results for burnt
flints from a mixed context provided age estimates in
accordance with the result of the lithic analysis, suggesting
an Upper as well as a Middle Paleolithic occupation based
on diagnostic artifacts (Richter et al., 2013).
Conclusions
Luminescence dating of flint and stone is an important
tool, especially in establishing chronologies for the Paleo-
lithic period by determining the age of lithic assemblages
through dating the last heating of artifacts. Thermolumi-
nescence dating of heated flint has a major impact in our
understanding of the later human evolution, especially
the spread of early modern humans and the replacement
of Neanderthals. The most useful time range lies beyond
the calibration range of radiocarbon dating, because the
precision achieved with luminescence methods is often
not high enough to answer chronological or chronostra-
tigraphic questions related to Late Upper Pleistocene and
464 LUMINESCENCE, FLINTS AND STONES
later period research. In the light of achievable uncer-
tainties, it always has to be considered if these techniques
actually can provide results capable of answering the
question raised. Individual uncertainties in ages at 1-s
between 5 % and 10 %, which is 10–20 % at 95 % proba-
bility, can be achieved, and precisions of about 3 % (1-s)
can be obtained for weighted average ages when the same
context is dated by multiple samples. Such a precision is
comparable to calibrated radiocarbon data at the limits of
the 14C method. Dating of objects larger than twice the
range of b-radiation includes a stable dose-rate component
which increases confidence in age result. The various
luminescence techniques employed in dating flint and
stone have been shown to agree very well with other dat-
ing evidence, while often having the advantage that
a human action is directly dated.
Bibliography
Aitken, M. J., 1985. Thermoluminescence Dating. London:
Academic Press.
Alperson-Afil, N., Richter, D., and Goren-Inbar, N., 2007. Phantom
hearths and the use of fire at Gesher Benot Ya‘aqov, Israel.
PaleoAnthropology, 1, 1–15.
Armitage, S. J., and King, G. E., 2013. Optically stimulated lumi-
nescence dating of hearths from the Fazzan Basin, Libya:
a tool for determining the timing and pattern of Holocene occu-
pation of the Sahara. Quaternary Geochronology, 15, 88–97.
Chapot, M. S., Sohbati, R., Murray, A. S., Pederson, J. L., and
Rittenour, T. M., 2012. Constraining the age of rock art by dating
a rockfall event using sediment and rock-surface luminescence
dating techniques. Quaternary Geochronology, 13, 18–25.
Dean, J. S., 1978. Independent dating in archaeological analysis.
In Schiffer, M. B. (ed.), Advances in Archaeological Method
and Theory. New York: Academic Press, Vol. 1, pp. 223–255.
Debenham, N. C., 1994. A guide to TL dating flint assemblages.
In Ashton, N., and David, A. (eds.), Stories in Stone. London:
Lithic Studies Society. Lithic Studies Society Occasional Paper,
Vol. 4, pp. 4–6.
Greilich, S., and Wagner, G. A., 2009. Light thrown on history – the
dating of stone surfaces at the geoglyphs of Palpa using optically
stimulated luminescence. In Reindel, M., and Wagner, G. A.
(eds.), New Technologies for Archaeology. Multidisciplinary
Investigations in Palpa and Nasca, Peru. Natural Science in
Archaeology. Berlin: Springer, pp. 271–286.
Guérin, G., 2012. Numerical simulations and modeling of dosimet-
ric effects in Quaternary sediments: application to luminescence
dating methods. Ancient TL, 30(1), 30.
Guérin, G., and Mercier, N., 2012. Field gamma spectrometry,
Monte Carlo simulations and potential of non-invasive measure-
ments. Geochronometria, 39(1), 40–47.
Guibert, P., Bechtel, F., Schvoerer, M., Müller, P., and Balescu, S.,
1998. A new method for gamma dose-rate estimation of hetero-
geneous media in TL dating. Radiation Measurements, 29(5),
561–572.
Higham, T., Basell, L., Jacobi, R., Wood, R., Ramsey, C. B., and
Conard, N. J., 2012. esting models for the beginnings of the
Aurignacian and the advent of figurative art and music: the
radiocarbon chronology of Geißenklösterle. Journal of Human
Evolution, 62(6), 664–676.
Kouremenos, P., 2008. Testing the Use of OSL on Cobbles from the
Raised Beaches of King George Island, Antarctica. M.S., Ann
Arbor: Oklahoma State University
Mercier, N., Valladas, H., Bar-Yosef, O., Vandermeersch, B.,
Stringer, C. B., and Joron, J. L., 1993. Thermoluminescence date
for the mousterian burial site of Es-Skhul, Mt. Carmel. Journal
of Archaeological Science, 20(2), 169–174.
Mercier, N., Froget, L., Miallier, D., Pilleyre, T., Sanzelle, S., and
Tribolo, C., 2004. Nouvelles données chronologiques pour le
site de Menez-Dregan 1 (Bretagne); l’apport de la thermolumi-
nescence. New chronological data for the Menez-Dregan I site
in Brittany; thermoluminescence data. Quaternaire, 15(3),
253–261.
Polymeris, G. S., Gogou, D., Afouxenidis, D., Rapti, S., Tsirliganis,
N. C., and Kitis, G., 2010. Preliminary TL and OSL Investiga-
tion of a naturally and artificially irradiated obsidian. Mediterra-
nean Archaeology and Archaeometry, 10(4), 83–91.
Poolton, N. R. J., Bøtter-Jensen, L., and Rink, W. J., 1995. An opti-
cally stimulated luminescence study of flint related to radiation
dosimetry. Radiation Measurements, 24(4), 551–555.
Richter, D., 2007. Advantages and limitations of thermolumines-
cence dating of heated flint from Paleolithic sites.
Geoarchaeology, 22(6), 671–683.
Richter, D., 2011. Dating small heated flint artifacts: a new thermo-
luminescence technique. In Conard, N. J., and Richter, J. (eds.),
Neanderthal Lifeways, Subsistence and Technology. Dordrecht:
Springer. Vertebrate Paleobiology and Paleoanthropology, Vol.
19, pp. 53–57.
Richter, D., and Krbetschek, M., 2006. A new thermoluminescence
dating technique for heated flint. Archaeometry, 48(4), 695–705.
Richter, D., and Krbetschek, M., 2014a. Luminescence dating of the
Lower Palaeolithic occupation at Schöningen. Journal of
Human Evolution (in press)
Richter, D., and Krbetschek, M., 2014b. Preliminary luminescence
dating results for two middle Palaeolithic occupations at
Neumark-Nord 2. In Gaudzinski-Windheuser, S., and
Roebroeks, W. (eds.), Multidisciplinary Studies of the Middle
Palaeolithic record from Neumark-Nord (Germany).
volume 1. Veröffentlichungen des Landesamtes für
Denkmalpflege und Archäologie Sachsen-Anhalt. Halle:
Landesamt für Denkmalpflege und Archäologie Sachsen-
Anhalt, Vol. 68 (in press).
Richter, D., and Temming, H., 2006. Testing heated flint palaeodose
protocols using dose recovery procedures. Radiation Measure-
ments, 41(7–8), 819–825.
Richter, D., Waiblinger, J., Rink, W. J., and Wagner, G. A., 2000.
Thermoluminescence, electron spin resonance, and 14C-dating
of the Late Middle and Early Upper Palaeolithic site
Geißenklösterle Cave in Southern Germany. Journal of Archae-
ological Science, 27(1), 71–89.
Richter, D., Hublin, J.-J., Jaubert, J., McPherron, S. P., Soressi, M.,
and Texier, J.-P., 2013. Thermoluminescence dates for the Mid-
dle Palaeolithic site of Chez-Pinaud Jonzac (France). Journal
of Archaeological Science, 40, 1176–1185.
Richter, D., Angelucci, D. E., Burbidge, C. I., Dias, M. I., Gouveia,
M. A., Prudencio, M. I., and Zilhão, J., 2014. Heated flint from
Gruta da Oliveira (Portugal): dosimetric challenges and compar-
ison of TL-dating results with radiocarbon and U-series dating.
Journal of Archaeological Science, 41, 705–715.
Roque, C., Guibert, P., Vartanian, E., Bechtel, F., and Schvoerer, M.,
2001. Thermoluminescence-dating of calcite: study of heated
limestone fragments from Upper Paleolithic Layers at Combe
Sauniere, Dordogne, France. Quaternary Science Reviews,
20(5–9), 935–938.
Schmidt, C., and Kreutzer, S., 2013. Optically stimulated lumines-
cence of amorphous/microcrystalline SiO2 (silex): basic investi-
gations and potential in archeological dosimetry. Quaternary
Geochronology, 15, 1–10.
Schmidt, C., Rufer, D., Preusser, F., Krbetschek, M., and Hilgers,
A., 2013. The assessment of radionuclide distribution in silex
 LUMINESCENCE, FLUVIAL SEDIMENTS
by autoradiography in the context of dose rate determination for
thermoluminescence dating. Archaeometry, 55(3), 407–422.
Sier, M. J., Roebroeks, W., Bakels, C. C., Dekkers, M. J., Brühl, E.,
De Loecker, D., Gaudzinski-Windheuser, S., Hesse, N., Jagich,
A., Kindler, L., Kuijper, W. J., Laurat, T., Mücher, H. J.,
Penkman, K. E. H., Richter, D., and van Hinsbergen, D. J. J.,
2011. Direct terrestrial-marine correlation demonstrates surpris-
ingly late onset of the last interglacial in central Europe. Quater-
nary Research, 75(1), 213–218.
Sohbati, R., Murray, A. S., Buylaert, J.-P., Almeida, N. A. C., and
Cunha, P. P., 2012. Optically stimulated luminescence (OSL)
dating of quartzite cobbles from the Tapada do Montinho archae-
ological site (east-central Portugal). Boreas, 41(3), 452–462.
Tribolo, C., Mercier, N., Selo, M., Valladas, H., Joron, J. L., Reyss,
J. L., Henshilwood, C., Sealy, J., and Yates, R., 2006. TL dating
of burnt lithics from Blombos Cave (South Africa): further evi-
dence for the antiquity of modern human behaviour.
Archaeometry, 48(2), 341–357.
Valladas, H., Joron, J. L., Valladas, G., Arensburg, B., Bar-Yosef,
O., Belfer-Cohen, A., Goldberg, P., Laville, H., Meignen, L.,
Rak, Y., Tchernov, E., Tillier, A. M., and Vandermeersch, B.,
1987. Thermoluminescence dates for the neanderthal burial site
at Kebara in Israel. Nature, 330, 159–160.
Valladas, H., Reyss, J. L., Joron, J. L., Valladas, G., Bar-Yosef, O.,
and Vandermeersch, B., 1988. Thermoluminescence dating of
mousterian “Proto-Cro-Magnon” remains from Israel and the
origin of modern man. Nature, 331, 614–616.
Cross-references
Chert
Luminescence Dating
Luminescence Dating, Dose Rates
Luminescence Dating, Uncertainties and Age Range
Luminescence, Pottery and Bricks
Luminescence, Rock Surfaces
Neutron Activation Analysis
Quartz
Quartz Defects, Optically Stimulated Luminescence and Thermo-
luminescence
Radiation and Radioactivity
Radiation Dose Rate
Terrestrial Cosmogenic Nuclide Dating
LUMINESCENCE, FLUVIAL SEDIMENTS
Amanda Keen-Zebert
Division of Earth and Ecosystem Sciences, Desert
Research Institute, Reno, NV, USA
Definition
The use of luminescence techniques to determine the
depositional age of sediments transported and deposited
by rivers and other fluvial systems.
Introduction
Dating fluvial sediments with luminescence techniques
has utility to a wide range of applications such as investi-
gating fluvial processes and alluvial stratigraphy or dating
fluvial sediments as proxies of climate, environmental and
sea-level change, tectonic activity, and the age of
465
archaeological sites. Thorough reviews of luminescence
applications to fluvial sediments and the typical problems
encountered in doing so are provided by Wallinga (2002),
Jain et al. (2004), and Rittenour (2008). The earliest
attempts to date fluvial sediments using luminescence
were published by Murray et al. (1995). Since then, equip-
ment and procedural advances, particularly the single ali-
quot regenerative dose (SAR) technique (Murray and
Wintle, 2000) and the ability to measure single grains
(see Bøtter-Jensen et al., 2000), have improved the capa-
bility to measure multiple equivalent doses (De) from
smaller and single grain aliquots quickly, thereby improv-
ing the ability to detect heterogeneous bleaching and to
date fluvial sediments.
Heterogeneous bleaching (also referred to as partial or
incomplete zeroing, or resetting) of the luminescence sig-
nals during transport is a common problem in the accurate
determination of De in fluvial systems because of the
attenuation of light through water and water-sediment
mixtures. Fluvial sediments are particularly susceptible
to type B partial bleaching (Duller, 1994), in which grains
in a sample are bleached to different degrees, resulting in
significant scatter in the distribution of De and a high like-
lihood of overestimating of the true burial age. Contribut-
ing factors include varied modes of transport, whether by
suspension, saltation, or bedload; rapid erosion associated
with floods; and inputs from non-bleached and older
deposits such as alluvial channel banks and beds, bedrock
banks and beds, and the breakdown of larger clasts into
sand sized and smaller particles through mechanical wear
and abrasion during transport.
Although partial bleaching is likely in fluvial environ-
ments, some studies report well-bleached sediments
evidenced by either strong agreement with independent
age control or tightly clustered Gaussian De distributions
(e.g., Murray, 1996; Kale et al., 2000; Stokes et al.,
2001; Kar et al., 2001; Rittenour et al., 2003; Leigh
et al., 2004; Rittenour et al., 2005). Possible explanations
for well-bleached sediments in fluvial environments
include transport during seasons or in systems with low
turbidity and slack water or overbank deposition in envi-
ronments where sediments may lay at the surface exposed
to sunlight for some time after deposition before burial.
In dating fluvial sediments, it is necessary to test for the
presence of partial bleaching and, when present, to apply
a strategy to determine the burial dose used in the final
age calculation. Evidence of age overestimation owing
to partial bleaching has been shown by measuring reset-
ting rates as a function of attenuation of light in water,
for example, in varied turbidity or transport distance; dat-
ing samples with independent age control; and dating
modern samples (see Wallinga, 2002; Jain et al., 2004
and references therein). Analytical approaches to identify-
ing partial bleaching include examining the scatter in De
distributions, comparing the form of different components
of decay curves to isolate the most sensitive and well-
bleached portion of luminescence signals, and comparing
different luminescence signals, for example, by
466 LUMINESCENCE, FLUVIAL SEDIMENTS
comparing TL and OSL results from a single sample (see
Wallinga, 2002 and references therein) or by comparing
the differential bleaching rates of quartz and feldspar
(Murray et al., 2012).
Approaches to determining the burial age from
a population of De from a sample that has experienced par-
tial bleaching include using small aliquots or single grains
to isolate and disregard the partially and/or unbleached
portion of the population of De and the application
a variety of statistical methods (see Wallinga, 2002;
Rittenour, 2008 and references therein). While using sin-
gle grain aliquots is often the preferred method to cope
with partially bleached samples, the use of small aliquots
is statistically viable and more efficient in many situations.
The luminescence signal from individual grains is highly
variable, and in multiple grain aliquots of quartz,
90–95 % of the total light emitted originates from only
5–10 % of the grains (Duller et al., 2000; Jacobs et al.,
2003). In those cases, small aliquots of < 100 grains are
statistically equivalent to the single grain approach and
improve the efficiency of analysis (Duller, 2008). One
approach is to test a small number single grains (e.g.,
one disk of 100 grains) to determine the percentage of
grains that produce a luminescence signal and then adjust
the number of grains per aliquot (i.e., the size of the ali-
quot) accordingly (e.g., Rodnight et al., 2006).
The use of single grains or small aliquots is particularly
important for young samples (< 1 ka), and Jain et al.
(2004) suggest that age offsets associated with partial
bleaching are sufficiently masked by the analytical uncer-
tainty associated with older samples. Nonetheless, hetero-
geneous bleaching has been observed in fluvial samples
>1 ka (e.g., Folz et al., 2001; Thomas et al., 2005; Brook
et al., 2006; Srivastava et al., 2006; Eitel et al., 2006;
Rodnight et al., 2006; Keen-Zebert et al., 2013), and
disregarding the heterogeneously bleached condition can
result in age overestimation greater than analytical error.
Regardless of the expected age, an assessment of the degree
of bleaching should be conducted taking into account the
characteristics particular to each sample and site.
Samples with heterogeneous bleaching tend to have De
populations with a wide range (high over-dispersion).
Several statistical approaches to estimating burial doses
have been posed including calculating the mean of the
lowest 5–10 % of population of De (Olley et al., 1998),
fitting the “leading edge” of the distribution after scatter
from experimental errors is removed (Lepper et al.,
2000), and fitting the distribution of De to statistical
models such as the minimum age model (MAM) or finite
mixture model (FMM) (Gailbraith et al., 1999; Galbraith,
2005). For widely ranging, complex, multimodal distribu-
tions of De, for example, arising from fluvial sediment
sources of mixed age, F-ratio filtering (a ratio of two inde-
pendent estimates of the variance in a normal distribution)
can be applied to discriminate between discrete
populations within a distribution (Spencer et al., 2003;
Spencer and Robinson, 2008). In some partially bleached
sedimentary environments, ages of samples in vertical
succession may overlap owing to high over-dispersion
and error. In these kinds of cases, rather than define
a single age for a sample, it may be advantageous to define
a probability distribution of ages from the population of
De using a bootstrapping approach that enables the identi-
fication of trends in sedimentation rates down-profile
(Cunningham and Wallinga, 2012). The choice of statisti-
cal approach to modeling the burial dose should be made
according to the characteristics of the site, considering fac-
tors that contribute to large dispersion in the population of
De such as partial bleaching, uneven dosimetry, and post-
depositional mixing, for example, owing to bioturbation
and other soil forming processes. Though inputting
a smaller number of replicates of De into statistical models
may be satisfactory for Gaussian distributions, for more
complex distributions, such as those exhibiting high
over-dispersion, at least 50–60 acceptable replicates of
De may be necessary to generate reproducible results
(Rodnight, 2008; Woda and Fuchs, 2008).
Current gaps in knowledge related to fluvial
processes
Early assumptions held that because fine grains are
transported higher in the water column, they should be
better bleached than coarser grains; however, studies have
shown that coarser grains tend to be better bleached (Olley
et al., 1998; Truelsen and Wallinga, 2003; Hu et al., 2010).
Although thermal transfer may be impacted by grain size,
Truelsen and Wallinga (2003) suggest that better
bleaching of coarse grains is not an artifact of thermal
transfer effects related to grain size. Some possible expla-
nations for this phenomenon include (1) channel bed
armoring at lower flows or during receding flood flows
in which finer sediments work into pore spaces in the
bed through kinematic sorting leaving a coarser layer on
top of the channel bed that would be exposed to sunlight
during lower base flows; (2) for rivers with flashy and/or
perennial flows, which are common in dryland rivers,
equal mobility of all grain sizes during floods which
would result in equal exposure to sunlight; (3) overbank
sand dunes being deposited on the floodplain that are well
exposed to sunlight then buried in subsequent flows;
(4) fine sediment being transported as flocs (larger aggre-
gates of fine sediment) the interiors of which would not be
exposed to light during transport; and (5) in basins with
heterogeneous lithology, differences in the source rock
of fine and course fractions may give rise to fundamental
differences in luminescence behavior or characteristics.
These processes do not occur in all rivers, and other fluvial
processes may contribute to a relationship between grain
size and bleaching, but this phenomenon remains untested
and more applications of luminescence could yield insight
into fundamental fluvial processes related to grain size and
transport.
Other issues associated with grain size and dating flu-
vial sediments relate to sorting, or the mixture of grain
sizes within a landform or sample. Potential problems
 LUMINESCENCE, FLUVIAL SEDIMENTS 467

related to poor sorting (large variation of grain sizes within agreement between OSL and soil age estimates,
a feature) involve both the high potential for partial morphostratigraphy, and well-dated lacustrine shorelines
bleaching and uneven dosimetry, for example, from mixed for the age of fan abandonment.
fine, sand, gravel, and cobble sizes within a single feature
such as a gravel bar. Outside of the findings of Murray Radiocarbon
et al. (1995) that floodplain deposits are better bleached
Because the applicable age range of luminescence over-
than channel deposits, observations of the impact of
laps that of radiocarbon, results from the two techniques
sorting on luminescence ages are lacking.
are often used to validate each other. In comparing radio-
A relationship between better bleaching with greater
carbon with OSL results and stratigraphic position in
transport distance has been observed in some studies
Holocene arroyo sediments in southern Utah, Summa-
(Stokes et al., 2001; Pietsch et al., 2008; Summa-Nelson
Nelson and Rittenour (2012) used both single grain and
and Rittenour, 2012). Though the phenomenon has not
small aliquots of quartz to determine De distributions
been thoroughly investigated in a variety of settings, possi-
and the minimum age model to determine the burial dose
ble explanations include increased sorting downstream,
used to calculate the age. They find good agreement
modification of grains during transport, and increased sen-
within errors between OSL and radiocarbon results except
sitivity owing to repeated bleaching and dosing cycles fol-
for one sample that had no positive skew and low over-dis-
lowing detachment from the parent material through
persion, and they suggest therefore that there was
erosion and redeposition of sediment. More work is needed
a problem with the dose rate estimation rather than with
to validate these observations in other fluvial systems.
the determination of De. Very small aliquots (0.3 mm)
Some basic understanding about fluvial processes may
were used with success by Cheetham et al. (2010b) to
be discovered through the use of portable OSL readers
complete a Holocene alluvial chronology in southeastern
even though they are currently limited to measuring pho-
Australia. Radiocarbon and OSL results were chronostra-
ton counts and do not enable age determination. In partic-
tigraphically consistent, but for a few samples, evidence of
ular, where multiple samples are collected across
bioturbation resulted in poor results from both techniques
a gradient, insight on sedimentary characteristics may be
(Cheetham et al., 2010b). Compared to the larger aliquots
discovered. For example, changes in sedimentological
used by Cheetham et al. (2010a), results from the small ali-
processes, variation in provenance, the likelihood of
quots used by Cheetham et al. (2010b) were more accu-
post-depositional mixing, and the suitability of sediments
rate, but were within two standard deviations of the
to dating with OSL can be assessed with photon counts
earlier measurements of larger aliquots. Briant and
from portable OSL readers (Muñoz-Salinas et al., 2011).
Bateman (2009) conducted a cross-comparison of paired
OSL and radiocarbon samples of fluvial sediments of the
Recent applications of luminescence to dating Nene and Welland rivers in eastern England and
fluvial sediments that have independent age reconfirmed results of an earlier study (Briant et al.,
control 2004a; Briant et al., 2004b, 2005) that had found offsets
Historical and sedimentary age controls between OSL and radiocarbon results in older samples.
In the later study, dose recovery tests and the SAR proto-
For case studies using historical information sedimentary
col were used on small (2 mm) aliquots to improve the
stratigraphy, and radiocarbon as independent age controls,
confidence in the results of the earlier study that had used
only publications that postdate Rittenour’s review (2008)
large aliquots, bringing into question the possibility of par-
are presented here. On terraces on the Middle Thames
tial bleaching. Reconfirming Briant et al. (2004a, b, 2005),
with a chronology well-constrained by OSL, U/Th, and
the results show good age agreement between OSL and
TL on artifacts, Pawley et al. (2010) found good compar-
radiocarbon ages before 35 14C ka BP, but age underesti-
ison of quartz OSL with the expected 450 ka age of the
mates by radiocarbon later (Briant and Bateman, 2009).
site. Initial results found systematic underestimation of
The authors attribute the result to the contamination of
10 % owing to contamination of the OSL signal by a ther-
older organic material with modern carbon and suggest
mally unstable medium component. The fast component
that confidence in conventionally pretreated radiocarbon
was isolated by curve fitting and by early subtraction of
ages be limited to < 35 14C ka BP.
the background, the latter of which was found to provide
the best separation of the thermally unstable signal. For
historical fluvial deposits on the Danube and Ebro, Fiebig Cosmogenic radionuclides
et al. (2009) used 2 mm aliquots and the finite mixture Owen et al. (2006) and DeLong and Arnold (2007) com-
model (FMM) on partially bleached sediments to find pared OSL ages of alluvial sediments within fans to 10Be
young ages (< 500 a) that were stratigraphically consis- cosmogenic radionuclide (CRN) exposure ages of boul-
tent and agreed with well-constrained chronologies from ders that were partially buried in alluvial fan surfaces.
dendrochronology and archaeology at the site. In the Both studies find good stratigraphic agreement among
hyperarid Negev, Porat et al. (2010) used the minimum the methods, but acknowledge scatter in the CRN results
age model on partially bleached sediments just underlying owing to inheritance, post-depositional erosion and
the abandoned surface of an alluvial fan and found good weathering, a limited number of CRN samples, and/or
468 LUMINESCENCE, FLUVIAL SEDIMENTS
shielding. Hetzel et al. (2004) and Nissen et al. (2009)
directly compared depth averaged 10Be CRN ages to
OSL ages of sediments within alluvial fans and found that
OSL underestimated the ages, citing feldspar contamina-
tion and abnormal quartz characteristics. However,
because the signal resetting mechanisms are fundamen-
tally different (OSL being reset by exposure to light or
heat and cosmogenic depth profiles being reset by
a number of processes including sediment mixing, deep
surface truncation, and inheritance), direct comparison of
OSL and CRN ages is somewhat questionable (Guralnik
et al., 2011). To cope with this discrepancy in resetting
mechanisms, Guralnik et al. (2011) fit cosmogenic and
luminescence data to a model that utilizes both OSL and
CRN data to provide a more nuanced platform for inter-
pretation of alluvial terrace history than is possible with
results from one technique or from directly comparing
both techniques on single sample sites.
Summary
Luminescence techniques can be used to date fluvial sed-
iments for a variety of applications from understanding
fluvial processes and alluvial stratigraphy to dating other
landscape features. The high potential for partial
bleaching prior to burial is the main concern regarding
the accuracy and precision of luminescence ages of fluvial
sediments. However, a combination of small aliquot or
single grain techniques and the use of appropriate statisti-
cal models have yielded successful age determination of
fluvial sediments with luminescence. Independent age
dating techniques including historical ages, radiocarbon,
and CRNs has repeatedly validated luminescence ages of
fluvial sediments in a wide age range environmental set-
tings and ages from the modern to the Pleistocene.
Bibliography
Bøtter-Jensen, L., Bulur, E., Duller, G. A. T., and Murray, A. S.,
2000. Advances in luminescence instrument systems. Radiation
Measurements, 32, 523–528.
Briant R. M., Coope G. R., Preece R. C., Gibbard P. L., 2004a. The
Upper Pleistocene deposits at Deeping St James, Lincolnshire:
evidence for Early Devensian fluvial sedimentation.
Quaternaire, 15, 5–15.
Briant R. M. , Coope G. R., Preece R. C. , Keen D. H., Boreham S.,
Griffiths H. I., Seddon M. B., Gibbard P. L., 2004b. Fluvial
response to Late Devensian aridity, Baston, Lincolnshire,
England. Journal of Quaternary Science, 19, 479–495.
Briant R. M., Bateman M. D., Coope G. R., Gibbard P. L., 2005. Cli-
matic control on Quaternary fluvial sedimentology: a Fenland
Basin river in England. Sedimentology, 52, 1397–1424.
Briant, R. M., and Bateman, M. D., 2009. Luminescence dating
indicates radiocarbon age underestimation in late Pleistocene
fluvial deposits from eastern England. Journal of Quaternary
Science, 24, 916–927.
Brook, G. A., Srivastava, P., and Marais, E., 2006. Characteristics
and OSL minimum ages of relict fluvial deposits near Sossus
Vlei, Tsauchab River, Namibia, and a regional climate record
for the last 30 ka. Journal of Quaternary Science, 21, 347–362.
Cheetham, M. D., Keene, A. F., Erskine, W. D., Bush, R. T.,
Fitzsimmons, K., Jacobsen, G. E., and Fallon, S. J., 2010a.
Resolving the Holocene alluvial record in southeastern Australia
using luminescence and radiocarbon techniques. Journal of
Quaternary Science, 25(7), 1160–1168. ISSN 0267–8179.
Cheetham, M., Bush, R., Keene, A., Erskine, W., and Fitzsimmons,
K., 2010b. Longitudinal correlation of later Quaternary terrace
sequences of Widden Brook, southeastern Australia. Australian
Journal of Earth Sciences, 57, 97–109.
Cunningham, A. C., and Wallinga, J., 2012. Realizing the potential
of fluvial archives using robust OSL chronologies. Quaternary
Geochronology, 12, 98–106.
DeLong, S. B., and Arnold, L. J., 2007. Dating alluvial deposits
with optically stimulated luminescence, AMS 14C, and cosmo-
genic techniques, western Transverse Ranges, California, USA.
Quaternary Geochronology, 2, 129–136.
Duller, G. A. T., 1994. Luminescence dating of poorly bleached sed-
iments from Scotland. Quaternary Science Reviews, 13,
521–524.
Duller, G. A. T., 2008. Single-grain optical dating of Quaternary
sediments: why aliquot size matters in luminescence dating.
Boreas, 37, 589–612.
Duller, G. A. T., Botter-Jensen, L., Kohsiek, P., and Murray, A. S.,
2000. Optical dating of single sand-sized grains of quartz:
sources of variability. Radiation Measurements, 32, 453–457.
Eitel, B., Kaderiet, A., Blümel, W.-D., Hüser, K., Lomax, J., and
Hilgers, A., 2006. Environmental changes at the eastern Namib
Desert margin before and after the Last Glacial Maximum: new
evidence from fluvial deposits in the upper Hoanib River catch-
ment, northwest Namibia. Palaeogeography, Palaeocli-
matology, Palaeoecology, 234, 201–222.
Fiebig, M., Preusser, F., Steffen, D., Thamó-Bozsó, E., Grabner, M.,
Lair, G. J., and Gerzabek, M. H., 2009. Luminescence dating of
historical fluvial deposits from the Danube and Ebro.
Geoarchaeology, 24, 224–241.
Folz, E., Bodu, P., Bonte, P., Joron, J. L., Mercier, N., and Reyss,
J. L., 2001. OSL dating of fluvial quartz from Le Closeau,
a Late Paleolithic site near Paris -comparison with 14C chronol-
ogy. Quaternary Science Reviews, 20, 927–933.
Gailbraith, R. F., Roberts, R. G., Laslett, G. M., Yoshida, H., and
Olley, J. M., 1999. Optical dating of single and multiple grains
of quartz from Jinmium rock shelter, northern Australia. Part I:
experimental design and statistical models. Archaeometry,
41, 339–364.
Galbraith, R. F., 2005. Statistics for Fission Tract Analysis. London:
Chapman and Hall. 240 pp.
Guralnik, B., Matmon, A., Avni, Y., Porat, N., and Fink, D., 2011.
Constraining the evolution of river terraces with integrated
OSL and cosmogenic nuclide data. Quaternary Geochronology,
6, 22–32.
Hetzel, R., Tao, M., Stokes, S., Niedermann, S., Ivy-Ochs, S., Gao,
B., Strecker, M. R., and Kubik, P. W., 2004. Late Pleistocene/
Holocene slip rate of the Zhangye thrust (Qilian Shan, China)
and implications for the active growth of the northeastern
Tibetan Plateau. Tectonics, 23, TC6006.
Hu, G., Zhang, J., Qiu, W., and Zhou, L., 2010. Residual OSL sig-
nals in modern fluvial sediments from the Yellow River (Huang
He) and the implications for dating young sediments. Quater-
nary Geochronology, 5, 187–193.
Jacobs, Z., Duller, G. A. T., and Wintle, A. G., 2003. Optical dating
of dune sand from Blombos Cave, South Africa: II. Single grain
data. Journal of Human Evolution, 44, 613–625.
Jain, M., Murray, A. S., and Botter-Jensen, L., 2004. Optically stim-
ulated luminescence dating: how significant is incomplete light
exposure in fluvial environments? Quaternaire, 15, 143–157.
 LUMINESCENCE, FLUVIAL SEDIMENTS
Kale, V. S., Singhvi, A. K., Mishra, P. K., and Banerjee, D., 2000.
Sedimentary records and luminescence chronology of Late
Holocene palaeofloods in the Luni River, Thar Desert, northwest
India. Catena, 40, 337–358.
Kar, A., Singhvi, A. K., Rajaguru, S. N., Juyal, N., Thomas, J. V.,
Banerjee, D., and Dhir, R. P., 2001. Reconstruction of the late
quaternary environment of the lower Luni plains, Thar Desert,
India. Journal of Quaternary Science, 16, 61–68.
Keen-Zebert, A., Tooth, S., Rodnight, H., Duller, G. A. T., Roberts,
H. M., and Grenfell, M., 2013. Late quaternary floodplain
reworking and the preservation of alluvial sedimentary archives
in unconfined and confined valleys of mixed bedrock-alluvial
rivers in the eastern interior of South Africa. Geomorphology,
185, 54–66.
Leigh, D. S., Srivastava, P., and Brook, G. A., 2004. Late Pleisto-
cene braided rivers of the Atlantic Coastal Plain, USA. Quater-
nary Science Reviews, 23, 65–84.
Lepper, K., Larsen, N. A., and McKeever, S. W. S., 2000. Equiva-
lent dose distribution analysis of Holocene eolian and fluvial
quartz sands from central Oklahoma. Radiation Measurements,
32, 603–608.
Muñoz-Salinas, E., Bishop, P., Sanderson, D. C. W., and Zamorano,
J., 2011. Interpreting luminescence data from a portable OSL
reader: three case studies in fluvial settings. Earth Surface Pro-
cesses and Landforms, 36, 651–660.
Murray, A. S., Olley, J. M. & Caitcheon, G. G. 1995: Measurement
of equivalent doses in quartz from contemporary water-lain sed-
iments using optically stimulated luminescence . Quaternary
Science Reviews, 14, 365–371.
Murray, A. S., 1996. Developments in optically stimulated
luminescence and photo-transferred thermoluminescence
dating of young sediments: application to a 2000 year sequence
of flood deposits. Geochemica et Cosmochimicha Acta, 60,
565–576.
Murray, A. S., and Wintle, A. G., 2000. Luminescence dating of
quartz using an improved single-aliquot regenerative-dose pro-
tocol. Radiation Measurements, 32, 57–73.
Murray, A. S., Thomsen, K. J., Masuda, N., Buylaert, J. P., and Jain,
M., 2012. Identifying well-bleached quartz using the different
bleaching rates of quartz and feldspar luminescence signals.
Radiation Measurements, 47, 688–695.
Nissen, E., Walker, R. T., Bayasgalan, A., Carter, A., Fattahi, M.,
Molor, E., Schnabel, C., West, A. J., and Xug, S., 2009. The late
Quaternary slip-rate of the Har-Us-Nuur fault (Mongolian Altai)
from cosmogenic 10Be and luminescence dating. Earth Plane-
tary Science Letters, 286(3–4), 467–478, doi:10.1016/j.
espl.2009.06.048.
Olley, J. M., Caitcheon, G., and Murray, A. S., 1998. The distribu-
tion of apparent dose as determined by optically stimulated lumi-
nescence in small aliquots of fluvial quartz: implications for
dating young sediments. Quaternary Science Reviews, 17,
1033–1040.
Owen, L. A., Finkel, R. C., Haizhou, M., and Barnard, P. L., 2006.
Late Quaternary landscape evolution in the Kunlun Mountains
and Qaidam Basin, Northern Tibet: a framework for examining
the links between lake level changes and alluvial fan formation.
Quaternary International, 154–155, 73–86.
Pawley, S. M., Toms, P., Armitage, S. J., and Rose, J., 2010. Quartz
luminescence dating of Anglian stage (MIS 12) fluvial sedi-
ments; comparison of SAR age estimates to the terrace chronol-
ogy of the Middle Thames valley, UK. Quaternary
Geochronology, 5, 569–582.
Pietsch, T. J., Olley, J. M., and Nanson, G. C., 2008. Fluvial trans-
port as a natural luminescence sensitizer of quartz. Quaternary
Geochronology, 3, 365–376.
469
Porat, N., Amit, R., Enzel, Y., Zilberman, E., Avni, Y., Ginat, H.,
and Gluck, D., 2010. Abandonment ages of alluvial landforms
in the hyperarid Negev determined by luminescence dating.
Journal of Arid Environments, 74, 861–869.
Rittenour, T. M., 2008. Luminescence dating of fluvial deposits:
applications to geomorphic, palaeoseismic, and archaeological
research. Boreas, 37, 613–635.
Rittenour, T. M., Goble, R. J., and Blum, M. D., 2003. An optical
age chronology of Late Pleistocene fluvial deposits in the north-
ern lower Mississippi valley. Quaternary Science Reviews,
22, 1105–1110.
Rittenour, T. M., Goble, R. J., and Blum, M. D., 2005. Development
of an OSL chronology for Late Pleistocene channel belts in the
lower Mississippi valley, USA. Quaternary Science Reviews,
24, 2539–2554.
Rodnight, H., 2008. How many equivalent dose values are needed
to obtain a reproducible distribution? Ancient TL, 26, 3–9.
Rodnight, H., Duller, G. A. T., Wintle, A. G., and Tooth, S., 2006.
Assessing the reproducibility and accuracy of optical dating of
fluvial deposits. Quaternary Geochronology, 1, 109–120.
Spencer, J. Q. G., and Robinson, R. A. J., 2008. Dating intermon-
tane alluvial deposits from NW Argentina using luminescence
techniques: problems and potential. Geomorphology, 93,
144–155.
Spencer, J. Q. G., Sanderson, D. C. W., Deckers, K., and
Sommerville, A. A., 2003. Assessing mixed dose distributions
in young sediments using small aliquots and a simple two-step
SAR procedure: the F-statistic as a diagnostic tool. Radiation
Measurements, 37, 315–321.
Srivastava, P., Brook, G. A., Marais, E., Morthekai, P., and Singhvi,
A. K., 2006. Depositional environment and OSL chronology of
the Homeb silt deposits, Kuiseb River, Namibia. Quaternary
Research, 65, 478–491.
Stokes, S., Bray, H. E., and Blum, M. D., 2001. Optical resetting
in large drainage basins: tests of zeroing assumptions using sin-
gle-aliquot procedures. Quaternary Science Reviews, 20,
879–885.
Summa-Nelson, M. C., and Rittenour, T. M., 2012. Application of
OSL dating to middle to late Holocene arroyo sediments in
Kanab Creek, southern Utah, USA. Quaternary Geochronology,
10, 167–174.
Thomas, P. J., Jain, M., Juyal, N. and Singhvi, A. K., 2005. Compar-
ison of single-grain and small-aliquot OSL dose estimates in
<3000 years old river sediments from South India. Radiation
Measurements, 39, 457–469.
Truelsen, J. L., and Wallinga, J., 2003. Zeroing of the OSL signal as
a function of grain size: investigating bleaching and thermal
transfer for a young fluvial sample. Geochronometria, 22, 1–8.
Wallinga, J., 2002. Optically stimulated luminescence dating in flu-
vial deposits: a review. Boreas, 31, 303–322.
Woda, C., and Fuchs, M., 2008. On the applicability of the leading
edge method to obtain equivalent doses in OSL dating and
dosimetry. Radiation Measurements, 43, 26–37.
Cross-references
Luminescence Dating
Luminescence Dating, Dose Rates
Luminescence, Geomorphological Processes
Luminescence Dating, Single-Grain Dose Distribution
Luminescence Dating, Uncertainties and Age Range
Quartz
Radiocarbon Dating
Terrestrial Cosmogenic Nuclide Dating
470 LUMINESCENCE, GEOMORPHOLOGICAL PROCESSES
LUMINESCENCE, GEOMORPHOLOGICAL
PROCESSES
Stephen Tooth
Department of Geography and Earth Sciences,
Aberystwyth University, Aberystwyth, Ceredigion, UK
Definition
Geomorphology is the science that studies the origin and
development of landforms and how those landforms com-
bine to form landscapes. Landforms are shaped by geo-
morphological processes, many of which involve the
weathering, erosion, transportation, and deposition of sur-
face materials (rock, sediment) by gravity, ice, wind, or
water. Both erosional and depositional landforms can be
identified, and study objectives may include establishing
(1) the initial timing of surface material movement (hence,
landform age), (2) the timings of subsequent surface mate-
rial movement (hence, landform development rate), and
(3) the nature of surface material movement (sediment
dynamics). Using a similar threefold breakdown, the
focus here is on how luminescence dating can contribute
to these objectives. Luminescence dating is a family of
techniques most suitable for investigating fine-grained
(typically silt, sand) depositional landforms. The basic
principles of luminescence dating and the various tech-
niques are explained in other entries (e.g., see “Lumines-
cence Dating”), and the emphasis here is mainly
on optically stimulated luminescence (OSL) dating,
which typically can be applied to quartz-rich sediment
ranging in age from a few years to several hundred
thousand years.
Principal applications of luminescence dating in
geomorphology
Luminescence dating relies on the exposure of transported
sediment to light prior to deposition and burial of that sed-
iment. Consequently, OSL and other luminescence tech-
niques are best suited to the investigation of depositional
landforms originating and/or developing as a result of
one or more of the following processes, all of which
involve varying degrees of sediment exposure to light:
aeolian (e.g., loess sheets, dunes), colluvial (e.g., aprons,
glacis), fluvial (e.g., alluvial fans, river terraces), and wave
and tidal action (e.g., beach ridges, spits). Luminescence
techniques have been less widely applied to the investiga-
tion of glacial, deeper marine, tectonic and volcanic pro-
cesses because of reduced exposure of sediment to light
and/or less suitable (coarser or finer) grain sizes or min-
erals, although techniques are being developed to cater
for these issues (see “Luminescence Dating, Deep-Sea
Marine and Lacustrine”, “Luminescence, Glacial Sedi-
ments”, “Luminescence, Earthquake and Tectonic Activ-
ity”). Numerous recent reviews have outlined the
applications of luminescence dating in specific areas of
geomorphological research (e.g., Duller, 2004; Singhvi
and Porat, 2008; Fuchs and Lang, 2009; Cunningham
and Wallinga, 2012), and the following provides only
a brief outline of three main, possibly interrelated,
applications.
The initial timing of sediment deposition
Commonly, luminescence dating is used to establish the
initial timing of sediment deposition, thereby providing
information on landform age. Examples include OSL dat-
ing of the timing of aeolian deposition in deserts (e.g.,
Hollands et al., 2006); in such instances, the ages for the
basal aeolian sediments provide maximum ages for the
overlying dune landforms (Figure 1a). Similarly, lumines-
cence dating of basal sediment can help establish the max-
imum ages of a variety of other landforms including
colluvial aprons (e.g., Botha et al., 1994), lacustrine beach
ridges (e.g., Nanson et al., 1998), and river floodplains and
terraces (e.g., Keen-Zebert et al., 2013). In addition, dating
of the initial timings of deposition in adjacent landforms
may be used to provide bracketing ages for geomorpho-
logical processes that partly involve erosion; for example,
the timing of river incision can be bracketed by
establishing an age for the alluvial sediments into which
a river has incised, and an age for the sediments subse-
quently deposited within the incision slot (Figure 1b).
The former provides a maximum limiting age for incision
and the latter a minimum limiting age for incision, thereby
bracketing the actual timing of incision (e.g., Keen-Zebert
et al., 2013).
The timings of subsequent sediment deposition
Typically, establishment of the initial timing of deposition
is coupled with the investigation of the timings of subse-
quent deposition, thereby providing information on land-
form development rate. Examples include using multiple
luminescence ages from sediment bodies to establish ver-
tical accretion rates (Figure 1), but sampling strategies
also can be designed to establish dune lateral migration
and/or forward extension rates (Figure 2a), meander bend
growth rates (Figure 2b), or coastal progradation rates
(Figure 2c) (e.g., Bristow et al., 2005; Rodnight et al.,
2005; Hollands et al., 2006; Roberts and Plater, 2007;
Rink and López, 2010; Telfer, 2011). In cases where single
or multiple bracketing ages for river incision can be
established (e.g., across a flight of alluvial terraces), these
can be coupled with the depth of incision to provide
a mean long-term incision rate (Figure 1b). Establishing
OSL ages for relatively unmodified sediments above and
below paleosols can also help to bracket soil formation
rates (Figure 2d), as has been demonstrated in studies of
loess, aeolian dunes, and sand ramps (e.g., Roberts,
2008; Fujioka et al., 2009; Telfer et al., 2012).
The nature of sediment movement
Some recent applications of luminescence techniques
have focused less on the timing and rates of landform
development and more on sediment dynamics. For
instance, OSL dating has been used in combination with
 LUMINESCENCE, GEOMORPHOLOGICAL PROCESSES 471

Luminescence, Geomorphological Processes, Figure 1 Examples of possible strategies for luminescence sample collection and
the insights into geomorphological processes and landform dynamics that may result: (a) Cross-sectional view of aeolian linear dunes
(b) Cross-sectional view of an incised river and flanking alluvial deposits.

hydraulic modelling to provide fundamental insights into
the nature of storage and flushing of fine-grained matrix
sediments in small mountain streams in southeast Austra-
lia (Thompson et al., 2007), and investigations of down-
stream quartz sensitivity (brightness) change in larger,
lower gradient Australian rivers potentially may provide
information on river style and associated fluvial processes,
including bedload transport rate, and water depth and tur-
bidity (Pietsch et al., 2008). In other geomorphological
contexts, a variety of analytical approaches have been
employed to investigate other aspects of local sediment
dynamics, including bioturbation and its implications for
depositional stratigraphy and luminescence chronologies
(Bateman et al., 2007).
Key issues in applications of luminescence dating
in geomorphology
The following outlines some overarching considerations
that arise from applications of luminescence dating in
geomorphology.
Sampling design and age consistency
Careful sampling designs can help to provide an internal
check on luminescence age consistency. For instance,
luminescence ages for the underlying sediments upon
which the basal aeolian or fluvial sediments have been
deposited provide a maximum limiting age for subsequent
development of the dune or floodplain and clearly should
be older than the maximum ages established by dating of
the basal aeolian or fluvial sediments underlying the land-
form of interest (Figure 1). In addition, where several sam-
ples have been collected in vertical, lateral, or longitudinal
sequence within or across landforms, normal stratigraphic
principles apply; the luminescence ages for sediment that
is superimposed upon, laterally adjacent to, or cross-
cutting pre-existing sediment should be younger than the
ages for the preexisting sediment (Figure 2b).
Comparison with independent geochronometers
Confidence in the veracity of luminescence ages also can
be provided by comparisons between different
472 LUMINESCENCE, GEOMORPHOLOGICAL PROCESSES

Luminescence, Geomorphological Processes, Figure 2 Examples of possible strategies for luminescence sample collection where
the focus is on establishing landform development rates: (a) Cross-sectional and plan view of a laterally migrating and longitudinally
extending aeolian linear dune (b) Cross-sectional and plan view of a laterally migrating meander, with scroll bars formed on the
inside bend. OSL ages for scroll bars and the oxbow lake infill are derived from Rodnight et al. (2005) and Tooth et al. (2009) (c) Plan
view of a prograding coastal spit and back barrier complex (d) Side view of a vertically accumulated loess-paleosol succession.

luminescence techniques and by comparison with
other geochronometers. For Australian fluvial sediments,
comparisons have been made between OSL and TL
ages (see “Luminescence Dating, History”, for TL); in
these instances, the ages commonly overlap within error
and have been used to argue for the veracity of TL ages
in those settings (Nanson et al., 2005; Maroulis et al.,
2007). More commonly, luminescence ages have been
compared with chronologies obtained using different
techniques on the same sediment layers or from
underlying or overlying sediment layers (e.g., Madsen
et al., 2005; Rodnight et al., 2006). For instance, in
the Klip River wetlands, South Africa, OSL dating was
used to establish the age of sandy paleochannel sedi-
ments, while 14C dating (see “Radiocarbon Dating”)
was used to establish the age of post-abandonment,
organic-rich infills (Rodnight et al., 2006). As expected,
the 14C ages tended to be younger than the OSL
ages (Figure 3), thus providing support for the
OSL analyses.
 LUMINESCENCE, GEOMORPHOLOGICAL PROCESSES
Luminescence, Geomorphological Processes, Figure 3 Schematic comparison between OSL ages for bedload sand in
paleochannels of the Klip River, South Africa, and the generally younger 14C ages for overlying organic-rich sandy muds (modified
from Rodnight et al. 2006).
Comparison with other paleoenvironmental
datasets
Luminescence dating can establish the timing, rates, and
nature of sediment movement, but a common aim of geo-
morphological enquiry is to identify the drivers of change.
Drivers of change may be intrinsic to the system dynam-
ics; for example, vertical accretion, lateral migration,
and/or forward extension of aeolian dunes, meander
bends, or coastal spits (Figure 2a–c) can occur under con-
ditions of constant sediment supply and steady wind
regime, discharge, or wave energy. Commonly, however,
drivers of change are extrinsic to the system and result
from wider paleoenvironmental changes, including those
related to climatic, sea level, tectonic, volcanic, or anthro-
pogenic perturbations. Where paleoenvironmental
changes lead to marked increases in the magnitude and
frequency of river flooding, for instance, this may lead to
increased rates of floodplain vertical accumulation or
meander bend migration. These changed rates may be
established by OSL dating, but inferences regarding the
drivers of change require comparison with independently
dated paleoenvironmental datasets. Uranium-thorium-
dated (see “U-Series Dating”) or 14C-dated lacustrine,
cave, or marine deposits, for instance, are commonly used
to derive proxy records of paleoclimate change (e.g., past
precipitation, temperature, and/or vegetation, or some
combination thereof, such as indices of moisture availabil-
ity). A lack of close correspondence between the timing
and rates of landform changes established by lumines-
cence dating and the changes in these proxy records might
suggest that the dynamics have been driven independently
of paleoclimate, such as by intrinsic system dynamics or
by some other paleoenvironmental factor (e.g., past tec-
tonics). Where close correspondence exists, however, this
provides a basis for paleoclimatic interpretation of the
landform changes, and this line of reasoning has been
widely adopted in geomorphological enquiries.
Limitations of luminescence dating in
geomorphology
Although the comparison of luminescence chronologies
with other paleoenvironmental datasets is common in
473
geomorphological enquiry, potential problems may
impinge on this objective. Two common problems are
outlined below.
An inability to establish finite luminescence ages
Where the sediment being dated has a depositional age
that approaches or exceeds the age range of the lumines-
cence technique, then saturation of the luminescence sig-
nal may occur (see “Luminescence Dating, Uncertainties
and Age Range”). In such instances, it may not be possible
to determine a finite age, and only a minimum age for
deposition can be reported (e.g., >XX years). Although
this may provide useful information – for instance, in
confirming that basal sediments have minimum ages older
than finite ages for overlying sediments – it does not allow
calculation of finite vertical accumulation, lateral migra-
tion, or forward extension rates and precludes detailed
comparison with other paleoenvironmental datasets.
Limited precision of luminescence ages
Luminescence ages are reported with errors that reflect ran-
dom and systematic uncertainties; for OSL ages, errors at
one sigma (68 % confidence interval) are typically around
5–10 % of the sample age (e.g., see Figures 2b and 3).
Hence, although the central age is the most likely, there is
actually a 68 % chance that the sample age lies within the
reported error range. In some instances, this means that
OSL ages for adjacent landforms will overlap within error,
causing uncertainty when calculating rates of change. For
instance, across a scroll bar sequence in the Klip River
floodplain, South Africa, most OSL ages were in the correct
order and did not overlap, thereby indicating an approxi-
mately steady lateral migration rate of ~0.16 m/year over
the last ~1 kyr. Two ages in the middle of the sequence, how-
ever, overlapped within error (Figure 2b – samples 3 and 4).
In such situations, it may be unclear whether the overlapping
ages reflect increased analytical uncertainty associated with
particular ages or whether they reflect a short-term increase
in the lateral migration rate, although the latter explanation
was preferred in the Klip River case (see Rodnight et al.,
2005). The limited precision of OSL ages also limits
the potential for comparison with high-resolution
474 LUMINESCENCE, GEOMORPHOLOGICAL PROCESSES

Luminescence, Geomorphological Processes, Figure 4 Hypothetical but realistic illustration of the problems in attempting to
correlate luminescence ages (in this instance, from fluvial samples) and associated error bars with the paleoenvironmental changes
(paleomoisture index) recorded in higher-resolution datasets for the last 50 kyr. Errors associated with OSL ages generally increase
farther back in time and so may encompass one or more peaks and troughs in the paleomoisture index. Consequently, it may not be
clear whether periods of fluvial activity identified by the OSL ages have been driven primarily by relatively wet, relatively dry, or
transitional (wet-dry, dry-wet) conditions.

paleoenvironmental datasets derived from, say, lacustrine,
cave, or marine sediments; errors on OSL ages may overlap
with identified fluctuations in past precipitation, tempera-
ture, vegetation, or moisture availability (see Figure 4), even
those occurring over multimillennial scales, such that
paleoenvironmental drivers are hard to establish. In the case
of alluvial fans in Death Valley, California, USA, Dorn
(1996) has argued that imprecision in many
geochronometers means that climatic hypotheses of fan for-
mation are not testable; with some caveats, this argument
could be extended to OSL analysis of other landforms.
Commonly, however, the lack of precision is acknowledged,
and correlation with paleoenvironmental datasets is made
using the central ages (e.g., Macklin et al., 2010).
Awareness of these limitations is important when
considering the applicability of luminescence dating in
geomorphological enquiry, and some may be partially
overcome by future developments. For instance, ongoing
analytical work using thermally transferred OSL (TT-
OSL) is attempting to extend the luminescence dating
age range (Rhodes, 2011); in future, this may enable com-
parisons with longer paleoenvironmental datasets, possi-
bly including those extending back ~500 kyr or farther,
although any finite ages are still likely to be associated
with large errors. In other instances, the use of lumines-
cence dating in combination with other geochronometers
may provide the best approach. In central Australia, for
instance, relatively young luminescence saturation ages
(typically <100 ka) mean that OSL cannot be used to
determine the age of aeolian linear dunes that may be
many hundreds of thousands or even millions of years
old. Hence, Fujioka et al. (2009) combined cosmogenic
burial dating (see “Terrestrial Cosmogenic Nuclide
Dating”) and OSL dating to investigate histories of dune
building; cosmogenic (10Be–26Al) burial dating of the
basal sediments helped to determine that the first dunes
formed in the western Simpson Desert ~1 Ma, while dat-
ing of younger, overlying sediments using OSL indicated
that episodic dune reworking and building occurred dur-
ing late Quaternary glacial intervals. Similarly innovative
sampling designs that employ complementary geochrono-
logical techniques may be a hallmark of future geomor-
phological enquiries.
Conclusions
Over the last few decades, luminescence dating has contrib-
uted to a revolution in our understanding of the timing, rates
and nature of many geomorphological processes, and their
implications for landform and landscape development.
Despite some limitations, this increased understanding has
occurred in parallel with increasing insight into the
paleoenvironmental drivers of change. Further innovations
in luminescence sampling approaches (e.g., adoption of
“range-finder” OSL dating – Durcan et al., 2010) and ongo-
ing analytical advances (e.g., wider application of single
grain OSL dating, the increasing use of feldspars as chro-
nometers, and the development of TT-OSL – Rhodes,
2011) mean that luminescence dating is likely to remain
an essential part of the geomorphologist’s “toolkit” in the
decades to come. This will provide fertile ground
for interdisciplinary work in scientific and applied contexts,
such as in archaeology, paleoanthropology, paleoseismicity,
and environmental management (e.g., Rittenour, 2008;
McCarthy et al., 2010; Barré et al., 2012; see also
“Luminescence Dating of Archaeological Sediments”).
Bibliography
Barré, M., Lamothe, M., Backwell, L., and McCarthy, T. S., 2012.
Optical dating of quartz and feldspars: a comparative study from
Wonderkrater, a Middle Stone Age site of South Africa. Quater-
nary Geochronology, 10, 374–379.
Bateman, M. D., Boulter, C. H., Carr, A. S., Frederick, C. D., Peter,
D., and Wilder, M., 2007. Preserving the palaeoenvironmental
record in drylands: bioturbation and its significance for lumines-
cence-derived chronologies. Sedimentary Geology, 195, 5–19.
Botha, G. A., Wintle, A. G., and Vogel, J. C., 1994. Episodic late
Quaternary gully erosion in northern KwaZulu-Natal, South
Africa. Catena, 23, 327–340.
Bristow, C. S., Lancaster, N., and Duller, G. A. T., 2005. Combining
ground penetrating radar surveys and optical dating to determine
 LUMINESCENCE, GLACIAL SEDIMENTS
dune migration in Namibia. Journal of the Geological Society
(London), 162, 315–321.
Cunningham, A. C., and Wallinga, J., 2012. Realizing the potential
of fluvial archives using robust OSL chronologies. Quaternary
Geochronology, 12, 98–106.
Dorn, R. I., 1996. Climatic hypotheses of alluvial-fan formation in
Death Valley are not testable. In Rhoads, B. L., and Thorn,
C. E. (eds.), The Scientific Nature of Geomorphology: Proceed-
ings of the 27th Binghamton Symposium in Geomorphology held
27–29 September 1996. Chichester: Wiley, pp. 191–220.
Duller, G. A. T., 2004. Luminescence dating of Quaternary sedi-
ments: recent advances. Journal of Quaternary Science,
19, 183–192.
Durcan, J. A., Roberts, H. M., Duller, G. A. T., and Alizai, A. H.,
2010. Testing the use of range-finder OSL dating to inform field
sampling and laboratory processing strategies. Quaternary Geo-
chronology, 5, 86–90.
Fuchs, M., and Lang, A., 2009. Luminescence dating of hillslope
deposits – a review. Geomorphology, 109, 17–26.
Fujioka, T., Chappell, J., Fifield, L. K., and Rhodes, E. J., 2009.
Australian desert dune fields initiated with Pliocene-Pleistocene
global climatic shift. Geology, 37, 51–54.
Hollands, C. B., Nanson, G. C., Jones, B. G., Bristow, C. S., Price,
D. M., and Pietsch, T. J., 2006. Aeolian-fluvial interaction: evi-
dence for late Quaternary channel change and wind-rift linear
dune formation in the northwestern Simpson Desert, Australia.
Quaternary Science Reviews, 25, 142–162.
Keen-Zebert, A., Tooth, S., Rodnight, H., Duller, G. A. T., Roberts,
H. M., and Grenfell, M., 2013. Late Quaternary floodplain
reworking and the preservation of alluvial sedimentary archives
in unconfined and confined valleys in the eastern interior of
South Africa. Geomorphology, 185, 54–66.
Macklin, M. G., Tooth, S., Brewer, P. A., Noble, P. L., and Duller,
G. A. T., 2010. Holocene flooding and river development in
a Mediterranean steepland catchment: the Anapodaris Gorge,
south-central Crete, Greece. Global and Planetary Change,
70, 35–52.
Madsen, A. T., Murray, A. S., Andersen, T. J., Pejrup, M., and
Breuning-Madsen, H., 2005. Optically stimulated luminescence
dating of young estuarine sediments: a comparison with 210Pb
and 137Cs dating. Marine Geology, 214, 251–268.
Maroulis, J. C., Nanson, G. C., Price, D. M., and Pietsch, T., 2007.
Aeolian-fluvial interaction and climate change: source-bordering
dune development over the past 100 ka on Cooper Creek, central
Australia. Quaternary Science Reviews, 26, 386–404.
McCarthy, T. S., Tooth, S., Kotze, D. C., Collins, N., Wandrag, G.,
and Pike, T., 2010. The role of geomorphology in evaluating
remediation options for floodplain wetlands: the case of
Ramsar-listed Seekoeivlei, eastern South Africa. Wetlands Ecol-
ogy and Management, 18, 119–134.
Nanson, G. C., Callen, R. A., and Price, D. M., 1998. Hydroclimatic
interpretation of Quaternary shorelines on South Australian
playas. Palaeogeography, Palaeoclimatology, Palaeoecology,
144, 281–305.
Nanson, G. C., Jones, B. G., Price, D. M., and Pietsch, T. J., 2005.
Rivers turned to rock: late Quaternary alluvial induration
influencing the behaviour and morphology of an anabranching
river in the Australian monsoon tropics. Geomorphology,
70, 398–420.
Pietsch, T. J., Olley, J. M., and Nanson, G. C., 2008. Fluvial trans-
port as a natural luminescence sensitiser of quartz. Quaternary
Geochronology, 3, 365–376.
Rhodes, E. J., 2011. Optically stimulated luminescence dating of
sediments over the past 200,000 years. Annual Review of Earth
and Planetary Sciences, 39, 461–488.
Rink, W. J., and López, G. I., 2010. OSL-based lateral progradation
and aeolian sediment accumulation rates for the Apalachicola
475
Barrier Island Complex, North Gulf of Mexico, Florida.
Geomorphology, 123, 330–342.
Rittenour, T., 2008. Luminescence dating of fluvial deposits: appli-
cations to geomorphic, palaeoseismic and archaeological
research. Boreas, 37, 613–635.
Roberts, H. M., 2008. The development and application of lumines-
cence dating to loess deposits: a perspective on the past, present
and future. Boreas, 37, 483–507.
Roberts, H. M., and Plater, A. J., 2007. Reconstruction of Holocene
foreland progradation using optically stimulated luminescence
(OSL) dating: an example from Dungeness, UK. The Holocene,
17, 495–505.
Rodnight, H., Duller, G. A. T., Tooth, S., and Wintle, A. G., 2005.
Optical dating of a scroll-bar sequence on the Klip River, South
Africa, to derive the lateral migration rate of a meander bend. The
Holocene, 15, 802–811.
Rodnight, H., Duller, G. A. T., Wintle, A. G., and Tooth, S., 2006.
Assessing the reproducibility and accuracy of optical dating of
fluvial deposits. Quaternary Geochronology, 1, 109–120.
Singhvi, A. K., and Porat, N., 2008. Impact of luminescence dating
on geomorphological and palaeoclimate research in drylands.
Boreas, 37, 536–558.
Telfer, M. W., 2011. Growth by extension, and reworking, of
a south-western Kalahari linear dune. Earth Surface Processes
and Landforms, 36, 1125–1135.
Telfer, M. W., Thomas, Z. A., and Breman, E., 2012. Sand ramps in
the Golden Gate Highlands National Park, South Africa: evidence
of periglacial aeolian activity during the last glacial.
Palaeogeography, Palaeoclimatology, Palaeoecology, 313–314,
59–69.
Thompson, C., Rhodes, E., and Croke, J., 2007. The storage of bed
material in mountain stream channels as assessed using Optically
Stimulated Luminescence dating. Geomorphology, 83, 307–321.
Tooth, S., Rodnight, H., McCarthy, T. S., Duller, G. A. T., and
Grundling, A., 2009. Late Quaternary dynamics of a South Afri-
can floodplain wetland and the implications for assessing recent
human impacts. Geomorphology, 106, 278–291.
Cross-references
Luminescence Dating of Archaeological Sediments
Luminescence Dating
Luminescence Dating, History
Luminescence Dating, Deep-Sea Marine and Lacustrine
Luminescence, Earthquake and Tectonic Activity
Luminescence, Glacial Sediments
Luminescence Dating, Uncertainties and Age Range
Radiocarbon Dating
Terrestrial Cosmogenic Nuclide Dating
U-Series Dating
LUMINESCENCE, GLACIAL SEDIMENTS
Geoffrey A.T. Duller
Department of Geography and Earth Sciences,
Aberystwyth University, Aberystwyth, Wales, UK
Definition
Glacial. Of, or pertaining to, ice masses of sufficient
magnitude that they are able to deform under their
own mass.
476 LUMINESCENCE, GLACIAL SEDIMENTS

Glacial sediments. Detrital lithic materials deposited as a of work was undertaken to try to define the extent to which
result either of direct glacial movement, such as a diamict, these TL signals were reset. Gemmell (1988) studied the
or an indirect consequence of glacial processes, such as change in the TL signal of suspended sediment collected
fluvioglacial sediments. along a 4 km long reach of a channel draining the base
Luminescence dating. The use of the luminescence signal of the Austerdalsbreen glacier in Norway. Although he
emitted from a mineral for dating. For geological sedi- could observe that the TL signal decreased with distance
ments, the event being dated is the last exposure to day- downstream of the glacier, this decrease was irregular
light, and this normally corresponds with deposition of because of the incorporation of older material from the
the sediment. banks of the channel. Forman (1988) chose to study a
range of different depositional environments around gla-
Introduction ciers in Spitsbergen and found that tills and ice-proximal
glaciomarine sediments appeared to have been exposed
Luminescence dating of sediments associated with former
to light the least, while littoral and ice-distal marine sedi-
glacial activity has been undertaken for over 40 years.
ments had been exposed for the longest. These studies of
Some of the earliest work undertaken on the application
proximal glacial sediments implied that it was not possible
of luminescence to dating geological sediments was car-
to obtain accurate luminescence ages using TL methods.
ried out in the former Soviet Union and included analysis
Optically stimulated methods, both using visible
of sediments resulting from a variety of different glacial
(optically stimulated luminescence, OSL) and infrared
processes. This work was not widely known outside the
stimulation (infrared stimulated luminescence, IRSL),
Soviet Union (see “Luminescence Dating, History”), and
were introduced in the 1980s and 1990s. They measure
in later reviews (e.g., Dreimanis et al., 1978; Wintle and
signals that are reset more rapidly by exposure to daylight
Huntley, 1982), it was criticized for some important errors.
than TL and thus may be more suitable for dating glacial
A major challenge that has been common to all lumi-
sediments where exposure to daylight at deposition is lim-
nescence studies of glacially derived sediments is
ited. A second important innovation at this time was the
assessing whether they were exposed to sufficient light
development of single aliquot methods which made it pos-
at deposition to reset the signal used for dating. Early work
sible to measure equivalent dose (DE) using a single sub-
used thermoluminescence (TL) signals, and a great deal of
sample (called an aliquot) and made it practical to make
effort was expended trying to assess what proportion of
replicate measurements. Duller et al. (1995) was one of
the TL signal remained at deposition so that ages could
the first to use single aliquot IRSL measurements in gla-
be calculated. Optically stimulated luminescence signals
cial environments and analyzed a variety of glacially
(OSL) tend to be reset by exposure to daylight more rap-
derived sediments from Scotland. Some of these sites
idly than TL, and when these methods became available,
had independent age control, and these showed that while
they were rapidly adopted for analysis of glacial
some ages were consistent with the other geochronologi-
sediments.
cal methods, others were clearly inconsistent. This was
There are many different types of glacially derived sed-
progress over the use of TL methods, but major hurdles
iments, and the likelihood of exposure to daylight prior to
still existed. The methods available for analysis of
deposition varies considerably. It is clear that selection of
the DE estimates was crude at that stage, and the
depositional facies where exposure to daylight at deposi-
measurements of IRSL made at that time were prone to
tion is most likely is important. The most suitable environ-
anomalous fading (see “Feldspar, Infrared-Stimulated
ments are probably those involving fluvioglacial or
Luminescence”).
fluviolacustrine processes, though in both cases shallow
Development of the single aliquot regenerative dose
water facies are much preferable to deepwater ones.
(SAR) method for OSL measurement of quartz, and
For some glacial sediments, OSL methods can be
improved understanding of the causes of variations in DE
applied in a relatively straightforward manner, while in
between different aliquots, led to renewed efforts to apply
environments where the exposure to daylight is more lim-
luminescence to glacial sediments. Of particular impor-
ited, complex methods involving measurement of single
tance was the development of statistical models such as
grains are required in order to be able to identify the small
the minimum age model (see “Luminescence Dating,
proportion of grains within the sample that did receive suf-
Single-Grain Dose Distribution”) which can be used to
ficient daylight exposure to reset their signal and hence
identify within a set of replicate DE measurements the
give an accurate age. Recent reviews of the application
population of aliquots which were best bleached at depo-
of OSL methods to glacially related sediments are pro-
sition. The development of instruments that were able to
vided by Fuchs and Owen (2008) and Thrasher
make measurements of the OSL signal from single grains
et al. (2009).
(see “Luminescence Dating”) was also important for some
studies as changing the number of grains in an aliquot can
History be valuable in ensuring that the best bleached grains are
The limited exposure of many types of glacially derived seen (Olley et al., 1999). However, there is still debate
sediments to daylight was realized in early studies that within the academic community about whether measure-
studied the thermoluminescence (TL) signals. A great deal ments of single grains are more appropriate than using
 LUMINESCENCE, GLACIAL SEDIMENTS
“small aliquots” where perhaps 10–30 grains are placed
on each aliquot (Duller, 2008). A common challenge when
looking at the OSL signal from quartz isolated from gla-
cial sediments is the low signal intensity, and this can
make single grain measurements difficult.
The application of luminescence dating to glacial sedi-
ments can usefully be divided into those applications that
have studied mountain glaciers and those that have studied
sediments derived from large continental-scale ice sheets
associated with Pleistocene expansions of ice cover.
Sediments from mountain glaciers
In the context of luminescence dating, the critical aspects
of glaciers in mountainous terrains are that the transport
distance between erosion and deposition of sedimentary
material tends to be short, and the source geology is often
dominated by intrusive igneous rocks. The impact of these
characteristics is to give quartz whose OSL sensitivity is
commonly very low. This means that the OSL signal emit-
ted per unit of radiation (Gy) tends to be low and is often
near the detection limit of available instruments. This
has been a major challenge in areas as diverse as the
Himalaya, Scotland, Chile, the Alps in Europe, and the
Southern Alps in New Zealand (Preusser et al., 2006).
Pietsch et al. (2008) measured the OSL sensitivity of
quartz along a river course and showed that it increased
downstream, presumably due to repeated cycles of erosion
and burial. Such repeated transport cycles are unlikely in
mountain glaciers.
These characteristics obviously do not apply to all
mountain glaciers, but they are commonly observed. In
some studies the quartz has had such low OSL sensitivity
that it has not been possible to obtain ages (e.g., Lukas
et al., 2007). In other areas it has been possible to obtain
ages that are both geomorphologically consistent and
which agree with Cosmogenic Nuclide Dating (Glasser
et al., 2006). Many studies make measurements of aliquots
containing many grains because of the low OSL sensitiv-
ity (e.g., Spencer and Owen, 2004). Single grain measure-
ments may be possible, though only a very limited
proportion of grains may give sufficient OSL signal to
be useful (~0.5 % of the grains in the samples from Chile
studied by Duller (2006)).
Sediments from continental-scale ice sheets
In contrast to mountain glaciers, large continental-scale
ice sheets typically have greater opportunities to incorpo-
rate material from a variety of geological sources, and
detrital materials travel longer distances. The spatial scale
of the sedimentary deposits also tends to be greater, and
this may provide more opportunities for exposure to day-
light. This is especially likely for glaciofluvial sediments
and glaciolacustrine sediments. One of the largest bodies
of work undertaken using luminescence dating was for
the QUEEN project (Quaternary Environment of the Eur-
asian North; 1996–2002), where over 600 OSL ages were
measured on quartz extracted from a wide range of
477
sediments associated with former glaciation of the Russian
arctic (Svendsen et al., 2004). Measurements on these
samples generally showed that the OSL signal from quartz
had been effectively reset at deposition (Thomas
et al., 2006) presumably because of the long transport dis-
tances in such a large ice sheet. Furthermore, low OSL
sensitivity is not reported to be a problem in this extensive
study.
Conclusions
Luminescence dating of sediments associated with glacial
activity is possible, provided that care is taken when sam-
pling to choose facies with the greatest likelihood of expo-
sure to daylight. Luminescence is often the only
geochronological method that can be applied to glacial
sediments. The quartz OSL signal is generally the most
appropriate signal for dating because it is reset by expo-
sure to daylight so rapidly. However, in some areas the
quartz OSL signal is very dim; this may be a challenge,
and in extreme cases it may make it impossible to obtain
an age. Replicate measurements of DE, whether on small
aliquots or single grains, are essential when studying gla-
cially derived sediments because it provides a means of
assessing whether the luminescence signal from the sam-
ple was effectively set to zero at deposition or not. Where
there is evidence that not all the grains in a sample were
reset at deposition, statistical methods such as the mini-
mum age model (see “Luminescence Dating, Single-Grain
Dose Distribution”) may be used to identify the population
of grains that were bleached and hence enable accurate
dating.
Bibliography
Dreimanis, A., Hutt, G., Raukas, A., and Whippey, P. W., 1978. Dat-
ing methods of Pleistocene deposits and their problems. 1. Ther-
moluminescence dating. Geoscience Canada, 5, 55–60.
Duller, G. A. T., 2006. Single grain optical dating of glacigenic sed-
iments. Quaternary Geochronology, 1, 296–304.
Duller, G. A. T., 2008. Single grain optical dating of Quaternary
sediments: why aliquot size matters in luminescence dating.
Boreas, 37, 589–612.
Duller, G. A. T., Wintle, A. G., and Hall, A. M., 1995. Lumines-
cence dating and its application to key pre-Late Devensian sites
in Scotland. Quaternary Science Reviews, 14, 495–519.
Forman, S. L., 1988. The solar resetting of thermoluminescence of
sediments in a glacier-dominated fiord environment in Spitsber-
gen; geochronologic implications. Arctic and Alpine Research,
20, 243–253.
Fuchs, M., and Owen, L. A., 2008. Luminescence dating of glacial
and associated sediments: review, recommendations and future
directions. Boreas, 37, 636–659.
Gemmell, A. M. D., 1988. Zeroing of the TL signal in sediment
undergoing glaciofluvial transport. An example from
Austerdalen, western Norway. Quaternary Science Reviews, 7,
339–345.
Glasser, N. F., Harrison, S., Ivy-Ochs, S., Duller, G. A. T., and
Kubik, P. W., 2006. Evidence from the Rio Bayo valley on the
extent of the North Patagonian Icefield during the Late
Pleistocene-Holocene transition. Quaternary Research, 65,
70–77.
478 LUMINESCENCE, MARTIAN SEDIMENTS
Lukas, S., Spencer, J. Q. G., Robinson, R. A. J., and Benn, D. I.,
2007. Problems associated with luminescence dating of Late
Quaternary glacial sediments in the NW Scottish Highlands.
Quaternary Geochronology, 2, 243–248.
Olley, J. M., Caitcheon, G. G., and Roberts, R. G., 1999. The origin
of dose distributions in fluvial sediments, and the prospect of
dating single grains from fluvial deposits using optically stimu-
lated luminescence. Radiation Measurements, 30, 207–217.
Pietsch, T., Olley, J. M., and Nanson, G. C., 2008. Fluvial transport
as a natural luminescence sensitiser of quartz. Quaternary Geo-
chronology, 3, 365–376.
Preusser, F., Ramseyer, K., and Schluchter, C., 2006. Characterisa-
tion of low OSL intensity quartz from the New Zealand Alps.
Radiation Measurements, 41, 871–877.
Spencer, J. Q., and Owen, L. A., 2004. Optically stimulated lumi-
nescence dating of Late Quaternary glaciogenic sediments in
the upper Hunza valley: validating the timing of glaciation and
assessing dating methods. Quaternary Science Reviews, 23,
175–191.
Svendsen, J. I., Alexanderson, H., Astakhov, V. I., Demidov, I.,
Dowdeswell, J. A., Funder, S., Gataullin, V., Henriksen, M.,
Hjort, C., Houmark-Nielsen, M., Hubberten, H. W., Ingolfsson,
O., Jakobsson, M., Kjaer, K. H., Larsen, E., Lokrantz, H.,
Lunkka, J. P., Lysa, A., Mangerud, J., Matiouchkov, A., Murray,
A., Moller, P., Niessen, F., Nikolskaya, O., Polyak, L., Saarnisto,
M., Siegert, C., Siegert, M. J., Spielhagen, R. F., and Stein, R.,
2004. Late Quaternary ice sheet history of northern Eurasia.
Quaternary Science Reviews, 23, 1229–1271.
Thomas, P. J., Murray, A. S., Kjaer, K. H., Funder, S., and Larsen,
E., 2006. Optically stimulated luminescence (OSL) dating of
glacial sediments from Arctic Russia – depositional bleaching
and methodological aspects. Boreas, 35, 587–599.
Thrasher, I. M., Mauz, B., Chiverrell, R. C., and Lang, A., 2009.
Luminescence dating of glaciofluvial deposits: a review. Earth-
Science Reviews, 97, 133–146.
Wintle, A. G., and Huntley, D. J., 1982. Thermoluminescence
dating of sediments. Quaternary Science Reviews, 1, 31–53.
Cross-references
Feldspar, Infrared-Stimulated Luminescence
Luminescence Dating
Luminescence Dating, History
Luminescence Dating, Single-Grain Dose Distribution
Terrestrial Cosmogenic Nuclide Dating
LUMINESCENCE, MARTIAN SEDIMENTS
Regina DeWitt
Department of Physics, East Carolina University,
Greenville, NC, USA
Synonyms
Martian regolith
Introduction
Mars is the most Earthlike planet of the solar system and
the planet most likely to have hosted life. The Martian sur-
face exhibits abundant evidence of having undergone
a complex geologic and climatic evolution. Aeolian, flu-
vial, periglacial, and volcanic processes appear to have
remained active through to the present day, and evidence
even points to active global climate change (Carr, 2006).
Luminescence dating has been suggested as one method
to provide a chronology for Martian surface regolith, and
efforts are underway to develop a robotic instrument for
an in situ mission. Differences between terrestrial and
Martian environments influence the dating methodology
and the instrumentation required.
Mineral composition
Unlike with terrestrial applications, an in situ lumines-
cence dating instrument for Martian regolith samples will
likely not allow chemical treatment and mineral separa-
tion. Work with polymineralic samples is a prerequisite.
Data obtained by the thermal emission spectrometer dis-
tinguish two different types of regolith on Mars, namely,
basaltic and andesitic, which are composed of plagioclase
feldspars, pyroxenes, and hematite (Bandfield et al., 2000;
Bandfield, 2002). Plagioclase feldspars are found to have
a calcium content of 30–70 % (Milam et al., 2004). Obser-
vations by the Spirit and Opportunity rovers further
revealed a wide variety of minerals (Chevrier and Mathe,
2007), some of which are believed to have formed under
the influence of water. Luminescence studies used
a variety of Martian simulants to approximate the type of
polymineralic samples that may be found on the Martian
surface, including the Martian soil simulant JSC Mars-1
(Lepper and McKeever, 2000; Banerjee et al., 2002), var-
ious minerals and mineral mixtures (Kalchgruber et al.,
2006; Jain et al., 2006; Blair et al., 2007; DeWitt and
McKeever, 2011; O’Connor et al., 2011), or basalt
(Tsukamoto et al., 2011).
Measurement procedures and fading
In situ luminescence dating procedures for Martian rego-
lith are limited by the polymineralic nature of the samples
and the small amounts of sample material that can be mea-
sured under the given power constraints. Kalchgruber
et al. (2006) and Blair et al. (2007) found the “single-
aliquot regenerative-dose” (SAR) technique suitable for
polymineralic samples. DeWitt and McKeever (2011)
using mineral mixtures found that over 80 % of the blue-
stimulated OSL and almost 100 % of the IR-stimulated
OSL signals come from the feldspar components, so that
anomalous fading is of concern. Fading can be minimized
by using thermoluminescence, IR stimulation and detec-
tion in the blue wavelength range, detection of red emis-
sions, pulsed stimulation, or IR stimulation at elevated
temperatures (Jain et al., 2006; Tsukamoto and Duller,
2008; Morthekai et al., 2008).
Solar resetting
A basic premise of OSL dating is that the sediments to be
dated have been exposed to sufficient amounts of light at
the time of deposition to erase any previously accumulated
signal. In terrestrial applications, this “zeroing” of the sig-
nal is accomplished within a few minutes of exposure to
 LUMINESCENCE, MARTIAN SEDIMENTS
sunlight (Aitken, 1998). Modeling of radiation transport in
the Martian atmosphere suggested a more intense UV
component at the Martian surface than is found on Earth
(McKeever et al., 2006), and this may lead to different
bleaching efficiencies. Carrying out experiments with
a Mars solar simulator and detecting the UV lumines-
cence, Kalchgruber et al. (2007) found efficient bleaching
of the IR-stimulated signal from mineral mixtures after
exposure for 1 h. When blue stimulation was used, the
authors observed a non-bleachable residual signal, which
was approximately 5 % of the original signal. Detschel
and Lepper (2009) tested the influence of UV
and a simulated Martian spectrum on the luminescence
of various samples. They found that Na-feldspar (albite)
tends to acquire an OSL signal with increasing UV expo-
sure and might pose a challenge to optical dating.
Ca- and K-feldspars did not acquire a significant signal
under UV exposure and could be bleached when exposed
to the simulated Martian spectrum. Both authors conclude
that luminescence dating of Martian regolith should not be
compromised by the UV-enhanced solar spectrum.
Temperature dependence
On Earth, most minerals that are dated by OSL are found
in conditions where the temperatures vary annually over
a maximum range of perhaps 50 K for the entirety of their
storage period. On Mars, the average ambient temperature
is much lower with much larger diurnal and annual varia-
tions. Temperatures can fluctuate by 80 K within 1 day
(Kieffer, 1976; Hansson, 1997). The global variation is
reported to be from 140 to 300 K (Kieffer et al., 1992).
Low-temperature OSL measurements and computer simu-
lations indicate that known doses delivered at low temper-
atures can be effectively estimated if the stimulation
temperature is greater than the highest temperature during
irradiation (Blair et al., 2006; McKeever et al., 2010). An
in situ OSL instrument must therefore provide a heating
option to temperatures at least as high as 380 K.
Radiation environment
Another challenge is the radiation environment encoun-
tered at the surface and in the regolith of Mars. Radioac-
tive nuclides (U, Th, and K) play only a minor role in
the near-surface deposits on Mars, and available data from
the Mars Odyssey Gamma-Ray Spectrometer (Boynton
et al., 2004; Wänke et al., 2005) will be sufficient to deter-
mine the resulting dose rate background. The primary
radiation field on Mars is from galactic cosmic rays
(GCR) and solar energetic particles (SEP). The GCR spec-
trum consists of approximately 85 % protons, 12 % alpha
particles, 1 % heavier nuclei, and about 2 % electrons and
positrons. The energies range from several tens up to
1012 MeV with a broad peak between 0.1 and 1 GeV. Solar
energetic particles include electrons, protons, and heavier
ions, and ~50 events can be expected in the 11-year solar
cycle (for a detailed description, see, e.g., Benton and
479
Benton, 2001). This is a significant difference compared
to Earth.
The efficiency of OSL signal production varies with the
type of radiation. Specifically, low-energy high-Z parti-
cles have a high ionization power or large linear energy
transfer (LET). Such particles are not as efficient in pro-
ducing OSL signals. The high ionization density in the
track of the particle as it traverses through the material
leads to local saturation of the signal (Jain et al., 2006,
2007; Kalchgruber et al., 2007). The composition of the
radiation and, therefore, LET change with depth due to
scattering and absorption of the particles.
In terrestrial applications, 90Sr beta sources are com-
monly used for determination of the dose response.
Corrections are applied to allow for the smaller lumines-
cence efficiency of alpha radiation compared to beta radia-
tion. Similar corrections will be necessary when the
luminescence produced by energetic particles in the cosmic
radiation is compared with the calibration radiation source
onboard the robotic instrument. However, cosmic ray spec-
tra are dominated by the lower LET parts of the spectrum
such that the absorbed dose is primarily (95 %) from
high-energy protons and He ions. For such low-LET irradi-
ations, the luminescence efficiency is ~1, with respect to
60
Co or 90Sr radiation, introducing only a small error.
Dose rate
Dose rates on the Martian surface depend upon altitude.
Numbers quoted by authors vary but are of the order
~80–120 mGy/year (Saganti et al., 2004) and
43–51 mGy/year (McKeever et al., 2003; Banerjee and
Dewangan, 2008). The intensity and composition of the
cosmic radiation field will change with increasing depth
in the regolith due to absorption and scattering. At
a depth of 25 g/cm2, the dose rate is estimated to be
200 mGy/year, while at 700 g/cm2 the dose rate is
expected to be reduced to approximately 0.5 mGy/year
(Pavlov et al., 2002; see also McKeever et al., 2003),
which is an order of magnitude lower than the
2–12 mGy/year natural radioactivity on Earth. Various
authors have carried out model calculations to determine
dose rate versus depth (Dartnell et al., 2007; Morthekai
et al., 2007; Banerjee and Dewangan, 2008).
Anticipated age limits and uncertainties
Assuming a lower measurable dose limit of 5 Gy
(Kalchgruber et al., 2007) and an average surface dose rate
of 100 mGy/year (Saganti et al., 2004), the lower age limit
for OSL dating can be expected to be a few decades. With
an upper measurable dose limit of 1,500 Gy, the upper
age limit would be only 10,000 years and, thus, consider-
ably lower than on Earth. However, the dose rate decreases
rapidly with increasing burial depth. At 3–4 m depth and
a background dose rate of 0.5 mGy/year, the upper age limit
would be 3 Ma. Depending on the burial rate, the upper age
limit is, therefore, expected to be in the region of several 105
years. Dose rate variations with depth depend on the model
480 LUMINESCENCE, MARTIAN SEDIMENTS
used but are in the range of 50 % for 2 m burial depth
(McKeever et al., 2003; Morthekai et al., 2007). For an ini-
tial approximation, the age can be bracketed by calculating
a minimum age based on the surface dose rate and
a maximum age based on the dose rate at the depth from
which the sample is recovered. An improved age value
can then be determined using burial models.
Instrumentation
Robotic instrumentation for a space mission has to meet
strong specifications, some of them regarding volume
(max. 1,000 cm2), weight (max. 0.6 kg), power consump-
tion (max. 10 W), and temperature range (20  C to
40  C). Designs for in situ instrumentation have been
suggested by Botter-Jensen et al. (2010) and DeWitt and
McKeever (2011).
Both instruments use photomultipliers for detection,
various light sources for stimulation, a heating element,
and a miniature X-ray source for calibration of the
measurements.
Summary
Luminescence dating is a suitable method to provide
a chronology for Martian surface materials. The use of
a robotic luminescence dating instrument for Mars
requires the development of new measurement techniques
in order to deal with the multiple new challenges of the
Martian environment not usually found when using OSL
to date terrestrial sediments. The challenges are posed,
among others, by the mineral composition, the sunlight
spectrum, and the radiation environment.
Bibliography
Aitken, M. J., 1998. An introduction to optical dating. Washington,
DC: Academic Press.
Bandfield, J. L., 2002. Global mineral distribution on Mars. Journal
of Geophysical Research-Planets, 107(E6), Art. No. 5042.
Bandfield, J. L., Hamilton, V. E., and Christensen, P. R., 2000.
A global view of Martian surface compositions from MGS-
TES. Science, 287, 1626–1630.
Banerjee, D., and Dewangan, A., 2008. Simulation of the cosmic-
ray induced dose-rate within a Martian soil profile. Radiation
Measurements, 43(2–6), 797–801.
Banerjee, D., Blair, M. W., Lepper, K., and McKeever, S. W. S.,
2002. Optically stimulated luminescence signals of polymineral
fine grains in the JSC Mars-1 soil stimulant samples. Radiation
Protection Dosimetry, 101, 321–326.
Benton, E. R., and Benton, E. V., 2001. Space radiation dosimetry in
low-earth orbit and beyond. Nuclear Instruments and Methods in
Physics Research B, 184, 255–294.
Blair, M. W., Yukihara, E. G., and McKeever, S. W. S., 2006.
A system to irradiate and measure luminescence at low tempera-
tures. Radiation Protection Dosimetry, 119, 454–457.
Blair, M. W., Kalchgruber, R., and McKeever, S. W. S., 2007.
Developing luminescence dating for extraterrestrial applica-
tions: characterization of Martian simulants and minerals. Radi-
ation Measurements, 42, 392–399.
Botter-Jensen, L., Thomsen, K. J., and Jain, M., 2010. Review of
optically stimulated luminescence (OSL) instrumental develop-
ments for retrospective dosimetry. Radiation Measurements,
45, 253–257.
Boynton, W. V., Feldman, W. C., Mitrofanov, I. G., Evans, L. G.,
Reedy, R. C., Squyres, S. W., Starr, R., Trombka, J. I., D’uston,
C., Arnold, J. R., Englert, P. A. J., Metzger, A. E., Wänke, H.,
Brückner, J., Drake, D. M., Shinohara, C., Fellows, C., Hamara,
D. K., Harshman, K., Kerry, K., Turner, C., Ward, M., Barthe,
H., Fuller, K. R., Storms, S. A., Thornton, G. W., Longmire,
J. L., Litvak, M. L., and Ton’chev, A. K., 2004. The mars odys-
sey gamma-Ray spectrometer instrument suite. Space Science
Reviews, 110, 37–83.
Carr, M. H., 2006. The surface of Mars. New York: Cambridge Uni-
versity Press.
Chevrier, V., and Mathe, P. E., 2007. Mineralogy and evolution of
the surface of Mars: a review. Planetary and Space Science,
55, 289–314.
Dartnell, L. R., Desorgher, L., Ward, J. M., and Coates, A. J., 2007.
Modelling the surface and subsurface Martian radiation environ-
ment: implications for astrobiology. Geophysical Research Let-
ters, 34, L02207, doi:10.1029/2006GL027494.
Detschel, M. J., and Lepper, K., 2009. Optically stimulated lumines-
cence dating properties of Martian sediment analogue materials
exposed to a simulated Martian solar spectral environment. Jour-
nal of Luminescence, 129, 393–400.
DeWitt, R., and McKeever, S. W. S., 2011. ODIN: a prototype lumi-
nescence reader for in situ dating of Martian regolith. In Pro-
ceedings of 2011 IEEE Aerospace Conference. IEEE
Aerospace Conference Proceedings, #2.1304.
Hansson, A., 1997. Mars and the development of life, 2nd edn. New
York: Wiley.
Jain, M., Andersen, C. E., Bøtter-Jensen, L., Murray, A. S., Haack,
H., and Bridges, J. C., 2006. Luminescence dating on Mars: OSL
characteristics of Martian analogue materials and GCR dosime-
try. Radiation Measurements, 41, 755–761.
Jain, M., Andersen, C. E., Hajdas, W. M., Edmund, J. M., and
Bøtter-Jensen, L., 2007. OSL response to proton irradiation in
some natural dosemeters: implications for Martian sediment dat-
ing. Nuclear Instruments and Methods in Physics Research A,
580, 652–655.
Kalchgruber, R., Blair, M. W., and McKeever, S. W. S., 2006. Dose
recovery with plagioclase and pyroxene samples as surrogates for
Martian surface sediments. Radiation Measurements, 41, 762–767.
Kalchgruber, R., Blair, M. W., McKeever, S. W. S., Benton, E. R.,
and Reust, D. K., 2007. Progress towards robotic in-situ dating
of Martian sediments using optically stimulated luminescence.
Planetary and Space Science, 55, 2203–2217.
Kieffer, H. H., 1976. Soil and surface temperatures at the Viking
landing site. Science, 194, 1344–1346.
Kieffer, H. H., Jakosky, B. M., and Snyder, C. W., 1992. The planet
Mars: from antiquity to present. In Kieffer, H. H., Jakosky,
B. M., Snyder, C. W., and Mathews, M. S. (eds.), Mars. Tucson:
University of Arizona Press.
Lepper, K., and McKeever, S. W. S., 2000. Characterization of
fundamental luminescence properties of the Mars soil simulant
JSC Mars-1 and their relevance to absolute dating of Martian
sediments. Icarus, 144, 295–301.
McKeever, S. W. S., Banerjee, D., Blair, M. W., Clifford, S. M.,
Clowdsley, M. S., Kim, S. S., Lamothe, M., Lepper, K.,
Leuschen, M., McKeever, K. J., Prather, M., Rowland, A.,
Reust, D., Sears, D. W. G., and Wilson, J. W., 2003. Concepts
and approaches to in-situ luminescence dating of Martian sedi-
ments. Radiation Measurements, 37, 527–534.
McKeever, S. W. S., Kalchgruber, R., Blair, M. W., and Deo, S.,
2006. Development of methods for in situ dating of Martian sed-
iments. Radiation Measurements, 41, 750–754.
McKeever, S. W. S., Blair, M. W., Yukihara, E. G., and DeWitt, R.,
2010. The effects of low ambient temperatures on optically stim-
ulated luminescence (OSL) processes: relevance to OSL dating
of Martian sediments. Radiation Measurements, 45, 60–70.
 LUMINESCENCE, POTTERY AND BRICKS
Milam, K. A., McSween, H. Y., Hamilton, V. E., Moersch, J. M.,
and Christensen, P. R., 2004. Accuracy of plagioclase composi-
tions from laboratory and Mars spacecraft thermal emission
spectra. Journal of Geophysical Research-Planets, 109(E4),
Art. No. E04001.
Morthekai, P., Jain, M., Dartnell, L., Murray, A. S., Botter-Jensen,
L., and Desorgher, L., 2007. Modelling of the dose-rate varia-
tions with depth in the Martian regolith using GEANT4. Nuclear
Instruments and Methods in Physics Research B, 580, 667–670.
Morthekai, P., Jain, M., Murray, A. S., Thomsen, K. J., and Botter-
Jensen, L., 2008. Fading characteristics of Martian analogue
materials and the applicability of a correction procedure. Radia-
tion Measurements, 43, 672–678.
O’Connor, V. A., Lepper, K., Morken, T. O., Thorstad, D. J., Podoll,
A., and Giles, M. J., 2011. A survey of the signal stability and
radiation dose response of sulfates in the context of adapting
optical dating for Mars. Journal of Luminescence, 131,
2762–2768.
Pavlov, A. K., Blinov, A. V., and Konstantinov, A. N., 2002. Steril-
ization of Martian surface by cosmic radiation. Planetary and
Space Science, 50, 669–673.
Saganti, P. B., Cucinotta, F. A., Wilson, J. W., Simonsen, L. C., and
Zeitlin, C., 2004. Radiation climate map for analyzing risks to
astronauts on the Mars surface from galactic cosmic rays. Space
Science Reviews, 110, 143–156.
Tsukamoto, S., and Duller, G. A. T., 2008. Anomalous fading of
various luminescence signals from terrestrial basaltic samples
as Martian analogues. Radiation Measurements, 43, 721–725.
Tsukamoto, S., Duller, G. A. T., Wintle, A. G., and Muhs, D., 2011.
Assessing the potential for luminescence dating of basalts. Qua-
ternary Geochronology, 6, 61–70.
Wänke, H., Boynton, W. V., Brückner, J., Dreibus, G., Taylor, G. J.,
Evans, L., James, B., Keller, J., Kerry, K., Starr, R., and GRS
Team, 2005. Sulfuric acid all over early Mars? In Proceedings
36th Lunar and Planetary Science Conference, Abstract #1389.
Cross-references
Luminescence Dating
Luminescence Dating, Dose Rates
Luminescence Dating, Uncertainties and Age Range
Radiation and Radioactivity
LUMINESCENCE, POTTERY AND BRICKS
Ian Bailiff
Department of Archaeology, University of Durham,
Durham, UK
Definition
The luminescence dating of ceramic materials, including
pottery and bricks, is an experimental technique for deter-
mining the time elapsed in years since the ceramic mate-
rial was produced by firing.
Introduction
The luminescence techniques for dating ceramic artifacts
have been described in detail by Aitken (1985, 1998).
Guidelines produced by English Heritage (Duller, 2008)
provide a concise overview of applying the method.
481
Although the dating of sedimentary deposits currently
forms the main focus of application, the dating of ceramics
is discussed in a number of review papers (Feathers, 2003;
Preusser et al., 2008; Wintle, 2008; Liritzis et al., 2013)
and also by Wagner (1998).
Historical perspective
When thermoluminescence (TL) dating was proposed as
a new scientific dating method in the 1960s (Aitken,
1985; Wintle, 2008), the principle, based on the accumula-
tion and storage of trapped charge in crystalline lumines-
cent minerals, was demonstrated using pottery from
archaeological contexts. Application to other types of
ceramic artifacts, such as curated works of art (Fleming,
1979) and ceramic building materials (CBMs), soon
followed. The introduction of optically stimulated lumi-
nescence (OSL) techniques during the 1980s led to
a focusing of research effort on the dating of processes
associated with the deposition of (unheated) sediments.
This type of application now dominates the field.
Although testing of the earliest pottery of Late Pleistocene
origin is within the chronological range of the method
(Kuzmin et al., 2001), adoption of the method for dating
pottery was muted, despite the common occurrence of pot-
tery on many sites from the Neolithic onwards (Orton
et al., 1993). For general application, the specialist advice
needed for sampling made its deployment less convenient
compared with radiocarbon testing, and moreover, the
levels of uncertainty in luminescence dates were consid-
ered too high to improve on chronologies developed on
the basis of pottery typology. Where suitable organic sam-
ples are unavailable or where temporal resolution is lim-
ited as a result of calibration issues (e.g., the flat spot ca
2500 BP or multiple ranges during the last 500 BP), the
dating of many sites has relied on pottery typology, partic-
ularly fabric composition. It is in the testing of fabric chro-
nologies where a particular role for luminescence dating
of pottery has emerged. Also, after years of dormancy,
a role has lately emerged for the dating of brick in standing
buildings, and the application to medieval buildings and
monuments lacking diagnostic features has become an
active area of research.
Technical issues
The luminescence age equation (Aitken, 1985, 1998),
expressed in its most compact form,
Age ¼ Paleodose=Dose rate,
is used to calculate the time elapsed since the chronometer
mechanism was last reset (zeroing event), which occurs
during the firing process in the case of a ceramic. The
age, given in years before the year of testing, is accompa-
nied by an overall uncertainty, given at the 68 % level of
confidence (1s), that is typically between 5 and
10 % of the age. The paleodose corresponds to the
cumulative absorbed dose (in units of gray, Gy) received
by grains of luminescent crystalline minerals within the
482 LUMINESCENCE, POTTERY AND BRICKS
ceramic sample (e.g., quartz or feldspar) since a zeroing or
resetting event that is achieved by firing. Since resetting
can be achieved for temperatures as low as ca 400  C
(i.e., baked or burnt clay), most well-fired archaeological
pottery will have met this condition. Within the relatively
short lifetime of a cooking pot, the effect of reheating on
a fire has a negligible effect on the luminescence age,
but if sherds were redeposited onto a ground surface long
after burial and subjected to prolonged heating during
a conflagration to temperatures in excess of those indi-
cated above, the date of the destruction event would be
obtained. The paleodose is determined experimentally
(the basic techniques are discussed in Duller, 2008) by
applying luminescence techniques to many individual
portions of grains of luminescent minerals extracted from
the ceramic. The dose rate is usually determined by mea-
suring the concentration of radionuclides present in the
sample and the surrounding burial medium, and then
applying conversion factors (Aitken, 1985, 1998).
Paleodose
Most ceramic materials contain grains of luminescent
minerals, such as quartz and feldspar, within the fine
(e.g., <20 mm dia.) or coarse (e.g., 100–300 mm dia.) frac-
tions of clays and temper, respectively. Such grains can be
extracted in the laboratory by mechanical crushing and,
sieving and the application of acid treatments. Typically
a minimum of 25 g of ceramic is sufficient for dating mea-
surements, although for some applications this can be
reduced significantly by drilling (Fleming, 1979). There
are important differences in the luminescence properties
of the two commonly present mineral types, quartz and
feldspar, and this affects their use for dating measure-
ments. In particular, trapped charge, which forms the basis
of the chronometer mechanism, is retained without signif-
icant loss in quartz over archaeological timescales,
whereas for most feldspars, there is a small but continuous
loss due to a physical effect referred to as anomalous fad-
ing, which causes the age to be underestimated. Although
a correction can be applied for the loss, this increases the
uncertainty in the luminescence age, and each sample
must be tested to establish the extent of fading since the
rate of loss varies according to the geological source of
the feldspar (Lamothe, 2004). Hence, coarse quartz grains
are usually preferred when testing ceramic materials, pro-
viding they are present in the fabric. Where this is not the
case, extraction of the “fine-grain” fraction that usually
contains both quartz and feldspar provides an alternative
approach. Although feldspars can be selectively removed
using acid treatment, the fine-grain technique for
paleodose determination tends to be applied routinely
without such additional treatment unless part of
a methodological study.
Dose rate
There are two physical volumes of interest when assessing
the dose rate to grains in pottery: (a) the ceramic sample;
(b) the surrounding medium to a distance of about 50 cm
(Aitken, 1985). Since the material within the latter con-
tains radionuclides emitting gamma rays contributing ca
25–45 % of the dose rate, where the proportion depends
on the grain size selected for paleodose measurements,
an assessment of the radionuclide content of both the
ceramic artifact and burial medium is required.
Such assessment is more complex for samples adjacent
to layers of soil with differing radionuclide composition or
within surface deposits, for example, and for this reason
the longstanding advice given (Aitken, 1985) to select
ceramic samples located within a uniform burial medium
remains technically sound (Duller, 2008).
For more complex situations, such as shallow or sur-
face contexts, laboratories may use the fine-grain fraction
for paleodose determination because a higher proportion
(ca 75 %) of the dose rate is due to radioactive sources
located within the pottery, and this reduces the depen-
dence of the composition of the burial medium and the
extent of overburden on the luminescence age. A further
factor that has a bearing on sampling is the moisture con-
tent of both ceramic and the burial medium because it has
a moderating effect on the dose rate. An average value for
the moisture content during the burial period is estimated
by the laboratory, and a conservative allowance is usually
made for the uncertainty (typically 20 %) in the value
adopted. Although the maximum moisture uptake of the
ceramic, controlled by the porosity, can be measured in
the laboratory, it is desirable to obtain samples of both pot-
tery and soil stored with moisture content that is represen-
tative of the contemporary conditions within the burial
environment. The overall precision in the luminescence
age increases (lower uncertainty) in regions where there
has been long-term aridity and decreases where the pottery
is porous and the moisture content relatively high.
Luminescence age
There is no internationally agreed specification for the pre-
sentation of luminescence dating results, but a listing of
technical parameters related to the evaluation of the
paleodose and the dose rate normally accompanies the
luminescence age in publications and formal reports.
The Ancient TL Date List, although no longer actively
compiled, provided a systematic compilation of dating
results and the last two issues produced included listings
of luminescence dates for pottery from Britain (Barnett,
1999) and the Southwest of North America (Feathers,
2000); more recent guidance concerning aspects of date
quotation is included in Duller (2008).
Pottery
Applications of the method to dating pottery can be
broadly grouped into those: (a) establishing the date of
the deposition of a layer or feature from which the pottery
was recovered; (b) testing chronologies developed for pot-
tery typologies; and (c) examining methodological issues,
such as the feasibility of dating a previously untested type
 LUMINESCENCE, POTTERY AND BRICKS
of pottery. There are several published large-scale dating
studies, where the term “large scale” applied here refers
to several tens of determinations rather than the several
hundreds that could be envisaged with radiocarbon.
The accuracy of luminescence has been tested against
independent dating evidence with pottery recovered from
sealed contexts where its use before deposition was judged
to be short lived. From a statistical analysis of the results
for 24 sherds of diagnostic pottery of the British Iron
Age (first millennium BC), an average difference of

100 years was obtained between the luminescence dates
and the mid-point of the assigned age range in each case
(Barnett, 2000). Similarly, comparisons of luminescence
dates for pottery from short-lived Navajo sites of ca AD
1700 in North America with independent dating evidence
provided by dendrochronology produced a favorable out-
come (Feathers, 2000, 2003).
On many sites, pottery is frequently undiagnostic and
found scattered both spatially and temporally within the
stratigraphy. Following fragmentation and discard, sherds
may be redeposited in the filling of large features such as
ditches, pits, and middens. Luminescence dating of pot-
tery from sites in Britain and North America has con-
firmed that indiscriminate testing of redeposited pottery
in large features will produce misleading results where
the processes of manufacture and deposition are signifi-
cantly displaced in time. However, luminescence has
a constructive role to play by providing a means of testing
chronologies for the onset and persistence of particular
fabric types. The luminescence dates obtained in several
studies on pottery from sites in Britain (Barnett, 2000;
Cramp, 2006) and North America (Feathers, 2009) chal-
lenge aspects of previously established fabric chronolo-
gies. In one region of Britain (Northampton), the
incorporation of fine or coarse shell fragments in temper
had been used to distinguish Early and Late Iron Age man-
ufacture, corresponding to a separation of several hundred
years, yet the luminescence dates showed that each type of
temper was not confined to one of these periods and that
the two groups strongly overlapped. In related work, the
luminescence dates obtained for sherds of scored ware,
although considered to be diagnostic pottery and thought
to have been introduced during the 4th century BC, sug-
gest much earlier introduction, in the 9th century BC and
therefore a much longer chronological span of manufac-
ture. Comparable issues have been raised by the testing
of shell-tempered pottery from North American native
sites in the Lower Mississippi Valley (Feathers, 2009)
and also sand-tempered fabrics from sites in southern Flor-
ida (Feathers, 2003). While the majority of luminescence
dates obtained for shell-tempered pottery from two mound
sites were consistent with the assigned date range in the
early eleventh century AD on the basis of the cultural
assemblage and radiocarbon dates, luminescence dates
of up to nearly a millennium earlier were obtained for sev-
eral sherds from each site. These early dates, although
considered enigmatic, flagged the possibility that relict
pottery may have been brought to the sites by migrants.
483
AMS radiocarbon dating applied to the carbonate temper
to cross-check the reliability of the luminescence dates
has since indicated that further work is required to evalu-
ate the extent of a reservoir offset (Peacock and Feathers,
2009) correction to the radiocarbon ages. Resolution of
this issue represents an important test of both the capabil-
ity of luminescence to provide a reliable tool for investi-
gating the full temporal ranges of pottery chronologies
and to further develop the combined application of inde-
pendent chronometric techniques when dating pottery
(Bonsall et al., 2002). While this aspect is being more
actively investigated for North American pottery, work
has also progressed sporadically on various sites across
Eurasia and East Asia.
Due to the nature of the native North American mound
sites, pottery is sampled from contexts that are very shallow
or within erosional surfaces, and, as indicated earlier,
the fine-grain technique can be applied to reduce the
uncertainty in the dose rate, albeit with the attendant issue
of anomalous fading to deal with if the feldspar minerals
are not removed. The further development of such work
has wider methodological implications for sites in other
parts of the world where pottery surface scatters form
a key component of the surviving archaeological evidence
(Dunnell and Feathers, 1994; Sampson et al., 1997) and
where the testing of organic matter by radiocarbon is highly
problematic because of postdepositional disturbance.
While evidence in the literature of research related to
the dating of curated artifacts that builds on the early
developmental work of the 1970s is sparse, its reactivation
in a museum environment is leading to the incorporation
of recent advances in experimental and statistical tech-
niques (Zink and Porto, 2005).
Ceramic building materials
The dating of ceramic building materials (CBM) is
a natural extension of the dating of pottery in terms of
the experimental techniques applied. Brick has been used
extensively in the construction of buildings during the last
two millennia. Although the literature is currently limited,
examples of dating applications can be found for individ-
ual buildings and structures in Europe (e.g., the Czech
Republic, Denmark, England, Finland, France, Germany,
Italy, and Poland), in Asia (e.g., Cambodia, India, Sri
Lanka, Thailand, and Uzbekistan), and in South America
(Brazil). While precise dating of high-status buildings of
the last millennium can be obtained to within a couple
of decades or better, lower-status “vernacular” buildings
of the Late Medieval and Early Modern periods in Europe
are problematic; dating the construction and phasing of
structural changes to better than 50–100 years are often
difficult to achieve. In these circumstances, there is the
opportunity for luminescence to make a contribution to
the dating of brick buildings where an overall uncertainty
in the date of ca 5 % of the age (25 years for a 500-year-
old sample) can be approached. When applied to four
buildings with reliable documentary and stylistic dating
484 LUMINESCENCE, POTTERY AND BRICKS
evidence in England, luminescence dating of brick pro-
duced extremely encouraging agreement (Bailiff, 2007);
for six samples taken from these dating control buildings
(within the range ca AD 1400–1720), the mean difference
between the central values of luminescence and assigned
ages was 5  10 years (s.d., n ¼ 6).
The value of applying an absolute dating method is
illustrated by studies that address wider research questions
of chronology that are not bound by regional typologies
and also extend beyond application to a single building.
For the period following the Roman withdrawal from
NW Europe during the early fifth century AD, in particu-
lar from Britain and Gaul, there has been a longstanding
uncertainty regarding when the manufacture of CBMs
was resumed. Recent work with bricks from a group of
ecclesiastical buildings in NW France (Blain
et al., 2007) and England (Bailiff et al., 2010) indicates
that, while in France the practice of brickmaking was
maintained during the Early Medieval period, it did not
resume until the eleventh century AD in England. The
testing of brick from Late Medieval buildings in England
has also provided an initially unexpected outcome.
Whereas it had been assumed that new bricks were pro-
duced for a specific building, the dates obtained for multi-
ple bricks from a single building suggest that recycling
was practiced by Late Medieval builders. This is of partic-
ular interest to building historians when assessing changes
to, and loss of, buildings in urban and rural landscapes.
However, the possibility of recycling CBM also provides
a note of caution when sampling; in addition to
a structural examination of the phasing of a building, the
brickwork requires careful inspection for the use of recycled
brick. In response to this issue, the possibility of dating the
emplacement, rather than manufacture, of bricks has been
explored by testing luminescent grains located within either
the surface of a brick (Veillevigne, 2007) or incorporated in
lime mortar (Goedicke, 2011) as long as the grains were suf-
ficiently exposed to sunlight before use in construction and
subsequently stored in dark conditions until tested in the
laboratory.
Summary
After a prolonged period of gestation, two main methodo-
logical roles for the application of luminescence dating to
archaeological ceramic materials have emerged; the build-
ing and testing of pottery fabric chronologies and dating
the manufacture of ceramic building materials. When
luminescence dates for pottery and brick have been tested
against independent dating evidence in well-designed
studies, the performance of the method has been found
to be reliable and accurate within the estimated experi-
mental uncertainties.
Bibliography
Aitken, M. J., 1985. Thermoluminescence Dating. London:
Academic Press.
Aitken, M. J., 1998. An Introduction to Optical Dating. Oxford:
Oxford University Press.
Bailiff, I. K., 2007. Methodological developments in the lumines-
cence dating of brick from English late-medieval and post-
medieval buildings. Archaeometry, 49, 827–851.
Bailiff, I. K., Blain, S., Graves, C. P., Gurling, T., and Semple, S.,
2010. Uses and recycling of brick in medieval English buildings:
insights from the application of luminescence dating and new
avenues for further research. The Archaeological Journal, 167,
165–196.
Barnett, S. M., 1999. Date list 6: luminescence dates for Late
Bronze Age and Iron Age pottery assemblages in eastern and
northern Britain. Ancient TL, 17, 23–40.
Barnett, S. M., 2000. Luminescence dating pottery from later
prehistoric Britain. Archaeometry, 42, 431–457.
Blain, S., Guibert, P., Bouvier, A., Vieillevigne, E., Bechtel, F.,
Sapin, C., and Baylé, M., 2007. TL dating applied to building
archaeology: the case of the medieval church Notre-Dame-Sous-
Terre (Mont-Saint-Michel, France). Radiation Measurements,
42, 1483–1491.
Bonsall, C., Cook, G., Manson, J. L., and Sanderson, D., 2002.
Direct dating of Neolithic pottery: progress and prospects.
Documenta Praehistorica, 29, 47–59.
Cramp, R. J., Bettess, G., and Bettess, F., 2006. Wearmouth and
Jarrow Monastic Sites Vol. II. Swindon: English Heritage.
Duller, G. A. T., 2008. Luminescence Dating: Guidelines on Using
Luminescence Dating in Archaeology. Swindon: English
Heritage.
Dunnell, R. C., and Feathers, J. K., 1994. Thermoluminescence dat-
ing of surficial archaeological material. In Beck, C. (ed.), Dating
in Exposed and Surface Contexts. Albuquerque: University of
New Mexico Press, pp. 115–137.
Feathers, J. K., 2000. Date list 7: luminescence dates for prehistoric
and protohistoric pottery from the American southwest. Ancient
TL, 18, 51–61.
Feathers, J. K., 2003. Use of luminescence dating in archaeology.
Measurement Science and Technology, 14, 1493–1509.
Feathers, J. K., 2009. Problems of ceramic chronology in the South-
east: does shell-tempered pottery appear earlier than we think?
American Antiquity, 74, 113–142.
Fleming, S. J., 1979. Thermoluminescence Techniques in Archaeol-
ogy. Oxford: Clarendon Press.
Goedicke, C., 2011. Dating mortar by optically stimulated lumines-
cence: a feasibility study. Geochronometria, 38, 42–49.
Kuzmin, Y. V., Hall, S., Tite, M. S., Bailey, R., O’Malley, J. M., and
Medvedev, V. E., 2001. Radiocarbon and thermoluminescence
dating of the pottery from the early Neolithic site of Gasya
(Russian Far East): initial results. Quaternary Science Reviews,
20, 945–948.
Lamothe, M., 2004. Optical dating of pottery, burnt stones, and sed-
iments from selected Quebec archaeological sites. Canadian
Journal of Earth Science, 41, 659–667.
Liritzis, I., Singhvi, A. K., Feathers, J. K., Wagner, G.A., Kadereit,
A., Zacharias, N., Li, S-H., 2013. Luminescence Dating in
Archaeology, Anthropology, and Geoarchaeology. An Over-
view. SpringerBriefs in Earth System Sciences, 97 pp.
doi:10.1007/978-3-319-00170-8.
Orton, C., Tyers, P., and Vince, A., 1993. Pottery in Archaeology.
Cambridge Manuals in Archaeology. Cambridge: Cambridge
University Press.
Peacock, E., and Feathers, J. K., 2009. Accelerator mass spectrom-
etry radiocarbon dating of temper in shell-tempered ceramics:
test cases from Mississippi, Southeastern United States. Society
for American Archaeology, 74, 351–369.
Preusser, F., Degering, D., Fuchs, M., Hilgers, A., Kadereit, A.,
Klasen, N., Krbetschek, M., Richter, D., and Spencer, J. Q. G.,
2008. Luminescence dating: basics, methods and applications.
Eiszeitalter und Gegenwart Quaternary Science Journal, 57,
95–149.
 LUMINESCENCE, ROCK SURFACES
Sampson, C. G., Bailiff, I., and Barnett, S., 1997. Thermolumines-
cence dates from Later Stone Age pottery on surface sites in
the Upper Karoo. The South African Archaeological Bulletin,
52, 38–42.
Vieillevigne, E., Guibert, P., and Bechtel, F., 2007. Luminescence
chronology of the medieval citadel of Termez, Uzbekistan: TL
dating of brick masonries. Journal of Archaeological Science,
34, 1402–1416.
Wagner, G. A., 1998. Age determination of young rocks and arti-
facts: physical and chemical clocks in quaternary geology and
archaeology. In Herrmann, B., and Wagner, G. A. (eds.), Natural
Sciences in Archaeology. Berlin: Springer.
Wintle, A. G., 2008. Fifty years of luminescence dating.
Archaeometry, 50, 276–312.
Zink, A., and Porto, E., 2005. Luminescence dating of the Tanagra
terracottas of the Louvre collections. Geochronometria, 24,
21–26.
Cross-references
Luminescence Dating
Luminescence Dating of Archaeological Sediments
LUMINESCENCE, ROCK SURFACES
Reza Sohbati
Department of Geoscience, Aarhus University, Aarhus,
Denmark
Center for Nuclear Technologies, Technical University of
Denmark, Roskilde, Denmark
Definition
Luminescence is the light emitted following the release of
stored energy (in the form of trapped charge) accumulated
in crystalline materials; this energy accumulates in natural
minerals such as quartz and feldspar through the absorp-
tion of ionizing radiation, either cosmic rays or resulting
from the decay of naturally occurring radionuclides. This
trapped charge can be released or reset by heat or light;
if reset by heat, the light emitted from the mineral is called
thermoluminescence (TL), and if released by photon stim-
ulation, it is called optically stimulated luminescence
(OSL). Thus, luminescence dating provides an estimate
of the time elapsed since the mineral grains were last
heated or exposed to daylight (Aitken, 1998).
Introduction
In geology and archaeology, there are many examples of
rock surfaces, rock art, and stone structures of unknown
age. In archaeology, megaliths, buildings, chambered
burial mounds, field walls, and cairns are very important
to understanding the way in which people have used the
landscape (Liritzis et al., 2013). In geology, there are
examples of ice-scoured bedrock, ice-transported rocks
(erratics), and cobble fans from extreme fluvial events
whose ages are essential for understanding the evolution
of the driving climate-related phenomena.
485
Luminescence dating is a well-established method of
absolute chronology that has been successfully applied
to a wide range of fine-grained sediments to provide depo-
sitional ages from a few years (Madsen and Murray, 2009)
to several tens of thousands (e.g., Murray and Olley, 2002;
Duller, 2004) and even several hundred thousand years
(e.g., Watanuki et al., 2005; Wang et al., 2006; Porat
et al., 2010; Buylaert et al., 2012). The luminescence age
of a sample is calculated by dividing the amount of ioniz-
ing radiation the sample absorbed during burial (measured
as the equivalent dose, De) by the rate of energy absorption
(the dose rate) from the environment. The single-aliquot
regenerative dose (SAR) procedure is the standard
approach to De determination in quartz and feldspar
(Murray and Wintle, 2000; Wallinga et al., 2000; Auclair
et al., 2003; Buylaert et al., 2012). Dose rate calculations
are dependent on the spatial distribution of energy deposi-
tion rates by the various radiations (a, b, g). Aitken (1985)
outlines an approach to such calculations near interfaces
(including buried rock surfaces), although later authors
suggest that the details of this approach may require mod-
ification (Yang et al., 1998).
Development of the luminescence dating
of rock surfaces
Early attempts to date buried rock surfaces used the TL
signal from calcitic rocks; this approach was based on
the idea that the signal was lost as a result of exposure to
daylight before burial. Liritzis (1994) proposed the use
of TL to date the construction of a megalithic limestone
building. Similarly, attempts were made to determine the
ages of the Apollo Temple at Delphi (Liritzis et al.,
1997) and of limestone pyramids in Greece (Theocaris
et al., 1997). However, even after long exposures to sun-
light, a significant residual TL signal remains (Liritzis
and Galloway, 1999); this significantly limits the youngest
ages that can be determined using this method. To circum-
vent this problem, later studies investigated the potential
of the OSL signal; it was known from the earliest days
of sediment dating that this signal bleaches much more
quickly and to a much lower level than TL signals (e.g.,
Huntley et al., 1985). First attempts were reported by
Huntley and Richards (1997) in which they tried to deter-
mine the burial age of quartzite pebbles using OSL. They
showed that there can be sufficient light penetration for
OSL dating to be used on a surface layer. However, for
most of their samples, the quartz gave so little lumines-
cence in response to optical stimulation that their attempts
to obtain reliable equivalent doses did not succeed. Trying
to develop a nondestructive surface dating method,
Habermann et al. (2000) built a new measurement instru-
ment with which they measured an infrared-stimulated
luminescence (IRSL) signal from the surface of a granitic
rock; they related this to the dose accumulated since
the last exposure to light. They also showed that almost
complete bleaching to a depth of at least 2 mm can result
from only a few minutes of exposure to sunlight.
486 LUMINESCENCE, ROCK SURFACES
Morgenstein et al. (2003) used IRSL to date the fine-
grained feldspar fraction scraped off the uppermost part
from the bleached surface of lithic artifacts. However, the
ages they obtained were not in agreement with the available
age controls. They attributed this to anomalous fading
(unexpected loss of signal due to quantum tunneling phe-
nomenon in feldspars) and partial bleaching of feldspar
grains. Following the idea proposed by Habermann et al.
(2000), Greilich et al. (2005) used a high-spatial-resolution
detection technique (HR-OSL) for OSL based on a CCD
camera system to date a stone wall of the medieval castle
of Lindenfels in southwestern Germany and the pre-
Columbian Nasca lines (geoglyphs) around Palpa in south-
ern Peru. Using the infrared-stimulated yellow emission
from their samples, they obtained some ages in agreement
with archaeological age control. However, they did not take
into account anomalous fading which is known to be very
common (if not universal; Huntley and Lian, 2006) in the
IRSL signals from feldspars. Vafiadou et al. (2007)
obtained useful signals from granitic rocks using the blue
stimulated ultraviolet (UV) signal originating from whole
rock slices and reported complete daylight bleaching up to
a depth of 5 mm after 14 days of exposure to sunlight for
all of their samples. More studies on the application of TL
and OSL signals to rock surfaces from archaeological sites
have been carried out by Liritzis et al. (2007, 2008, 2010);
however, none of these are supported by independent age
control. Liritzis (2011) reviews most of these studies of
luminescence rock surface dating. Simms et al. (2011) tried
to reconstruct sea levels in Antarctica by measuring OSL
from quartz grains obtained from the underside of cobbles
within raised beaches and boulder pavements. The ages
they obtained were internally consistent and agreed with
radiocarbon ages. However, their quartz grains were con-
taminated with feldspar and they had to take into account
both fading and internal dose rate (Simms et al., 2011).
In a very different approach, Polikreti et al. (2003),
Polikreti (2007) suggested the possibility of using lumi-
nescence to estimate the length of time a rock surface
was exposed to daylight; this would give similar informa-
tion to that obtained from cosmogenic nuclide (CN) dat-
ing. While looking for a method to determine the
authenticity of marble artifacts of disputed age, they inves-
tigated the bleaching of the TL signal with depth in several
samples and developed a model to describe the depen-
dence of TL intensity on exposure time and depth; how-
ever, they were unable to estimate values for parameters
in their model and so could not quantify the exposure time.
Sohbati et al. (2011) studied the depth dependence of the
bleaching of the IRSL signal from granitic rocks. Using
some simplifying assumptions, they presented a model
of this dependence and showed it to be a good descriptor
of the remaining luminescence in their naturally exposed
samples. According to their model, the longer the expo-
sure time, the further into a rock surface the luminescence
is reset. However, they were unable to determine realistic
estimates for some of the parameters in the model from
first principles, and they did not have any sample of
known exposure history with which to calibrate their
model. Shortly afterward, Laskaris and Liritzis (2011)
proposed a mathematical function to describe the attenua-
tion of daylight into rock surfaces. However, their
suggested approximation depends on the assumption that
the residual luminescence as a function of depth into rock
surfaces has a log-normal distribution, an assumption for
which there does not appear to be any physical basis.
Sohbati et al. (2012a) overcame the problem of param-
eter estimation by using a known-age roadcut sample for
calibration. They then used the resulting parameter values
to estimate the exposure times of samples of unknown age.
One of their samples was a buried rock surface on which
pigment from an extant rock painting was found. They
were able to model the distortion in the OSL-depth reset-
ting profile resulting from subsequent burial. This allowed
them to determine a pre-burial exposure period age for
their sample. This study was the first to report credible
exposure ages obtained using OSL surface exposure dat-
ing. In a related study, Chapot et al. (2012) derived
a burial age for the same sample by dating the outer 1-
mm surface of the buried boulder. This age was supported
by the OSL age of single grains of sediment buried under
the boulder and by the14C age of a leaf buried between the
boulder and the underlying sediment. Together, the two
studies constrained the age of the Great Gallery, an archae-
ologically significant Barrier Canyon Style (BCS) rock art
in Canyonlands National Park, Utah, USA.
As a part of an investigation into the burial age of an
alluvial cobble pavement at an archaeological site in Por-
tugal, Sohbati et al. (2012b) further developed their model
to include the environmental dose rate. The new model
successfully described the variation of dose with depth
and time into a quartzite cobble and allowed them to iden-
tify and quantify four events (two light exposures of dif-
ferent durations and two sequential burial periods) in the
dose record contained within a single clast. Sohbati et al.
(2012b) showed that, for the first time, prolonged periods
of exposure can leave a distinct bleaching record in the
luminescence/dose profile with depth into a rock surface
and, thus, that it may be possible to identify and date mul-
tiple deposition events in one sample.
In their model, Sohbati et al. (2012b) assume that, dur-
ing daylight exposure, the rate of trapped charge accumu-
lation from natural ionizing radiation was negligible
compared to the rate of detrapping due to daylight
bleaching. However, they discuss that this assumption is
only valid for dating those terrestrial objects which have
a short exposure history and a low environmental dose
rate. In considering the potential application of their tech-
nique in dating non-terrestrial surfaces, where environ-
mental dose rates can be orders of magnitude greater
than that on Earth, Sohbati et al. (2012c) further developed
the OSL surface exposure dating model by including the
simultaneous effect of daylight bleaching and environ-
mental dose rate. This now represents the most complete
model of OSL resetting with depth so far presented and
has the potential to date non-terrestrial geomorphological
 LUMINESCENCE, ROCK SURFACES
features such as fault movements, rockfalls, landslides,
volcanic eruptions, and crater impacts, with an upper limit
to the method between 10 and 100 ka on, for example, the
Martian surface.
Discussion
Despite the recent progress with luminescence dating of
rock surfaces, especially the potential application of this
method in surface exposure dating, a widely applicable
method has not yet been agreed upon. In sample prepara-
tion, for example, some favor working with grains
extracted from rock surfaces (e.g., Simms et al., 2011;
Chapot et al., 2012; Sohbati et al., 2012a), while others
prefer to measure 1- to 1.5-mm-thick intact slices cut from
the cores drilled into the rock surfaces (e.g., Greilich et al.,
2005; Vafiadou et al., 2007; Sohbati et al., 2012b). In part,
this is imposed by the nature of the samples and the fact
that solid slices cannot be recovered from friable rock
types (e.g., Chapot et al., 2012; Sohbati et al., 2012a).
However, the significance of the issue goes beyond the
physical properties of rocks and sampling methods and
appears to be of dosimetric importance; Sohbati et al.
(2011, 2012b) report a difference between doses measured
from grains and slices recovered from the same surface for
two different rock types. Unfortunately, their investigation
into the source of this discrepancy did not result in
a conclusive answer.
The above problem also relates to the choice of
a luminescence mineral. Although quartz is often the pre-
ferred dosimeter in sediment dating, it is often not suffi-
ciently sensitive when extracted from solid rocks (e.g.,
Huntley and Richards, 1997; Tsukamoto et al., 2011;
Sohbati et al., 2011). The intensity of signals from feld-
spars tends to be much less dependent on geological origin
and erosion history, but the dosimetry of K-rich feldspar
grains extracted from rocks is complicated because the
internal dose rate is very dependent on the original feld-
spar grain size. The in situ grain size information is lost
during the crushing process used to separate the grains
for measurement. There have been attempts to circumvent
this problem by using a nondestructive technique based on
high-resolution spatially resolved luminescence detection
from whole rock slices at a microscale level (e.g.,
Habermann et al., 2000; Greilich et al., 2002, 2005). How-
ever, dose measurement at the microscale level requires
dose rate determination on the same scale for age deriva-
tion (Guérin et al., 2013); this is unlikely ever to be routine
because it requires complex microdosimetry modeling in
rock surfaces to include beta dose rate heterogeneity
(see Nathan et al., 2003). This problem does not apply to
Na-rich feldspar to the same degree because of the absence
of internal radioactivity. Sohbati et al. (2013) investigated
the potential application of Na-rich feldspar as a potential
luminescence dosimeter for the IRSL dating of rock sur-
faces. The elevated-temperature IRSL signal at 290  C
that they measured after an IR bleach at 50  C (pIRIR290
signal) in the yellow emission seemed to be the most
487
reliable signal for dating rock surfaces in their study; how-
ever, they speculated that higher preheat and stimulation
temperatures (i.e., >320  C and 290  C for preheat and
stimulation temperatures, respectively) may result in an
even more stable yellow emission signal from Na-rich
feldspar.
Conclusions
Direct luminescence dating of rock surfaces is a viable
method for dating various geological and archaeological
phenomena, including some that cannot be dated using
sediments. Depending on the rock type and the sensitivity
of quartz extracted from rock surfaces, quartz OSL and
Na-rich feldspar pIRIR yellow signals can be used as reli-
able and dosimetrically tractable signals in the OSL and
IRSL dating of rock surfaces. Unlike most sediments, rock
surfaces can be exposed for prolonged periods; such long
exposure periods leave a distinct bleaching record in the
luminescence/dose profile with depth into a cobble. Thus,
the degree of bleaching of a rock surface is recorded
within that surface, and in favorable circumstances, it is
possible to identify and date multiple burial and exposure
events in one sample.
The new technique of OSL surface exposure dating has
considerable potential in the dating of many archaeologi-
cal and geological features, such as the time of construc-
tion of megaliths, agricultural land clearance and
enclosure, glacial advances and retreats, phases of erosion,
and other geological processes and hazards such as mass
wasting and fault scarp movement. The method is also ide-
ally suited to the problem of in situ measurement of the
age of young (<105 years) exposed surfaces on non-
terrestrial bodies.
Bibliography
Aitken, M. J., 1985. Thermoluminescence Dating. London:
Academic.
Aitken, M. J., 1998. An Introduction to Optical Dating. Oxford:
Oxford University Press.
Auclair, M., Lamothe, M., and Huot, S., 2003. Measurement of
anomalous fading for feldspar IRSL using SAR. Radiation
Measurements, 37, 487–492.
Buylaert, J.-P., Jain, M., Murray, A. S., Thomsen, K. J., Thiel, C.,
and Sohbati, R., 2012. A robust feldspar luminescence dating
method for Middle and Late Pleistocene sediments. Boreas,
41, 435–451.
Chapot, M. S., Sohbati, R., Murray, A. S., Pederson, J. L., and
Rittenour, T. M., 2012. Quaternary Geochronology Constraining
the age of rock art by dating a rockfall event using sediment and
rock-surface luminescence dating techniques. Quaternary
Geochronology, 13, 18–25.
Duller, G. A. T., 2004. Luminescence dating of quaternary sedi-
ments: recent advances. Journal of Quaternary Science,
19, 183–192.
Greilich, S., Glasmacher, U. A., and Wagner, G. A., 2002. Spatially
resolved detection of luminescence: a unique tool for
archaeochronometry. Die Naturwissenschaften, 89, 371–375.
Greilich, S., Glasmacher, U. A., and Wagner, G. A., 2005. Optical
dating of granitic stone surfaces. Archaeometry, 47, 645–665.
Guérin, G., Murray, A. S., Jain, M., Thomsen, K. J., and Mercier,
N., 2013. How confident are we in the chronology of the
488 LUMINESCENCE, ROCK SURFACES
transition between Howieson’s Poort and Still Bay? Journal of
Human Evolution, 64, 314–317.
Habermann, J., Schilles, T., Kalchgruber, R., and Wagner, G. A.,
2000. Steps towards surface dating using luminescence. Radia-
tion Measurements, 32, 847–851.
Huntley, D. J., and Lian, O. B., 2006. Some observations on
tunnelling of trapped electrons in feldspars and their implications
for optical dating. Quaternary Science Reviews, 25, 2503–2512.
Huntley, D. J., and Richards, M., 1997. The age of the Diring
Quriakh archaeological site. Ancient TL, 15, 48–51.
Huntley, D. J., Godfrey-Smith, D. I., and Thewalt, M. L. W., 1985.
Optical dating of sediments. Nature, 313, 105–108.
Laskaris, N., and Liritzis, I., 2011. A new mathematical approxima-
tion of sunlight attenuation in rocks for surface luminescence
dating. Journal of Luminescence, 131, 1874–1884.
Liritzis, I., 1994. A new dating method by thermoluminescence of
carved megalithic stone building. Comptes Rendus de l’Acadé
mie des Sciences-Série II, 319, 603–610.
Liritzis, I., 2010. Strofilas (Andros Island, Greece): new evidence
for the cycladic final neolithic period through novel dating
methods using luminescence and obsidian hydration. Journal
of Archaeological Science, 37, 1367–1377.
Liritzis, I., 2011. Surface dating by luminescence: an overview.
Geochronometria, 38, 292–302.
Liritzis, I., and Galloway, R. B., 1999. Dating implications
from solar bleaching of thermoluminescence of ancient
marble. Journal of Radioanalytic and Nuclear Chemistry, 241,
361–368.
Liritzis, I., Guibert, P., Foti, F., and Schvoerer, M., 1997. The temple
of Apollo (Delphi) strengthens novel thermoluminescence dat-
ing method. Geoarchaeology, 12, 479–496.
Liritzis, I., Sideris, C., Vafiadou, A., and Mitsis, J., 2007. Mineral-
ogical, petrological and radioactivity aspects of some building
material from Egyptian Old Kingdom monuments. Journal of
Cultural Heritage, 9, 1–13.
Liritzis, I., Kitis, G., Galloway, R. B., Vafiadou, A., Tsirliganis,
N. C., and Polymeris, G., 2008. Probing luminescence dating
of archaeologically significant carved rock types. Mediterranean
Archaeology and Archaeometry, 8, 61–79.
Liritzis, I., Zacharias, N., and Polymeris, G., 2010. Surface lumines-
cence dating of ‘Dragon Houses’ and Armena Gate at Styra
(Euboea, Greece). Mediterranean Archaeology and
Archaeometry, 10, 65–81.
Liritzis, I., Singhvi, A. K., Feathers, J. K., Wagner, G. A., Kadereit,
A., Zacharias, N., and Li, S.-H., 2013. Luminescence Dating in
Archaeology, Anthropology, and Geoarchaeology: An Overview.
New York: Springer. Springer Briefs in Earth System Sciences.
Madsen, A. T., and Murray, A. S., 2009. Optically stimulated dating
of young sediments: a review. Geomorphology, 109, 3–16.
Morgenstein, M. E., Luo, S., Ku, T. L., and Feathers, J., 2003. Ura-
nium-series and luminescence dating of volcanic lithic artefacts.
Archaeometry, 45, 503–518.
Murray, A. S., and Olley, J. M., 2002. Precision and accuracy in the
optically stimulated luminescence dating of sedimentary quartz:
a status review. Geochronometria, 21, 1–16.
Murray, A. S., and Wintle, A. G., 2000. Luminescence dating of
quartz using an improved single-aliquot regenerative-dose pro-
tocol. Radiation Measurements, 32, 57–73.
Nathan, R., Thomas, P. J., Jain, M., Murray, A. S., and Rhodes, E. J.,
2003. Environmental dose rate heterogeneity of beta radiation
and its implications for luminescence dating: Monte Carlo
modelling and experimental validation. Radiation Measure-
ments, 37, 305–313.
Polikreti, K., 2007. Detection of ancient marble forgery: techniques
and limitations. Archaeometry, 49, 603–619.
Polikreti, K., Michael, C. T., and Maniatis, Y., 2003.
Thermoluminescence characteristics of marble and dating of
freshly excavated marble objects. Radiation Measurements, 37,
87–94.
Porat, N., Chazan, M., Grün, R., Aubert, M., Eisenmann, V., and
Horwitz, L. K., 2010. New radiometric ages for the Fauresmith
industry from Kathu Pan, southern Africa: Implications for the
Earlier to Middle Stone Age transition. Journal of Archaeologi-
cal Science, 37, 269–283.
Simms, A. R., DeWitt, R., Kouremenos, P., and Drewry, A., 2011.
A new approach to reconstructing sea levels in Antarctica using
optically stimulated luminescence of cobble surfaces. Quater-
nary Geochronology, 6, 50–60.
Sohbati, R., Murray, A. S., Jain, M., Buylaert, J.-P., and Thomsen,
K. J., 2011. Investigating the resetting of OSL signals in rock
surfaces. Geochronometria, 38, 249–258.
Sohbati, R., Murray, A. S., Chapot, M. S., Jain, M., and Pederson, J.,
2012a. Optically stimulated luminescence (OSL) as
a chronometer for surface exposure dating. Journal of Geophys-
ical Research, 117, B09202.
Sohbati, R., Murray, A. S., Buylaert, J.-P., Almeida, N. A. C.,
and Cunha, P. P., 2012b. Optically stimulated luminescence
(OSL) dating of quartzite cobbles from the Tapada do
Montinho archaeological site (east-central Portugal). Boreas,
41, 452–462.
Sohbati, R., Jain, M., and Murray, A., 2012c. Surface exposure dat-
ing of non-terrestrial bodies using optically stimulated lumines-
cence: a new method. Icarus, 221, 160–166.
Sohbati, R., Murray, A., Jain, M., Thomsen, K., Hong, S.–. C., Yi,
K., and Choi, J.–. H., 2013. Na-rich feldspar as a luminescence
dosimeter in infrared stimulated luminescence (IRSL) dating.
Radiation Measurements, 51–52, 67–82.
Theocaris, P. S., Liritzis, I., and Galloway, R. B., 1997.
Dating of two Hellenic Pyramids by a novel application of
thermoluminescence. Journal of Archaeological Science, 24,
399–405.
Tsukamoto, S., Nagashima, K., Murray, A. S., and Tada, R., 2011.
Variations in OSL components from quartz from Japan sea sed-
iments and the possibility of reconstructing provenance. Quater-
nary International, 234, 182–189.
Vafiadou, A., Murray, A. S., and Liritzis, I., 2007. Optically stimu-
lated luminescence (OSL) dating investigations of rock and
underlying soil from three case studies. Journal of Archaeologi-
cal Science, 34, 1659–1669.
Wallinga, J., Murray, A. S., and Wintle, A., 2000. The single aliquot
regenerative-dose (SAR) protocol applied to coarse-grain feld-
spar. Radiation Measurements, 32, 529–533.
Wang, X. L., Wintle, A. G., and Lu, Y. C., 2006. Thermally
transferred luminescence in fine-grained quartz from Chinese
loess: basic observations. Radiation Measurements, 41,
649–658.
Watanuki, T., Murray, A. S., and Tsukamoto, S., 2005. Quartz and
polymineral luminescence dating of Japanese loess over the last
0.6 Ma: comparison with an independent chronology. Earth and
Planetary Science Letters, 240, 774–789.
Yang, Q., Rink, W. J., and Brennan, B. J., 1998. Experimental deter-
minations of beta attenuation in planar dose geometry and appli-
cation to ESR dating of tooth enamel. Radiation Measurements,
29, 663–671.
Cross-references
Luminescence Dating
Luminescence Dating, Dose Rates
Luminescence Dating, History
Luminescence Dating, Meteorites
Luminescence, Flints and Stones
Luminescence, Martian Sediments
Luminescence, Volcanic Rocks
 LUMINESCENCE, SOILS 489

with the amount of soil moving being linearly propor-

LUMINESCENCE, SOILS
Arjun M. Heimsath
School of Earth and Space Exploration, Arizona State
University, Tempe, AZ, USA
Definition
Luminescence dating measures the time since quartz and
feldspar mineral grains were last exposed to light. Dating
soils with this method quantify the last time mineral grains
sampled from within soil profiles were at the ground
surface.
Soils can be defined in many different ways depending
on the environmental context. For the purposes of this
application, soils are defined as colluvial material pro-
duced from underlying parent material and being
transported by various physical processes. Windblown
soils, agricultural soils, and deep chemical weathering
profiles are not considered here.
Introduction
Hilly upland landscapes are cloaked in a thin layer of soil
derived primarily from the underlying parent material
(Carson and Kirkby, 1972; Dietrich et al., 1995; Heimsath
et al., 1997) (Figure 1). Understanding how the Earth’s
surface evolves under changing climatic, tectonic, or
human land-use forces thus includes understanding how
these physically mobile upland soils are produced and
transported. Setting aside landslides, which are likely to
dominate soil erosion in steep and mountainous terrain,
soil moves slowly downhill by creep and slope wash.
Observations from gently rounded hills show that soil
thickness increases with distance downslope, supporting
the view that creep occurs by diffusion-like processes
tional to surface slope (Young, 1960; Carson and Kirkby,
1972; Heimsath et al., 1999). Such movement of individ-
ual soil grains could be caused by burrowing creatures
(worms, ants, gophers, etc.) and by tree throw (trees
uprooted, typically by high winds, often carry soil and
bedrock attached to the root wad) coupled with localized
slope wash. Field observations and laboratory studies also
show that other processes may also transport the soil,
including shear and depth-dependent viscous-like flow
(Young, 1960; Carson and Kirkby, 1972; Fleming and
Johnson, 1975; Selby, 1993). Tracking the transport paths
of soil particles helps build our understanding of these
processes (Figure 2). Dating the last time soil minerals
were at the ground surface can enable further quantifica-
tion and establishing mathematical relationships captur-
ing the essence of these processes. These relationships
can then be used in models that can help predict soil sus-
tainability, landslide frequency, and magnitude and also
help develop water management and land-use develop-
ment plans.
Other methods for dating soils
Early thinking about how hilly, upland landscapes evolve
with time recognized that landscape morphology likely
reflects the long-term dominant process shaping the land-
scape (Davis, 1892; Gilbert, 1909). Specifically, the
observation that smooth hillslopes tended to steepen with
increasing distance from the drainage divide led to the
assumption that sediment transport flux, which must
increase with increasing distance from the divide, is pro-
portional to topographic gradient. This led to the writing
of a linear transport relationship, which became known
as the “diffusion equation” because of its similar form to
Fick’s law of heat diffusion. Because creep processes were

Luminescence, Soils, Figure 1 (a) Photograph of typical soil-mantled hillslope used for studies such as the ones covered in this
chapter. This one is from Point Reyes, CA, during the late summer dry season. (b) Photograph of the well-mixed colluvial soil that
mantles the same landscape enriched with organic matter and quite moist during the winter rains. Note the well-defined boundary
between the physically mobile soil above (dark layer) and the underlying weathered bedrock that retains relict rock structure and is
considered to be in place. Holes are from bulk density sampling.
490 LUMINESCENCE, SOILS
Luminescence, Soils, Figure 2 (Upper Panel): The conceptual
framework as used by Heimsath et al. (1997) and similar studies
for vertical soil depth, h indicating that the slope-normal depth,
H, is equal to the vertical depth times the secant of the slope
angle. The change in soil mass in a column of soil with time is
equal to the production of soil from the underlying bedrock
minus the balance of soil transported past the column. (Lower
Panel): Idealized displacement profile (a) showing how soil
particles might move from a to b in a “laminar flow” path
envisioned by Fleming and Johnson (1975) and later sketched in
by Selby (1993). The right-hand side (b) shows Selby’s sketch of
the commonly recognized “chaotic” flow path of soil particles
that may be mixed while the soil profile is shrinking and swelling
from changing water content.
thought responsible for transport that could be modeled as
“diffusive,” the term “diffusion” became synonymous
with creep. Recently, more accurate terms such as
“diffusion-like” or “linear transport relationship” are
favored since soil particles are not actually diffusing
downhill. They are, however, moving in a creep-like
way as gravitational force imparts a downslope preference
to any physical disruption of the soil. Quantifying the pro-
cess necessitates dating or tracking the particles.
Physical tracers
An innovative and not often repeated study attempting to
quantify soil movement using physical tracers was
published in Nature over half a century ago (Young,
1960). The study recognized the importance of quantifying
rates and processes of hillslope denudation and set about
doing it by initially inserting metal pegs into the ground sur-
face at regular intervals downhill from a bedrock outcrop,
noting the distances carefully and tracking surface move-
ment of clasts over time. The uncertainty involved with
peg surface displacement was huge, and the problem with
discerning surface transport from downslope movement
occurring within the soil profile led to a modification of
the method to develop what became known as a “Young
Pit.” Despite two publications in Nature (Young, 1960,
1963a), few studies have followed the lead of this challenge
(e.g., Schumm, 1967; Finlayson, 1981), including
a modified approach (Richards and Humphreys, 2010) that
yielded results no less equivocal. Even the results of
Schumm (1967), often cited as one of the few field-based
studies quantifying soil transport processes, show consider-
able scatter and were collected over a 7-year period, longer
than most are willing to devote to a single study. Given the
results (or lack thereof) from the two most recent studies
(Clarke et al., 1999; Richards and Humphreys, 2010), few
others have attempted to follow Andrew Young’s
inspirational lead.
Tephra deposits
Fortunately, geochemical methodological advances have
enabled us to directly quantify processes and rates with
significantly less uncertainty and physical effort. Another
innovative approach that involves using a physical tracer
that does not involve human disturbance (i.e., the digging
of numerous Young Pits) is worth summarizing. Of
course, this physical tracer method also involves using
geochemistry, as the studies do rely on relatively precise
geochemical dating (using calibrated 14C dating) of tephra
deposits in upland soils. Two recent studies documented
profiles of tephra concentrations in hillslope soils and
combined a detailed mapping of these concentrations with
topographic derivatives to quantify soil transport pro-
cesses (Roering et al., 2002; Walther et al., 2009). In both
cases, one on a series of incised fluvial terraces along the
Charwell River, South Island, New Zealand (Roering
et al., 2002), and the other along a hillslope transect in
the Blue Mountains, SE Washington state, USA (Walther
et al., 2009), the tephra deposits are well dated and are
treated as marker beds in the hillslope soils. Both studies
involved extensive sampling of soil pits and auger profiles
to quantify tephra distributions in the hillslope soils.
Because the sample locations were located along profiles
characterized by the topographic properties supporting
linear transport relationships, the topographic derivatives
were used with tephra distribution and concentration to
provide calibrated transport relationships for each field
site. Given the paucity of such field-based data supporting
transport relationships, these studies are especially impor-
tant. Given how difficult it is to find such ideal field loca-
tions where there are both well-dated tephras and
hillslopes of characteristic form, such an innovative
method is unlikely to be broadly used.
 LUMINESCENCE, SOILS
Short-lived isotopes (210Pb and 137Cs)
A technically less cumbersome approach involves using
short-lived isotopes measured in bulk soil samples that
can be collected without any disruption (i.e., by inserting
tracers) and without any special site characteristics (i.e.,
dated tephra layers). These fallout-derived radionuclides
are used extensively in erosion and sediment transport
studies on both agricultural and forested landscapes.
Atmospherically delivered 210Pb and weapons-derived
137
Cs can be used alone or simultaneously to quantify
and trace erosional processes (Wallbrink and Murray,
1996; Walling and He, 1999a; Walling and He, 1999b;
Whiting et al., 2001). Relative rates of soil mixing can
be evaluated through measurements of nuclide activities
in soil profiles (Dorr, 1995; Tyler et al., 2001; Kaste
et al., 2007; Dixon et al., 2009). Because these isotopes
are deposited at the soil surface, mixing processes can
increase the dispersion of nuclides with depth and the
overall downward transport rate. For example, on agricul-
tural landscapes, where tilling homogenizes the soil to the
depth of the plow layer, a unique, well-mixed 137Cs profile
captures the process (Walling and He, 1999a). By measur-
ing the vertical distribution of fallout radionuclides in soils
and calculating diffusion-like coefficients, the sediment
transport mechanisms and mixing rates are quantified over
10- to 100-year timescales.
Meteoric 10Be
In addition to short-lived isotopes, fallout-delivered
(or garden variety) 10Be can be used to quantify hillslope
sediment transport processes over longer timescales
(Monaghan et al., 1983; McKean et al., 1993; Jungers
et al., 2009). Meteoric 10Be is a cosmogenic radionuclide
well suited for use as a tracer of soil movement over geo-
morphic timescales. Spallation of oxygen and nitrogen in
Earth’s atmosphere produces the majority of 10Be present
in natural systems (Lal and Peters, 1967). Following pro-
duction, 10Be is delivered to the Earth’s surface at a rate
most significantly proportional to regional precipitation
fluxes. A smaller, but not insignificant, inventory of 10Be
falls from the atmosphere adsorbed to terrestrial dust. The
timescale over which any radioactive tracer is useful is
a function of the rate at which the isotope is produced
and/or delivered to the natural system of interest and of
the rate at which the radionuclide decays (10Be has a half-
life of 1.39 106 years) making it especially useful as a tool
for quantifying rates of Earth surface processes that take
place over thousands to millions of years. Additionally,
10
Be adsorbs quickly and strongly to soil particles and
a soil mantle’s total 10Be inventory can reliably be
accounted for if samples integrate from the soil-atmosphere
interface to the contact between soil/saprolite and
unweathered bedrock. Naturally, it follows that landscapes
with relatively thin (0–150 cm) soil mantles are best suited
to the task of capturing complete 10Be inventories. Mea-
surement of meteoric 10Be is not trivial or inexpensive,
however, and application of this methodology is limited.
491
Luminescence dating of soils
Rods, blocks, and flexible tubes inserted in the soil are too
clumsy to examine grain scale sediment transport, which
requires the use of labeled particles (Selby, 1993). Inser-
tion and detection of labeled grains throughout a soil man-
tle with microscale positional measurements would be
a massive undertaking (Cox and Allen, 1987; Cox, 1990;
Nichols, 2004). Assuming, however, that grains brought
to the surface by bioturbation and tree throw are buried
again, vertical mixing at the grain scale may be deter-
mined by measuring the time since individual soil grains
last visited the surface. This was first done by single-grain
optical dating, which is based on an optically stimulated
luminescence (OSL) signal that accumulates within buried
quartz grains and is reset to zero by exposure to sunlight
(Heimsath et al., 2002).
OSL became the established successor to thermolumi-
nescence (TL) as the means for measuring the time
elapsed since a mineral grain’s last exposure to sunlight
(e.g., Huntley et al., 1985; Aitken, 1985, 1994; Duller,
1996; Roberts et al., 1990, 1997). The OSL measurements
from quartz and feldspar grains focus on the most light-
sensitive electron traps, less readily measurable than
TL. While exposure ages can be measured from both min-
erals, quartz is the dosimeter of choice because the
microdosimetry in feldspar grains is complex and the
stored dose decays unpredictably (Murray et al., 1995).
The OSL dating initially used gram-sized samples of thou-
sands of sand grains. The second stage of development
involved the “small aliquot” methods, which used a few
tens to a few hundred mineral grains per sample. Several
recent reviews provide details of the methodology
(Fuchs and Lang, 2009; Madsen and Murray, 2009;
Rhodes, 2011; Stockmann et al., 2013). The essence of
the methodology applied to dating soils involves follow-
ing detailed steps that are expanded upon in the above
references.
While most applications of OSL dating of soils have
been toward dating human artifacts and sedimentary
deposits, the methodology can readily be used to quantify
hillslope sediment transport rates and to distinguish
between different transport processes. Optical ages of
grains from different depths depend on transport pro-
cesses: for example, grains moving in a surface layer
would intermittently be exposed to sunlight, whereas
grains never exposed at the surface should show “infinite”
ages. Alternatively, mixing throughout the entire soil col-
umn would lead to finite-aged grains from the soil surface
to its base, the boundary with the underlying saprolite, and
the source of the mobile soil.
Downhill movement of any soil grain resolves into
slope-normal and slope-parallel components. Only the
slope-normal velocity of a grain can be derived from its
optical age. Such methods cannot assess the slope-parallel
velocity of individual grains, but the depth-averaged
velocity can be calculated if the rate of soil production
from rock weathering is known at all points and if the
492 LUMINESCENCE, SOILS
topography of the site is well constrained. Soil-production
rates were previously determined from measurements of
the in situ cosmogenic nuclides 10Be and 26Al (Heimsath
et al., 2000) and were then used in a new way by Heimsath
et al. (2002) to derive slope-parallel velocities. Addition-
ally, modeling of the landscape evolution of the site
(Braun et al., 2001) highlighted the need for more detailed
field understanding of sediment transport processes active
on soil-mantled landscapes. Unfortunately, the resource
intensity of using OSL to date individual soil grains and
attempt to fully quantify soil transport processes proves
to be a significant hurdle.
More recent work expanded upon the methodology of
Heimsath et al. (2002) by using OSL to derive soil-mixing
rates for a set of soil pits from hilltop to base (Stockmann
et al., 2013). Similar results led the authors to discern
between soil grains that had been mixed throughout the
soil column, ones that had never been to the surface and
ones that were very recently bleached by the sun. The
authors did not use the ages to determine or verify
a transport relationship for the sediment transport at the
site, but were able to quantify mixing rates and suggest
that faunal mixing was less dominant than at other studied
sites.
Summary and conclusions
Establishing chronologies for profiles of soils and for
young sedimentary deposits is an essential aspect of quanti-
fying sediment transport processes and rates. Dating soils
also enjoys wide application to archaeology, land-use man-
agement, tectonic geology, and sedimentary geology. It is,
however, exceedingly difficult to do and decades of work
involving scores of researchers and a wide variety of meth-
odologies have been devoted to it. Luminescence dating,
specifically OSL and especially single-grain OSL, shows
great promise as the leading methodology that enables both
an accurate and precise dating tool that is ideally suited to
the task. There are, however, essentially only two studies
that have applied single-grain OSL dating to the rich prob-
lem of dating soils and using the dates to better understand
erosional processes. The other applications of dating soils
(sediments) have focused on the age of the deposit for
archaeological needs rather than geomorphic. There are
many challenges to overcome before this methodology
becomes more widespread. First and foremost is the fact
that sample collection is challenging and needs to be care-
fully tailored to the process geomorphology of the site in
question. Second and related to the first because of the
limits on sample numbers is that the method is relatively
expensive, with each analysis approaching the cost of
obtaining a cosmogenic nuclide age analysis. Third, there
remain sources of significant uncertainty in establishing
the particle flow paths such that interpreting the OSL age
can be a challenge. Given the nearly endless list of exciting
scientific problems that can be tackled using luminescence
dating of soils, it is likely that these challenges will soon be
overcome and this method will truly shine.
Bibliography
Aitken, M. J., 1985. Thermoluminescence Dating. London: Aca-
demic Press.
Aitken, M. J., 1994. Optical dating: a non-specialist review. Quater-
nary Science Reviews, 13, 503–508.
Braun, J., Heimsath, A. M., and Chappell, J., 2001. Sediment trans-
port mechanisms on soil-mantled hillslopes. Geology, 29, 683–
686.
Carson, M., and Kirkby, M., 1972. Hillslope Form and Process.
Cambridge, UK: Cambridge University Press, 475 pp.
Clarke, M. F., Williams, M. A. J., and Stokes, T., 1999. Soil creep:
problems raised by a 23 year study in Australia. Earth Surface
Processes & Landforms, 24, 151–175.
Cox, G. W., 1990. Soil mining by pocket gophers along
topographic gradients in a mima moundfield. Ecology, 71(3),
837–843.
Cox, G. W., and Allen, D. W., 1987. Soil translocation by pocket
gophers in a mima moundfield. Oecologia, 72, 207–210.
Davis, W. M., 1892. The convex profile of badland divides. Science,
20, 245.
Dietrich, W. E., Reiss, R., Hsu, M.-L., and Montgomery, D. R.,
1995. A process-based model for colluvial soil depth and shal-
low landsliding using digital elevation data. Hydrological Pro-
cesses, 9, 383–400.
Dixon, J. L., Heimsath, A. M., and Amundson, R., 2008. Saprolite
and the evolution of upland landscapes – links between erosion
and weathering in Sierra Nevada, CA. Geochimica et
Cosmochimica Acta, 72(12), A220–A220.
Dixon, J. L., Heimsath, A. M., Kaste, J., and Amundson, R., 2009.
Climate-driven processes of hillslope weathering. Geology,
37(11), 975–978.
Dorr, H., 1995. Application of 210Pb in soils. Journal of
Paleolimnology, 13, 157–168.
Duller, G. A. T., 1996. Recent developments in luminescence dating
of Quaternary sediments. Progress in Physical Geography,
20, 127–145.
Finlayson, B., 1981. Field measurements of soil creep. Earth Sur-
face Processes and Landforms, 6, 35–48.
Fleming, R. W., and Johnson, A. M., 1975. Rates of seasonal creep
of silty clay soil. Quarterly Journal of Engineering Geology,
8, 1–29.
Fuchs, M., and Lang, A., 2009. Luminescence dating of hillslope
deposits – a review. Geomorphology, 109, 17–26.
Gilbert, G. K., 1909. The convexity of hilltops. Journal of Geology,
17(4), 344–350.
Heimsath, A. M., 2006. Eroding the land: steady state and stochastic
rates and processes through a cosmogenic lens. GSA Special
Paper 415, pp. 111–129.
Heimsath, A. M., Chappell, J., Dietrich, W. E., Nishiizumi, K., and
Finkel, R. C., 2000. Soil production on a retreating escarpment in
southeastern Australia. Geology, 28, 787–790.
Heimsath, A. M., Chappell, J., Spooner, N. A., and Questiaux,
D. G., 2002. Creeping soil. Geology, 30(2), 111–114.
Heimsath, A. M., Dietrich, W. E., Nishiizumi, K., and Finkel, R. C.,
1997. The soil production function and landscape equilibrium.
Nature, 388, 358–361.
Heimsath, A. M., Dietrich, W. E., Nishiizumi, K., and Finkel, R. C.,
1999. Cosmogenic nuclides, topography, and the spatial varia-
tion of soil depth. Geomorphology, 27, 151–172.
Huntley, D. J., Godfrey-Smith, D. I., and Thewalt, M. L. W., 1985.
Optical dating of sediments. Nature, 313, 105–107.
Jungers, M. C., Bierman, P. R., Matmon, A., Nichols, K., Larsen, J.,
and Finkel, R., 2009. Tracing hillslope sediment production and
transport with in situ and meteoric Be-10. Journal of Geophysi-
cal Research: Earth Surface, 114, F04020, doi:10.1029/
2008JF001086.
 LUMINESCENCE, VOLCANIC ROCKS
Kaste, J. M., Heimsath, A. M., and Hohmann, M., 2006. Quantify-
ing sediment transport across an undisturbed prairie landscape
using Caesium-137 and high-resolution topography. Geomor-
phology, 76, 430–440.
Kaste, J., Heimsath, A. M., and Bostick, B. C., 2007. Short-term soil
mixing quantified with fallout radionuclides. Geology, 35(3),
243–246.
Lal, D., and Peters, B., 1967. Cosmic ray produced radioactivity on
the earth. In Encyclopedia of Physics. Vol. XLVI/2,
pp. 551–612.
Madsen, A. T., and Murray, A. S., 2009. Optically stimulated lumi-
nescence dating of young sediments: a review. Geomorphology,
109, 3–16.
McKean, J. A., Dietrich, W. E., Finkel, R. C., Southon, J. R., and
Caffee, M. W., 1993. Quantification of soil production and
downslope creep rates from cosmogenic Be-10 accumulations
on a hillslope profile. Geology, 21(4), 343–346.
Monaghan, M. C., Krishnaswami, S., and Thomas, J., 1983. 10Be
concentration and long-term fate of particle-reactive nuclides in
five soil profiles from California. Earth and Planetary Science
Letters, 65, 51–60.
Murray, A. S., Olley, J. M., and Caitcheon, G. G., 1995. Measure-
ment of equivalent doses in quartz from contemporary water-lain
sediments using optically stimulated luminescence. Quaternary
Science Reviews, 14(4), 365–371.
Nichols, M. H., 2004. A radio frequency identification system for
monitoring coarse sediment particle displacement. Applied Engi-
neering in Agriculture, 20, 783–787.
O’Farrell, C. R., Heimsath, A. M., and Kaste, J. M., 2007. Quanti-
fying hillslope erosion rates and processes for a coastal Califor-
nia landscape over varying timescales. Earth Surface
Processes and Landforms, 32, 544–560.
Rhodes, E. J., 2011. Optically stimulated luminescence dating of
sediments over the past 200,000 years. In Jeanloz, R., and Free-
man, K. H. (eds.), Annual Review of Earth and Planetary Sci-
ences. Annual Review of Earth and Planetary Sciences, Vol.
39, pp. 461–488.
Richards, P. J., and Humphreys, G. S., 2010. Burial and turbulent
transport by bioturbation: a 27-year experiment in southeast
Australia. Earth Surface Processes and Landforms, 35(7),
856–862.
Roberts, R. G., Jones, R., and Smith, M. A., 1990. Thermolumines-
cence dating of a 50,000-year-old human occupation site in
northern Australia. Nature (London), 345(6271), 153–156.
Roberts, R. G., Walsh, G., Murray, A., Olley, J., Jones, R.,
Morwood, M., Tuniz, C., Lawson, E., Macphail, M., Bowdery,
D., and Naumann, I., 1997. Luminescence dating of rock art
and past environments using mud-wasp nests in northern Austra-
lia. Nature, 387, 696–699.
Roering, J. J., Almond, P., Tonkin, P., and McKean, J., 2002. Soil
transport driven by biological processes over millennial time
scales. Geology, 30(12), 1115–1118.
Schumm, S. A., 1967. Rates of surficial rock creep on hillslopes in
Western Colorado. Science, 155, 560–562.
Selby, M. J., 1993. Hillslope Materials and Processes. Oxford:
Oxford University Press. 451 pp.
Stockmann, U., Minasny, B., Pietsch, T. J., and McBratney, A. B.,
2013. Quantifying processes of pedogenesis using optically
stimulated luminescence. European Journal of Soil Science,
64(1), 145–160.
Tyler, A., Carter, S., Davidson, D., Long, D., and Tipping, R., 2001.
The extent and significance of bioturbation on 137Cs distribu-
tions in upland soils. Catena, 43, 81–99.
Wallbrink, P. J., and Murray, A. S., 1996. Distribution and variabil-
ity of Be-7 in soils under different surface cover conditions and
its potential for describing soil redistribution processes. Water
Resources Research, 32(2), 467–476.
493
Walling, D. E., and He, Q., 1999a. Improved models for estimating
soil erosion rates from cesium-137 measurements. Journal of
Environmental Quality, 28, 611–622.
Walling, D. E., and He, Q., 1999b. Using fallout lead-210 measure-
ments to estimate soil erosion on cultivated land. Soil Science
Society of America Journal, 63(5), 1404–1412.
Walther, S. C., Roering, J. J., Almond, P. C., and Hughes, M. W.,
2009. Long-term biogenic soil mixing and transport in a hilly,
loess-mantled landscape: Blue Mountains of southeastern Wash-
ington. Catena, 79(2), 170–178.
Whiting, P. J., Bonniwell, E. C., and Matisoff, G., 2001. Depth and
areal extent of sheet and rill erosion based on radionuclides in
soils and suspended sediment. Geology, 29(12), 1131–1134.
Young, A., 1960. Soil movement by denudational processes on
slopes. Nature, 188, 120–122.
Young, A., 1963a. Soil movement on slopes. Nature, 200, 129–130.
Cross-references
Luminescence Dating
Luminescence Dating of Archaeological Sediments
Luminescence Dating, Dose Rates
Luminescence Dating, History
Luminescence Dating, Instrumentation
Luminescence Dating, Single-Grain Dose Distribution
Luminescence, Colluvial Sediments
Luminescence, Geomorphological Processes
LUMINESCENCE, VOLCANIC ROCKS
Sumiko Tsukamoto
Leibniz Institute for Applied Geophysics (LIAG),
Hannover, Germany
Definition
Luminescence dating methods are applicable to estimate
ages of past volcanic eruptions using (1) volcanic pheno-
crysts (quartz and feldspar) and glass within volcanic
rocks (including unconsolidated tephra) and (2) rocks
heated by lava or pyroclastic flow. Unlike luminescence
dating of light-exposed sediments, the luminescence clock
is reset by heating or is naturally zero at the time of crys-
tallization. Another resetting mechanism of temperature-
assisted hydrostatic pressure by phreatic eruptions has
been also proposed (Zöller et al., 2009).
Introduction
The first attempt of luminescence dating of volcanic rocks
has been done by Wintle (1973) using thermolumines-
cence (TL) of rhyolites and basalts containing plagioclase
phenocrysts. She found, however, a significant signal loss
in artificially irradiated samples after a storage, which is
known as anomalous fading, leading to an age underesti-
mation. Since then, luminescence dating studies using
quartz, feldspar, and volcanic glass have been done both
using TL and optically stimulated luminescence (OSL).
A comprehensive review of luminescence dating of volca-
nic products was published by Fattahi and Stokes (2003).
494 LUMINESCENCE, VOLCANIC ROCKS
Quartz
Dating applications using quartz is possible for both
quartz-bearing tephra having rhyolitic composition and
rocks heated by lava or pyroclastic flows. Quartz-bearing
tephras were dated initially by TL in the blue detection
wavelength (e.g., Ichikawa et al., 1982; Takamiya and
Nishimura, 1986). Later, Hashimoto et al. (1986) found
that volcanic quartz has a TL peak in the orange-red region
(
620 nm) at around 380  C. This peak has a higher satu-
ration dose than that of the blue TL peak at
470 nm
(Hashimoto et al., 1987; Miallier et al., 1991). Since then,
red TL has been used to date volcanic quartz successfully
both using multiple-aliquot (e.g., Miallier et al., 1994;
Pilleyre et al., 1992) and single-aliquot (e.g., Toyoda
et al., 2006) approaches. The high saturation dose (charac-
teristic saturation dose, D0
6,500 Gy) made it possible
to date a tephra older than 1 Ma (Fattahi and Stokes,
2000). However, the measurement of red TL needs a PM
tube which is more sensitive to longer wavelength
(trialkali or GaAs tubes), and the tube has to be cooled
to reduce dark counts. A high blackbody radiation is also
expected. In order to subtract the blackbody radiation
more easily from the red TL signal, the use of isothermal
TL, by keeping temperature the same for a prolonged
time, has been proposed (Tsukamoto et al., 2007;
Ganzawa and Maeda, 2009). Dating of tephra using red
TL has been also done using quartz single grains from
tephra (Ganzawa and Ike, 2011). OSL using single-aliquot
regenerative dose (SAR) protocol has been tested for dat-
ing volcanic rocks (Bonde et al., 2001). However, OSL of
volcanic quartz has been found to fade and has lower ther-
mal stability than sedimentary quartz (Tsukamoto et al.,
2007). But these problems do not exist in the case of
quartz which originated from nonvolcanic bedrocks.
OSL dating was successfully applied for samples from
Eifel volcanic field, Germany, where high-pressure shock
waves during phreatic eruptions had probably reset the
OSL signals (Preusser et al., 2011).
Feldspar
After the finding of anomalous fading by Wintle (1973),
there have not been many studies on luminescence dating
of volcanic feldspars until very recently, because of the
difficulty in dealing with high fading rates. May (1977)
investigated TL of plagioclase feldspars from alkali and
tholeiitic basalts with known ages from Hawaii and found
a correlation between normalized natural signal intensities
and expected ages up to 200 ka, despite the presence of
significant anomalous fading. Guérin and Valladas
(1980) studied TL in the UV detection wavelength at high
temperature (>580  C) from volcanic plagioclase from
France and concluded that the high-temperature TL peak
is not significantly affected by anomalous fading. TL from
sanidine in the far-red region (
710 nm) was also found to
be not affected by anomalous fading (Zink et al., 1995;
Visocekas and Zink, 1999). Recently, the possibility of
luminescence dating on the Martian surface has been
suggested (McKeever et al., 2003), and since then
a growing number of studies have been published (e.g.,
Kalchgruber et al., 2006; Jain et al., 2006). Morthekai
et al. (2008) and Tsukamoto and Duller (2008) compared
fading rates of various luminescence signals from Martian
analog materials on Earth. While Morthekai et al. (2008)
obtained the lowest fading rate from red TL
(630–670 nm) signal from andesite and basalt samples,
Tsukamoto and Duller (2008) reported relatively uniform
fading rates of 5–7 %/decade from IRSL signal at
200  C from basalts in the blue detection wavelength. Fol-
lowing this, Tsukamoto et al. (2010, 2011) dated andesitic
tephra and basalts using IRSL at 200  C and obtained con-
sistent ages with independent age controls up to
70 ka.
Scoria samples from Vesuvius containing leucite and
sanidine have been dated using post-IR IRSL method
(Tsukamoto et al., 2014), which has been developed to
minimize anomalous fading (Thomsen et al., 2008). The
fading rate was successfully reduced from
30 % to less
than 5 % by increasing both preheat and post-IR stimula-
tion temperatures, and the ages after fading correction
agreed with known eruption ages of Vesuvius
(Tsukamoto et al., 2014).
Volcanic glass
TL dating of volcanic glass has been first reported by
Berger and Huntley (1983) using 2–11 mm fraction of
glass. Fine grain volcanic glass was selected, because it
was homogeneous and it was certain that it originated pri-
marily from the eruption (without lithic fragments). This
work was followed by Berger (1985, 1992) who dated vol-
canic ashes between 8 and 400 ka using TL in blue-green
detection wavelength using multiple-aliquot additive dose
technique. TL spectra were measured from volcanic glass
from tephra in Japan by Kanemaki et al. (1991). They
found that the TL spectra from volcanic glass are similar
to those of quartz and plagioclase phenocrysts, and there-
fore, the TL from volcanic glass originates from quartz
and/or plagioclase microcrystals in the glass. Optical dat-
ing of volcanic glass has been tested by Berger and
Huntley (1994). The OSL and IRSL signals from volcanic
glass from 15 samples are generally dim, and they con-
cluded that the optical dating of volcanic glass is only
occasionally feasible. However, Biswas et al. (2013) used
fine grain volcanic ashes (4–11 mm) which are presumably
dominated by volcanic glass and applied post-IR IRSL
protocol. The intensity of post-IR IRSL signal was much
larger than IRSL at 50  C and it showed negligible anom-
alous fading.
Summary
Red TL dating using volcanic quartz has been proven to be
a powerful dating technique to date rhyolitic tephra. Dif-
ferent measurement methods (red TL, UV-TL, and IRSL)
have been suggested to be useful in dating volcanic feld-
spars. Since the post-IR IRSL technique has been success-
fully applied in dating sedimentary feldspar without
 LUMINESCENCE, VOLCANIC ROCKS
a fading correction, this method should be tested more for
volcanic feldspars. Volcanic glass has a problem of low
luminescence sensitivity, but the post-IR IRSL is helpful
in both reducing anomalous fading and increasing lumi-
nescence intensity.
Bibliography
Berger, G. W., 1985. Thermoluminescence dating of volcanic ash.
Journal of Volcanology and Geothermal Research, 25, 333–347.
Berger, G. W., 1992. Dating volcanic ash by use of thermolumines-
cence. Geology, 20, 11–14.
Berger, G. W., and Huntley, D. J., 1983. Dating volcanic ash by ther-
moluminescence. PACT, 9, 581–592.
Berger, G. W., and Huntley, D. J., 1994. Tests for optically stimu-
lated luminescence from tephra glass. Quaternary Geochronol-
ogy, 13, 509–511.
Biswas, R. H., Williams, M. A. J., Raj, R., Juyal, N., and Singhvi,
A. K., 2013. Methodological studies on luminescence dating of
volcanic ashes. Quaternary Geochronology, 17, 14–25,
doi:10.1016/j.quageo.2013.03.004.
Bonde, A., Murray, A. S., and Friedlich, W. L., 2001. Santorini:
luminescence dating of a volcanic province using quartz? Qua-
ternary Science Reviews, 20, 789–793.
Fattahi, M., and Stokes, S., 2000. Extending the time range of lumi-
nescence dating using red TL (RTL) from volcanic quartz. Radi-
ation Measurements, 32, 479–485.
Fattahi, M., and Stokes, S., 2003. Dating volcanic and related sedi-
ments by luminescence methods: a review. Earth-Science
Reviews, 62, 229–264.
Ganzawa, Y., and Ike, M., 2011. SAR-RTL dating of single grains of
volcanic quartz from the late Pleistocene Toya Caldera. Quater-
nary Geochronology, 6, 42–49.
Ganzawa, Y., and Maeda, M., 2009. 390–410  C isothermal red
thermoluminescence (IRTL) dating of volcanic quartz using the
SAR method. Radiation Measurements, 44, 517–522.
Guérin, G., and Valladas, G., 1980. Thermoluminescence dating of
volcanic plagioclases. Nature, 286, 697–699.
Hashimoto, T., Hayashi, Y., Koyanagi, A., Yokosaka, K., and
Kimura, K., 1986. Red and blue colouration of thermolumines-
cence from natural quartz sands. Nuclear Tracks and Radiation
Measurements, 11, 229–235.
Hashimoto, T., Yokosaka, K., and Habuki, H., 1987. Emission prop-
erties of thermoluminescence from natural quartz – blue and red
TL response to absorbed dose. Nuclear Tracks and Radiation
Measurements, 13, 57–66.
Ichikawa, Y., Higihara, N., and Nagatomo, T., 1982. Dating of pyro-
clastic flow deposits by means of the quartz inclusion method.
PACT, 6, 409–416.
Jain, M., Andersen, C. E., Bøtter-Jensen, L., Murray, A. S., Haack,
H., and Bridges, J. C., 2006. Luminescence dating on Mars: OSL
characteristics of Martian analogue materials and GCR dosime-
try. Radiation Measurements, 41, 755–761.
Kalchgruber, R., Blair, M. W., and McKeever, S. W. S., 2006. Dose
recovery with plagioclase and pyroxene samples as surrogates
for Martian surface sediments. Radiation Measurements,
41, 762–767.
Kanemaki, M., Ninagawa, K., Yamamoto, I., Nakagawa, M., Wada,
T., Yamashita, Y., and Endo, K., 1991. Red thermoluminescence
of volcanic glass fractions from tephras. Nuclear Tracks and
Radiation Measurements, 18, 81–88.
May, R., 1977. Thermoluminescence dating of Hawaiian basalt.
Geological Survey Professional Paper, 1095, 1–47.
McKeever, S. W. S., Banerjee, D., Blair, M. W., Clifford, S. M.,
Clowdsley, M. S., Kim, S. S., Lamothe, M., Lepper, K.,
495
Leuschen, M., McKeever, K. J., Prather, M., Rowland, A.,
Reust, D., Sears, D. W. G., and Wilson, J. W., 2003. Concepts
and approaches to in-situ luminescence dating of Martian sedi-
ments. Radiation Measurements, 37, 527–534.
Miallier, D., Fain, J., Montret, M., Pilleyer, T., Sanzelle, S., and
Soumanam, S., 1991. Properties of the red TL peak of quartz rel-
evant to thermoluminescence dating. Nuclear Tracks and Radia-
tion Measurements, 18, 89–94.
Miallier, D., et al., 1994. Intercomparisons of red TL and ESR sig-
nals from heated quartz grains. Radiation Measurements,
23, 143–154.
Morthekai, P., Jain, M., Murray, A. S., Thomsen, K. J., and Botter-
Jensen, L., 2008. Fading characteristics of Martian analogue
materials and the applicability of a correction procedure. Radia-
tion Measurements, 43, 672–678.
Pilleyre, T., Montret, M., Fain, J., Miallier, D., and Sanzelle, S.,
1992. Attempts at dating ancient volcanoes using the red TL of
quartz. Quaternary Science Reviews, 11, 13–17.
Preusser, F., Rufer, D., and Schreuers, G., 2011. Direct dating of
quaternary phreatic maar eruptions by luminescence methods.
Geology, 39, 1135–1138.
Takamiya, H., and Nishimura, S., 1986. Thermoluminescence ages
of some volcanic materials. Nuclear Tracks and Radiation Mea-
surements, 11, 251–257.
Thomsen, K. J., Murray, A. S., Jain, M., and Bøtter-Jensen, L.,
2008. Laboratory fading rates of various luminescence signals
from feldspar-rich sediment extracts. Radiation Measurements,
43, 1474–1486.
Toyoda, S., Tsukamoto, S., Usui, H., Hameau, S., and Suzuki, T.,
2006. Dating of Japanese Quaternary tephras by ESR and lumi-
nescence methods. Quaternary Geochronology, 1, 320–326.
Tsukamoto, S., Kataoka, S., Oguchi, T., Murray, AS., and Komatsu,
G., 2014. Luminescence dating of scoria fall and lahar deposits
from Somma-Vesuvius, Italy. Quaternary Geochronology,
20, 39–50.
Tsukamoto, S., and Duller, G. A. T., 2008. Anomalous fading
of various luminescence signals from terrestrial basaltic samples
as Martian analogues. Radiation Measurements, 43, 721–725.
Tsukamoto, S., Murray, A. S., Huot, S., Watanuki, T., Denby, P. M.,
and Bøtter-Jensen, L., 2007. Luminescence property of volcanic
quartz and the use of red isothermal TL for dating tephras. Radi-
ation Measurements, 42, 190–197.
Tsukamoto, S., Duller, G. A. T., Wintle, A. G., and Frechen, M.,
2010. Optical dating of a Japanese marker tephra using plagio-
clase. Quaternary Geochronology, 5, 274–278.
Tsukamoto, S., Duller, G. A. T., Wintle, A. G., and Muhs, D., 2011.
Assessing the potential for luminescence dating of basalts.
Quaternary Geochronology, 6, 61–70.
Visocekas, R., and Zink, A., 1999. Use of the far red TL emission
band of alkali feldspars for dosimetry and dating. Quaternary
Science Reviews, 18, 271–278.
Wintle, A. G., 1973. Anomalous fading of thermoluminescence in
mineral samples. Nature, 245, 143–144.
Zink, A., Visocekas, R., and Bos, A. J. J., 1995. Comparison of
‘blue’ and ‘infrared’ emission bands in thermoluminescence of
alkali feldspars. Radiation Measurements, 24, 513–518.
Zöller, L., Blanchard, H., and McCammon, C., 2009. Can tempera-
ture assisted hydrostatic pressure reset the ambient TL of
rocks? – a note on the TL of partially heated country rock from
volcanic eruptions. Ancient TL, 27, 15–22.
Cross-references
Luminescence Dating
Luminescence, Martian Sediments
 M
MAGNETIC ANOMALIES
Roi Granot
Department of Geological and Environmental Sciences,
Ben Gurion University of the Negev, Beer Sheva, Israel
Definition
A magnetic anomaly is the magnetic field remaining after
the Earth’s magnetic field has been removed from the
observed amplitude of the local magnetic field. The rema-
nent magnetization of rocks records past variations of the
geomagnetic field whereby spatial changes in their mag-
netization give rise to magnetic anomalies. We can date
the oceanic crust by studying the unique pattern of
lineated positive and negative magnetic anomalies that
arise due to past reversals of the geomagnetic field with
known age. Besides the primary signal that is related to
past polarity reversals, the shape and tiny variations
(wiggles) of the anomalies can be used to further constrain
the age of the crust. Altogether, magnetic anomalies pro-
vide the main source of dating for the oceanic basins and
lay the foundations for global plate reconstructions which
place important constraints on the development of the lith-
osphere, biosphere, hydrosphere, cryosphere, and global
climate.
Introduction
The Earth’s magnetic field (the geomagnetic field) is gen-
erated by convection motions in the liquid outer core. The
field is not steady but varies with time due to the dynamic
of motions within the outer core leading to the ever-
changing strength and position of its poles. In extreme
cases and in irregular occurrences, hundreds of rapid
reversals in the polarity of the field have taken place
J.W. Rink, J.W. Thompson (eds.), Encyclopedia of Scientific Dating Methods, DOI 10.1007/978-94-007-6304-3,
© Springer Science+Business Media Dordrecht 2015
(Amit et al., 2010). The direction and strength of the
geomagnetic field may get permanently imprinted into
iron-rich rocks during their formation. The resultant rem-
anent magnetization gives rise to local magnetic fields that
generate “magnetic anomalies.” Although continental
magnetic anomalies may, in certain circumstances, pro-
vide some age constraints on the magnetized rocks,
it is the marine magnetic anomalies that have crucial
implications for the dating of the oceanic crust and serve
as the basis for the Mesozoic and Cenozoic global plate
reconstruction models (e.g., Seton et al., 2012).
The oceanic crust is formed along mid-ocean ridges,
where melt rises and cools to form the crust (Vine and
Matthews, 1963). The cooling crust becomes magnetized
in the direction of the geomagnetic field and retains mag-
netization proportional to its strength. After initial forma-
tion, the newly formed crust moves away and ages
progressively from the ridge in a process known as sea-
floor spreading. In that way, a symmetric and unique mag-
netization pattern is formed across the spreading ridges.
Although the oceanic crust is formed in a complex accre-
tion process that varies in space and time, a remarkably
consistent pattern of lineated anomalies are found globally
(Maus et al., 2009). This pattern of anomalies predomi-
nantly reflects past reversals of the dipolar geomagnetic
field and has been calibrated with radioisotopic age con-
trol at certain tie points, or astronomical cyclicity in strat-
igraphic sections, to form the basis for the geomagnetic
polarity timescale (GPTS, Cande and Kent, 1995;
Channell et al., 1995). Comparison of the predicted pat-
tern of anomalies (based on the GPTS and a simple
two-dimensional magnetic structure) with the observed
marine magnetic anomalies (for example see Figure 1)
allows researchers to globally map lines of equal and
known age (i.e., isochrones) (Figure 2).
498 MAGNETIC ANOMALIES

a 300
nT
0
b
A13 A6 A5 A3 A1 A3 A5 A6 A13 Anomalies
c
30 20 10 0 10 20 30 Age [Myr]

Depth [km]
200 km
d 4

e Pacific-Antarctic Ridge

NBP
9707
˚
−46 Menard Fracture zone

˚ ˚
−48 −50

−100˚
−120

−110˚
−13
0˚

Magnetic Anomalies, Figure 1 A representative magnetic anomaly profile (NBP9707 cruise) across the Pacific–Antarctic Ridge.
(a) Observed magnetic anomalies along the NBP9707 track. The synthetic magnetic anomaly profile (b) is calculated based on the
geomagnetic polarity timescale of Cande and Kent (1995) where the positive chrons used are shown in the bar graph (c). The
numbers beneath the graph are ages (in millions of years) and the key polarity chrons are identified above the graph. The forward
synthetic model uses this polarity pattern in a 0.5-km-thick source layer that has a mean magnetization of 10 Am
1. Spreading rates
that were used to construct the model follow Croon et al. (2008). Seafloor depth (d) was used as the top of the magnetic source layer.
The location of the NBP9707 track is shown in panel (e), projected on top of a satellite-derived free-air gravity anomaly map (Sandwell
and Smith, 2009).

0 20 40 60 80 100 120 140 160

Age [Myr]
Magnetic Anomalies, Figure 2 Identified locations of marine magnetic anomalies with known age (Compiled by Seton et al., 2014).
Note the absence of identifications in the equatorial regions and in the Cretaceous quiet zones.

Magnetic anomaly data and their physical the magnetic field. To avoid contamination of the signal
properties by the survey platform (e.g., submarine vehicles, ships,
The collection of marine magnetic data is traditionally helicopters, and airplanes), the magnetometers are usually
made with magnetometers that measure the magnitude of towed some distance away from the towing platform.
 MAGNETIC ANOMALIES
The measured values are the summation of the strength of
the present-day geomagnetic field and the magnetic fields
induced by the igneous crust in the direction of the geo-
magnetic field. The contribution of the geomagnetic field
is computed and subtracted from the total value using
the appropriate International Geomagnetic Reference
Field model (e.g., Finlay et al., 2010). Minor additions
are induced by external magnetic fields but these are usu-
ally rather negligible and contain shorter wavelengths as
compared to the typical geological signature. These addi-
tions can be accounted for using gradiometer data or data
collected in a reference magnetic base station. Modern
magnetometers have an absolute accuracy of ~1 nanotesla
(nT), which is two to three orders of magnitude smaller
than the target magnetic anomalies.
The shape of a magnetic anomaly depends on the orien-
tation of the spreading axis, the direction of the ambient
geomagnetic field, the magnetization direction of the
source rocks, and the shape of the magnetic boundaries.
Therefore, in most places, the anomalies are skewed and
spatially shifted from the actual location of the magnetiza-
tion contrasts that generate them. To assign an accurate
location for each of the magnetic contrasts with known
age, it is essential to take into consideration the various
parameters affecting the shape of the anomalies through
a comparison against a two-dimensional forward mag-
netic modeling computed with the appropriate parameters
(Blakely, 1995 and Figure 1). The resultant magnetic
anomaly picks (Figure 2) delineate isochrons of the igne-
ous oceanic crust. These isochrons, found on both sides
of the spreading ridges, together with fracture zone loca-
tions, provide the basis for the computation of relative
motions between adjacent plates and serve as the basis
for the construction of global plate reconstruction models.
Outstanding issues
Marine magnetic anomalies have outstanding issues that
stem from the physical principles that govern their forma-
tion and from limitations that relate to the recorded geo-
magnetic field. In the equatorial region, where spreading
ridges are oriented in a north–south direction parallel to
the direction of the geomagnetic field, no obvious (i.e.,
large amplitude) magnetic anomalies exist. Hence, locat-
ing magnetic anomalies there is virtually impossible, lead-
ing to the absence of accurate age models for the
equatorial Atlantic and Pacific (Figure 2). Oriented
three-axial magnetometers that measure the components
of the anomalous field allow the detection of the magneti-
zation boundaries with better confidence (Gee and Cande,
2002). Although efforts have recently been made to con-
strain the age of the oceanic crust in certain equatorial
regions (Barckhausen et al., 2008, 2013), the age of the
oceanic crust there remains poorly defined and is the focus
of geophysical investigations.
Thus far, dating of the oceanic crust has been based on
reversals-related anomalies. Therefore, the spatial resolu-
tion of which magnetic anomalies can date the crust
499
depends on past reversal frequency of the geomagnetic
field. Periods that had frequent reversals, such as during
the Cenozoic period, allow for the construction of well-
constrained crustal age models (Figure 2). However, dur-
ing the Cretaceous period, the reversal frequency of the
geomagnetic field was low (less than once every million
years), with the extreme case of the Cretaceous Normal
Superchron that lasted between 121 to 83.5 million
years ago (Cande and Kent, 1995; He et al., 2008). There-
fore, the age of the crust that was formed during the
superchron period (termed “quiet zone,” Figure 2) remains
poorly resolved. Luckily, past fluctuations in the intensity
of the dipolar geomagnetic field are also recorded by
the magnetized rocks, resulting in globally correlatable
wiggles found between the primary, reversals-related
magnetic anomalies (Cande and Kent, 1992; Gee et al.,
2000; Bouligand et al., 2006; Granot et al., 2012). Ongo-
ing efforts are being made to refine the existing crustal
age models using these magnetic wiggles as a means
to establish internal isochrones between the primary
anomalies.
Remanent magnetization directions of the magnetic
source layer influence the shape of magnetic anomalies.
Therefore, if a plate drifted during its geological history
in a known fashion, one could potentially utilize the
predicted remanent magnetization directions to constrain
the age of the crust. “Paleomagnetic poles,” the magnetic
poles of a certain plate, are calculated from independent
paleomagnetic sites with known ages (see entry
“Palaeomagnetism, Continental Drift”). These poles place
constraints on past motions of the plates, thereby allowing
the prediction of the directions of magnetization for every
location on a given plate. Forward modeling of the
expected magnetic anomalies for different periods of time
based on the paleomagnetic poles of the appropriate plate
until a reasonable match is found provides a crude, yet
important, age constraint on the source rocks. This meth-
odology may, under certain conditions, help to constrain
the age of continental magnetic anomalies.
Bibliography
Amit, H., Leonhardt, R., and Wicht, J., 2010. Polarity reversals from
paleomagnetic observations and numerical dynamo simulations.
Space Science Reviews, 155, 293–335.
Barckhausen, U., Ranero, C. R., Cande, S. C., Engels, M., and
Weinrebe, W., 2008. Birth of an intraoceanic spreading center.
Geology, 36, 767–770.
Barckhausen, U., Bagge, M., and Wilson, D. S., 2013. Seafloor
spreading anomalies and crustal ages of the Clarion-Clipperton
Zone. Marine Geophysical Research, 34, 79–88.
Blakely, R., 1995. Potential Theory in Gravity and Magnetic Appli-
cations. Cambridge, UK: Cambridge University Press.
Bouligand, C., Dyment, J., Gallet, Y., and Hulot, G., 2006. Geomag-
netic field variations between chrons 33r and 19r (83–41 Ma)
from sea-surface magnetic anomaly profiles. Earth and Plane-
tary Science Letters, 250, 541–560.
Cande, S. C., and Kent, D. V., 1992. Ultrahigh resolution marine
magnetic anomaly profiles: a record of continuous paleointensity
variations? Journal of Geophysical Research, 97, 15075–15083.
500 MAGNETIC CHRONOLOGY
Cande, S. C., and Kent, D. V., 1995. Revised calibration of the geo-
magnetic polarity timescale for the Late Cretaceous and
Cenozoic. Journal of Geophysical Research, 100, 6093–6095.
Channell, J. E. T., Erba, E., Nakanishi, M., and Tamaki, K., 1995.
Late Jurassic-Early Cretaceous time scales and oceanic magnetic
anomaly block models. In Berggren, W. A., Kent, D. V., Aubry,
M.-P., and Hardenbol, J. (eds.), Geochronology, Time Scales and
Global Stratigraphic Correlation, SEPM Special Publication
54. Society of Economic Mineralogy and Petrology, Tulsa,
pp. 51–63.
Croon, M. B., Cande, S. C., and Stock, J. M., 2008. Revised Pacific-
Antarctic plate motions and geophysics of the Menard Fracture
Zone. Geochemistry Geophysics Geosystems, 9, 1–20.
Finlay, C. C., Maus, S., Beggan, C. D., Bondar, T. N., Chambodut,
A., Chernova, T. A., Chulliat, A., Golovkov, V. P., Hamilton, B.,
Hamoudi, M., Holme, R., Hulot, G., Kuang, W., Langlais, B.,
Lesur, V., Lowes, F. J., Luhr, H., Macmillan, S., Mandea, M.,
McLean, S., Manoj, C., Menvielle, M., Michaelis, I., Olsen,
N., Rauberg, J., Rother, M., Sabaka, T. J., Tangborn, A.,
Toffner-Clausen, L., Thebault, E., Thomson, A. W. P.,
Wardinski, I., Wei, Z., Zvereva, T. I., and Wo, I. A. G. A.,
2010. International geomagnetic reference field: the eleventh
generation. Geophysical Journal International, 183,
1216–1230.
Gee, J. S., and Cande, S. C., 2002. A surface-towed vector magne-
tometer. Geophysical Research Letters, 29, 1–4.
Gee, J. S., Cande, S. C., Hildebrand, J. A., Donnelly, K., and Parker,
R. L., 2000. Geomagnetic intensity variations over the past
780 kyr obtained from near-seafloor magnetic anomalies.
Nature, 408, 827–832.
Granot, R., Dyment, J., and Gallet, Y., 2012. Geomagnetic field var-
iability during the Cretaceous Normal Superchron. Nature Geo-
science, 5, 220–223.
He, H. Y., Pan, Y. X., Tauxe, L., Qin, H. F., and Zhu, R. X., 2008.
Toward age determination of the M0r (Barremian-Aptian bound-
ary) of the Early Cretaceous. Physics of the Earth and Planetary
Interiors, 169, 41–48.
Maus, S., Barckhausen, U., Berkenbosch, H., Bournas, N.,
Brozena, J., Childers, V., Dostaler, F., Fairhead, J. D., Finn, C.,
von Frese, R. R. B., Gaina, C., Golynsky, S., Kucks, R.,
Luhr, H., Milligan, P., Mogren, S., Muller, R. D., Olesen, O.,
Pilkington, M., Saltus, R., Schreckenberger, B., Thebault, E.,
and Tontini, F. C., 2009. EMAG2: a 2-arc min resolution earth
magnetic anomaly grid compiled from satellite, airborne, and
marine magnetic measurements. Geochemistry Geophysics
Geosystems, 10, 1–12.
Sandwell, D. T., and Smith, W. H. F., 2009. Global marine gravity from
retracked Geosat and ERS-1 altimetry: ridge segmentation versus
spreading rate. Journal of Geophysical Research, 114, 1–18.
Seton, M., Muller, R. D., Zahirovic, S., Gaina, C., Torsvik, T. H.,
Shephard, G., Talsma, A., Gurnis, M., Turner, M., Maus, S.,
and Chandler, M., 2012. Global continental and ocean basin
reconstructions since 200 Ma. Earth-Science Reviews, 113,
212–270.
Seton, M., Whittaker, J., Wessel, P., Muller, R. D., DeMets, C.,
Merkouriev, S., Cande, S., Gaina, C., Eagles, G., Granot, R.,
Stock, J., Wright, N., and Williams, S., 2014. Community infra-
structure and repository for marine magnetic identifications.
Geochemistry Geophysics Geosystems, 15, 1–13.
Vine, F. J., and Matthews, D. H., 1963. Magnetic anomalies over
oceanic ridges. Nature, 199, 947–949.
Cross-references
Ar–Ar and K–Ar Dating
Continental Drift (Paleomagnetism)
Magnetometer
MAGNETIC CHRONOLOGY
Manuel Calvo-Rathert
Departamento de Física, Escuela Politécnica Superior,
Universidad de Burgos, Avenida de Cantabria s/n, Burgos,
Spain
Definition
Study of the variations with time of the Earth’s magnetic
field by the analysis of direction and intensity of magneti-
zation recorded by rocks, sediments, and archaeological
materials and its use for dating and chronological correla-
tion purposes.
Variations of the earth’s magnetic field
The present-day magnetic field at the surface of the Earth
can be well approximated by an inclined magnetic dipole
placed at the Earth’s center which forms an angle of
~11.5 with the rotation axis. Such a dipole accounts for
nearly 90 % of the field at the Earth’s surface. The still sig-
nificant amount remaining can be described by the
non-dipole field. The Earth’s magnetic field (EMF) is,
however, not static, as its direction and strength vary with
time. This field variability spans time scales ranging from
fractions of a second to millions of years. The short-term
variations arise mostly from currents flowing in the iono-
sphere and magnetosphere. The slower changes are pro-
duced by the so-called main field, which is generated by
magnetohydrodynamic processes in the Earth’s liquid iron
outer core, and display periods of a year to millions of
years. Periods of internal origin of an order shorter than
a year cannot be detected at the Earth’s surface because
of the screening effect of the lowermost mantle (e.g.,
Merril et al., 1998). Since a magnetic field has existed
on the Earth since early in its history (e.g., Tarduno
et al., 2010) and its direction and intensity vary with time
with a broad spectrum of frequencies, it can be used as
a dating tool on very different time ranges.
According to their magnitude, length, and global or
regional character, different types of variations of the
EMF can be distinguished. Variations of both direction
and magnitude with periods ranging between years and
tens of thousands of years are termed secular variation
(SV). SV can be described as produced by the combined
effect of variations in the dipole field and changes in the
strength and location of the non-dipole field components.
The former are synchronous on a global scale and are char-
acterized by longer periods, while the latter vary over
smaller distances and display shorter periods. As
a result, patterns of SV are similar over subcontinental
regions but differ markedly between continents.
Besides secular variation, one of the main characteris-
tics of the EMF is that it switches its polarity,
interchanging the positions of magnetic north and south.
The duration of geomagnetic polarity intervals is rather
variable, ranging between some tens of thousands and
several millions of years. The term polarity chrons is used
 MAGNETIC CHRONOLOGY
for the main subdivisions of time recognized on the basis
of polarity, while short (0.1 Ma) polarity intervals occur-
ring within a chron are termed subchrons (Harland
et al., 1990). Periods of stable polarity are interrupted in
several occasions by short-lived geomagnetic “events” of
a few thousands of years of duration (e.g., Laj and
Channell, 2007) called excursions.
Paleomagnetism
Detailed records of geomagnetic secular variation are only
available for the last 400 years (Jackson et al., 2000), but
there are no direct surviving observations of the EMF prior
to this period of historical observations. Thus, information
about characteristics and variations of the ancient geomag-
netic field has to be obtained indirectly. Both rocks and
archaeological materials can carry a remanent magnetiza-
tion that is in most cases parallel to the field in which they
were magnetized, so that its measurement can supply the
direction of the EMF vector in the past. The determination
of the ancient magnetic field by studying the magnetiza-
tion recorded in rocks or other materials led to the devel-
opment of paleomagnetism (e.g., Butler, 1992; Tauxe,
2010, which are good introductory books on this subject).
Rocks (or archaeological materials like ceramics or
bricks) are assemblages in which fine-grained ferromag-
netic minerals are dispersed within a matrix of diamag-
netic and paramagnetic minerals. The stability of
remanent magnetization is largely dependent on the grain
size of these ferromagnetic minerals. Thermal oscillation
leads to magnetic relaxation, in which remanent magneti-
zation of an assemblage of grains decays with time. The
most stable magnetization occurs in particles with
single-domain (SD) state, characterized by a uniform
magnetization. Less stable magnetization is carried by
larger particles which are not uniformly magnetized and
are referred to as multidomain (MD) grains. Very small
uniformly magnetized ferromagnetic grains are, however,
in a superparamagnetic state and are only able to produce
a rapidly decaying short-lived remanent magnetization.
On the other hand, remanent magnetization carried by
SD grains can remain stable without decaying for time
spans of the order of magnitude of the age of the Earth.
There are different mechanisms leading to the acquisi-
tion of a remanent magnetization. A thermoremanent mag-
netization (TRM) is acquired when a material is cooled
down from above its Curie temperature in the presence of
a magnetic field. The Curie temperature is a temperature
characteristic of each ferromagnetic mineral type above
which it loses its ability to preserve a remanent magnetiza-
tion. A TRM is parallel to the ambient geomagnetic field
direction during remanence acquisition and is typically
found in igneous rocks or archaeological materials which
have been subjected to firing (e.g., pottery, bricks, or kilns).
A detrital remanent magnetization (DRM) is acquired
during deposition of sedimentary rocks. During deposi-
tion in a quite environment, magnetized particles that
can rotate freely will turn into the direction of the applied
501
geomagnetic field like compass needles and will be subse-
quently locked in place during lithification. The formation
of sedimentary rocks is, however, rather complex, and
remanence acquisition may be affected by a variety
of conditions and processes, like a diversity of initial
mineralogies, perturbations during sediment deposition,
Brownian motion of magnetized particles, flattening into
the bedding plane by compaction, post-depositional mod-
ification through the action of organisms, or diagenesis.
A remanent magnetization acquired due to physical align-
ment processes after deposition but before consolidation is
usually termed post-depositional remanent magnetization
(pDRM).
Relaxation time of small superparamagnetic grains can
be dramatically increased by augmenting their grain size.
Chemical changes that form ferromagnetic minerals
which grow large enough in a magnetizing field create
a chemical remanent magnetization (CRM) if this chemi-
cal change takes place at temperatures not as high as to
produce a TRM. Chemical reactions giving rise to new
ferromagnetic minerals include the alteration to
a ferromagnetic mineral of a preexisting one and the pre-
cipitation of a ferromagnetic mineral from a solution. Very
often CRM is a secondary magnetization, not acquired
during rock formation but long after.
Paleointensity
Although in many paleomagnetic studies only directional
information about the geomagnetic field is needed, its
intensity also contains valuable information concerning
field variability. Therefore, several investigations also
demand the knowledge of the strength of the paleofield
vector, which is called the paleointensity. Its determina-
tion is more difficult, since the magnitude of the paleofield
vector cannot be directly read from the magnetization vec-
tor. While the direction of magnetization is, in most cases,
parallel to the direction of the applied field, its magnitude
is only proportional to the field intensity and only if the
latter is low, like in the case of the geomagnetic field. In
order to retrieve the strength of the magnetizing field,
experiments which try to duplicate the acquisition of rem-
anent magnetization in a known field have to be carried
out. Complex remanence acquisition mechanisms like
the ones creating CRM or DRM are, however, almost
impossible to reproduce exactly in laboratory experi-
ments. TRM acquisition, on the other hand, is easier to
reproduce in laboratory experiments and well supported
by a theoretical basis (Néel, 1949, 1955). If a TRMa was
acquired by a rock in an ancient geomagnetic field Ba,
TRMa should be proportional to Ba. In the laboratory,
a new TRMlab can be imparted to the rock in a known field
Blab. If the proportionality constant between magnetizing
field and magnetization is in both cases the same, the
strength of the ancient geomagnetic field Ba ¼
Blab(TRMa/TRMlab). This experimental procedure is the
foundation of the initial and most reliable technique for
absolute paleointensity determination (Thellier and
502 MAGNETIC CHRONOLOGY
Magnetic Chronology, Figure 1 Geomagnetic polarity time scale with magnetic anomaly numbers for the past 160 Myr (Figure from
Merril et al., 1998).
Thellier, 1959). Fired archaeological materials and volca-
nic rocks should be therefore most suitable to determine
absolute paleointensity.
DRM or pDRM acquisition is a complex process that
cannot be reproduced exactly in laboratory experiments.
Non-field effects like mineralogy and amount or magnetic
domain state of the ferromagnetic materials present in a sed-
imentary sample must be removed from the magnetization
signal in order to isolate the dependence of DRM on the
ancient geomagnetic field intensity. Typically, the intensity
of the remanent magnetization of a sample is divided by
a factor that is sensitive to the amount of remanence-
carrying grains present in the sample, like magnetic suscep-
tibility or specific laboratory-induced remanence types. The
normalized DRM or pDRM obtained this way will at best
reflect the relative intensity of the applied geomagnetic field
and is termed relative paleointensity. Results from globally
distributed records allowed the construction of weighted
averages of many relative paleointensity records which
can be used as dating and millennial-scale correlation tools
(e.g., Guyodo and Valet, 1999), and ongoing studies are
extending paleointensity stacks further back in time (e.g.,
Valet et al., 2005).
Magnetic chronology
Time variations of the EMF take place in a wide range of
periodicities, and the determination of polarities, direc-
tions, and intensities of remanent magnetization in paleo-
magnetic and paleointensity studies can be used as
a powerful chronological tool. Magnetic chronology
methods and techniques differ, however, according to the
age and thus the time resolution needed.
Geomagnetic polarity reversals
One of the main characteristics of the Earth’s magnetic
field is that it has reversed its polarity many hundreds of
times. A polarity reversal is a global event, experienced
simultaneously all over the Earth. Thus, geomagnetic
reversals provide a convenient means of stratigraphic cor-
relation and dating. The correlation of polarity intervals
with biostratigraphic stage boundaries associated to radio-
metric ages yields a dated geomagnetic polarity time scale
(GPTS) (Figure 1). The development of the GPTS was
initiated in the 1960s combining polarity data from
globally distributed Pliocene-Pleistocene igneous
rocks with ages obtained from the newly developed
 MAGNETIC CHRONOLOGY
potassium-argon (K–Ar) dating method (e.g., Cox
et al., 1964). The precision of the radiometric dating put,
however, a practical limit to the application of this tech-
nique, as the error in determining the age of a lava sample
that is about 5 Ma old was longer than the duration of many
short events. The paleomagnetic study of deep-sea cores, in
which sedimentary horizons could be paleontologically
dated, allowed extending the polarity time scale into the
Miocene (e.g., Opdyke et al., 1974). More recently, strati-
graphic sequences have also been dated using climatically
induced changes in lithology or stable isotopic records in
sediments that are caused by variations in the Earth’s orbit
around the sun. This method is known as astrochronology,
and astronomical tuning has allowed for a much more
detailed and accurate chronology for the younger Cenozoic
polarity reversals (Shackleton et al., 1990).
Seafloor spreading has created sequences of lineated
marine magnetic anomalies in the ocean basins which pro-
vide the most complete and detailed record of geomagnetic
polarity in the time interval between Middle Jurassic – the
oldest seafloor – and present. Investigations of magnetic
anomalies in all major oceanic areas yield a clear, consistent
record of the history of geomagnetic polarity during the past
155–160 Ma. (e.g., Lowrie, 2007a, b and references
therein; Figure 1). Because the age of the oldest seafloor
is about 180 Ma and magnetic anomalies in these areas dis-
play subdued amplitudes, early polarity intervals have to be
verified using magnetostratigraphic sections obtained from
land exposures (e.g., Channell et al., 1995). Knowledge of
geomagnetic polarity history older than late Jurassic is,
however, still irregular.
The GPTS is a useful tool for dating stratigraphic
sequences. By measuring the magnetization recorded in
samples from a section and correlating the obtained
polarity pattern to the GPTS, a precise temporal frame-
work for the studied sequence can be obtained. This pro-
cedure is called magnetostratigraphy (e.g., Opdyke
and Channell, 1996), and it allows correlating strati-
graphic information on a global basis. Although the num-
ber of polarity reversals occurred during the Quaternary
is small, their study has also been useful by providing
a temporal reference in studies concerning hominin
presence (e.g., Parés and Pérez-Gonzlez, 1995;
Calvo-Rathert et al., 2008).
Geomagnetic excursions
Recognition of brief polarity excursions as a part of the
Earth’s paleomagnetic field behavior has developed dur-
ing the last two decades together with astrochronological
calibration of the polarity time scale. During this time,
numerous geomagnetic excursions have been discovered
in the previously believed stable last two chrons, Brunhes
and Matuyama (e.g., Laj and Channell, 2007), although
their number is still a matter of debate. Singer et al. (2002)
propose the development of a geomagnetic instability time
scale (GITS) for the Brunhes and late Matuyama chrons
(Figure 2).
503
Although theoretically geomagnetic excursions could
be a useful correlation and dating tool in Quaternary stra-
tigraphy, their utilization can be rather problematical.
Questions like if they are local or global in extent or if they
are mainly due to dipolar or non-dipolar field variations
(e.g., Roberts, 2008) can impose severe limitations to their
applicability to chronological purposes. Merril and
McFadden (2005) recommend the use of excursions as
a correlation tool only for locations separated less than
30 on the Earth’s surface, as they are not global field fea-
tures. On the other hand, as features of the geomagnetic
field like the Laschamp or Iceland Basin excursions repre-
sent full polarity reversals and all adequately defined
excursions are characterized by a decrease in intensity
(e.g., Laj and Channell, 2007), they can be valuable in
Quaternary geochronology if used with caution. Never-
theless, their short duration (less than 5 ky, e.g., Laj and
Channell, 2007) also imposes limitations to their applica-
bility to chronological purposes, as they can be filtered out
of sedimentary records (Merril and McFadden, 2005).
Archaeomagnetism and paleosecular variation
The study of the magnetization found in archaeological
objects – and mainly of the TRM recorded in materials
which have undergone firing at the time of their fabrica-
tion or use – is called archaeomagnetism. Retrieval of the
direction of magnetization requires that the studied objects
are in situ or their original position is known. Other arti-
facts may have been fired in an unknown position and
only their paleointensity (called archaeointensity) can be
determined. Also lava flows whose age is known from his-
torical records can be used for archaeomagnetic studies.
When the age of the object or its last firing is known –
by methods such as radiocarbon dating, thermolumines-
cence, and the use of historical records – the
archaeomagnetic and archaeointensity data obtained can
be incorporated into a master curve for the region where
the remanence was acquired (Figure 3). The establishment
of regional archaeomagnetic records needs that individual
determinations from different locations are reduced to
a common one, following the approximation that the geo-
magnetic field is described by a tilted dipole (Noël and
Batt, 1990). Due to the importance of non-dipole field
components in secular variation, master curves for direc-
tional studies can be only constructed for areas of subcon-
tinental size, of not more than about 1,000 km radius.
Spatial variations in intensity are usually modeled using
an axial geocentric dipole model of the geomagnetic field.
Archaeomagnetic dating is performed by comparing
directional and intensity data of an archaeological artifact
of unknown age with a reference curve (Figure 3).
Secular variation that predates magnetic observatories
is often referred to as paleosecular variation (PSV). The
most suitable material for PSV studies are sediments
which can provide a high resolution and a good continu-
ity, like oceanic sediments of continental margins or lake
sediments. In PSV studies in lakes, several cores are
504 MAGNETIC CHRONOLOGY

Magnetic Chronology, Figure 2 The geomagnetic instability time scale (GITS) for the Brunhes chron. Column at left shows the
relative paleointensity SINT-2000 stack (Valet et al., 2005). Ages in the GITS are from 40Ar/39Ar dated lava flows and
astrochronologically dated Ocean Drilling Project (ODP) cores (Singer et al., 2014 and references therein, figure modified from Singer
et al., 2014).

usually obtained and correlated using physical character-
istics of the sediments like magnetic susceptibility or
other lithological markers. Age control is provided
by varve counting, isotopic dating on organic material,
palynology, and identification of tephra levels. Regional
declination, inclination, and intensity of PSV master
curves can be obtained by stacking of PSV records of
several lakes of the same region (e.g., Snowball
et al., 2007).
Several global models using spherical harmonic analy-
sis have been constructed using archaeomagnetic data,
lava flows, and lake sediment records as input data during
the last decade (e.g., Korte and Constable, 2009). Specific
regional models of secular variation have been developed
from archaeomagnetic and sedimentary data, which are
fitted to series of mathematical functions (e.g., Pavón-
Carrasco et al., 2010). This kind of model allows deter-
mining a secular variation curve for a chosen location
within the study region.
Magnetic paleointensity stratigraphy
The dominantly dipolar nature of geomagnetic
paleointensity variations provides a global signal not only
 MAGNETIC CHRONOLOGY 505

Magnetic Chronology, Figure 3 Secular variation curve for Iberia obtained by Bayesian modeling of data coming from localities less
than 900 km from Madrid: (a) stereoplot of declination and inclination, where orange and red indicate century time scales. A thin line
indicates curve sectors based on a small number of data, (b) declination versus time, and (c) inclination versus time (Figure from
Gómez-Paccard et al., 2006, this figure is reproduced with permission of John Wiley & Sons, Inc.).

during polarity reversals but also within periods of stable Summary

polarity. This signal can be used to supply a high-
resolution (millennial-scale) time scale for Quaternary
chronostratigraphy (e.g., Roberts et al., 2013). Sedimen-
tary sequences allow obtaining continuous records of rel-
ative geomagnetic paleointensity variations (Figure 2).
Detailed relative paleointensity stacks now span the entire
Quaternary, and some records or stacks of records from
rapidly deposited sediments provide millennial-scale
chronological resolution (e.g., Roberts et al., 2013 and ref-
erences therein). Recognition that precisely dated, detailed
paleointensity records are coherent on a global scale is
leading to the use of relative paleointensity as a means to
constrain the chronology of a sedimentary sequence, often
termed paleointensity-assisted chronology (e.g., Roberts
et al., 2013).
Direction and strength of the Earth’s magnetic field vary
with time. The changes produced by the main field, which
is generated by magnetohydrodynamic processes in the
Earth’s liquid iron outer core, display periods of the order
of a year to millions of years. Analysis of the remanent
magnetization recorded by rocks, sediments, and archaeo-
logical materials allows retrieving direction and intensity
of the ancient magnetic field which produced the
remanence.
In order to be useful as a dating tool, the changes of
direction and intensity of the EMF both with time and with
location – if variations are of regional character – have to
be known. If EMF variation reference curves or data have
been obtained and dated with other methods (radiometric
506 MAGNETIC CHRONOLOGY
dating, fossil record, etc.), they can be compared with
the polarity, directional, or intensity record of the studied
geological units or archaeological artifacts to be dated.
Magnetic chronology methods and techniques differ,
however, according to the age and the time resolution
needed.
Bibliography
Brachfeld, S., 2007. Paleointensity, relative, in sediments. In
Gubbins, D., and Herrero-Bervera, E. (eds.), Encyclopedia of
Geomagnetism and Paleomagnetism. Dordrecht: Springer, pp.
758–765.
Butler, R. F., 1992. Paleomagnetism: Magnetic Domains to Geo-
logic Terranes. Boston: Blackwell Scientific, p. 319.
Calvo-Rathert, M., Goguitchaichvili, A., Sologashvili, D., Villalaín,
J. J., Bógalo, M. F., Carrancho, A., and Maissuradze, G., 2008.
New paleomagnetic data from the hominin bearing Dmanisi
paleo-anthropologic site (southern Georgia, Caucasus). Quater-
nary Research, 69(1), 91–96.
Channell, J. E. T., Erba, E., Nakanishi, M., and Tamaki, K., 1995.
Late Jurassic-early cretaceous time scales and oceanic magnetic
anomaly block models. In Berggren,W. A., Kent, D. V., Aubry,
M. P., and Hardenbol, J. (eds.), Geochronology Time Scales
and Global Stratigraphic Correlation. Tulsa: SEPM. SEPM
(Society for Sedimentary Geology) Special Publication No. 54,
51–63.
Cox, A. V., Doell, R. R., and Dalrymple, G. B., 1964. Reversals of
the Earth’s magnetic field. Science, 144, 1537–1543.
Gómez-Paccard, M., Chauvin, A., Lanos, P., McIntosh, G.,
Osete, M. L., Catanzariti, G., Ruiz-Martínez, V. C., and
Nuñez, J. I., 2006. First archaeomagnetic secular variation for
the Iberian peninsula: comparison with other data from western
Europe and with global geomagnetic field models. Geochemis-
try, Geophysics, Geosystems, 7, 12, doi:10.1029/2006
GC001476.
Guyodo, Y., and Valet, J. P., 1999. Global changes in intensity of the
Earth’s magnetic field during the past 800 kyr. Nature, 399,
249–252.
Harland, W. B., Armstrong, R. L., Cox, A. V., Craig, L. E., Smith,
A. G., and Smith, D. G., 1990. A Geological Time Scale 1989.
Cambridge, UK: Cambridge University Press, p. 263.
Jackson, A., Jonkers, A. R. T., and Walker, M. R., 2000. Four cen-
turies of geomagnetic secular variations from historical records.
Philosophical Transactions of the Royal Society A, 358,
957–990.
Korte, M. F. D., and Constable, C. G., 2009. Geomagnetic field for
0–3 ka: 2. A new series of time varying global models. Geo-
chemistry, Geophysics, Geosystems, 10, Q06008.
Laj, C., and Channell, J. E. T., 2007. Geomagnetic excursions. In
Kono, M. (ed.), Treatise in Geophysics. Amsterdam: Elsevier,
Vol. 5, pp. 373–416.
Lowrie, W., 2007a. Geomagnetic polarity timescales. In Gubbins, D.,
and Herrero-Bervera, E. (eds.), Encyclopedia of Geomagnetism
and Paleomagnetism. Dordrecht: Springer, pp. 328–333.
Lowrie, W., 2007b. Magnetostratigraphy. In Gubbins, D., and
Herrero-Bervera, E. (eds.), Encyclopedia of Geomagnetism and
Paleomagnetism. Dordrecht: Springer, pp. 664–670.
Merril, R. T., and McFadden, M. W., 2005. The use of magnetic
field excursions in stratigraphy. Quaternary Research, 63,
232–237.
Merril, R. T., McElhinny, M. W., and McFadden, P. L., 1998. The
Magnetic Field of the Earth: Paleomagnetism, the Core and
the Deep Mantle. San Diego: Academic Press. International
Geophysics Series, 63, p. 531.
Néel, L., 1949. Théorie du traînage magnétique des ferromag-
nétiques en grains fins avec applications aux terres cuites.
Annales de Geophysique, 5, 99–136.
Néel, L., 1955. Some theoretical aspects of rock magnetism. Philo-
sophical Magazine. 4, 191–242.
Noël, M., and Batt, C. M., 1990. A method for correcting geograph-
ically separated remanence directions for the purpose of
archaeomagnetic dating. Geophysical Journal International,
102, 753–756.
Opdyke, N. D., and Channell, J. E. T., 1996. Magnetic Stratigraphy.
San Diego: Academic. International Geophysics Series, Vol. 64.
Opdyke, N. D., Burkle, L. H., and Todd, A., 1974. The extension of
the magnetic time scale in sediments of the Central Pacific
Ocean. Earth and Planetary Science Letters, 22, 300–306.
Parés, J. M., and Pérez-Gonzlez, A., 1995. Paleomagnetic age for
hominid fossils at Atapuerca archaeological site, Spain. Science,
269(5225), 830–832.
Pavón-Carrasco, F. J., Osete, M. L. and Torta, J. M., 2010. Regional
modeling of the geomagnetic field in Europe from 6000 to
1000B.C. Geochemistry, Geophysics, Geosystems, 11(11), 1–20.
Roberts, A. P., 2008. Geomagnetic excursions: knowns and
unknowns. Geophysical Research Letters, 35, L17307.
Roberts, A. P., Tauxe, L., and Heslop, D., 2013. Magnetic
paleointensity stratigraphy and high-resolution quaternary geo-
chronology: successes and future challenges. Quaternary Sci-
ence Reviews, 61, 1–16.
Shackleton, N. J., Berger, A., and Peltier, W. R., 1990. An alterna-
tive astronomical calibration of the lower Pleistocene timescale
based on ODP site 677. Transactions of the Royal Society of
Edinburgh: Earth Sciences, 81, 251–261.
Singer, B. S., Relle, M. R., Hoffman, K. A., Battle, A., Guillou, H.,
Laj, C., and Carracedo, J. C., 2002. Ar/Ar ages of transitionally
magnetized lavas on La Palma, Canary Islands, and the geomag-
netic instability timescale. Journal of Geophysical Research,
107(B11), 2307.
Singer, B. S., Guillou, H., Jicha, B. R., Zanella, E. and Camps, P.,
2014. Refining the quaternary geomagnetic instability timescale
(GITS): lava flow recordings of the Blake and Post-Blake excur-
sions. Quaternary Geochronology, 21, 16–28.
Snowball, I., Zillén, L., Ojala, A., Saarinen, T., and Sandgren, P.,
2007. Fennostack and Fennorpis: varve dated Holocene
palaeomagnetic secular variation and relative palaeointensity
stacks for fennoscandia. Earth and Planetary Science Letters,
255, 106–116.
Tarduno, J. A., Cottrell, R. D., Watkeys, M. K., Hofmann, A.,
Doubrovine, P. V., Mamajek, E. E., Liu, D. J., Sibeck, D. G.,
Neukirch, L. P., and Usui, Y., 2010. Geodynamo. Solar wind,
and magnetopause 3.4 to 3.45 billion years ago. Science,
327(5970), 1238–1240, doi:10.1126/science.1183445.
Tauxe, L., 2010. Essentials of Paleomagnetism. Berkeley: Univer-
sity of California Press.
Thellier, E., and Thellier, O., 1959. Sur l’intensité du champ
magnétique terrestre dans le passé historique et géologique.
Ann. Geophys, 15, 285–376.
Valet, J. P., Meynadier, L., and Guyodo, Y., 2005. Geomagnetic
dipole strength and reversal rate over the past two million years.
Nature, 435, 802–805.
Cross-references
Archaeomagnetic Dating
Geomagnetism
Magnetic Anomalies
Magnetostratigraphic Dating
Sedimentary Rocks (Rb-Sr Geochronology)
Sediments, Terrestrial (Paleomagnetism)
Volcanogenic Sedimentary Rocks (K/Ar, 40Ar/39Ar)
 MAGNETOSTRATIGRAPHIC DATING
MAGNETOMETER
Vicente Soler Javaloyes
Estación Volcanológica de Canarias, IPNA-CSIC,
La Laguna, Spain
Definition
An instrument used to measure the intensity and direction
of the remanent magnetization of specimens.
In the middle of the last century, the astatic magnetom-
eter was widely used in paleomagnetic laboratories for the
measurement of the natural remanent magnetization
(NRM). It is based on a magnet pair that is equally and
oppositely magnetized, and this magnet system twists in
response to the direction and intensity of magnetization
of the specimen beneath it. The measurement limit of these
instruments was about 10
3 amperes per meter (Am
1), for
a standard 2.5  2.1 cm cylinder (Collinson, 1983).
Today, the most common instruments used for the mea-
surement of the NRM are spinner and cryogenic magne-
tometer, like those manufactured by Agico, 2G
Enterprises, and Cryogenic Limited, among others.
Many different spinner magnetometers have been
developed, but they all involve spinning a rock specimen
within or adjacent to a magnetic field sensor (typically
a conductive pickup coil or fluxgate element). This rapid
rotation places significant demands on the physical com-
petence of specimens. The models available commercially
can measure magnetizations less than 10
5 Am
1
(McElhinny and McFadden, 2000).
Cryogenic magnetometers use a detector called
a SQUID (superconducting quantum interference device)
that is maintained at liquid helium temperatures and are
based on the properties of a superconducting ring. This
instrument has greater sensitivity and speed of measure-
ment than spinner magnetometers and can measure
magnetizations of 10
6 Am
1.
Pass-through cryogenic magnetometers are now avail-
able that can be used to measure “U-channels” and either
whole cores or split half-cores, without the necessity of
physically sampling the cores.
The main disadvantage of cryogenic magnetometers is
the requirement of having to keep the sensing coil super-
conductive, since this normally requires immersion in liq-
uid helium.
Recently, 4.2 K cryocoolers have been used without the
need of liquid helium. High-temperature SQUIDs are able
to operate at liquid nitrogen temperatures, but the
increased thermal noise prevents them from attaining the
sensitivities available for conventional SQUIDs
(Williams, 2007).
A new magnetometer is now under development, the
spin-exchange relaxation-free (SERF) atomic magnetom-
eter, based on the precession of the spins of alkali atoms
in the vapor phase. It is anticipated that SERF atomic mag-
netometers could achieve sensitivities exceeding those of
SQUID devices (Shah et al., 2007).
507
Bibliography
Collinson, D. W., 1983. Methods in Rock Magnetism and
Palaeomagnetism: Techniques and Instrumentation. London:
Chapman & Hall. 503 pp.
McElhinny, M. W., and McFadden, P. L., 2000. Paleomagnetism:
Continents and Oceans. San Diego: Academic, p. 386.
Shah, V., Knappe, S., Schwindt, D. D., and Kitching, J., 2007.
Subpicotesla atomic magnetometry with a microfabricated
vapour cell. Nature Photonics, 1, 649–652.
Williams, W., 2007. Magnetometers, laboratory. In Gubbins, D.,
and Herrero-Bervera, E. (eds.), Encyclopedia of Geomagnetism
and Paleomagnetism. Dordrecht: Springer. 1054 pp.
MAGNETOSTRATIGRAPHIC DATING
Miguel Garces
University of Barcelona, Barcelona, Spain
Definition
Magnetostratigraphy relies on the ability of sedimentary
rocks to acquire a remanent magnetization when they
form, which parallels the direction of the Earth’s ambient
magnetic field. Since the geomagnetic field has undergone
multiple nonperiodic reversals through Earth history,
a magnetic zonation of sedimentary sequences is feasible
according to the polarity of the rock magnetization.
A magnetostratigraphic zonation allows dividing the strat-
igraphic record into time slices which can be correlated
worldwide. Magnetostratigraphic dating refers to the iden-
tification in the stratigraphic record of magnetozones,
which can be correlated to age-equivalent geomagnetic
chrons. The compiled absolute ages of geomagnetic
chrons form the basis for the geomagnetic polarity time
scale (GPTS), which is routinely revised with new radio-
metric or astronomic calibrations (Figure 1).
Essential concepts
The magnetization of sedimentary rocks
Some minerals in nature can behave as magnets, that is,
they exhibit spontaneous magnetization. They rarely rep-
resent a significant volume in sedimentary rocks, with
abundances typically below detection limits of X-ray dif-
fraction and grain size too small to be spotted by optical
microscopy. Their magnetization, however, can be nor-
mally measured in the laboratory using a state-of-the-art
rock magnetometer. Most common magnetic minerals in
sedimentary rocks belong to the group of iron oxides and
sulfides (Table 1).
The magnetization of an assemblage of magnetic parti-
cles (M) decays with geologic time (t) according to the
following equation:
M ðt Þ ¼ M 0 eð
t=tÞ , ð1Þ
where M0 is the initial magnetization and t (relaxation
time) is the time elapsed to decay to 1/e of the
508 MAGNETOSTRATIGRAPHIC DATING

Magnetostratigraphic Dating, Figure 1 Development of the geomagnetic polarity time scale of the last 12 Ma since marine magnetic
anomaly patterns was first studied (Heirtzler et al., 1968). The convention is that periods of normal polarity are represented in black and
reversed in white. Short lines to the right of the Cande and Kent (1995) polarity column correspond to cryptochrons. Event names are
J (Jaramillo), CB (Cobb Mountain), R (Reuinion), K (Kaena), M (Mammoth), C (Cochiti), N (Nunivak), S (Sidufjall), T (Thvera).

Magnetostratigraphic Dating, Table 1 Common magnetic initial magnetization. Following the Néel theory, relaxa-
minerals of application in paleomagnetism. Ms Saturation tion time is defined as
magnetization, Tc Curie temperature, Hcmax maximum coercivity
(Lowrie, 1990; Tauxe, 1998; Evans and Heller, 2003) 1 ½Kv=½kT 
t¼ e , ð2Þ
Mineral Formula Ms(kA/m) Tc ( C) Hcmax (T) C
where C is a frequency constant (~1010 s
1) and K is the
Magnetite Fe3O4 480 580 0.3
Hematite a-Fe2O3 ~2.5 675 >5
magnetic anisotropy factor (Tauxe, 1998). From this equa-
Maghemite g-Fe2O3 380 590–675 0.3 tion, it is derived that relaxation time is dependent on par-
Goethite a-FeOOH ~2 120 >5 ticle size (volume, v) and the thermal energy (kT). In the
Pyrrhotite Fe7S8 ~80 320 >0.1 case of equidimensional pure magnetite with size range
Greigite Fe3S4 ~125 ~330 >0.1 of 0.1–0.08 mm, relaxation time is greater than the age
of the Earth, but it sharply decreases to seconds for particle
 MAGNETOSTRATIGRAPHIC DATING
sizes lower than 0.06 mm. On the other hand, magnetite
particles larger than 1 mm gradually reduce their relaxation
time as their internal magnetic ordering shifts from a stable
single domain to a multi-domain state. Thus, to be useful
for paleomagnetic studies, magnetic minerals must occur
in the appropriate size range, so that magnetic remanence
remains stable over time scales comparable to the age of
the rocks.
The Natural Remanent Magnetization (NRM) of rocks
typically results from the addition of different magnetic
components acquired since the time of rock formation.
Primary magnetic components record the polarity of the
geomagnetic field at the time of rock formation and thus
provide the basic information for a magnetostratigraphic
zonation. Secondary magnetizations acquired in later
stage during geological history have no stratigraphic rele-
vance and need to be identified and clearly separated from
primary magnetizations.
The NRM signal in sedimentary rocks comprises three
main types: Detrital Remanent Magnetization (DRM),
Chemical Remanent Magnetization (CRM), and Thermal
Remanent Magnetization (TRM). A DRM is carried by
magnetic minerals incorporated into the sedimentary
rock as detrital particles, either as part of the terrigenous
fraction or from biogenic sources within the water col-
umn. If the magnetic grains are sufficiently small, their
magnetic moments counterbalance the gravitational
torques, producing a statistical particle alignment with
the ambient field. A Detrital Remanent Magnetization
(DRM) is blocked as early compaction reduces porosity
and prevents further mechanical rotation of magnetic
particles. The lock-in depth at which a DRM is fixed
varies from millimeter to meter scale and depends on
both magnetic mineralogy, rock type, and the sedimen-
tary environment. In pelagic sediments, a reasonable-
estimate of the lock-in depth is 10–20 cm below the
sediment-water interface (Opdyke and Channell, 1996).
If the time elapsed between sedimentation and DRM
acquisition is considered of relevance, the term post-
Depositional Remanent Magnetization (pDRM) may
be used. Sediment bioturbation of different kinds may
influence DRM, either allowing reorientation of
magnetic particles giving rise to a new delayed pDRM
or causing randomization of magnetic particles and eras-
ing a preexisting DRM. Common carriers of DRM are the
iron oxides magnetite and hematite, which resist chemi-
cal alteration as they are transported along the sediment
transfer system. Once a DRM is locked-in, its preserva-
tion in sediments is determined by burial conditions:
reducing environments that are associated with the pres-
ence of abundant organic matter can cause the dissolution
of preexisting magnetic particles. Despite the lock-in
depth uncertainty, DRM is taken as a reliable recorder
of the geomagnetic field at the time of rock formation
for most magnetostratigraphic purposes.
A Chemical Remanent Magnetization (CRM) is
acquired due to authigenic formation of magnetic min-
erals, with the magnetic moment being blocked when
509
the crystal grows through a critical size. Processes
leading to a CRM acquisition can occur in an early
post-depositional stage, but also in later stages of the rock
history. A CRM is suitable for magnetostratigraphy
as long as the acquisition takes place in an early
post-depositional stage. A typical CRM carrier in terres-
trial environments is hematite, which occurs as very
fine-grained cement filling the porosity of siliciclastic
sediments (redbeds). In marine environments, authigenic
magnetite may form as a product of oxidation of iron
sulfides, giving rise to a stable CRM which is relatively
frequent in environments influenced by changing redox
conditions. Finally, an early CRM carried by authigenic
iron sulfides can occur under specific burial conditions
(Vasiliev et al., 2005), but their occurrence in the geolog-
ical record is rare because these minerals easily undergo
later oxidation.
AThermal Remanent Magnetization (TRM) is acquired
when magnetic minerals cool below their Curie tempera-
ture (Table 1), the temperature at which magnetic minerals
change from paramagnetic into ferromagnetic (s.l.) behav-
ior. Magnetization is then free to align with the ambient
field until the minerals cool through their blocking tem-
perature, representing the temperature transition at which
relaxation time sharply increases from seconds to millions
of years, efficiently locking magnetic moments. The TRM
is typically carried by volcanic rocks forming the ocean
crust, as well as stratified lava flows. The net magnetic
moment of rocks carrying a TRM is several orders of mag-
nitude greater than sediments having a DRM. Most com-
mon carriers of TRM in igneous rocks belong to the
titanomagnetite solid solution series.
A time dependence exist for unblocking temperatures
such that a thermoviscous magnetization may be acquired
by rocks subjected to moderate temperatures over millions
of years. Occurrence of a thermoviscous remanent magne-
tization is not uncommon if long-term deep burial occurs
after sedimentation, particularly in regions of elevated
geothermal gradients. A partial to total resetting of earlier
(primary) magnetizations can result from thermoviscous
magnetization processes.
As shown above, common processes of rock magneti-
zation in nature are not uniquely related to the time of
rock formation. Therefore, the NRM of rocks is an aggre-
gate of multiple components of different age, each carried
by a characteristic population of magnetic particles. Iden-
tification of the different NRM components is crucial for
all paleomagnetic studies, and this is achieved by means
of diverse stepwise demagnetization techniques
(Langereis et al., 1989). Stepwise demagnetization
is based on the assumption that each component is carried
by a magnetic phase which has its own thermal stability
and/or coercivity spectra (Table 1). Ideally, the stability
fields of magnetic components do not overlap
completely, so that the direction of every component
can be determined. A wide overlap between components
may lead to errors in the interpretation of paleomagnetic
directional data.
510 MAGNETOSTRATIGRAPHIC DATING
The geomagnetic polarity time scale
The fact that the Earth’s magnetic field has undergone
reversals does not rely on direct observations. When Ber-
nard Brunhes first discovered in 1906 the existence of
reversely magnetized volcanic rocks, the finding was
interpreted as a self-reversal artifact related to the thermo-
remanence acquisition process. It was Motonori
Matuyama who later observed that a stratigraphic ordering
existed in the magnetic polarity of successive lava flows,
such as age dependence supporting the existence of past
geomagnetic reversals. The parallel development of the
radioisotope dating techniques allowed the birth of a first
K–Ar calibrated geomagnetic polarity time scale (GPTS)
back to 4 Ma (Cox et al., 1964). This first time scale was
based on pairs of radiometric ages and magnetic polarity
of lava flows distributed globally, a method that could
not be extended to older ages because at that time the dat-
ing error exceeded the duration of chrons. The use of bio-
stratigraphic correlation in deep sea cores allowed the
construction of composite magnetostratigraphic
sequences and further extension of the time scale to older
ages (Opdyke et al., 1966). But, it was the discovery of
seafloor magnetic anomalies which provided the unprece-
dented record of datable geomagnetic reversals back to
Late Jurassic (Heirtzler et al., 1968; Cande and Kent,
1992, 1995; Gee and Kent, 2007).
The transformation of seafloor magnetic anomaly pro-
files into absolute ages relies on age interpolation from
a selected number of radiometric dates (Figure 1). Cande
and Kent (1992) used nine tie points for age calibration
of the GPTS for the Late Cretaceous and
Cenozoic. More recent calibrations have included astro-
nomical ages of geomagnetic reversals of the Pliocene
(Cande and Kent, 1995), the Neogene (Gradstein
et al., 2004), and Paleogene (Gradstein et al., 2012)
(Figure 2). It is expected that future developments of the
GPTS will include astrochronological information of pro-
gressively older ages (Hinnov and Ogg, 2007). In order to
improve on both accuracy and precision of age calibration,
important efforts for synchronization of the radiometric
and astronomic time are needed (Kuiper et al., 2008).
Seafloor magnetic anomalies provide a quasi-
continuous record of geomagnetic reversals. However, in
the deconvolution of the magnetic anomaly signal into a
polarity sequence, a resolution limit exists which depends
on the depth and thickness of the magnetized layer in the
oceanic crust. In general terms, geomagnetic chrons of
105 kyr duration are well represented in anomaly profiles
acquired from the sea surface. But even better resolution
may be achieved in high-resolution profiles from fast-
spreading ridges (>90 km/Myr), where short polarity
intervals of few 104 year duration can be recognized
(Gee and Kent, 2007). Cande and Kent (1992) coined
the term “cryptochron” to refer to magnetic anomalies
which are too short to be unequivocally interpreted as true
reversals. Further evidence for short chrons not yet present
in the time scale can be obtained from high-resolution
magnetostratigraphic records (Channell et al., 2003;
Krijgsman and Kent, 2004). At present, stability of the
GPTS is not reached at this level of resolution, and it is
expected that more cryptochrons will be elevated in the
future to the category of new subchrons.
The occurrence of geomagnetic reversals follows
a random pattern with no signs of periodicity. This indicates
that geomagnetic reversal is essentially a process free of
memory. And it is precisely the random and global character
of geomagnetic reversal which makes magnetostratigraphic
zonations useful for long-distance correlations. The age
resolution of magnetostratigraphy depends on the average
frequency of geomagnetic reversals, which has not remained
stable through geological time (Gee and Kent, 2007). The
best resolution is achievable for the last 30 Ma, when the
mean duration of geomagnetic chrons was 0.25 Ma, increas-
ing to 0.75 Ma for ages between 30 and 80 Ma. On the other
hand, magnetostratigraphy is inadequate for some quiet
long periods when no reversals occurred, or one polarity
dominated, such as the Cretaceous Normal Superchron
(121–83 Ma) and the Carboniferous-Permian Reversed
Superchron (310–265 Ma) (Langereis et al., 2010).
The local magnetostratigraphy
Sampling
A magnetostratigraphic study requires that suitable lithol-
ogies are sampled at approximately regular intervals. In
principle, clay/silty sediments are best candidates to carry
a primary remanence because stable magnetization is car-
ried by small magnetic particles within the sub-mm size
range. Coarser sandy sediments may yield good results
as long as they contain significant amounts of clay matrix.
Low-permeability fine-grained sediments are also pre-
ferred because they are less affected by later fluid flow
and related remagnetization processes. As rock alteration
processes can add secondary spurious magnetizations,
fresh outcrops are highly recommended for paleomagnetic
sampling. Some lithologies, such as organic-rich soft sed-
iments, are prone to undergo alteration after sampling. In
these cases, sample refrigeration may help preservation
of their magnetic properties.
Paleomagnetic samples are taken in the field using
either a water/air cooled portable drill or a hand sample.
In soft sediments, the use of a hand pick is often required
to dig down to fresh material. Drilling cores in situ is
recommended because it is faster and simplifies sample
preparation work in the laboratory. Electric-, rather than
gasoline, powered drills are more appropriate for sampling
unconsolidated soft sediments. In all cases, samples must
be oriented before they are removed from the outcrop in
order to calculate directions in geographic coordinates.
In the case of tilted strata, bedding orientation data is
needed in order to restore paleomagnetic vectors into tilt-
corrected coordinates.
An optimal sampling resolution for magnetostra-
tigraphy is such that allows recovery of all geomagnetic
 MAGNETOSTRATIGRAPHIC DATING 511

Magnetostratigraphic Dating, Figure 2 Geomagnetic polarity chron zonation included in the last version of the geological time
scale of Gradstein et al. (2012) and detailed nomenclature adopted from Cande and Kent (1992).
512 MAGNETOSTRATIGRAPHIC DATING
reversals recorded in the rock sequence. Available age
constraints help constraining the most appropriate sam-
pling density, although other factors such as field logistics,
outcrop accessibility, and availability of suitable litholo-
gies determine the sampling resolution finally achieved.
Special care must be taken at this stage, since an incom-
plete sampling may give a polarity zonation that is
unrepresentative of the true pattern of reversals.
Drill cores from marine and lacustrine unconsolidated
sediments often provide unaltered material suitable for
paleomagnetic measurements. Leading technologies used
in scientific drilling have made possible full core recovery
in soft sediments, providing the best material for very
detailed magnetostratigraphic studies, where the resolu-
tion limit is marked by the sample size (7 cm3) or, for con-
tinuous core measurements, the width of the response
function of the magnetometer pick-up coils. In addition,
advanced tools exist that provide azimuthal orientation
to core samples, thus allowing full correction of paleo-
magnetic vectors into geographic coordinates.
Laboratory processing
The NRM of samples is measured in the laboratory with
rock magnetometers specially designed for paleomagnetic
applications. Superconducting rock magnetometers are
best suited for magnetostratigraphy because of their adapt-
ability to standard paleomagnetic samples, cores, and
U-channel samples. They provide the highest sensitivity
and best performance with in-line automated measuring
routines. An alternative instrument is the inexpensive
spinner magnetometer, which has sufficient sensitivity to
measure many sedimentary rocks, but requires more care-
ful sample preparation and handling.
Treatment of paleomagnetic samples in the laboratory
is oriented at identifying the different components contrib-
uting to the NRM. Widely applied techniques include
stepwise thermal (TH) and alternating field (AF) demag-
netization (Langereis et al., 1989; Butler, 1992). Sample
demagnetization is achieved by keeping the samples in a -
low-field environment, typically less than 10 nTesla,
while subjected to TH or AF treatment. Complete thermal
demagnetization is accomplished after exposure to
a number of increasing temperature steps, until the maxi-
mum unblocking temperature of magnetic carriers is
reached. Assuming that every NRM component is carried
by a different population of magnetic grains, an effective
isolation of components is achieved if their unblocking
temperature spectra do not overlap significantly. TH
demagnetization is optimal for removing secondary com-
ponents carried by hydroxides such as goethite, because
of its low unblocking temperature (Table 1). Thermal sep-
aration of fine-grained hematite and magnetite compo-
nents needs closely spaced temperature steps because of
the overlap of their stability fields. Alternatively, AF
demagnetization successfully isolates magnetite from
hematite components because they exhibit a large contrast
in coercivity. Ideally, a set of pilot samples of
representative lithologies is both TH and AF
demagnetized, so that a specific demagnetization protocol
can be designed for each case study.
Among the NRM components isolated after stepwise
demagnetization, the most stable and consistent compo-
nent is referred to as the Characteristic Remanent Magne-
tization (ChRM). Principal component analysis
(Kirschvink, 1980) is used to calculate the orientation of
the ChRM, which is expressed with their values of decli-
nation (azimuth) and inclination (dip). By application of
the geocentric axial dipole hypothesis, and given a site
location on the Earth’s surface and the orientation of
a paleomagnetic component, the latitude of the
corresponding virtual geomagnetic pole (VGP) can be cal-
culated (Butler, 1992; Tauxe, 1998). Northern hemisphere
VGP latitudes (positive) correspond to normal polarity
and southern hemisphere latitudes (negative) to reversed
polarity. A local magnetic polarity stratigraphy can be
constructed after plotting VGP latitudes as a function of
stratigraphic position of samples (Figure 3).
Paleomagnetic data from drill cores not azimuthally
oriented can be used for magnetostratigraphic purposes
as long as the drill hole is perpendicular to bedding. Infor-
mation of magnetic polarity is then given by the vertical
component (Z) of magnetization. In this case, special
attention must be given to the latitude at which the sedi-
ments originated; the Z component tends to zero at equato-
rial latitudes, and its sign is no longer a good indicator of
polarity. At low latitudes, azimuthal orientation of cores
is needed because information of magnetic polarity is best
extracted from the paleomagnetic declination.
The reliability of a given magnetostratigraphy relies on,
first, the demonstrable primary nature of the ChRM and,
second, the completeness of the polarity sequence. To
assess the age of the magnetization, some stability tests
(fold test, conglomerate test, consistency test, and reversal
test) can be carried out (Butler, 1992; Opdyke and
Channell, 1996; Tauxe, 1998). The fold test provides
information on the timing of magnetization relative to
folding. A fold test is positive if paleomagnetic directions
from opposing flanks of a fold converge after tilt correc-
tion. A positive fold test indicates that remanence predates
folding, but does not strictly demonstrate a primary
(synsedimentary) origin as required for magnetostra-
tigraphy. Only in cases of early synsedimentary folding,
or slumped sediments, a positive fold test provides greater
confidence on the primary origin of the magnetization.
The conglomerate test assesses the stability of the magne-
tization measured from intraformational clasts. Random
paleomagnetic directions from different clasts indicate
that the magnetization has not been reset since the clasts
were reworked from its original strata, thus yielding
a positive test. The consistency test is applied for multiple
overlapping magnetostratigraphic sections. A sequence of
reversals that can be traced laterally, and cross-checked
with lithostratigraphy, gives the strongest evidence of
a primary magnetization. The reversal test is used to check
for antipodality of the average normal and reversed
 MAGNETOSTRATIGRAPHIC DATING
Magnetostratigraphic Dating, Figure 3 Magnetostratigraphy
of the Librilla section, Fortuna basin, Spain (Garcés et al., 2001).
A polarity sequence is determined from the Virtual Geomagnetic
Pole latitude, positive VGP latitudes represent normal polarity,
and negative VGP latitudes reverse polarity.
paleomagnetic directions. A failed reversal test indicates
that stepwise demagnetization has not succeeded in fully
isolating discrete magnetic components. Passing the
reversal test is critical for studies in which the angular
deviation from a reference paleomagnetic direction is used
to interpreted tectonic rotation of rock units. On the other
hand, the reversal test is not so crucial for magnetostra-
tigraphy since the magnetic polarity can be often assessed
from partially overlapped paleomagnetic directions.
513
Magnetostratigraphic Dating, Figure 4 Magnetostratigraphic
jackknife of the Librilla section. A J value between 0 and
0.5
indicates that the probability of missing magnetic reversals due
to sampling resolution is low.
Assessing completeness of local
magnetostratigraphy
A magnetostratigraphic section is considered complete
when the sampling density is enough to support that no
geomagnetic reversals are missed between sampled
horizons. Note that, in this sense, magnetostratigraphic
completeness is different from stratigraphic completeness,
the latter referring to the occurrence and duration of
sedimentary (¼time) gaps. Completeness of local
magnetostratigraphy can be statistically assessed by differ-
ent means. The dependence of a given magnetic polarity
sequence on sampling strategy was investigated by Johnson
and McGee (1983), who concluded that polarity sequences
yielding an average of 8 sites/magnetozone are insignifi-
cantly affected by sampling bias. Tauxe and Gallet (1991)
proposed a statistical jackknife in order to test magnetostra-
tigraphic completeness. The sensitivity to sample resolution
is described by a parameter J, which represents the slope
relating the percentage of magnetozones missed after
a random and progressive deletion of samples up to 20 %
of the total number of samples (Figure 4). These authors
recommend J ranges between 0 and
0.5, lower values
indicating that the section was insufficiently sampled.
Magnetostratigraphic correlation
The random character of geomagnetic reversals originates
a distinctive polarity sequence when a sufficient length of
time is observed, which depends on the average frequency
of reversals. For Neogene times, it can be estimated that
a stratigraphic sequence spanning about 3 Myr will reveal
514 MAGNETOSTRATIGRAPHIC DATING

Magnetostratigraphic Dating, Figure 5 (a) Average sedimentation rates in continental sediments as a function of time span of
observation (duration of chron) (Data from magnetostratigraphic studies in Cenozoic basins of Spain). Note that sedimentation rates
cluster to an order of magnitude when data from a single basin is considered. (b) Ratio of duration of adjacent chrons in the Cenozoic
compared to ratios of sedimentation rates of adjacent chrons (Data from the Ebro Basin).

a magnetic polarity sequence of 10–15 reversals, an ade-
quate number to assess a correlation. A correlation with
the geomagnetic polarity time scale (GPTS) is achieved
if the pattern of magnetozones finds a positive match with
a sequence of reversals in the GPTS. In achieving this,
absolute ages of geomagnetic reversals can be assigned
to every magnetozone boundary in the local stratigraphy.
Magnetostratigraphic correlation with the GPTS is
often assisted by absolute age constraints, for example,
radiometric ages from interstratified ash layers or lava
flows, or biostratigraphic data which may provide
chronostratigraphic information at stage level or better.
But attention must be paid to the diverse sources of error
associated to external age constraints. In the case of radio-
isotopic dating, only the analytical precision (often less
than 0.5 %) is propagated into the error reported (Renne,
1998). Radioisotope decay constant uncertainties and
other sources of systematic errors, which are often
unnoticed by non specialists, are particularly problematic
when different isotopic systems are combined, as occurs
in the calibration of the GPTS. A recommended practice
is that best correlation is searched within a time range only
approximated by external age constrains, but not forced by
a single tie point of presumed high accuracy. A correlation
which is not supported by a positive match of the
magnetostratigraphic pattern and is solely substantiated
on an external age constraint must be taken with caution.
Achieving a robust correlation between a local
magnetostratigraphy and the GPTS requires that sedimen-
tation rates do not change significantly at the time scale of
geomagnetic chrons, so that the relative thickness of
magnetozones maintains proportionality with the duration
of corresponding geomagnetic chrons. From
a sedimentological perspective, sedimentation steadiness
may appear as severe restriction for magnetostratigraphic
correlation, particularly in continental settings dominated
by alluvial/fluvial processes of clearly episodic nature.
A compilation by Sadler (1981) showed that for typical
durations of chrons (104–106 year), average accumulation
rates of fluvial sediments may vary by three orders of mag-
nitude. The scattering, however, partially responds to the
diverse geologic (tectonic, paleogeographic, climatic) set-
tings from which the database was constructed. Observa-
tions restricted to a specific setting and location result in
sedimentation rates which cluster to within an order of
magnitude (Figure 5a), as already observed by (Johnson
et al., 1985) in the Siwaliks Himalayan foredeep. More
important for correlation purposes is the relative change
of sedimentation rate between adjacent magnetozones.
Observations from the alluvial sequences of the Ebro
Basin shows that sedimentation rates rarely change by
a factor greater than 2 between two adjacent chrons, while
the relative duration of adjacent chrons of the GPTS varies
over an order of magnitude (Figure 5b). These observa-
tions lead to the conclusion that changes in sedimentation
rate rarely originate an excessive mismatch between the
local magnetostratigraphy and its equivalent reversal
sequence in the GPTS (Figure 6).
In the long term (>107 year), average accumulation rates
are fundamentally constrained by basin forming mecha-
nisms and the resulting balance between sediment yield
and accommodation space. At this scale, significant changes
of the tectonic, paleogeographic, and/or paleoclimatic set-
ting may lead to gradual or even abrupt changes in the sed-
imentation trends. Fortunately, periods of sedimentary
reorganization, which impact on a specific site location,
are few compared to times of stability. The result is that evo-
lution of basin infill often reveals long periods of steady sed-
imentation, these being interrupted by short periods of
change (Figure 7). Under these circumstances, unambigu-
ous magnetostratigraphic correlation with the GPTS, inde-
pendent of external age controls, is still an achievable goal.
 MAGNETOSTRATIGRAPHIC DATING 515

Magnetostratigraphic Dating, Figure 6 A magnetochronology of the Fortuna basin infill based on multiple magnetostratigraphic
sections (Garcés et al., 2001). Basin scale correlations are guided by both lithostratigraphic boundaries (diachrony allowed) and
biostratigraphic markers (MN12, MN13, MN14 correspond to European mammal zones). A magnetostratigraphic correlation between
overlapping sections yields a positive consistency test, supporting a primary magnetization. The resulting composite magnetostratigraphy
is made of 19 magnetozones (10 Normal, 9 Reverse), with a pattern of reversals which allows unambiguous and independent correlation
with the GPTS. Note that some external (biostratigraphic and radiometric) age constrains do not match with the proposed correlation. In
particular, K/Ar data of the Barqueros volcanism (Bellon et al., 1983) and calcareous nanoplankton from the Rio Chicamo (Lancis et al., 2010)
suggest an age >1Myr younger for the lower marine units. On the other hand, the Ar/Ar dating of the Cabezos Negros volcanism (Kuiper
et al., 2006) supports the proposed magnetostratigraphic correlation. The Tortonian Salinity Crisis (TSC) corresponds to a tectonic-driven
basin restriction event which caused the precipitation of evaporates in the Fortuna and Lorca basins (Krijgsman et al., 2000). The Messinian
Salinity Crisis (MSC) of the Mediterranean culminated with a base level drop which caused erosion of the Mediterranean peripheral
margins. The MSC in the Fortuna basin caused fluvial incision which was later flooded during the Pliocene transgression (Garcés
et al., 2001).
516 MAGNETOSTRATIGRAPHIC DATING
Magnetostratigraphic Dating, Figure 7 Sedimentation rates in
the Fortuna basin derived from magnetostratigraphic
correlation (Figure 6). The basin underwent a complex evolution,
with a paleogeography which rapidly changed from open
marine to a restricted elongated gulf and, finally, continental
environments eventually altered by the sea level drop of the
Mediterranean during the MSC. Despite this unfavorable
scenario, periods of steady sedimentation in the Fortuna basin
are long enough to support a preferred magnetostratigraphic
correlation.
Assessing overall reliability
Assessment of the overall reliability of magnetostra-
tigraphic dating was addressed by others (Talling and
Burbank, 1993), who analyzed all sources of error associ-
ated with absolute ages derived from magnetostratigraphy
and quantified error magnitudes related to the accuracy of
the GPTS and the completeness of the local magnetostra-
tigraphy. But, uncertainties involved in nonunique corre-
lations could only be qualitatively treated. In this
respect, it is recommend that an explicit discrimination is
made between correlations based on an external
(biostratigraphic or radiometric) age control and those
based on the smoothest-derived sediment-accumulation
rates. There are cases that a best-fit magnetostratigraphic
correlation does not meet other external age constraints
(Figure 6). At this point, magnetostratigraphic correlation
becomes a multidisciplinary approach which involves
a proper analysis of all sources of uncertainty.
A reliability index for magnetostratigraphic data was
put forward by Opdyke and Channell (1996) based on
criteria listed below, with the recommendation that ratings
of 5 (out of 10) should be achieved:
1. Biostratigraphic data provides an age control at the
level of stage.
2. Sampling localities placed in a measured stratigraphic
section.
3. Complete stepwise NRM demagnetization of all
samples.
4. Paleomagnetic directions determined by least-square
analysis.
5. Numerical data fully documented with plots of VGP
latitude against stratigraphic position (or declination
and inclination).
6. Determination of magnetic mineralogy.
7. Use of field tests (fold test, conglomerate test) to
determine the stability and age of the magnetization.
8. Reversal test, to check for the overlap with secondary
magnetizations.
9. Radiometric ages are available for the section and
fully documented with their associated errors.
10. Consistency test, to check for the stratigraphic control
of polarity reversals, and the primary nature of the
magnetization.
Among these, criteria 2, 3, 4, 5, and 6 are standardized
procedures in modern magnetostratigraphic studies and
should be mandatory for proper publication of results.
Other criteria, such as availability of external age controls
(1 and 9) and stability tests (7, 8, and 10), are case depen-
dent and cannot always be met.
Conclusions
Magnetostratigraphy is challenging among the different
dating methods of sedimentary rocks, because the out-
come is not as simple as determining a numerical age
with its associated error. A correlation is established
between a local magnetic polarity zonation and the glob-
ally isochronous geomagnetic reversals of the GPTS,
whose age calibration relies on absolute chronometers,
such as radioactive dating and astrochronology. Thus,
a recommended practice is that magnetostratigraphic
ages are not only expressed with its derived numerical
value but as a correspondence with geomagnetic chron
boundaries. Popularity of magnetostratigraphy as
a dating tool relies on the fact that many sedimentary
rocks are faithful recorders of the ancient magnetic field
and that modern magnetometers are capable of measur-
ing very weakly magnetized samples. On the other hand,
the inadequacy of magnetostratigraphy to be applied to
discontinuous or very short sedimentary records remains
as the fundamental limitation. A full comprehension of
the regional stratigraphic context and accessibility to
long and continuous sedimentary sequences is crucial
for magnetostratigraphy to contribute robust and inde-
pendent age constraints.
Bibliography
Bellon, H., Bordet, P., and Montenat, C., 1983. Chronologie du
magmatisme néogène des Cordillères bétiques (Espagne
méridionale). Bulletin de la Société géologique de France, 25,
205–217.
Butler, R. F., 1992. Paleomagnetism: Magnetic Domains to
Geologic Terranes. Boston, MA: Blackwell, p. 238.
 MARINE ISOTOPE STRATIGRAPHY
Cande, S. C., and Kent, D. V., 1992. A new geomagnetic polarity
time scale for the Late Cretaceous and Cenozoic. Journal of
Geophysical Research, 97(B10), 13917–13951, doi:10.1029/
92JB01202.
Cande, S. C., and Kent, D. V., 1995. Revised calibration of the geo-
magnetic polarity time scale for the Late Cretaceous and
Cenozoic. Journal of Geophysical Research, 100, 6093–6095.
Channell, J. E. T., Galeotti, S., Martin, E. E., Billups, K., Scher,
H. D., and Stoner, J. S., 2003. Eocene to Miocene magnetostra-
tigraphy, biostratigraphy, and chemostratigraphy at ODP Site
1090 (sub-Antarctic South Atlantic). GSA Bulletin, 115(5),
607–623.
Cox, A., Doell, R., and Dalrymple, G., 1964. Reversals of the
Earth’s magnetic field. Science, 144, 1537–1543.
Evans, M., and Heller, F., 2003. Environmental Magnetism: Princi-
ples and Applications of Enviromagnetics. London: Elsevier,
p. 299.
Garcés, M., Krijgsman, W., and Agustí, J., 2001. Chronostratigraphic
framework and evolution of the Fortuna basin (Eastern Betics)
since the Late Miocene. Basin Research, 13(2), 199–216.
Gee, J. S., and Kent, D. V., 2007. Source of oceanic magnetic anom-
alies and the geomagnetic polarity timescale. In Kono, M. (ed.),
Geomagnetism, volume 5 of Treatise on Geophysics. Amster-
dam: Elsevier, pp. 455–507.
Gradstein, F., Ogg, J., and Smith, A., 2004. A Geologic Time Scale
2004. Cambridge, UK: Cambridge University Press.
Gradstein, F., Ogg, J., Schmitz, M., and Ogg, G., 2012. The Geo-
logic Time Scale 2012. Amsterdam: Elsevier.
Heirtzler, J. R., Dickson, G. O., Herron, E. M., Pitman, W. C., and
Le Pichon, X., 1968. Marine magnetic anomalies, geomagnetic
field reversals, and motions of the ocean floor and continents.
Journal of Geophysical Research, 73(6), 2119–2136.
Hinnov, L. A., and Ogg, J. G., 2007. Cyclostratigraphy and the
astronomical time scale. Stratigraphy, 4(2/3), 239–251.
Johnson, N. M., and McGee, V. E., 1983. Magnetic polarity stratig-
raphy: stochastic properties of data, sampling problems, and the
evaluation of interpretations. Journal of Geophysical Research,
88(B2), 1213–1221.
Johnson, N. M., Stix, J., Tauxe, L., Cerveny, P. F., and Tahirkeli,
R. A. K., 1985. Paleomagnetic chronology, fluvial processes
and tectonic implications of the Siwalik deposits near Chinji
Village, Pakistan. Journal of Geology, 93, 27–40.
Kirschvink, J. L., 1980. The least squares lines and plane analysis of
paleomagnetic data. Geophysical Journal of the Royal Astro-
nomical Society, 62, 699–718.
Krijgsman, W., and Kent, D. V., 2004. Non-uniform occurrence of
short-term polarity fluctuations in the geomagnetic field? New
results from middle to late Miocene sediments of the North Atlan-
tic (DSDP site 608). In Channell, J. E. T., Kent, D. V., Lowrie, W.,
and Meert, J. G. (eds.), Timescales of the Paleomagnetic Field.
Washington, DC: American Geophysical Union, pp. 1–14.
Krijgsman, W., Garcés, M., Agustí, J., Raffi, I., Taberner, C., and
Zachariasse, W. J., 2000. The ‘Tortonian salinity crisis’ of the
eastern Betics (Spain). Earth and Planetary Science Letters,
181, 497–511.
Kuiper, K. F., Krijgsman, W., Garcés, M., and Wijbrans, J. R. R.,
2006. Revised isotopic (40Ar/39Ar) age for the lamproite volcano
of Cabezos Negros, Fortuna Basin (Eastern Betics, SE Spain).
Palaeogeography Palaeoclimatology Palaeoecology, 238
(1–4), 53–63.
Kuiper, K. F., Deino, A., Hilgen, F. J., Krijgsman, W., Renne, P. R.,
and Wijbrans, J. R., 2008. Synchronizing rock clocks of Earth
history. Science, 320(5875), 500–504.
Lancis, C., Tent-Manclús, J.-E., Soria, J.-M., Caracuel, J.-E., Corbí,
H., Dinarès-Turell, J., Estévez, A., et al., 2010. Nanoplankton
biostratigraphic calibration of the evaporitic events in the Neo-
gene Fortuna Basin (SE Spain). Geobios, 43(2), 201–217.
517
Langereis, C. G., Linssen, J. H., Mullender, T. A. T., and Zijderveld,
J. D. A., 1989. Demagnetisation. In James, D. E. (ed.), The Ency-
clopedia of Solid Earth Geo- physics. New York: Van Nostrand
Reinhold, pp. 201–211.
Langereis, C. G., Krijgsman, W., Muttoni, G., and Menning, M.,
2010. Magnetostratigraphy – concepts, definitions, and applica-
tions. Newsletters on Stratigraphy, 43(3), 207–233.
Lowrie, W., 1990. Identification of ferromagnetic minerals in a rock
by coercivity and un- blocking temperature properties. Geophys-
ical Research Letters, 17, 159–162.
Opdyke, N. D., and Channell, J. E. T., 1996. Magnetic Stratigraphy.
San Diego, CA: Academic Press.
Opdyke, N. D., Glass, B., Hays, P. J., and Foster, J., 1966. Paleo-
magnetic study of Antarctica deep-sea cores. Science, 154,
349–357.
Renne, P. R., 1998. Absolute ages aren’t exactly. Science,
282(5395), 1840–1841.
Sadler, P. M., 1981. Sediment accumulation rates and the complete-
ness of stratigraphic sections. Journal of Geology, 89, 569–584.
Talling, P. J., and Burbank, D. W., 1993. Assessment of uncer-
tainties in magnetostratigraphic dating of sedimentary strata. In
Aissaoui, D., McNeill, D., and Hurley, N. (eds.), Applications
of Paleomagnetism to Sedimentary Geology. Tulsa, OK: SEPM,
pp. 59–69. SEPM Special Publication No. 49.
Tauxe, L., 1998. Paleomagnetic Principles and Practice. Dor-
drecht: Kluwer, p. 299.
Tauxe, L., and Gallet, Y., 1991. A jacknife for magnetostratigraphy.
Geophysical Research Letters, 18(9), 1783–1786.
Vasiliev, I., Krijgsman, W., Stoica, M., and Langereis, C. G., 2005.
Mio-Pliocene magnetostratigraphy in the southern Carpathian
foredeep and Mediterranean-Paratethys correlations. Terra
Nova, 17(4), 376–384.
Cross-references
Magnetic Anomalies
Magnetic Chronology
Magnetometer
MARINE ISOTOPE STRATIGRAPHY
Galen P. Halverson
Department of Earth and Planetary Sciences/Geotop,
McGill University, Montréal, QC, Canada
Definition
Marine isotope stratigraphy is the application of proxies of
seawater isotopic composition, as preserved in marine
sediments and sedimentary rocks, for correlation and
dating purposes.
Introduction
Chemical stratigraphy, or chemostratigraphy, is the appli-
cation of chemical signatures preserved in sediments and
sedimentary rocks for the purpose of correlation, dating,
or interpreting past environments and environmental
change (Weissert et al., 2008; Halverson et al., 2010).
Marine isotope stratigraphy is the most important subdivi-
sion of chemostratigraphy and encapsulates diverse stable
and radiogenic isotope systems that have been applied to
518 MARINE ISOTOPE STRATIGRAPHY

studies of marine strata spanning from Archean to Recent

in age. The list of isotope systems that have been investi-
gated is large and growing, spurred in part by significant
recent developments in gas source (IRMS) and multi-
collector inductively coupled plasma (MC-ICP-MS) mass
spectrometry (see also entry on Mass Spectrometry).
However, the focus here is on a small subset of established
isotope proxies that are widely utilized and particularly
useful for chronostratigraphic purposes. Specifically,
the light stable isotope systems of oxygen (d18O), carbon
(d13C), and sulfur (d34S) and the radiogenic isotope sys-
tems of neodymium (143Nd/144Nd), strontium (87Sr/86Sr),
and osmium (187Os/188Os) are briefly reviewed.

Underlying principles
An overarching principle in the application of marine iso-
tope stratigraphy is that targeted materials for analysis,
such as shells or organic matter in marine sediments,
can be regarded as reliable proxies of the isotopic compo-
sition of the seawater in which they formed. Where these
data are to be applied for correlation or dating, it is para-
mount that the residence time of the isotope system in
seawater is sufficiently large relative to the mixing time
of the oceans (~103 years) that the isotope proxy retains
a global seawater signature. If these conditions are met
and the isotope systems have not been significantly
disturbed by subsequent diagenesis or other alteration,
then fluctuations, or in some cases absolute values, in
the isotope ratios can be used to correlate strata deposited
in different parts of the global ocean or assign an approx-
imate age based on comparison with a reference marine
isotope curve (see also entry on Seawater Sr Curve).
Because seawater compositions of most commonly
applied isotope ratios have varied secularly over Earth
history, a simple isotope ratio or profile is typically insuf-
ficient to assign an age, and other chronostratigraphic
tools, such as biostratigraphy, magnetostratigraphy, or
absolute ages on bounding strata, must be used first to
narrow down the possible age range of a given sample
or stratigraphic section. The most useful proxies and time
intervals in which they are applied are those that show
rapid changes in seawater composition (McArthur et al.,
2012). Fortuitously, isotopic shifts and excursions
(Figure 1) are common across major geological events,
such as mass extinctions, glaciations, and episodes of
abrupt warming.
Stable isotopes
Temporal fluctuations in stable isotope compositions of
seawater result from the tendency of isotopes to fraction-
ate at low temperatures due to their relative mass differ-
ences. These mass discrepancies give rise to subtle but
important differences in the strength of chemical bonds
formed in molecules by the different isotopes of a given
element. The result is the preferential diffusion, reaction,
or biological uptake of one isotope over another, resulting
in fractionation of isotope ratios between different
Marine Isotope Stratigraphy, Figure 1 Schematic illustration of
the difference between steady-state and non-steady-state
changes in marine isotope ratios. A steady-state shift (blue line)
results from a change in the isotope ratio of the input to or
output from the ocean of a given system. For example, a shift in
87
Sr/86Sr could result from the extrusion of a large continental
flood basalt (time A) and resulting decrease in average
weathering input. The timescale over which the shift occurs is
proportional to the residence time of that element in the ocean.
A non-steady-state perturbation (red line) results in an excursion,
commonly driven by a relatively rapid input of the element of
interest to the ocean reservoir (at time A), such as a catastrophic
release of low d13C methane from seafloor clathrate reservoirs.
The magnitude of the excursion is a measure of the relative size
of the non-steady-state addition (or subtraction) of the element
to seawater, and the timescale over which the system recovers is
governed by the residence time.
molecules or phases. This fractionation generates geolog-
ical or oceanic reservoirs with contrasting isotopic ratios.
For example, the exit channels of carbon from the seawa-
ter dissolved inorganic carbon (DIC) pool can be
subdivided into inorganic (i.e., carbonate; relatively 13C
enriched) and organic sedimentary carbon (13C-depleted)
sinks. The relative importance of these sinks controls the
13 12
C/ C ratio of the DIC pool at steady state.
Many biogeochemical and geological processes
can drive the secular variation in isotope ratios in sea-
water, but those changes that occur at timescales useful
in stratigraphy are commonly the result of changing
paleoenvironments. Where these changes occur at steady
state (i.e., the mass of the seawater reservoir of the element
of interest does not change in time), the resulting isotopic
change is gradual. However, the most useful isotopic fluc-
tuations are commonly in the form of relatively brief
excursions (Figure 1) that result from sudden or cata-
strophic events, such as glaciation (see section on oxygen
isotopes below), flood basalt volcanism, and mass extinc-
tions. Such events often give rise to precipitous (non-steady
state) changes in multiple isotope proxies simultaneously
 MARINE ISOTOPE STRATIGRAPHY
(e.g., fluctuations in carbon and oxygen isotope ratios are
commonly correlated or anticorrelated). Importantly, the
timescale of these changes may differ due to variable oce-
anic residence times and the proximal driving mechanism
for the perturbation.
Radiogenic isotopes
Variations in the ratios of radiogenic isotopes in seawater
are fundamentally the result of the fractionation of two
elements that include the parent and daughter isotopes of
a given radiogenic system (e.g., Rb and Sr, where 87Rb
decays to 87Sr) between geological reservoirs, rather than
isotopic fractionation. Coupled with progressive decay of
the radiogenic isotope, this elemental fractionation pro-
duces reservoirs, such as the mantle and the continental
crust, with distinct radiogenic isotope fingerprints. Pro-
gressive decay of the radiogenic isotope results in a first-
order, long-term rise in the radiogenic isotope ratio over
Earth history. Changes in the relative weight of the flux
to the ocean of one of these elements from the different
reservoirs result in a steady-state shift in the isotope ratio
in seawater, over a timescale determined by the residence
time of that element in seawater (Figure 1), and the time
span over which the change in flux occurs. In some cases,
sharp excursions may occur.
Development and application of the method
The origin of marine isotope stratigraphy can be traced to
the pioneering work of Harold Urey and his research
group at the University of Chicago in the 1950s who ana-
lyzed carbon and oxygen isotope ratios in modern and
ancient carbonate shells and other biogenic materials.
The initial motivation of Urey’s group was in
reconstructing paleoenvironments, namely, past seawater
temperatures (Epstein et al., 1951; Urey et al., 1951).
However, Emiliani’s (1955) documentation of systematic
stratigraphic (hence, temporal) variations in oxygen iso-
tope ratios in foraminifera tests recovered from Pleisto-
cene sediment cores established the great potential of
oxygen isotope stratigraphy for correlation and dating pur-
poses (Emiliani, 1958). Oxygen isotopes would subse-
quently prove to be an invaluable and much more
effective means of reconstructing the tempo and magni-
tude of late Cenozoic glacial advances and retreats than
the continental record of glacial drift (e.g., Shackleton
and Opdyke, 1973; Hays et al., 1976; Lisiecki and Raymo,
2005).
Although Wickman (1948) first speculated that stron-
tium isotopes were a potentially useful dating tool in
marine carbonates and evaporites well before analytical
techniques became routine, the full potential of strontium
isotope stratigraphy was not realized until Burke et al.
(1982) published the hallmark record of Phanerozoic sea-
water Sr isotope compositions, which demonstrated sub-
stantial variations in 87Sr/86Sr. Similarly, carbon isotope
stratigraphy did not emerge as an effective and efficient
chronostratigraphic tool until the late 1970s and early
519
1980s (e.g., Berger et al., 1978; Scholle and Arthur,
1980), but is now a standard technique when working on
sedimentary carbonates of all ages.
Stable isotope systems
Due to the relatively small absolute variations in stable
isotope ratios, such as 13C/12C and 18O/16O, most com-
monly measured stable isotope ratios are expressed as
per mil (parts per thousand) units in delta (d) notation,
relative to an international standard:


RSAM
RSTD
d ðU wasyhÞ ¼ 1000
RSTD
where RSAM and RSTD refer to the isotope ratio of interest
of a measured sample and standard, respectively. Many
different isotope ratios are measured on sediments and
sedimentary rocks, but of these, few have proven consis-
tently effective in chronostratigraphy. Several other sys-
tems have very specific environmental applications, such
as boron isotopes (paleo-pH) and clumped C–O isotopes
(paleo-temperature) in carbonates. Other proxies, such as
Mg, Ca, and Li isotopes in carbonates, have emerged as
potentially useful chemostratigraphic tools, but are either
not yet well established or are not effective replacements
for established isotope systems.
Oxygen isotopes
Oxygen isotope stratigraphy has proven invaluable in
understanding Cenozoic paleoclimate (Emiliani, 1958;
Hays et al., 1978; Zachos et al., 2001; Lisiecki and
Raymo, 2005) and has enjoyed wide application to rocks
and sediments of all ages. Oxygen isotope ratios are nor-
malized to two different standards: PDB (named for bel-
emnites in the Cretaceous Peedee Formation, which was
the original laboratory carbonate standard developed by
Urey’s group at the University of Chicago, and subse-
quently recalibrated as VPDB) and SMOW (Standard
Mean Ocean Water standard, recalibrated as VSMOW).
Oxygen isotope values for a given material can be
converted between these two standardizations using the
equation
d18 OVSMOW ¼ 1:03091  d18 OVPDB þ 30:91
By convention, silicate and phosphate minerals are
standardized to VSMOW and carbonate minerals are stan-
dardized to VPDB. Here, we are only concerned with
those minerals deposited from seawater, most importantly
aragonite and calcite, although oxygen in phosphate
biominerals is also used for chemostratigraphy. The origi-
nal oxygen isotope composition of these minerals is deter-
mined by temperature at which the mineral formed and the
d18O of the oxygen source for that mineral, which, in the
case of carbonates, is the oxygen bound to carbon in
the dissolved inorganic carbon (DIC) pool (Grossman,
2012). The d18O of DIC, in turn, is governed by the
d18O of seawater. Hence, oxygen isotope ratios can be
520 MARINE ISOTOPE STRATIGRAPHY

used as a paleothermometer, assuming the composition of

the waters from which the carbonate was precipitated is
known (Epstein and Mayeda, 1953).
Carbonates are strongly enriched in 18O relative to the
waters from which they precipitate. Because this fraction-
ation is controlled by temperature, with larger fraction-
ations occurring at lower temperatures, if seawater
composition is held constant, then increasing temperatures
result in carbonates with lower d18O values, and vice
versa. However, seawater composition does vary, specifi-
cally as a consequence of the growth and decay of conti-
nental ice sheets, which are strongly depleted in 18O
compared to seawater. The magnitude of this shift
between the last glacial maximum and today was about
0.9 % (Schrag et al., 1996). Fortunately, this ice volume
effect drives the d18O of seawater to higher values during
glacial maxima and lower values during glacial minima –
that is, in the same direction as the temperature effect.
Hence, oxygen isotopes are invaluable in reconstructing
late Cenozoic (Pliocene–Pleistocene) glaciation, which
is now widely accepted to be modulated by orbital
(Milankovitch) cycles (Hays et al., 1976). Indeed, both
marine carbonate and ice core d18O records over the past
five million years are commonly tuned to oscillations in
insolation (typically June 21 insolation at 65 N; e.g.,
Shackleton, 2000; Zachos et al., 2001).
Figure 2 shows the marine benthic foraminifera (i.e.,
deep water) d18O record (LR04) for the past 3 m.y., as
compiled and orbitally tuned by Lisiecki and Raymo
(2005). Benthic forams are preferred over planktonic
forams for generating seawater curves because the latter
are sensitive to local temperature or salinity affects,
whereas benthic forams better capture whole ocean values
and trends. The LR04 record captures many key features,
including a general increase in d18O of marine carbonates
through the Pliocene–Pleistocene that is presumably
related to gradual intensification of continental glaciation;
the well-developed, asymmetric sawtooth 100 k.y.
glacial–interglacial cycles over the past 800 k.y.; and the Marine Isotope Stratigraphy, Figure 2 The benthic
foraminifera oxygen isotope record (LR04) for the past three
prevalence of 40 k.y. glacial cycles prior to 800 k.a. The billion years, as compiled by Lisiecki and Raymo (2005) from
peaks and troughs in the marine isotope record are identi- a stack of 57 sediment core records. Note that even-numbered
fied as numbered marine isotope stages, beginning with stages correspond to glacial maxima and odd-numbered stages
the present low (interglacial) stage (MIS 1) and extending to glacial minima. Red numbers refer to marine isotope stages
back through the Pliocene (Lisiecki and Raymo, 2005). (MIS), although note that only selected stages are labeled.
Correlating stages between a given oxygen isotope record Glacial maxima occur at higher d18O values (positive MIS; see
and the LR04 or other compilation provides a means of text for explanation), hence the inverse values on the x-axis. The
boxes on the right delineate normal (blue) and reversed (white)
dating sediment (or ice) cores and enables direct correla- magnetic polarity chrons. Mat. Matuyama, J Jaramillo, Old
tion and comparison of equivalently aged strata, which Olduvai.
remains valid even if the timescale of the compilation is
subsequently revised. This technique is most reliably
applied to deep-sea strata that experience consistent,
uninterrupted sedimentation. 2001). The utility and reliability of this proxy decreases
Oxygen isotope stratigraphy is also widely applied to back in time with diminished deep-sea sediment records
older strata and successfully delineates many key climatic and increasing degree of alteration of most carbonate sed-
events and trends over the Cenozoic Era, including the iments and rocks. Nevertheless, d18O anomalies and
Paleocene–Eocene Thermal Maximum (Figure 3a), the trends that are consistent with known events such as glaci-
onset of Antarctic glaciation (Eocene–Oligocene bound- ations have been documented in rocks spanning the
ary), and mid-Miocene climatic optimum (Zachos et al., Phanerozoic (Grossman, 2012).
 MARINE ISOTOPE STRATIGRAPHY 521

Marine Isotope Stratigraphy, Figure 3 Compilation of oxygen (a) and carbon (b) isotope data from benthic foraminifera, spanning
the Paleocene–Eocene Thermal Maximum (PETM) event, which is attributed to a catastrophic input of 13C-depleted carbon to the
ocean–atmosphere system (see Zachos et al. (2007) for a review of possible mechanisms and consequences of this event). Carbon
and oxygen isotope records (extracted from the database of Cramer et al., 2009) from individual cores from the North Atlantic
(ODP1051), Pacific (ODP865), and Southern Oceans (ODP690) are shown for comparison.

Inorganic carbon isotopes
The carbon isotope ratio (13C/12C) of the marine-dissolved
inorganic carbon (DIC) pool has varied secularly over
Earth history. The steady-state d13C value of DIC reflects
the composition of carbon input to the ocean (commonly
assumed to be
5 % to
6 % over long timescales;
Kump and Arthur, 1999), the relative partitioning of that
carbon into inorganic (carbonate; d13Ccarb) carbon
and sedimentary organic carbon (d13Corg) that is buried
within marine sediments (forg ¼ fraction of carbon
buried as organic matter), and the isotopic difference
between these two reservoirs (d13Ccarb
d13Corg 
d13CDIC
d13Corg) (Figure 4). Because d13Ccarb is
close to the value of the DIC reservoir and the fraction-
ation between the two species is effectively insensitive
to temperature, the carbon isotope composition of marine
carbonates is treated as an approximation of seawater
from which it derives.
This approximation is most applicable to pelagic
carbonates, which are deposited in the open ocean.
The predominant sources of pelagic carbonate are calcare-
ous nannoplankton (e.g., Cocolithophora) and planktonic
foraminifera. Although these two groups evolved earlier,
neither was widespread in the global ocean until the early
Cretaceous (Saltzman and Thomas, 2012). Consequently,
most middle Mesozoic–Cenozoic marine carbon isotope
compilations are derived from deep-sea sediment cores.
These may be either bulk carbonate records, which domi-
nantly comprise planktonic nannofossils (Milliman,
1993), or records developed from individual planktonic
or benthic foraminifera (e.g., Figure 5). Because the DIC
of the surface mixed layer may be heavily modified by
primary productivity (photosynthesis), planktonic and
bulk records may vary considerably from the benthic fora-
minifera records, which are a more faithful proxy for
whole ocean DIC composition. Combined planktonic
and benthic foraminifera records may be used to quantify
the effect of the biological pump, which produces a
surface-to-deep d13C gradient where surface waters are
relatively 13C enriched.
522 MARINE ISOTOPE STRATIGRAPHY
Marine Isotope Stratigraphy, Figure 4 A simple schematic
model of the coupled atmosphere–ocean carbon cycle at
present, which illustrates the carbon isotope composition of the
major fluxes and reservoirs that affect the d13C value of dissolved
inorganic carbon (DIC) in seawater and sedimentary carbon over
geological timescales. Steady-state changes in the d13C value of
DIC can be effected by (1) a change to the composition of the
input to the coupled ocean–atmosphere system, (2) a change in
the relative amount of carbon removed from the ocean–
atmosphere system and buried in sediments, and (3) the d13C
composition of organic carbon. A non-steady excursion can be
brought about by a sudden input of isotopically distinct carbon
to the DIC-atmosphere reservoir.
Even benthic foraminifera records may not precisely
record the carbon isotope composition of the integrated
oceanic DIC pool. As shown in Figure 5, the present
Pacific and North Atlantic basins are offset from one
another by about 1 %. This carbon isotope gradient
between the two ocean basins first developed in the mid-
dle Miocene. For the remainder of the Cenozoic Period,
the North Atlantic and Pacific records align reasonably
well, and both capture major shifts in d13C, including
a large decline across the Paleocene–Eocene boundary
(punctuated by the PETM event; see below), the ca. 17
Ma Monterrey carbon burial event, and a steady decline
starting about 8 million years ago and presumably related
to the expansion of C4 plants (Figure 5; Zachos et al.,
2001). Such salient trends are important from the perspec-
tive of chronostratigraphy because they permit correlation
and approximate dating of poorly age-constrained sec-
tions by comparison with a reference seawater d13C curve.
However, because both positive and negative steady-state
shifts are relatively common occurrences in the geological
record, they typically require additional chronological
constraints or a long record with enough structure that it
can be confidently correlated to the reference curve.
Non-steady-state perturbations to the seawater DIC iso-
topic composition, known as carbon isotope excursions,
may provide a unique solution when correlating and
dating marine sediments. Perhaps the best-studied exam-
ple of such a non-steady-state excursion is the carbon iso-
tope anomaly that spans the Paleocene–Eocene Thermal
Maximum (PETM) (Figure 3). This anomaly features an
abrupt decline in the d13C composition of benthic forami-
nifera of 2–3 %, followed by a gradual recovery over
a few hundreds of thousands of years (Figure 3b); this
timescale is governed by the c. 100–200 thousand-year
residence time of DIC in the oceans. It is widely accepted
that this excursion must have resulted from the geologi-
cally instantaneous addition of a large mass of
13
C-depleted carbon to the marine DIC pool. However,
the actual source of this carbon and its precise link to the
corresponding temperature change evinced by the oxygen
isotope record (Figure 3a) remain debated (Higgins and
Schrag, 2006).
Carbonates deposited on continental margins tend to
show greater variability than pelagic carbonate records.
One reason for this discrepancy in signals is that they are
more susceptible to local and regional affects that generate
heterogeneity in seawater DIC, such as input of
13
C-depleted surface and groundwater or elevated
bioproductivity and removal of 13C-depleted organic mat-
ter. This problem is particularly acute in restricted water
bodies, such as epeiric seas, where local carbon cycling
processes can strongly influence the d13C record pre-
served in carbonate sediments (e.g., Holmden et al.,
1998; Panchuk et al., 2005). Another reason for the vari-
ability is that platform carbonates are more susceptible to
overprinting by meteoric diagenesis, particularly during
glacial epochs where they are periodically exposed during
glacioeustatic lowstands (Swart, 2008). Nevertheless,
because deep-sea pelagic carbonates are absent from the
pre-Mesozoic record and other deepwater records are
sparse, carbon isotope stratigraphy is heavily dependent
on records derived from shallow-water carbonate plat-
forms (Saltzman and Thomas, 2012). The reliability of
these archives for chemical stratigraphy remains highly
debated (Derry, 2010).
Organic carbon isotopes
The carbon isotopic composition of marine sedimentary
organic matter (d13Corg) also yields important paleoenvir-
onmental information and in some cases is a useful chemical
stratigraphic tool. Most carbon isotope data produced on
sediments and sedimentary rocks are measured on
bulk total organic carbon (TOC). To the extent that this
TOC reservoir is largely derived from primary biomass
and has not been heavily altered isotopically during diagen-
esis and metamorphism, it should closely parallel
the original DIC pool from which that carbon was
fixed (Hayes et al., 1999). For this reason, the carbon isoto-
pic composition of bulk sedimentary organic carbon
(d13Corg 
20 to
30 %) has been widely applied in place
of or as a supplement to carbonate carbon isotopes
(where carbonate is scarce) to capture secular evolution
and excursions in the DIC reservoir (e.g., Knoll et al., 1986).
 MARINE ISOTOPE STRATIGRAPHY 523

Marine Isotope Stratigraphy, Figure 5 Compilation of carbon isotope composition of benthic foraminifera over the past 65 million
years (Cenozoic). Blue data points represent all data from deep-sea sediment cores, as compiled by Cramer et al. (2009), adjusted to
the timescale of Gradstein et al. (2004). White swaths indicate intervals with extensive data coverage. Black and grey lines are fits to
the North Atlantic and Pacific data, respectively (Cramer et al., 2009). Several carbon isotopic features are apparent in this curve,
including the Paleocene–Eocene Thermal Maximum (PETM) negative excursion (see also Figure 3), the Monterey Carbon Burial Event,
and a decline in marine d13C corresponding to the expansion of C4 plants beginning about eight million years ago (Zachos et al.,
2001). Note the gap between Pacific and North Atlantic d13C beginning in the middle Miocene, indicating an increasing isotopic
gradient between ocean basins.

Paired carbonate and organic carbon data can addi-
tionally be used to calculate the net isotopic fractionation
between the inorganic and organic carbonate fraction-
ations (eTOC), which is an approximation of the initial
fractionation between DIC and the primary (photosyn-
thetic) organic carbon pool (Hayes et al., 1999). This fig-
ure can vary considerably due to a several factors, but
mainly to differences in the initial fractionation between
dissolved CO2 and primary biomass (eP) which is
imparted by primary producers and principally con-
trolled by dissolved CO2 concentrations, growth rate,
and cell geometry (Popp et al., 1998). Not surprisingly,
eTOC has varied considerably over geological time, with
highs of up to 34 % during the Neoproterozoic
(100–541 Ma) and current values below 22 % (Hayes
et al., 1999). As a consequence, d13CTOC cannot simply
be used as a replacement for d13Ccarb, even though it
commonly captures the major excursions and trends in
the DIC reservoir and hence can be applied to
chronostratigraphy.
Sulfur isotopes
Sulfur isotope stratigraphy has been applied to a wide
range of paleoenvironmental questions, for which the
ratio 34S/32S is commonly used. Sulfur occurs in seawater
mainly in the oxidized form of sulfate (SO42-), but locally
and at times in the past has also been abundant in its
reduced form of hydrogen sulfide (HS-). Sulfate is deliv-
ered to the ocean mainly through continental weathering
(d34S ¼ 0–10 %) and removed as both sulfate and sul-
fide. The sulfate is sequestered in evaporite minerals
(gypsum and anhydrite), barite (CaSO4), and as a trace
constituent in carbonate minerals (carbonate associated
sulfate), all of which may be used as proxies for seawater
sulfate isotopic compositions. Under anoxic conditions,
sulfate is also utilized as the electron acceptor during bac-
terial sulfate reduction (BSR). This metabolic pathway,
along with complex redox recycling of intermediate sul-
fur species, produces sulfide that is depleted in 34S rela-
tive to seawater sulfate typically by ~30–60 %. The
highly variable figure results in a wide range of d34Ssulfide
values in the sedimentary record (Canfield, 1998). For
this reason, d34Ssulfide is not an especially useful tool in
chronostratigraphy, although distinct ranges in d34Ssulfide
are characteristic of certain time intervals over Earth his-
tory (e.g., Canfield, 1998).
The generally highly depleted values of sedimentary
sulfide are responsible for the relatively high d34S of
sulfate in seawater. Sulfate is the second most abundant
524 MARINE ISOTOPE STRATIGRAPHY

Marine Isotope Stratigraphy, Figure 6 A LOWESS fit sulfur isotope curve to the marine barite d34SSO4 record for the past 130 million
years (Paytan et al., 2004), modified from Paytan and Gray (2012). Barite, which is a trace constituent in many deep-sea sediments, is
presumed to preserve primary seawater sulfate isotope ratios and is a robust during diagenesis, which makes it a well-suited mineral
for reconstructing ancient seawater composition.

anion (after Cl
) in modern seawater and has a residence
time of nearly ten million years. Hence, sulfate is effec-
tively uniform in the modern ocean (d34SSO4 ¼ 21.0 
0.1 % in modern seawater; Rees et al., 1978), and its iso-
topic composition is expected to remain relatively stable
over timescales of millions of years (Paytan and Gray,
2012). Nevertheless, d34SSO4 has varied considerably
over Earth history, in large part as a result of significant
changes in the size of the sulfate reservoir and hence its
sensitivity to perturbations by changes in other parame-
ters controlling its isotopic composition, namely, the net
fractionation accompanying BSR and sulfide burial and
the fraction of sulfur buried as sulfide. Due to high intrin-
sic variability of d34SSO4 values recorded in pre-
Mesozoic sedimentary rocks and difficulty in
establishing their fidelity as seawater proxies, sulfur iso-
tope stratigraphy has limited application as a dating tool
throughout most of Earth history. However, in some
cases, sulfur isotope chemostratigraphy can be used to
apply rough age constraints. For example, distinctly
34
S-enriched values (d34SSO4 > 30 %) characterize the
latest Precambrian and Cambrian sulfur isotope record
(Kampschulte and Strauss, 2004). The sulfur isotope
record for the past 130 million years shows many distinct
features, notably a 5 % decline between 130 and 120 Ma
and a rise of similar magnitude around 55 Ma (Figure 6;
Paytan et al., 2004; Wortmann and Paytan, 2012). This
unique structure to the seawater d34SSO4 record presents
potential of using sulfur isotopes for age determinations
in sedimentary rocks of middle Mesozoic and younger
age, although in many cases, other proxies would be
more effective.
Radiogenic isotope systems
In addition to their applications in geochronology, several
different radiogenic isotopes systems serve as important
chemical stratigraphic tools. Although many different
radiogenic isotope systems have been applied to marine
sedimentary records for multiple purposes, only a limited
number of these techniques enjoy routine application with
chronostratigraphic relevance. Radiogenic isotope ratios
are conventionally reported with a stable isotope with
a radiogenic contribution of a given element as the numer-
ator (e.g., 87Sr) and a stable isotope with no or a negligible
radiogenic component as the denominator (e.g., 86Sr).
Neodymium isotopes
Nd isotopes, specifically the ratio 143Nd/144Nd, are
applied to both recent and ancient carbonate and clastic
sediments and sedimentary rocks. Neodymium isotope
ratios in sedimentary rocks are commonly reported in
epsilon notation (eNd(t)), where the 143Nd/144Nd ratio is
normalized against the chondritic uniform reservoir
(CHUR) and both are corrected for radiogenic ingrowth
of 143Nd (DePaolo and Wasserburg, 1976).
Neodymium isotope stratigraphy owes its utility to
the fact that Sm is preferentially retained in mafic
rocks relative to Nd, leading to a greater contribution of
radiogenic 143Nd resulting from decay of the parent
isotope 147Sm. This elemental fractionation gives rise to rel-
atively radiogenic (higher 143Nd/144Nd) mafic rocks and
less radiogenic felsic rocks, which makes the 143Nd/144Nd
ratio a powerful tracer of sediment and solute sources.
However, due to the low solubility of the rare earth elements
 MARINE ISOTOPE STRATIGRAPHY 525

Marine Isotope Stratigraphy, Figure 7 A LOWESS fit Sr isotope curve for the Ordovician to present (McArthur et al., 2012), plus
the Sr isotope fit to data for the Cambrian from Shields and Veizer (2002) and Halverson et al. (2010). See entry on “Seawater Sr Curve”
for a more detailed discussion of this record and its application to chronostratigraphy.

(which include both Sm and Nd) in seawater, their tendency
to adsorb to settling particles, and isotopic exchange
between sediments and seawater, the effective isotopic res-
idence time of Nd in the oceans is small: ~200–1,000 years
(Tachikawa et al., 1999). Consequently, seawater is not uni-
form with respect to 143Nd/144Nd ratios, and no unique sea-
water Nd curve can be generated. Rather, Nd isotopes, as
typically measured in carbonates, phosphates, and ferro-
manganese nodules, are more commonly used to track the
sources and circulation of ancient water masses (Banner,
2004). Neodymium isotopes measured in fine-grained
siliciclastic sediments are used to track tectonic, volcanic,
and geographic evolution of the drainage basins supplying
those sediments. To the extent that shifts in the initial
143
Nd/144Nd composition of sediments can be linked to spe-
cific events, such as basement uplift or an episode of flood
basalt magmatism, the Nd isotope record can have impor-
tant chronostratigraphic applications to poorly dated sedi-
mentary sequences.
Strontium isotopes
Strontium isotope stratigraphy is one of the most powerful
and widely applied chemostratigraphic tools to determine
ages in otherwise undated marine rocks (see also entry
on the Seawater Sr Curve). Similar to the Nd isotope sys-
tem, strontium isotope ratios, namely, 87Sr/86Sr, are useful
because of the elemental fractionation between felsic and
mafic rocks. In this case, Rb is concentrated in felsic rocks,
resulting in higher 87Sr/86Sr ratios in felsic continental
crust from the decay of 87Rb. In contrast, mafic continental
and oceanic crust have low (unradiogenic) 87Sr/86Sr. Dis-
tinct from Nd, however, the Sr budget of seawater includes
not only a riverine input reflecting typically radiogenic
continental sources but also a large unradiogenic compo-
nent from the hydrothermal alteration of oceanic crust.
The residence time of strontium is also much higher than
neodymium: 4–5 million years, meaning that oceans are
uniform with respect to 87Sr/86Sr, except in rare cases
where local water masses are heavily influenced by isoto-
pically distinct freshwater or hydrothermal input.
Because the Sr2+ cation readily substitutes for Ca2+,
marine strontium isotope ratios can be preserved in evap-
orite, carbonate, and phosphate minerals precipitated
directly from seawater. However, Sr isotope ratios are
highly susceptible to diagenesis, during which these min-
erals are commonly subjected to loss of total Sr and input
of radiogenic strontium from the detrital sediment fraction
or meteoric fluids. Hence, constructing the seawater Sr
curve (Figure 7) and using Sr isotopes as correlation and
dating tools requires the selection of high-quality samples
and a combination of petrographic and geochemical screen-
ing for the affects of diagenesis (McArthur et al., 2012).
The secular evolution of seawater 87Sr/86Sr (Figure 7)
is conventionally interpreted in terms of competing influ-
ence of continental weathering (riverine) and seafloor
526 MARINE ISOTOPE STRATIGRAPHY

Marine Isotope Stratigraphy, Figure 8 Compilations of seawater 87Sr/86Sr and 187Os/188Os data spanning from 70 million years ago
until the present [Modified from Misra and Froehlich (2012)]. Note the short-lived anomalies in 187Os/188Os across the Cretaceous–
Paleogene and Eocene–Oligocene boundaries (dashed vertical lines), which are not mirrored in the 87Sr/86Sr record.

hydrothermal Sr inputs. However, other factors also influ-
ence the long-term evolution of seawater Sr isotope ratios,
including the progressive decay of 87Rb, which imparts
a long-term rise in 87Sr/86Sr, and both tectonic (e.g., rifting
or mountain building) and climatic effects. The result is
a seawater Sr isotope curve with structure at multiple dif-
ferent timescales (Burke et al., 1982). Strontium isotope
stratigraphy is most useful as a correlation and dating tool
at time periods where 87Sr/86Sr changes rapidly (DePaolo
and Ingram, 1985), such as across the Permo-Triassic
boundary and in the latter half of the Cenozoic (Figures 7
and 8). Due to the considerable fluctuations in 87Sr/86Sr
over the past billion years (McArthur et al., 2012;
Figure 7), obtaining a single, precise and accurate
87
Sr/86Sr value from a marine sedimentary rock is not suf-
ficient to establish a date. Rather, additional age or
chemostratigraphic constraints, or a distinct 87Sr/86Sr pro-
file is necessary to link to the composite strontium refer-
ence curve (Figure 7). However, where this connection
can be made, strontium isotope stratigraphy can be used
in many cases to establish relatively precise (<1 m.y.)
ages (McArthur et al., 2012).
Osmium isotopes
Osmium isotope ratios, which owe their heterogeneity and
utility to ongoing radioactive decay of 187Re to 187Os, are
commonly reported as 187Os/188Os ratios. Osmium has
a residence time in the modern oceans of 10–50 thousand
years, and its isotopic composition (187Os/188Os ¼ 1.06 
0.005) reflects the balance between an unradiogenic man-
tle component (187Os/188Os  0.13) and a radiogenic con-
tinental component (187Os/188Os  0.14), with a small
contribution from the dissolution of unradiogenic cosmic
dust (Peuker-Ehrenbrink and Ravizza, 2000, 2012).
Hydrogenous osmium is removed from seawater domi-
nantly in metalliferous and organic-rich sediments.
Whereas ferromanganese nodules serve as an important
archive for Cenozoic Os isotope ratios (Figure 8),
organic-rich sediments are important in determining pre-
Cenozoic 187Os/188Os. However, because organic-rich
sediments also concentrate Re, this requires an isochron
approach to measure initial 187Os/188Os ratios.
As seen in Figure 8, the Cenozoic marine Os isotope
record (187Os/188Os) broadly resembles the Sr isotope
record with both displaying a prominent, first-order rise
toward more radiogenic ratios. However, due to the much
shorter residence time of Os in seawater, it is sensitive
to certain perturbations where the Sr isotope system is
buffered. Specifically, prominent low 187Os/188Os excur-
sion occurs across the Cretaceous–Paleogene and
Eocene–Oligocene boundaries. The sensitivity of the Os
isotope system to events such as flood basalt volcanism
(Cohen and Coe, 2002; Turgeon and Creaser, 2008),
 MARINE ISOTOPE STRATIGRAPHY
abrupt warming (Ravizza et al., 2001; Schmitz et al.,
2004), and large meteorite impacts (Paquay et al., 2008)
highlights its importance as a chronostratigraphic tool.
For example, Ravizza and Peuker-Ehrenbrink (2003) used
Os isotope stratigraphy on deep-sea cores to resolve the
timing of Deccan Trap volcanism from the Cretaceous–
Paleogene boundary and hence link the main mass extinc-
tion event to the Chicxulub bolide impact. Similarly, to the
extent that events that were large enough to leave an
imprint in the seawater Os isotope record are well dated,
then the Os isotope system provides an indirect but robust
way of dating sedimentary records.
Summary and conclusions
Marine isotope stratigraphy is a widely applied tool in cor-
relating and dating marine strata. It complements many
other dating techniques, such as direct radiometric dating,
biostratigraphy, and magnetostratigraphy. A large number
of stable and radiogenic isotope systems are used in
marine isotope stratigraphy, including both a core suite
of well-established systems (C, O, and Sr) and a growing
list of nontraditional isotope systems. Each isotope system
has specific benefits and drawbacks, and no single system
is universally applicable across sediment types and ages.
Indeed, in many studies, multiple isotope proxies, along
with other chemical stratigraphic tools, are applied, with
each providing distinct constraints on the timing and ori-
gin of geological events. Important considerations in the
application of isotope systems are the residence time and
behavior of the chemical species of the element in seawa-
ter, the types of sediments in which the element is prefer-
entially removed, and the potential of the isotope ratios
to be preserved against diagenetic and metamorphic dis-
turbances. Seawater isotope reference curves for the
Cenozoic, and in some cases for the entire Phanerozoic,
have been produced for many isotope systems, and these
provide an excellent framework for indirect dating sedi-
ments and sedimentary rocks whose precise ages are not
known.
Bibliography
Banner, J. L., 2004. Radiogenic isotopes: systematics and applica-
tions to earth surface processes and chemical stratigraphy.
Earth-Science Reviews, 65, 141–194.
Berger, W. H., Killingley, J. S., and Vincent, E., 1978. Stable iso-
topes in deep-sea carbonates: box core ERDC-92, west equato-
rial Pacific. Oceanologica Acta, 1, 203–216.
Burke, W. M., Denison, R. E., Hetherington, E. A., Koepnik, R. B.,
Nelson, M., and Omo, J., 1982. Variations of seawater 87Sr/86Sr
throughout Phanerozoic shales. Geology, 10, 516–519.
Canfield, D. E., 1998. A new model for Proterozoic ocean chemis-
try. Nature, 396, 450–453.
Cohen, A. S., and Coe, A. L., 2002. New geochemical evidence for
the onset of volcanism in the Central Atlantic Magmatic Prov-
ince and environmental change at the Triassic-Jurassic boundary.
Geology, 30, 267–270.
Cramer, B. S., Toggweiler, J. R., Wright, J. D., Katz, M. E., and
Miller, K. G., 2009. Ocean overturning since the Late Creta-
ceous: inferences from a new benthic foraminiferal isotope com-
pilation. Paleoceanography,24, doi:10.1029/2008PA001683.
527
DePaolo, D. J., and Ingram, B. L., 1985. High-resolution stratigra-
phy with strontium isotopes. Science, 227, 938–941.
DePaolo, D. J., and Wasserburg, G. J., 1976. Nd isotopic variations
and petrogenetic models. Geophysical Research Letters, 3,
249–252.
Derry, L. A., 2010. A burial diagenesis origin for the Ediacaran-
Shuram Wonoka carbon isotope anomaly. Earth and Planetary
Science Letters, 294, 152–162.
Emiliani, C., 1955. Pleistocene temperatures. The Journal of Geol-
ogy, 63, 538–578.
Emiliani, C., 1958. Paleotemperature analysis of Core 280 and
Pleistocene glaciations. The Journal of Geology, 66, 264–275.
Epstein, S., Buchsbaum, R., Lowenstam, H., and Urey, H. C., 1951.
Carbonate-water isotopic temperature scale. Geological Society
of America Bulletin, 62, 417–426.
Epstein, S., and Mayeda, T., 1953. Variation in O18 content of
waters from natural sources. Geochimica et Cosmochimica Acta,
4, 213–224.
Gradstein, F. M., Ogg, J. G., and Smith, A. G. (eds.), 2004. A Geo-
logic Time Scale 2004. Cambridge, UK: Cambridge University
Press, Vol. 86.
Grossman, E. L., 2012. Oxygen isotope stratigraphy. In: Gradstein,
F. M., Ogg, J. O., Schmitz, M., and Ogg, G. (eds.), The Geolog-
ical Time Scale 2012. Amsterdam: Elsevier, pp. 181–206.
Halverson, G. P., Wade, B. P., Hurtgen, M. T., and Barovich, K. M.,
2010. Neoproterozoic chemostratigraphy. Precambrian
Research, 182, 337–350.
Hayes, J. M., Strauss, H., and Kaufman, A. J., 1999. The abundance
of 13C in marine organic carbonate and isotopic fractionation in
the global biogeochemical cycle of carbon in the past 800 Ma.
Chemical Geology, 161, 103–125.
Hays, J. D., Imbrie, J., and Shackleton, N. J., 1976. Variations in
Earth’s orbit: pacemaker of the ice ages. Science, 194, 1121–1131.
Higgins, J. A., and Schrag, D. P., 2006. Beyond methane: towards
a theory for the Paleocene-Eocene thermal maximum. Earth
and Planetary Science Letters, 245, 523–537.
Holmden, C., Creaser, R. A., Muehlenbachs, K., Leslie, S. A., and
Bergstrom, S. M., 1998. Isotopic evidence for geochemical
decoupling between ancient epeiric seas and bordering oceans:
implications for secular curves. Geology, 26, 567–570.
Kampschulte, A., and Strauss, H., 2004. The sulfur isotopic evolu-
tion of Phanerozoic seawater based on the analysis of structur-
ally substituted sulfate in carbonates. Chemical Geology, 204,
255–286.
Knoll, A. H., Hayes, J. M., Kaufman, A. J., Swett, K., and Lambert,
I. B., 1986. Secular variations in carbon isotope ratios from
Upper Proterozoic successions of Svalbard and East Green- land.
Nature, 321, 832–838.
Kump, L. R., and Arthur, M. A., 1999. Interpreting carbon-isotope
excursions: carbonates and organic carbon. Chemical Geology,
161, 181–198.
Lisiecki, L. E., and Raymo, M. E., 2005. A Pliocene-Pleistocene
stack of 57 globally distributed benthic d18O. Paleocea-
nography, 20, PA1003, doi:10.1029/2004PA001071.
McArthur, J., Howarth, R. J., Shields, G. A., 2012. Strontium iso-
tope stratigraphy. The Geological Time Scale. Amsterdam:
Elsevier, Chap. 7, pp. 127–144.
Milliman, J. D., 1993. Production and accumulation of calcium car-
bonate in the ocean: budget of a nonsteady state. Global Biogeo-
chemical Cycles, 7, 927–957.
Misra, S., and Froelich, P. N., 2012. Lithium isotope history of
Cenozoic seawater: changes in silicate weathering and reverse
weathering. Science, 335, 818–823.
Panchuk, K. M., Holmden, C., and Kump, L. R., 2005. Sensitivity of
the epeiric sea carbon isotope record to local-scale carbon cycle
processes: tales from the Mohawkian sea. Palaeogeography,
Palaeoclimatology, Palaeoecology, 228, 320–337.
528 MARINE VARVES
Paquay, F. S., Ravizza, G. E., Dalai, T. K., and Peucker-Ehrenbrink,
B., 2008. Determining chondritic impactor size from the marine
osmium isotope record. Science, 320, 214–218.
Paytan, A., and Gray, E. T., 2012. Sulfur isotope stratigraphy. In
Gradstein, F. M., Ogg, J. G., Schmitz, M., and Ogg, G. (eds.),
The Geological Time Scale 2012. Amsterdam: Elsevier,
pp. 167–180.
Paytan, A., Martínez-Ruiz, F., Eagle, M., Ivy, A., and Wankel,
S. D., 2004. Using sulfur isotopes in barite to elucidate the
origin of high organic matter accumulation events in marine
sediments. Sulfur Biogeochemistry, GSA Special Paper 379,
pp. 151–160.
Peucker-Ehrenbrink, B., and Ravizza, G., 2000. The marine isotope
record. Terra Nova, 12, 205–219.
Peucker-Ehrenbrink, B., and Ravizza, G., 2012. Osmium isotope
stratigraphy. In Gradstein, F. M., Ogg, J. G., Schmitz, M., and
Ogg, G. (eds.), The Geological Time Scale 2012. Amsterdam:
Elsevier, pp. 145–166.
Popp, B. N., Laws, E. A., Bidigare, R. R., Dore, J. E., Hanson, K. L.,
and Wakeham, S. G., 1998. Effect of phytoplankton cell geome-
try on carbon isotopic fractionation. Geochimica et
Cosmochimica Acta, 62, 69–77.
Ravizza, G., and Peuker-Ehrenbrink, B., 2003. Chemostratigraphic
evidence of Deccan volcanism from the marine osmium isotope
record. Science, 302, 1392–1395.
Ravizza, G. E., Norris, R. N., Blusztajn, J., and Aubry, M.-P., 2001.
An osmium isotope excursion associated with the late Paleocene
thermal maximum: evidence of intensified chemical weathering.
Paleoceanography, 16, 155–163.
Rees, C. E., Jenkins, W. F., and Monster, J., 1978. The sulphur iso-
topic composition of ocean water sulphate. Geochimica et
Cosmochimica Acta, 42, 377–382.
Saltzman, M. R., and Thomas, E., 2012. Carbon isotope stratigra-
phy. In Gradstein, F. M., Ogg, J. G., Schmitz, M., and Ogg, G.
(eds.), The Geological Time Scale 2012. Amsterdam: Elsevier,
pp. 207–232.
Schmitz, B., Peucker-Ehrenbrink, B., Heilmann-Clausen, C.,
Åberg, G., Asaro, F., and Lee, C.-T. A., 2004. Basaltic explosive
volcanism, but no comet impact, at the Paleocene–Eocene
boundary: high-resolution chemical and isotopic records from
Egypt, Spain and Denmark. Earth and Planetary Science
Letters, 225, 1–17.
Scholle, P. A., and Arthur, M. A., 1980. Carbon isotope fluctuations
in Cretaceous pelagic limestone: potential stratigraphic and
petroleum exploration tool. American Association of Petroleum
Geologists Bulletin, 64, 67–87.
Schrag, D. P., Hampt, G., and Murray, D. W., 1996. Pore fluid con-
straints on the temperature and oxygen isotopic composition of
the glacial ocean. Science, 272, 1930–1932.
Shackleton, N. J., 2000. The 100,000-year ice-age cycle identified
and found to lag temperature, carbon dioxide, and orbital eccen-
tricity. Science, 289, 1897–1902.
Shackleton, N. J., and Opdyke, N. D., 1973. Oxygen isotope and
palaeomagnetic stratigraphy of equatorial Pacific core
V28-238: oxygen isotope temperatures and ice volumes on
a 105 Year and 106 year scale. Quaternary Research, 3, 39–55.
Shields, G., and Veizer, J., 2002. Precambrian marine carbonate iso-
tope database: Version 1.1. Geochemistry Geophysics
Geosystems, 3. doi:10.1029/2001GC000266.
Swart, P. K., 2008. Global synchronous changes in the carbon isoto-
pic composition of carbonate sediments unrelated to changes in
the global carbon cycle. Proceedings of the National Academy
of Sciences United States of America, 105, 13741–13745.
Tachikawa, K., Jeandel, C., and Roy-Barman, M., 1999. A new
approach to the Nd residence time in the ocean: the role of atmo-
spheric inputs. Earth and Planetary Science Letters, 170,
433–446.
Turgeon, S. C., and Creaser, R. A., 2008. Cretaceous oceanic anoxic
event 2 triggered by a massive magmatic episode. Nature, 454,
323–326.
Urey, H. C., Lowenstam, H. A., Epstein, S., and McKinney, C. R.,
1951. Measurement of paleotemperatures and temperatures of
the Upper Cretaceous of England, Denmark, and the southeast-
ern United States. Geological Society of America Bulletin, 62,
399–416.
Weissert, H., Joachimski, M., and Sarntheiin, M., 2008. Chemostra-
tigraphy. Newsletters on Stratigraphy, 42, 145–179.
Wickman, F. E., 1948. Isotope ratios: a clue to the age of certain
marine sediments. Journal of Geology, 56, 61–66.
Wortmann, E. G. and Paytan, A., 2012. Rapid variability of seawa-
ter chemistry over the past 130 million years. Science 337,
334–336.
Zachos, J., Pagani, M., Sloan, L., Thomas, E., and Billups, K., 2001.
Trends, rhythms, and aberrations in global climate 65 Ma to pre-
sent. Science, 292, 686–693.
Zachos, J. C., Bohaty, S. M., John, C. M., McCarren, H., Kelly,
D. C., and Nielsen, T., 2007. The Palaeocene-Eocene carbon iso-
tope excursion: constraints from individual shell planktonic for-
aminifer records. Philosophical Transactions of the Royal
Society A, 365, 1829–1842.
MARINE VARVES
Konrad A. Hughen
Department of Marine Chemistry and Geochemistry,
Woods Hole Oceanographic Institution,
Woods Hole, MA, USA
Synonyms
Annually laminated sediments; Laminated sediments;
Rhythmites
Definition
The term varve is a Swedish word that means cyclic layer.
It was first used in 1862 as Hvarfig lera by the Swedish
Geological Survey to describe rhythmically deposited
light and dark clay layers in a proglacial environment on
one of the first geological maps of Sweden.
Introduction and history
The study of varved sediments extends back more than
150 years, when alternating pale and dark laminations in
marine and lacustrine sediments in Europe and North
America were first mapped and interpreted as a potential
source of chronological information. The concept of annu-
ally laminated sediments was later developed by Swedish
geologist De Geer, who applied the chronological infor-
mation stored in varved clays deposited near the margin
of the retreating Scandinavian ice sheet to determine the
time elapsed since the termination of the last glacial
period. De Geer established a network of sites along the
Baltic coast of Sweden and linked them using recogniz-
able variations in varve thickness (De Geer, 1912).
In addition, he defined the term varve as being one
 MARINE VARVES
complete annual cycle consisting of a coarse-grained pale
summer layer and a fine-grained dark winter layer. Finally,
De Geer coined the term geochronology and for the first
time introduced a temporal unit (the year) to geology
(De Geer, 1912). Following his work, varve research
expanded to include a broad range of types with applica-
tions to different scientific disciplines.
Varve formation and preservation
Varved sediments are created primarily through the inter-
play of two processes: seasonal variations in the genesis,
transport, and deposition of sediments and the preserva-
tion of that signal after burial. A strong seasonal cycle that
influences sedimentation is the fundamental process con-
trolling varve deposition. In most settings this is not
a limiting factor because a dominant seasonal climate
cycle with large impacts on the environment, particularly
the hydrological cycle, can be found nearly everywhere.
In temperate climates, for example, surface water runs
freely in summer and freezes over in winter, greatly vary-
ing the amount of energy available for clastic sediment
transport. Similarly, in the tropics, monsoons create dra-
matically different runoff during wet and dry seasons. Of
course, biology also follows seasonal cycles, and this
influences sedimentation. Diatoms in freshwater lakes
bloom during spring and summer and die out in winter,
and production based on marine upwelling may fluctuate
seasonally following movements of zonal winds. Such
manifestations of seasonality result in different sedimen-
tary regimes, delivering biological or mineralogical mate-
rial to the sediment surface at different times of year as
distinct layers.
Seasonal sediment layers are produced and deposited
in many environments but are usually disturbed by bio-
turbation from benthic organisms, as well as sediment
mixing and resuspension due to slumping and current
action. In order for varves to be preserved, the fragile sea-
sonal layers must remain intact after deposition. In lakes,
morphometry of the basin is one of the most important
features controlling laminae preservation and the poten-
tial for sediment disturbance. A relationship exists
between lake area and depth in those sites with preserva-
tion of laminated sediments (O’Sullivan, 1983). In gen-
eral, very deep lakes or lakes that have great relative
depth will avoid deep wind mixing and allow develop-
ment of strong thermal stratification. Deep lakes, thus,
experience reduced resuspension of surface sediments
and, importantly, ventilation of bottom water. Oxidation
of organic matter falling through the water column then
depletes available oxygen and the deep waters become
anoxic, preventing the growth of benthic organisms and
the influence of bioturbation. Physical isolation of bot-
tom waters can also occur in the ocean, in shallow-silled
marine basins formed as deep fjords such as the Saanich
Inlet, British Columbia (Blais-Stevens et al., 1997), or
as tectonic basins such as the Santa Barbara Basin, Cali-
fornia (Behl and Kennett, 1996); Cariaco Basin,
529
Venezuela (Peterson et al., 2000); and Guaymas Basin,
Gulf of California (Calvert, 1966). Water column stratifi-
cation occurs due to chemical as well as thermal pro-
cesses, such as can be found in coastal tidewater lakes
where saltwater introduced at high tides is overlain by
a freshwater cap from continual stream input (Hughen
et al., 1996a). Dense, hypersaline bottom waters can also
form from dissolution of exposed salt formations on the
seafloor, as is found in the Orca Basin, Gulf of Mexico
(Kennett et al., 1985). Although not absolutely required
for varve formation, anoxic bottom waters are one of
the best indicators of potentially laminated sediments
and are often used to help guide coring expeditions. For
example, anoxic bottom waters occur where the oxygen
minimum zone intersects the seafloor on the continental
slopes, and annually laminated sediments have been
located and studied off the coast of Pakistan (Schultz
et al., 1998) as well as California (Anderson et al.,
1987) and Peru (De Vries and Schrader, 1981).
Classification and morphology
For convenience, varves can be classified into basic types
according to the most prominent seasonal feature or layer
components. The earliest recognized and most com-
monly studied varved sediments are clastic varves, com-
posed primarily of allochthonous mineral grains that
differ seasonally in grain size (De Geer, 1912;
O’Sullivan, 1983; Saarnisto, 1986). Clastic varves are
common in glacial marine and lacustrine settings where
mineral grain production and deposition rates are high,
and biological productivity is low. A magnified image
of clastic varves from an Arctic lake (Figure 1) illustrates
typical sedimentary structures and fabric in this type of
varve. The sediments consist of couplets of a light-col-
ored, coarser-grained layer covered by a thin dark layer.
Grains as large as fine sand mark the bottom of the coarse
layer and grade upward into silt-sized particles, whereas
a dark layer of fine clays caps the top of each couplet
(Moore et al., 2001). The silts grade gradually upward
into the fine clays, but the boundary between the clays
and the overlying coarse grains is typically sharp
(Figure 1).
The succession of fine sands to silts in a thick layer,
topped by a thin clay cap, is the classic structure of glacial
varves (De Geer, 1912; O’Sullivan, 1983; Saarnisto,
1986; Leeman and Niessen, 1994; Zolitschka, 1996;
Retelle and Child, 1996; Moore et al., 2001). The upward
gradation from relatively coarse to fine grains in the thick
layer results from a decrease in the energy available for
sediment transport over the course of the summer. During
the spring meltwater freshet, runoff is at a maximum and
sands and silts previously locked up in winter ice are first
available to be transported. Glacial runoff continues
throughout the summer, but with snow cover mostly
melted, the flux of runoff and energy of transport are
diminished, and the size of grains washed into the basin
is reduced (Retelle and Child, 1996). During winter in
530 MARINE VARVES
Marine Varves, Figure 1 Photomicrograph of clastic varve
couplets from glacial Donard Lake, Baffin Island, core 95DON-05,
47 cm. The sediment surface is toward the top of the photo.
Scale bar is 1 mm (Photograph: K. Hughen).
glacial environments, surface water freezes and transport
of sediment is greatly reduced or ceases altogether. In
the absence of continual mixing, fine, clay-sized particles
suspended in the water column settle out and form an inde-
pendent dark clay layer. This idealized sequence can be
interrupted by abrupt events such as intense rainstorms
during the late summer or fall (Retelle and Child, 1996;
Lamoureux, 1999). An increased pulse of runoff from
a rainstorm can result in a sub-lamina of larger grains,
often recognizable as an anomalous sand lens. The most
diagnostic part of the sediment fabric is the clay cap, usu-
ally used to distinguish consecutive years’ accumulation.
Another widely studied type of varved sediments are
termed biogenic (O’Sullivan, 1983; Saarnisto, 1986) or
biochemical varves (Lotter, 1991). Biogenic varves form
due to distinct pulses of aquatic biological productivity,
including seasonal plankton blooms. The biological layer
may interrupt and temporarily dilute a more constant rate
of background deposition, such as mineral grains from
steady stream flow or eolian input, or else alternate
with a different distinct pulse of sedimentation, for exam-
ple, mineral grains from monsoonal river runoff.
A magnified image of biogenic varves from a coastal
Arctic tidewater lake illustrates an example of these
Marine Varves, Figure 2 Photomicrograph of biogenic varve
couplets from Ogaq Lake, Baffin Island, core 91-2b, 73.5 cm. The
sediment surface is toward the top of the photo. Scale bar is
100 mm (Photograph: K. Hughen).
seasonal layers (Figure 2). The biogenic layers form dur-
ing the ice-free summer months, and during the winter,
suspended particles settle out of the frozen, low-energy
environment and are deposited as dark-colored, clay-rich
layers (Hughen et al., 1996a, 2000a). Biogenic laminae
form in a variety of marine and lacustrine environments,
and in many cases a succession of distinct sub-laminae
can be deposited as different diatom species bloom
throughout the growth season (Dean et al., 1999).
Varves may also form through seasonal changes in the
physical and chemical properties of the water column
and are usually restricted to lakes. Calcareous varves
may form in lakes with high CO3– content, when calcite
layers precipitate out of solution during summer warming
(O’Sullivan, 1983; Lotter, 1991). Ferrogenic varves may
form due to changes in redox conditions affecting the sol-
ubility of Fe species and the formation of sulfides versus
hydroxides throughout the year (O’Sullivan, 1983). Evap-
orite varves may form as interbedded laminae of precipi-
tates in shallow evaporite basins and can consist of
couplets of calcite and anhydrite (e.g., Castile Formation;
Anderson and Dean, 1995) or aragonite with eolian grains
(e.g., Lake Lisan; Katz and Kologny, 1989), depending on
water chemistry.
 MARINE VARVES
Analytical techniques
Independent dating and age control
Before laminated sediments can be counted and used to
construct calendar-age chronologies, it must be demon-
strated using independent dating techniques that the lami-
nae couplets are indeed annually deposited varves. Simply
identifying and counting laminae couplets can lead to mis-
interpretation of event deposits as seasonal layers, or the
counting of layers representing seasonal processes, but
ones that do not occur on a regular basis from year to year.
Numerous studies have used independent dating tech-
niques to demonstrate that supposed “varves” were not
in fact annually deposited (e.g., Crusius and Anderson,
1992). Whenever possible, a combination of several dat-
ing methods, such as lead-210 and anthropogenic
nuclides, should be used both to increase confidence and
to provide temporal coverage over a broader range of time
scales.
Chronology construction
By their nature, varves are a result of changing sediment
supply to a depositional basin and should be present and
recognizable throughout that basin. Subtle disturbances
may disrupt or distort the varve sequence recovered in sin-
gle sediment cores, however, and numerous steps are nec-
essary to create a reliable annual chronology from
laminated sediments. In constructing a varve chronology,
many of the principles used in dendrochronology must
be applied. For example, multiple measurements of varves
(tree rings) from a single core (tree) must be averaged,
followed by cross-correlation and averaging between mul-
tiple cores to build a chronology representative of an entire
sediment basin (stand of trees). Small fluctuations in varve
thickness as seen in thin sections require that each lamina-
tion be measured several times and averaged to provide an
objective, representative record from each sediment core
(Zolitschka, 1996). In addition, local disturbances such
as turbidites and erosional scours, as well as core breaks
or gaps arising from the sampling process, may limit the
physical integrity of a record from any single sediment
core. To minimize these sources of error, it is critical to
correlate varve chronologies constructed from multiple
different cores and produce a basin-wide average
(Zolitschka, 1996; Hughen et al., 2000a). Downcore pat-
terns of variation in varve thickness are usually quite dis-
tinct and can be recognized in cores across a basin.
Cross-correlation of these marker patterns allow the iden-
tification and bridging of missing or disturbed sections in
single cores. A larger number of cores used for cross-
correlation increases the likelihood of avoiding all core
gaps and constructing an accurate and reliable varve
chronology.
Varve applications
Geochronology
The geochronological applications of varved sediments
are too numerous to provide an exhaustive list. One
531
significant application that should be emphasized is the
calibration and development of other geochronological
tools, specifically, calibration of the radiocarbon time
scale. Varves from lakes and marine basins have been
one of the primary tools used to provide calendar-age cal-
ibration of 14C ages beyond the limit of tree rings,
>12,000 years ago (Hughen et al., 2000b; Ramsey et al.,
2012). Annual calendar-age varve chronologies also pro-
vide accurate dates for specific sedimentological events,
as well as age models for continuous sediment time series.
An example of a study investigating discrete events would
be paleoseismology. Earthquakes cause turbidity flows
and turbidite deposits that in the proper setting can be
dated by counting overlying varves (Hughen et al.,
1996b). Interbedded sequences of varved sediments and
massive turbidites can be used to determine recurrence
intervals of earthquakes in a specific region (Blais-Stevens
et al., 1997).
Varve chronologies offer an excellent means for
constructing age models for high-resolution paleoclimate
records. Varves are well suited to studies comparing the
relative timing and lead-lag relationships of rapid climate
changes in different regions, based on high-resolution
paleoclimate records with calendar-age chronologies from
varved sediments, tree rings, and ice cores (Hughen et al.,
2000b). In addition, varve chronologies provide excellent
precision at the annual scale and are particularly well
suited for studies involving spectral analysis at short
(sub-decadal) periodicities (Rittenour et al., 2000).
Applications other than dating
In addition to having geochronological applications,
varved sediments may themselves contain valuable
paleoenvironmental information. For example, hydrologi-
cal and sedimentological process studies in glacier-fed
lakes have shown that the fluxes of runoff and suspended
sediment, as well as varve thickness, vary logarithmically
as a function of average temperature during the snow or
glacier melting season (Leeman and Neissen, 1994;
Retelle and Child, 1996). Downcore records of clastic
varve or clastic laminae thickness from both glacial and
non-glacial lakes correlate well with average summer or
snowmelt season temperatures (Leeman and Neissen,
1994; Hardy et al., 1996; Wohlfarth et al., 1998; Hughen
et al., 2000a; Moore et al., 2001). For use as
a paleotemperature proxy, it is critical that the varve thick-
ness signal in any new location is calibrated for the recent
period using instrumental data from a nearby meteorolog-
ical station.
Varves have also been used to reconstruct changes in
other paleoenvironmental parameters. Biogenic laminae
in marine upwelling regimes reflect primary productivity
and therefore the intensity of upwelling and trade winds,
in the tropical North Atlantic during the last deglaciation
(Hughen et al., 1996c). Laminated sediments, and their
absence, have been used to reconstruct the ventilation his-
tory of intermediate-depth waters off the coast of California
(Behl and Kennett, 1996), Pakistan (Schultz et al., 1998)
532 MARINE VARVES
and northern South America (Peterson et al., 2000).
These studies and others have shown that abrupt climatic
shifts first seen in the high latitude North Atlantic region
were in fact global in scope. Finally, varves can be used
to study changes in past glacial activity and evolution of
sediment source regions, as well as evolution of sedimen-
tary basins (Lamoureux, 1999). The applications of varved
sediments to geochronological and paleoenvironmental
research are limitless and will continue to grow as increas-
ingly sophisticated methods are developed for obtaining
and analyzing them.
Conclusions
Annually laminated (varved) sediments are valuable tools
for research in geochronology, as well as other disciplines
including paleoclimatology, paleoseismology, and arche-
ology. Varved sediments are formed when distinct layers
of sediment are deposited during different seasons
throughout the year and occur in a wide variety of environ-
ments all over the globe. They are found in diverse settings
from alpine glaciers and coastal tidewater lakes to isolated
marine basins and continental slopes and occur at all lati-
tudes from the high Arctic to the tropics. Varves can be
counted and measured, similar to tree rings, in order to
provide accurate calendar-age chronologies. Varve chro-
nologies can be used in geochronological research in order
to study and develop sediment dating techniques, such as
210
Pb and 14C, and to date single events such as earth-
quakes and turbidity currents. Varves can also be applied
to the reconstruction of past environmental conditions.
At a minimum, varves provide annual-resolution calen-
dar-age models for records based on other climate proxies
(stable isotopes, pollen, etc.). In the best cases, however,
varves themselves can be shown to contain environmental
information and thus combine high temporal precision and
accuracy with paleoenvironmental reconstructions from
the same archive.
Bibliography
Anderson, R. Y., and Dean, W. E., 1995. Filling the Delaware Basin:
hydrologic and climatic controls on the Permian Castile Forma-
tion varved evaporite. In Scholle, P. A., et al. (eds.), The Permian
of Northern Pangea. New York: Springer, Vol. 2, pp. 61–78.
Anderson, R. Y., Hamphill-Halley, E., and Gardner, J. V., 1987. Per-
sistent late Pleistocene-Holocene seasonal upwelling and varves
of the coast of California. Quaternary Research, 28, 307–313.
Behl, R. J., and Kennett, J. P., 1996. Brief interstadial events in the
Santa Barbara Basin, NE Pacific, during the past 60 kyr. Nature,
379, 243–246.
Blais-Stevens, A., Clague, J. J., Bobrowsky, P. T., and Patterson,
R. T., 1997. Late Holocene sedimentation in Saanich Inlet, Brit-
ish Columbia, and its paleoseismic implications. Canadian Jour-
nal of Earth Sciences, 34, 1345–1357.
Bronk Ramsey, C., Staff, R. A., Bryant, C. L., Brock, F., Kitagawa,
H., der Plicht, J., Schlolaut, G., Marshall, M. H., et al., 2012.
A complete terrestrial radiocarbon record for 11.2 to 52.8 kyr
B.P. Science, 338(6105), 370–374.
Calvert, S. E., 1966. Origin of diatom-rich, varved sediments from
the Gulf of California. Journal of Geology, 74, 546–565.
Crusius, J., and Anderson, R. F., 1992. Inconsistencies in accumula-
tion rates of Black Sea sediments inferred from records of lami-
nae and 210Pb. Paleoceanography, 7, 215–227.
De Geer, G., 1912. A Geochronology of the past 12,000 years. In
Congrès Géologique International Compte rendu, Stockholm,
Vol. 11, pp. 241–253.
De Vries, T. J., and Schrader, H., 1981. Variation of upwelling/oce-
anic conditions during the latest Pleistocene through Holocene
off the central Peruvian coast; a diatom record. Marine Micropa-
leontology, 6, 157–167.
Dean, J. M., Kemp, A. E. S., Bull, D., Pike, J., Patterson, G., and
Zolitschka, B., 1999. Taking varves to bits: scanning electron
microscopy in the study of laminated sediments and varves.
Journal of Paleolimnology, 22, 121–136.
Hardy, D. R., Bradley, R. S., and Zolitschka, B., 1996. The climatic
signal in varved sediments from Lake C2, northern Ellesmere
Island, Canada. Journal of Paleolimnology, 16, 227–238.
Hughen, K. A., Overpeck, J. T., Anderson, R. F., and Williams,
K. M., 1996a. The potential for paleoclimatic records from
varved Arctic lake sediments: Baffin Island, Eastern Canadian
Arctic. In Kemp, A. E. S., (ed.), Geological Society Special Pub-
lication. London, Vol. 116, pp. 57–71.
Hughen, K. A., Overpeck, J. T., Peterson, L. C. and Anderson, R. F.,
1996b, The nature of varved sedimentation in the Cariaco Basin,
Venezuela, and its palaeoclimatic significance. In Kemp, A. E. S.
(ed.), Palaeoclimatology and Palaeoceanography from Lami-
nated Sediments, Special Publication, Geological Society of
London. Vol. 116, 171–183.
Hughen, K. A., Overpeck, J. T., Anderson, R. F. and Williams, K.
M., 1996c. The potential for palaeoclimate records from varved
Arctic lake sediments. In Kemp, A. E. S. (ed.), Palaeocli-
matology and Palaeoceanography from Laminated Sediments,
Special Publication, Geological Society of London. Vol. 116,
57–71.
Hughen, K. A., Overpeck, J. T., and Anderson, R. F., 2000a. Recent
warming in a 500-year palaeotemperature record from varved
sediments, Upper Soper Lake, Baffin Island, Canada. Holocene,
10, 9–19.
Hughen, K. A., Southon, J. R., Lehman, S. J., and Overpeck, J. T.,
2000b. Synchronous radiocarbon and climate shifts during the
last deglaciation. Science, 290, 1951–1954.
Katz, A., and Kolodny, N., 1989. Hypersaline brine diagenesis and
evolution in the Dead Sea-Lake Lisan system (Israel).
Geochimica et Cosmochimica Acta, 53, 59–67.
Kemp, A. E. S., (ed.), 1996. Palaeoclimatology and Palaeocea-
nography from Laminated Sediments, Special Publication, Geo-
logical Society of London, Vol. 116.
Kennett, J. P., Elmstrom, K. M., and Penrose, N., 1985. The last
deglaciation in Orca Basin, Gulf of Mexico; high-resolution
planktonic foraminiferal changes. Palaeogeography, Palaeocli-
matology, Palaeoecology, 50, 189–216.
Lamoureux, S. F., 1994. Embedding unfrozen lake sediments for thin
section preparation. Journal of Paleolimnology, 10, 141–146.
Lamoureux, S. F., 1999. Catchment and lake controls over the for-
mation of varves in monomictic Nicolay Lake, Cornwall Island,
Nunavut. Canadian Journal of Earth Sciences, 36, 1533–1546.
Leemann, A., and Neissen, F., 1994. Varve formation and the cli-
matic record in an Alpine proglacial lake: calibrating annually-
laminated sediments against hydrological and meteorological
data. Holocene, 4, 1–8.
Lotter, A. F., 1991. Absolute dating of the Late-Glacial period in
Switzerland using annually laminated sediments. Quaternary
Research, 35, 321–330.
Moore, J. J., Kughen, K. A., Miller, G. H., and Overpeck, J. T.,
2001. Little Ice Age recorded in summer temperature reconstruc-
tion from varved sediments of Donard Lake, Baffin Island, Can-
ada. Journal of Paleolimnology, 25, 484–499.
 MASS SPECTROMETRY
O’Sullivan, P. E., 1983. Annually-laminated lake sediments and the
study of Quaternary environmental changes; a review. Quater-
nary Science Reviews, 1, 245–313.
Peterson, L. C., Haug, G. H., Hughen, K. A., and Roehl, U., 2000.
Rapid changes in the hydrologic cycle of the tropical Atlantic
during the last glacial. Science, 290, 1947–1951.
Pike, J., and Kemp, A. E. S., 1996. Preparation and analysis tech-
niques for studies of laminated sediments. In Kemp, A. E. S.
(ed.), Palaeoclimatology and Palaeoceanography from Lami-
nated Sediments, Special Publication, Geological Society of
London. Vol. 116, pp. 37–48.
Retelle, M. J., and Child, J. K., 1996. Suspended sediment transport
and deposition in a high arctic meromictic lake. Journal of
Paleolimnology, 16, 151–167.
Rittenour, T. M., Brigham-Grette, J., and Mann, M. E., 2000. El
Nino-like climate teleconnections in New England during the
late Pleistocene. Science, 288, 1039–1042.
Saarnisto, M., 1986. Annually laminated lake sediments. In
Berglund, B. E. (ed.), Handbook of Holocene Palaeocology
and Palaeohydrology. John Wiley and Sons, pp. 343–370.
Schultz, H., von Rad, U., and Erlenkeuser, H., 1998. Correlation
between Arabian Sea and Greenland climate oscillations of the
past 110,000 years. Nature, 393, 54–57.
Wohlfarth et al. (1998). The Holocene, 8, 521–534.
Zolitschka, B., 1996. Recent sedimentation in a high arctic lake,
northern Ellesmere Island, Canada. Journal of Paleolimnology,
16, 169–186.
Cross-references
210
Pb Dating
Lacustrine Environments (14C)
Radiocarbon Dating
Sedimentary Rocks (Rb-Sr Geochronology)
MASS SPECTROMETRY
James B. Paces1, Dominique Weis2 and Trevor R. Ireland3
1
Geosciences and Environmental Change Science Center,
U.S. Geological Survey, Denver, CO, USA
2
Department of Earth, Ocean, and Atmospheric Sciences,
Pacific Centre for Isotopic and Geochemical Research,
The University of British Columbia, Vancouver, BC,
Canada
3
Research School of Earth Sciences, The Australian
National University, Canberra, Australia
Definition
Any of a number of methods used to determine the spec-
trum of isotopic abundances in a given material based on
the measurement of relative masses of atoms or molecules
present in that material. As applied to scientific dating,
mass spectrometry is most commonly used to determine
abundances of parent and progeny isotopes in naturally
radioactive decay systems that have half-lives of geologi-
cal relevance (years to billions of years).
Introduction
Scientific methods for dating materials of geological inter-
est commonly utilize natural radioactive isotopes that
533
spontaneously transform to progeny isotopes at constant
and well-known rates of decay. In order to use this prop-
erty to estimate absolute ages, it is essential to accurately
determine the abundances of both parent and progeny iso-
topes in a mineral or rock sample that has remained closed
to isotopic exchange with its surroundings since its forma-
tion. Decay counting can be used to quantify abundances
of short-lived radioactive isotopes. However, for radioac-
tive isotopes with half-lives longer than a few thousand
to tens of thousands of years, only a small number of
atoms undergo radioactive decay during the counting
period resulting in poor precision of isotope abundances.
Furthermore, a number of geochronometers rely on stable,
albeit radiogenic, progeny which cannot be measured by
decay counting. Therefore, isotope ratios are more com-
monly determined by mass spectrometry where poten-
tially all of the atoms present in the sample can be
counted. Geochronological methods using K-Ar, Rb-Sr,
Sm-Nd, Lu-Hf, Re-Os, U-Pb, and U-Th decay schemes
take advantage of atom counting by mass spectrometry
to provide precise measurements of relative isotope abun-
dance. Other methods for measuring mass ratios exist;
however, this entry focuses on mass spectrometric
methods that are most commonly employed in geochrono-
logical studies.
Determination of isotope ratios by mass spectrometry
involves:
(1) introduction of the sample into the mass spectrometer
either as a gas, liquid, or solid and ionization of the
element or elements of interest,
(2) separation of ions on the basis of their mass-to-charge
ratios, and
(3) detection of those ions by measuring currents for large
signals or by ion counting for smaller ones.
Methods by which material is introduced and ionized
allow mass spectrometers to be classified into different
types, each of which has strengths and weaknesses. How-
ever, all methods take advantage of the physical response
of moving charged particles through a strong electromag-
netic field and conversion of the beam of separated ions
into an electronic signal that is directly proportional to
the number of atoms detected. Output takes the form of a
mass spectrum of peak intensities that are indicative of
the relative abundances of isotopes present in the sample,
which can then be interpreted to represent age information
through application of various decay schemes and dating
models. More detailed explanations of the general princi-
ples and various methods of mass spectrometry are readily
available (Faure, 1986; Goldstein and Stirling, 2003;
Dickin, 2005; Faure and Mensing, 2005; Gross, 2011;
Ireland, 2013; White, 2013).
History
The initial development of mass spectrometry closely
followed the discovery of radioactivity in the middle to
late 1890s. The first demonstration that isotopes could
534 MASS SPECTROMETRY

b
ION SOURCE

a S2

TO PUMP
m
Pump .5 c 60° 60°
12
COLLECTOR
H
TO
E AMPLIFIER
F POLE S3 S5
PIECE

S1
S2 MAGNET

A A

D B

Mass Spectrometry, Figure 1 Historical mass spectrometer designs. (a) Cross-sectional schematic of Dempster’s first thermal
ionization instrument which included an analyzer chamber A, constructed from a brass tube fitted with iron plates B, an entrance slit
S1, exit slit S2, and defining aperture D. The source consisted of an electron-emitting filament F and a heating element H. Ion currents
were detected by the electrometer fitted in an ebonite plug, E (Figure modified after Dempster, 1918). (b) Cross-sectional schematic
of Nier’s 1940 thermal ionization mass spectrometer consisting of widely separated source and collector whose slits S2 and S3,
respectively, are collinear with the center of curvature of the flight tube (Modified after Figure 2 of Nier, 1940). The flight tube passed
through a 60 magnetic sector analyzer and had a length of 58 cm between S2 and S3. The basic design has remained similar after
more than 70 years of improvements.

be separated on the basis of their masses came in 1913
when J.J. Thomson constructed a spectrometer at the Uni-
versity of Cambridge Cavendish Laboratory that used a
photographic emulsion to show that neon gas produced
two spectral lines of differing intensities representing
20
Ne (90.48 %) and 22Ne (9.25 %). Modifications of the
original design made shortly thereafter by A.J. Dempster
at the University of Chicago in 1918 and F.W. Aston at
the Cavendish Laboratory in 1919 included addition of
180 sector magnetic fields enabling directional
refocusing, separate thermal ionization filaments to pro-
duce and accelerate ions, source and exit slits to better
refine the ion beam, and a separate electric field to remove
the velocity spread and yield point images (Figure 1a).
Even though early instruments had low-mass resolution
and sensitivity, they were capable of identifying all of
the naturally occurring isotopes of the light elements as
well as their relative abundances.
In 1940, Alfred Nier at the University of Minnesota
revolutionized mass spectrometry by developing a design
that utilized a 60 sector-field magnet and moved the ion
source and ion collectors away from the influence of the
electromagnet (Figure 1b; Nier, 1940; De Laeter and Kurz,
2006). Nier also completely redesigned the ion source and
introduced the electronic measurement of ion beams.
These fundamental changes made production and opera-
tion of magnetic sector mass spectrometers more efficient
and reliable and opened the field of isotope geosciences
far beyond the initial scope imagined by earlier physicists.
Many modern instruments still use Nier’s designs.
Subsequent improvements in mass spectrometer per-
formance are more related to incorporation of digital oper-
ation and control, refinement of various electronic and
vacuum technologies, and innovation in sample introduc-
tion into the ion source. Improvements in instrumentation
coupled with the use of clean-laboratory conditions to
minimize contamination were stimulated by the lunar
sample return missions in the late 1960s and early 1970s
and the accompanying interest in understanding solar sys-
tem evolution by studying meteorites with low elemental
abundances for many of the geochronological systems.
The development of the plasma source coupled to a
multicollector magnetic sector system represented another
major wave of improvements in the mid-1990s, allowing
the measurement of isotopic ratios for nearly any element
in the periodic table.
Principles of mass spectrometry
Mass spectrometers are designed to separate ions based on
their mass-to-charge ratio. To do this, atoms or molecules
of interest in a sample must become charged forming an
ion cloud, extracted from that cloud by applying an accel-
eration voltage (typically 1–10 kV), and focused into a
narrow beam using a series of source plates that constitute
the ion optics of the instrument. Ions in the accelerated
 MASS SPECTROMETRY 535

a
Magnet
pole Magnet pole pieces
b
pieces
V

v F

ION OPTICS F
ION m-1 m m+1
B SOURCE FARADAY COLLECTORS
Mass Spectrometry, Figure 2 Diagram depicting the Lorentz force law operating in magnetic sector mass analyzers. (a) Oblique
schematic view of the magnetic pole pieces designed to provide a uniform magnetic field B. The resulting force (F) operates at right
angles to both the ion velocity vector (v) and the magnetic field vector (B) at any point along the flight path. (b) Plan view showing
the paths of ions of different masses (m) affected by the Lorentz force to different degrees. Ions with heavier masses (m + 1) follow
less curved paths than ions with lighter masses (m
1).

beam have kinetic energy that is equal to the electrostatic
field produced in the source (½ mv2 ¼ qVE/d; where
m ¼ mass, v ¼ velocity, q ¼ charge, VE ¼ extraction volt-
age; and d ¼ distance over which the acceleration takes
place). The beam thus produced is introduced into a mass
analyzer where dispersion takes place using magnetic
fields (magnetic sector instruments), rapidly cycling elec-
tric fields (quadrupole instruments), or velocity selection
(time-of-flight instruments).
Magnetic sector instruments
An ion with mass and charge traveling at velocity through
a uniform magnetic field (B) will experience a centripetal
force that imparts a circular motion. As a result the ion will
follow a curved path having a radius (R) within a plane
perpendicular to the magnetic field (F ¼ qvB ¼ mv2/R).
Magnetic sector mass spectrometers take advantage of this
principle (Lorentz law) by using electromagnets specifi-
cally designed to produce a homogeneous field that is
positioned at right angles to the path of the ion beam
(Figure 2). Ions traveling through the magnetic field will
follow an arc with a radius proportional to its mass (R ¼
mv/qB). As other factors are held constant, ions of differ-
ent masses follow paths with different radii through the
magnetic field, with heavier ions being deflected less than
lighter ones.
To steer beams of lighter and heavier ions into the col-
lector slit on a single-collector spectrometer, either the
magnetic field or the velocity of the ions (via changes to
VE) can be varied to alter the radius of curvature via the
Lorentz law. In practice, the magnetic field is more pre-
cisely monitored using a Hall probe that senses the field
strength and adjusts the magnet power supply as necessary
(Dickin, 2005). In multicollector instruments, a number of
faraday cups and ion counters are positioned to simulta-
neously measure multiple dispersed beams, thus avoiding
temporal variability in the signal due to ion beam instabil-
ities. Recent advances in magnet design feature poles set
at oblique angles to the exiting beam in order to use fring-
ing fields for better focusing. Because this design also
increases the distance from the magnet to the ion focal
point, it is referred to as “extended geometry.” This design
allows increased ion transmission and greater dispersion,
which, in turn, facilitates additional collectors. Ion trans-
mission in magnetic sector instruments can be close to
100 %, whereas multiple-sector instruments commonly
have lower transmission.
High-precision measurement of isotope ratios needed for
geochronology requires that magnetic sector mass spec-
trometers maintain very high vacuum throughout the ion
path in order to avoid reduced velocities or loss caused by
inelastic collisions with ambient air molecules. Pressures
within the flight tube (ion path between the source and
collector) typically are between 10
9 and 10
8 mbar. High
vacuum is maintained by continuous pumping using a com-
bination of turbo pumps (momentum transfer pump that
uses high-speed rotating blades to capture gas molecules)
backed by positive displacement mechanical pumps,
cryotraps (cold fingers), and ion getters (entrapment pump
that uses electrical fields to ionize gas molecules that are
then captured on a charged solid substrate).
Quadrupole instruments
Although magnetic sector mass spectrometry is the most
common method used for measurements of high-precision
isotope ratios, dispersion of isotopes on the basis of mass-
to-charge ratios also can be accomplished by applying
electric fields to two pairs of metal electrodes forming a
quadrupole mass analyzer (Miller and Denton, 1986).
Opposing pairs of rods are electrically connected, and a
combination of fixed DC and alternating RF (radio fre-
quency) potentials are applied such that two rods have
536 MASS SPECTROMETRY
a ELECTROSTATIC
ANALYSER
MAGNET
FORWARD GEOMETRY
SOURCE
SLIT
COLLECTOR
SLIT
Mass Spectrometry, Figure 3 Schematic depictions showing the ion beam paths (in blue) passing through two types of double-
focusing mass spectrometers. Designs employ both magnetic sector analyzers for mass dispersion and electrostatic analyzers (ESA) to
compensate for energy dispersion inherent in inductively coupled plasma and secondary ion sources. (a) Forward geometry where
the magnetic sector precedes the ESA and (b) reverse geometry where the ESA precedes the magnetic sector (Modified after Figure 3
of Ireland (2013)).
oscillating positive and negative voltages. Ions are
extracted from an inductively coupled Ar plasma source
and introduced at one end of the quadrupole. Oscillations
of charges on the electrodes cause complex ion trajectories
where only ions with the proper mass-to-charge ratio for a
given set of voltages can traverse the length of the quadru-
pole. All other ions will have unstable trajectories and be
neutralized by collisions with the electrodes. Ions that
enter the collector slit at the far end of the quadrupole
are measured as described above.
Quadrupole mass spectrometers benefit from the ability
to switch masses very quickly allowing collection of data
from the entire periodic table during a single, rapid analy-
sis. They require only modest vacuum conditions and are
not as complex or costly as magnetic sector instruments.
These attributes allow quadrupole instruments to excel in
determination of elemental concentrations in either solu-
tion mode or attached to a laser ablation system. However,
quadrupole instruments are not as sensitive nor as stable as
magnetic sector instruments and are typically operated at
unit mass resolution, which typically restricts their use to
low-resolution applications (Gross, 2011). Therefore, they
cannot produce the same level of high-precision, isotope–
ratio data compared to magnetic sector instruments and
are not widely used in geochronological studies.
Time-of-flight instruments
Time-of-flight (TOF) mass analysis relies on the fact that
heavier ions are accelerated more slowly than lighter ions
of the same charge in the same potential field. TOF
instruments are designed to measure differences in
arrival times of ions that are generated during a discrete
event. In practice, a group of ions is typically generated
using a pulsed laser shot, a pulsed beam of primary ions,
or a pulse generator coupled with a plasma source. Mass
separation is accomplished in the flight tube as ions in the
pulse are separated into individual packets defined by
mass number. Additional mass dispersion can be
b
ELECTROSTATIC
ANALYSER
MAGNET
REVERSE GEOMETRY
SOURCE COLLECTOR
SLIT SLIT
accomplished by extending the path length through
incorporation of reflection geometries.
Time-of-flight analysis has been primarily used to pro-
duce elemental and molecular maps of solid surfaces as
well as depth profiles in material-science applications.
Benefits include a theoretically unrestricted mass limit,
analysis of a complete mass range for a single ionization
pulse, high ion transmission, and extremely small sample
requirements (<10
18 mol). However, time-of-flight ana-
lyses generally do not produce fully quantitative data and,
as such, have not yet been used to produce geochronolog-
ical information.
Double-focusing instruments
Production of ions in mass spectrometers using plasma
sources or secondary ion sources results in ions with a
range of kinetic energies as well as mass-to-charge ratios.
Dispersal of ion energy results in poor mass resolution due
to beam spread and increased tailing. In order to narrow
the spread of ion energies and improve mass resolution,
both an electrostatic analyzer (ESA) and a magnetic sector
analyzer are used in series, resulting in a double-focusing
mass spectrometer (Figure 3). The ESA consists of a set of
two concentric curved plates held at different potentials to
filter and focus the ions traveling through them, regardless
of their mass. The ESA may be placed before (forward
geometry) or after (reverse geometry) the magnetic sector.
The use of double-focusing designs is critical for
obtaining the high-precision measurements required for
geochronological studies when using plasma source
instruments or ion microprobes; however, an ESA is not
required for thermal ionization instruments (see descrip-
tions of different mass spectrometers below).
Other types of tandem geometries have been employed,
including double- or triple-focusing designs with multiple
magnetic sectors (de Hoffman, 1996). These designs
result in increased abundance sensitivity and high selec-
tivity but suffer from increased complexity, cost, and
transmission losses.
 MASS SPECTROMETRY 537

Mass Spectrometry, Table 1 Distribution of masses steered to different collectors that have been positioned at single atomic mass
unit spacing for three types of data collection (dynamic peak jumping, static multicollection, and multi-dynamic triple jumping) for
Sr-isotopic compositions (84Sr, 86Sr, 87Sr, and 88Sr) while monitoring 85Rb to correct mass 87 for 87Rb

Collector number
Analysis type Magnet jump Lo-3 Lo-2 Lo-1 Axial Hi-1 Hi-2 Hi-3

Dynamic peak jumping 1 – – – 84 – – –

2 – – – 85 – – –
3 – – – 86 – – –
4 – – – 87 – – –
5 – – – 88 – – –
Static multicollection 1 – – 84 85 86 87 88
Multi-dynamic triple jumping 1 – – 84 85 86 87 88
2 – 84 85 86 87 88 –
3 84 85 86 87 88 – –

Ion detection spectrometers (Sinha et al., 2011). However, these detec-

Two general types of detectors are commonly used to mea-
sure ion-beam intensity. For large beam currents up to
10
10 A, a deep, narrow faraday cup positioned behind
the collector slit captures the incoming ions. The incident
ions are neutralized by electrons from the ground that flow
through a large resistor (1010–1012 O). The potential
across the resistor is amplified and read by a digital volt-
meter. Signals of several millivolts up to 50 V can be
reliably measured using faraday cups.
At lower ion beam currents, electrical noise becomes a
substantial component of the total signal present in the
cups. In order to measure beam currents smaller than
about 10
13 A, collectors that use signal multiplication
are needed. The two types most commonly employed are
Daly detectors and secondary electron multipliers
(SEMs). A Daly detector consists of a polished metal
“knob” placed under a large negative potential (Daly,
1960). Incoming positively charged ions are attracted to
the knob and, upon impact, release a shower of secondary
electrons. These electrons are accelerated toward a nearby
positively charged phosphor screen where they generate
light pulses that are amplified and measured by a sensitive
photomultiplier. SEMs consist of either discrete-dynode
or continuous-dynode (channeltron) designs that rely on
secondary electron emission after initial collision of an
ion on the dynode surface (Allen, 1947; Poenisch,
1976). Release of additional electrons causes a cascade
effect, multiplying the initial collision by gain factors of
104–108 (Gross, 2011). Electrical output for Daly and
SEM detectors is typically given in counts per second that
must be calibrated (at least at the upper end of the range)
against voltages produced for the same beam on a
faraday cup.
Charge-coupled device (CCD) detectors, a relatively
new type of silicon-based electronic device capable of
detecting multichannel arrays of electromagnetic radia-
tion, are capable of measuring impinging ion currents
and have been used as detectors on miniaturized
tors have not yet been incorporated into instruments used
for high-precision geochronological work.
Data collection
Data collection in modern computer-controlled mass spec-
trometers involves measurement of multiple scan cycles in
order to attain high-precision results. Single-collector
instruments require magnet cycling from low to high-
mass, counting intensities at each mass (and background
positions) before returning to the low-mass peak to repeat
the cycle. This form of analysis is referred to as dynamic
peak-jumping mode and requires stable ion beam intensi-
ties to achieve precise isotope ratios. Multicollector instru-
ments running in static mode do not require changes in
magnet current and can therefore tolerate short-term beam
instability because ion beams for all masses of interest are
measured simultaneously. Static multicollection methods
still gather data in separate blocks that allow statistical
treatment and correction for instrument drift and beam
growth or decay. However, static multicollection requires
that corrections for electronic biases between detectors
and amplifiers must be made by running collector gain cal-
ibrations with some frequency. Alternatively, high-
precision measurements can be made on multicollector
instruments using a multi-dynamic analysis that involves
double or triple magnet jumps that measure the same iso-
tope ratios in sequential sets of collectors. This method
is capable of canceling out beam growth or decay,
correcting for collector/amplifier bias, and applying a
power law mass fractionation correction. An example of
these three types of data collection schemes intended to
determine the isotopic composition of Sr is shown in
Table 1. Total length of analysis time depends on signal
intensity and desired precision but can take from a few
minutes to several hours for the highest-precision results.
A variety of other corrections are necessary to produce
the most precise and accurate data including baseline sub-
traction, detector linearity, and isobaric interference.
538 MASS SPECTROMETRY
1.0
0.5 c 50%
Δm 10%
0.0 s and represents the ratio of the maximum ion current mea-
m
Mass Spectrometry, Figure 4 Schematic depiction of an
instrumental scan across an isotope of mass m, showing a
trapezoidal-shaped peak with a flat top at a fractional intensity
of 1 (thick, gray line). Peak width measured at 50 % of the peak
height represents the collector slit width (c) and the width of the
peak slope (s) represents the width of the demagnified source
slit. Mass resolution at the 10 % peak height is given by m/Dm
(Modified after Figure 6 of Ireland (2013).
Mass resolution
One of the main performance considerations in mass spec-
trometry is the ability of an instrument to distinguish two
peaks having different mass-to-charge ratios. This is easier
for lower-mass elements where differences in isotope
masses are large relative to total mass, but becomes
increasingly difficult for high-mass elements or for differ-
ent species having similar masses (for instance, 29Si
vs. 28Si1H or 87Rb vs. 87Sr). Mass resolution is specified
as the overall mass under consideration divided by the dif-
ference between masses being considered (m/Dm). For
instance, a mass spectrometer needs a mass resolution of
only 87 to resolve 86Sr from 87Sr (87/[86.9089-85.9083])
but a value of about 3,500 to resolve 29Si from 28Si1H
(29/[28.9765-28.9848]) and more than 200,000 to resolve
87
Rb from 87Sr (87/[86.9092-86.9089]). Modern magnetic
sector mass spectrometers have mass resolutions ranging
from 300 to more than 5,000.
Mass resolution can be determined by noting the Dm
for the closest spacing of two, equal-intensity peaks sepa-
rated by a minimum signal less than a specified fraction of
peak height (typically 10 % or 50 %). A more common
and easily measured determination of mass resolution uses
the width of a peak (in Dm) measured at a stated fraction of
the peak height (commonly at 1 %, 10 %, or 50 %, the lat-
ter of which yields full width at half maximum, or FWHM;
Figure 4). Ideally, peaks obtained by scanning the ion
beam across the collector slit will have flat tops and
steeply sloping sides that allow minor mass drift without
affecting ion beam intensity. Width at the base of the peak
is approximately the sum of the widths of the source and
collector slits, whereas the width of the peak flat is related
to the ratio of the collector slit width to the ion beam
width. A narrower collector slit will result in higher mass
resolution, but will reduce the width of the peak flat.
Reducing the width of the source slit will also result in
higher mass resolution along with increased peak flat,
but will cause a reduction of ion beam transmission.
Therefore, mass resolution is often a compromise between
the ability to effectively separate peaks with similar
masses and the ability to acquire high-precision isotope
data from those peaks.
A concept related to mass resolution is the amount that
the tail of one peak may contribute to the signal of an adja-
cent peak. This measure is called abundance sensitivity
sured at a specified mass divided by the maximum ion cur-
rent present for the same species measured at an adjacent
mass. It is commonly measured at mass 237 (no natural
isotope) while running a large 238U signal. Tails arise from
collision of ions in the flight tube with residual gas mole-
cules resulting in a loss of energy and a shift toward the
lower-mass side of the peak. Abundance sensitivity is
therefore related to the quality of the vacuum in the instru-
ment. Flight tube pressures in the low 10
9 mbar range
commonly yield abundance sensitivities of ~0.000002
(2 ppm), although this can be improved by up to two
orders of magnitude by addition of an energy filter in front
of the collector or by designs incorporating multiple mag-
netic sectors.
Chemical separations
In order to reduce the number of isobaric interferences and
to improve precision and reproducibility of the measure-
ments, samples used for geochronological purposes are
commonly reduced to pure chemical forms using cryo-
genic separation for noble gases (Ar) or liquid chromatog-
raphy after acid digestion for solids. Only the purified
elements of interest are then introduced into the ion source
of the mass spectrometer, reducing or eliminating interfer-
ences associated with complex matrices typical of geo-
logic material. Consequently, mass resolution
requirements are simplified to unit mass differences, and
isobaric interferences from combinations of lower-mass
molecules are dramatically reduced.
Chemical separations are usually performed under
cleanroom conditions involving highly filtered airflow
and the use highly refined reagents. Cleanroom conditions
are designed to reduce the potential contaminants added
during processing, which in turn results in reduced
amounts of materials not present in the original sample
(known as “blank”). Low chemical processing blanks
allow more accurate analysis of substantially smaller-
sized samples, which facilitates dating of rocks using
isochrons derived from mineral separates, or when total
sample size is very limited. Analysis of solid materials
(rocks and minerals) requires quantitative digestion using
high-purity hydrofluoric, nitric, and hydrochloric acids.
Complete dissolution of a known amount of sample
allows the addition of a “spike” consisting of a solution
containing a known amount of purified tracer isotope.
Once the tracer is isotopically equilibrated with the sam-
ple, it can be used to monitor and correct for mass
 MASS SPECTROMETRY
a VARIABLE
MULTICOLLECTOR
MAGNET
AMPLIFIER
HOUSING
BEAM
ISOLATION
VALVE
ION
OPTICS
MULTI-SAMPLE
FILAMENT MAGAZINE
Mass Spectrometry, Figure 5 Schematic diagram depicting designs of modern commercially available magnetic sector instruments
consisting of a single-focusing solid-source thermal ionization mass spectrometer (a), and a double-focusing inductively coupled
plasma mass spectrometer with forward geometry (b). Note that the two instruments are identical past the beam isolation value
(Diagrams courtesy of Thermo Fisher Scientific, Inc.).
fractionation that occurs during the isotope analysis or for
quantifying the concentration of the element of interest in
an unknown through a procedure known as isotope dilu-
tion (Dickin, 2005; Faure and Mensing, 2005).
Mass spectrometry used for dating
Principles described above are common to a variety of
mass spectrometers used to collect isotope ratios for geo-
chronological purposes. The following groupings are
largely based on differences in how ions are produced.
Comparisons between these mass spectrometric methods
applied to radiogenic elements used for dating rocks and
minerals can be found in a number of textbooks and
review articles (Goldstein and Stirling, 2003; Faure and
Mensing, 2005; Dickin, 2005; Siebel and van den Haute,
2007; Ireland, 2013).
Thermal ionization mass spectrometry
Thermal ionization mass spectrometry (TIMS) relies on
thermal energy to produce ions from a sample placed on
a thin ribbon (filament) of a highly purified refractory
metal. Solid samples are prepared by acid digestion
followed by chemical separation processes that are spe-
cific to the individual elements of interest. After separa-
tion, the resulting purified salt is loaded as a small drop
of liquid onto a filament of rhenium, tantalum, or tungsten,
either by itself or along with an activator such as Si-gel,
Ta-oxide, or carbon suspension that helps stabilize or
enhance ion emission. Depending on the ionization char-
acteristics of a given element, samples can be loaded onto
a single filament that produces both evaporation and ioni-
zation or on double- or triple-filament assemblies where
separate filaments control evaporation and ionization.
Loaded filaments are placed in the TIMS source, which
is evacuated to pressures of ~10-8 mbar, and then heated
by passing a current of about 1–5 A. Heating under these
539
b VARIABLE
MULTICOLLECTOR
MAGNET
AMPLIFIER
HOUSING
BEAM
ISOLATION
VALVE
PLASMA
INTERFACE
ELECTRO- TORCH
STATIC
ANALYSER TRANSFER
OPTICS
conditions yields filament temperatures of 800–3,500  C,
which is sufficient to cause sample evaporation and at
least partial ionization of a number of elements. Ions are
extracted from the evaporation cloud in front of the fila-
ments and accelerated through the source optics into the
mass analyzer (Figure 5a).
TIMS analyses of natural, nonvolatile radioactive iso-
tope systems have been the historical standard in geochro-
nology studies for more than 60 years. This includes
dating systems such as U-Pb, Rb-Sr, Sm-Nd, and more
recently Lu-Hf, Re-Os (in negative-ion mode), and U-Th
disequilibrium studies. Advantages of TIMS analyses are
in part derived from the careful chemical preparation which
reduces mass-resolution requirements. The fact that differ-
ent elements evaporate at different temperatures can also
be used as an advantage for eliminating unwanted interfer-
ences to a certain extent (i.e., removal of 87Rb at low tem-
perature prior to analysis of 87Sr at higher temperature) or
for combined runs from the same filament (i.e., single-grain
zircon or opal dating with no chemical separation where Pb
data are collected at low temperature, U at intermediate
temperature, and Th at high temperature). Thermal ioniza-
tion also results in ions with a low energy spread allowing
TIMS instruments to be relatively simple, relying only on
single-focusing designs (magnetic sector). The main disad-
vantage of TIMS is that many elements have low to very
low ionization efficiencies resulting in detection of only a
small fraction of the total atoms present on the filament.
Variations in matrix composition can also affect sample
chemistry, resulting in poor ionization for impure separates.
These shortcomings can be overcome by improving chem-
ical separations and, in some cases, by adding an enhancer
to the filament along with the sample load to increase the
ionization efficiency without adding blank. TIMS analysis
times are lengthy (several hours to obtain the highest-
precision results), and total throughput can be low due to
arduous chemical preparations.
540 MASS SPECTROMETRY

a RF coil
Liquid-
cooled
Fassel-type torch interface
Aerosol from laser Skimmer
Sample cone cone

Spray
chamber

Asperated Auxillary Plasma Port to Liquid-

liquid sample gas gas rough pump cooled
inter-
face Extraction plates
and ion-optics
assembly
b RF coil
Fassel-type torch
ION BEAM
PLASMA
Aerosol stream Skimmer
containing sample cone
(~1 l/m)
Sample Entrance
Auxillary cone slit
gas
(~1 l/m) Plasma Evacuate
gas – to rough Evacuate
coolant pump to turbo
(10-15 l/m) pump

Mass Spectrometry, Figure 6 (a) Photograph of the source area of an inductively coupled plasma mass spectrometer showing
the layout of major components (left) and the sample and skimmer cones (right) which are attached to the liquid-cooled interface.
(b) Schematic diagram of the main components shown in photograph (a).

Inductively coupled plasma mass spectrometry
Inductively coupled plasma–mass spectrometry (ICP–MS)
uses a plasma source utilizing Ar as the plasma support gas
(Figure 6). A Fassel-type torch with automated radio-
frequency matching allows the aerosol gas to punch a hole
through the plasma. The Fassel torch uses relatively low
gas flow and can be used for aqueous samples using either
a peristaltic pump or Meinhard nebulizer, or, for “dry”
samples from which most of the liquid has been removed,
by desolvation using a microconcentric nebulizer. The
nebulizer converts the solution into an aerosol that, while
passing through the plasma source, is desolvated, dissoci-
ated, and ionized. The interface region of the instrument
usually consists of two cones, separated by a very small dis-
tance that is being pumped by a rough pump. The whole
interface region is made from material that dissipates heat
easily such as copper or aluminum that is also cooled by
circulating chilled water to reduce the effects of the high-
temperature plasma (up to 8,000  K) on the cones. The first
cone, known as the sampler cone, typically has an orifice of
0.8–1.2 mm internal diameter, while the second cone, the
skimmer cone, has a much smaller orifice (0.4–0.8 mm).
The cones are usually made of Ni, although different mate-
rials, such a Pt, can be used depending on the element being
ionized and the aggressiveness of the matrix. The role of the
interface region is to transport the ions efficiently, consis-
tently, and with electrical integrity from the plasma, at atmo-
spheric pressure, to the mass spectrometer analyzer region,
where pressure is at ~10
7 or ~10
8 mbar. When the ions
emerge from the skimmer cones, they are directed through
the ion optics that will focus the beam into the analyzer
region of the mass spectrometer (Figure 5b).
The advantage of the ICP source is that it can ionize
nearly all elements in the periodic system, and when com-
bined with a multicollector system as described above, it
allows for the measurement of isotopic compositions of
many more elements than TIMS methods, especially those
that are difficult to ionize by thermal methods. This
opened up the analysis of many useful systems in geochro-
nology, such as Al-Mg, Fe-Ni, Mn-Cr, Pd-Ag, and Hf-W
as well as allowing for more precise analysis of Hf and
Pb isotopic compositions.
The ICP source can also be coupled to a laser ablation
(LA) system that allows in situ production of ions while
 MASS SPECTROMETRY
imaging the sampling process (Košler and Sylvester,
2003; Longerich, 2008; Koch and Günther, 2011).
LA–ICP–MS has been used extensively for U-Pb dating
of zircons, as well as on other mineral matrices when a
detailed understanding of sample texture and composition
is required. Laser ablation adds more complexity to ICP–
MS systems such as the effects of laser wavelength and
pulse length on formation of aerosols, processes of elemen-
tal fractionation during lasing and in the vapor stream, ade-
quacy of reference materials and calibration of the system
using age standards, sampling strategies and analytical pro-
tocols, and data collection and reduction strategies (see
numerous chapters in Sylvester, 2008). Despite these com-
plexities, LA–ICP–MS represents a successful and power-
ful dating tool with a growing number of applications.
There are strong matrix effects in ICP–MS that tend to
reduce the reproducibility and accuracy of the analysis. To
overcome this disadvantage, the element of interest can be
carefully separated from its matrix through ion exchange
while insuring a maximum yield (>95 %) through the pro-
cess. The major limiting factor on the precision of
multicollector-ICP–MS isotopic analyses stems from the
instability of the plasma. An ESA is used to clean up the
energy dispersion inherent in the plasma source and make
it match the dispersion of the sector analyzer. In addition,
mass fractionation associated with ionization in the ICP
source is significantly larger than that observed in TIMS.
However, instrument-induced mass biases are usually
constant during the analyses allowing for proper correc-
tion, either by sample-standard bracketing or by the addi-
tion of another element of comparable mass range such as
Tl to monitor Pb mass bias. Finally, because the ICP
source is capable of ionizing all constituents introduced
into the plasma, corrections for spectral interferences that
may overlap the masses of interest must be made.
Noble gas mass spectrometry
Radiogenic noble gases including 4He and 40Ar can reside
within certain minerals over geologic time scales and
are often used in research applying geochronology and
thermochronology. Noble gas mass spectrometry requires
extraction and purification of exceedingly small fractions
of gas from mineral grains under ultrahigh vacuum systems
and is generally done using a high-temperature furnace or
infrared laser. Heating can be done incrementally, such that
gas fractions having different isotopic compositions can be
separated. In some cases, step heating may provide informa-
tion constraining the thermal evolution of the sample. Gases
released from minerals during heating are isolated using
chemical and cryogenic traps. The noble gas of interest is
introduced into a gas-source mass spectrometer through
an inlet port. Positive ions are created when the gas interacts
with a heated filament causing ejection of an outer electron,
and the resulting ions are accelerated through the source
optics into a single-focusing magnetic sector mass spec-
trometer. The noble gas isotopes are typically measured in
static mode.
541
Secondary ion mass spectrometry
Secondary ion mass spectrometers (SIMS) were devel-
oped to provide in situ elemental and isotopic composi-
tions on samples at high spatial resolution (Ireland
et al., 2008). To do this, a primary beam of energetic ions
is focused onto a small spot on the sample (typically
10–50 mm). Bombardment of the sample surface causes
ablation and generates a variety of metal ions and more
complex species that can be extracted to form a secondary
ion beam. Chemically active elements, such as O
and
Cs+, are most commonly used for the primary beam
because of their enhanced ability to generate secondary
ions. Because secondary ions generated in this manner
have a large range of both energies and mass-to-charge
ratios, SIMS instruments employ a dual-focusing design
with either forward or reverse geometries (Figure 3). They
also require high-mass resolution in order to discriminate
between numerous secondary ion species that may have
similar masses. To accomplish this, these instruments
employ large magnetic and electrostatic sectors and long
flight paths.
In the last 30 years, SIMS analyses of the U and Pb iso-
topic compositions of zircon have become a critical tool
for dating igneous and metamorphic rocks and for provid-
ing age data used in sedimentary provenance studies.
Although individual ages are not as precisely determined
as those analyzed by TIMS, SIMS has the distinct advan-
tage of being able to analyze distinct spatial domains
within a single grain that may be related to separate geo-
logical events. Furthermore, in situ SIMS analyses elimi-
nate the involved and time-consuming step of chemical
processing required by TIMS. Additional elements can
be added as part of the same SIMS spot analysis (i.e.,
rare-earth and high-field-strength elements in zircon)
allowing petrologic interpretations to accompany age data
(petrochronology). Other minerals besides zircon have
been dated using SIMS by both U-Pb and U-series
methods (monazite, xenotime, titanite, baddeleyite, rutile,
and opal). Unlike LA–ICP–MS, the SIMS primary beam
is used both to sample the target and to simultaneously
generate ions from resulting volume. The SIMS
“sputtering” process is much less aggressive than vapori-
zation via laser ablation and evacuates sample volumes
that are only several micrometers rather than several tens
of micrometers deep. Consequently, SIMS analyses
require longer data acquisition times but are less likely to
incorporate zones of disparately aged material at depth.
Mass-dependent fractionation is also less pronounced in
SIMS analyses, and isotope ratios require less elaborate
correction procedures.
Summary
Over the last 100 years, mass spectrometry has become a
critical aspect of geological science by providing ages
from isotope–ratio measurements of natural radioactive
decay. Mass spectrometry has also provided isotopic com-
positions of a myriad of solids, liquids, and gases that have
542 METAMORPHIC TERRANES (K–Ar/40Ar/39Ar)
been invaluable for understanding earth science processes,
the rates at which they occur, and the characterization of
the compositions of numerous geochemical reservoirs.
Mass spectrometry is based on the principle that
charged particles moving through electromagnetic fields
are affected in ways that allow the dispersion of ions of
different masses into separate signals that can be measured
independently. All mass spectrometers consist of three
main components: (1) a source of ions and a means of
imparting velocity, (2) a mass analyzer that separates the
moving ions on the basis of their mass-to-charge ratios,
and (3) a collector system that can quantify the intensity
of separated ion beams. Most mass spectrometers used
for geochronological studies must be able to produce
high-precision isotope–ratio measurements of stable ion
beams and therefore make use of a magnetic sector mass
analyzer. Because it is possible to generate ions and intro-
duce them into the mass analyzer by various means, sev-
eral types of mass spectrometers have been developed.
These instruments can provide precise and accurate mea-
surements on a wide range of isotope systems using a
spectrum of different materials and analytical strategies.
Bibliography
Allen, J.S., 1947. An improved electron multiplier particle counter.
Review of Scientific Instruments, 18, 739–749, doi:10.1063/
1.1740838#blank. http://dx.doi.org/10.1063/1.1740838
Daly, N. R., 1960. Scintillation type mass spectrometer ion detector.
Review of Scientific Instruments, 31, 264–267.
de Hoffman, E., 1996. Tandem mass spectrometry: a primer. Jour-
nal of Mass Spectrometry, 31, 129–137.
De Laeter, J., and Kurz, M. D., 2006. Alfred Nier and the sector field
mass spectrometer. Journal of Mass Spectrometry, 41, 847–854.
Dempster, A. J., 1918. A new method of positive ray analysis. Phys-
ical Review, 11, 316–325.
Dickin, A. P., 2005. Radiogenic Isotope Geology, 2nd edn. Cam-
bridge, UK: Cambridge University Press, 452 p.
Faure, G., 1986. Principles of Isotope Geology, 2nd edn. New York:
Wiley, 589 p.
Faure, G., and Mensing, T. M., 2005. Isotopes – Principles and
Applications, 3rd edn. New York: Wiley, 897 p.
Goldstein, S. J., and Stirling, C. H., 2003. Techniques for measuring
uranium-series nuclides: 1992–2002. In Bourdon, B., Hender-
son, G. M., Lundstrom, C. C., and Turner, S. P. (eds.), Uranium-
Series Geochemistry. Washington, DC: Mineralogical Society of
America. Reviews of Mineralogy and Geochemistry, Vol.
52, pp. 533–576.
Gross, J. H., 2011. Mass Spectrometry, 2nd edn. Berlin/Heidelberg:
Springer, 754 p.
Ireland, T. R., 2013. Invited review article: recent developments in
isotope-ratio mass spectrometry for geochemistry and cosmo-
chemistry. Review of Scientific Instruments, 84, 011101, 21 p.
Ireland, T. R., Clement, S., Compston, W., Foster, J. J., Holden, P.,
Jenkins, B., Lanc, P., Schram, N., and Williams, I. S., 2008.
The development of SHRIMP. Australian Journal of Earth Sci-
ences, 55, 937–954.
Koch, J., and Günther, D., 2011. Review of the state-of the-art of
laser ablation inductively coupled plasma mass spectrometry.
Applied Spectroscopy, 65, 155A–162A.
Košler, J., and Sylvester, P. J., 2003. Present trends and future of zir-
con in geochronology: laser ablation ICPMS. In Hanchar, J. M.,
and Hoskin, P. W. O. (eds.), Zircon. Washington, DC:
Mineralogical Society of America. Reviews in Mineralogy and
Geochemistry, Vol. 53, pp. 243–275.
Longerich, H., 2008. Laser ablation-inductively coupled plasma-
mass spectrometry (LA-ICP-MS); an introduction. In Sylvester,
P. J. (ed.), Laser Ablation-ICP-MS in the Earth Sciences, Current
Practices and Outstanding Issues. Vancouver: Mineralogical
Association of Canada. Mineralogical Association of Canada,
Short Course Series, Vol. 40, pp. 1–18.
Miller, P. E., and Denton, M. B., 1986. The quadrupole mass filter:
basic operating concepts. Journal of Chemical Education, 63,
617–622.
Nier, A. O., 1940. A mass spectrometer for routine isotope abun-
dance measurements. Review of Scientific Instruments, 11,
212–216.
Poenisch, A., 1976. Mass-spectrometer electron-multiplier
improvement. Journal of Vacuum Science and Technology, 13,
1110–1112, doi:10.1116/1.569086#Link. http://dx.doi.org/
10.1116/1.569086
Siebel, W., and van den Haute, P., 2007. Radiometric dating and
tracing. In Nagy, S. (ed.), Radiochemistry and Nuclear Chemis-
try, Encyclopedia of Life Support Systems, Developed Under the
Auspices of the UNESCO. Oxford, UK: Eolss Publishers. http://
www.eolss.net/sample-chapters/c06/e6-104-02.pdf
Sinha, M. P., Neidholdt, E. L., Hurowitz, J., Sturhahn, W., Beard,
B., and Hecht, M. H., 2011. Laser ablation-miniature mass spec-
trometer for elemental and isotopic analysis of rocks. Review of
Scientific Instruments, 82, 094102, doi:10.1063/1.3626794, 7 p.
Sylvester, P. J. (ed.), 2008. Laser Ablation-ICP-MS in the Earth Sci-
ences, Current Practices and Outstanding Issues. Vancouver:
Mineralogical Association of Canada. Mineralogical Associa-
tion of Canada, Short Course Series, Vol. 40, 348 p.
White, W. M., 2013. Geochemistry. Chichester, UK: Wiley-
Blackwell, Wiley, 668 p.
Cross-references
Accelerator Mass Spectrometry
Ar–Ar and K–Ar Dating
Historical Development of Dating Methods
Laser Ablation Inductively Coupled Mass Spectrometer
(LA ICP-MS)
Lu-Hf Dating: The Lu-Hf Isotope System
Noble Gas Mass Spectrometer
Rhenium–Osmium Dating (Meteorites)
Rubidium–Strontium Dating, Hydrothermal Events
Secondary Ion Mass Spectrometry (SIMS)
Sm–Nd Dating
Thermal Ionization Mass Spectrometer (TIMS)
U-Series Dating
U–Th/He Dating
Uranium–Lead Dating
METAMORPHIC TERRANES (K–Ar/40Ar/39Ar)
Jan R. Wijbrans
Department of Earth Sciences, Faculty of Earth and Life
Sciences, VU University, Amsterdam, The Netherlands
Synonyms
Metamorphic basement; Metamorphic belts
 METAMORPHIC TERRANES (K–Ar/40Ar/39Ar)
Definition
Metamorphism in its widest sense refers to processes of
alteration of rocks within the Earth crust and mantle that
cause recrystallization of an original mineralogy to form
new minerals that are thermodynamically stable at depth
and thus reflect in their chemistry and mineralogy the
changed physical conditions of pressure, temperature, stress
regime, and partial pressures of volatile phases such as H2O,
CO2, CH4, and others during the time of recrystallization.
Metamorphism covers a spectrum from partial recrys-
tallization to complete recrystallization. Partial recrystalli-
zation may reflect processes within a single cycle
(pluri-facial metamorphism) or multiple cycles of
metamorphism spaced out in time (poly-metamorphism).
Retrograde metamorphism reflects often partial recrystal-
lization after peak metamorphism.
Terranes are units of rocks at the crustal scale that share
common geological features and a common geological
history.
Metamorphic terranes therefore are terranes that consist
largely of metamorphic rocks that share a common history
of burial, metamorphism, and exhumation.
Introduction
Much of the continental middle and lower crust consists of
metamorphic rocks. Much of the upper mantle, as far as
we can tell from mantle rocks exposed at the Earth surface
in ophiolite units and from xenoliths brought up from the
upper mantle by magma transport, is also metamorphosed.
In the oldest nuclei of continental crust, preserved
Archean cratons, metamorphism may range from very
low grade in the greenstone belts to granulite grade in
the basement gneiss terranes. From the early Proterozoic
onward, metamorphic terranes are often found in long
and narrow belts, reflecting the fact they formed along
the margins of plates of continental crust or by conti-
nent–continent collision processes (Figure 1).
The general objective of geochronological analysis of
metamorphic terranes is to shed light on the age of the
protoliths and to place time constraints on the prograde
(burial) path, peak metamorphic temperature (T) and pres-
sure (P) conditions, and cooling after the peak of the meta-
morphic event, often during exhumation of the rocks
(Figure 2). Metamorphism is the alteration of the mineral
content of rocks caused by either changing PT conditions
during burial caused by subduction or continental colli-
sion and associated rock deformation events or by heat
or fluid advection as is, for example, the case when gran-
ites intrude into higher crustal levels or when basaltic
magma underplates the continental crust.
Peak temperatures reached during metamorphism
may range widely from ca. 400  C for the prehnite–
pumpellyite facies to temperatures in excess of
900–1,000  C for the ultrahigh-temperature domain of
the granulite facies. Metamorphic pressures recorded in
crustal rocks in extreme cases are known to have exceeded
lithostatic pressures consistent with what can be expected
543
for normal crustal thicknesses by at least three times,
suggesting either that during subduction or collision
events, crustal rocks may have been pushed to great
depths or that tectonic stress may build up to values sub-
stantially exceeding the normal lithostatic pressure. Min-
erals diagnostic for ultrahigh pressure include coesite
and diamond that form at pressures in excess of 2.5 and
3 GPa, respectively.
For slowly cooling metamorphic terranes, the K-Ar and
40
Ar/39Ar geochronological techniques record ages usu-
ally interpreted as corresponding to the Ar closure temper-
atures (Dodson, 1973) of their constituent minerals.
Although closure temperature is dependent on the grain
size of the minerals, such ages typically reflect the
time at which the rocks cooled below ca. 500–550  C
for metamorphic hornblende, ca. 150–400  C for
K-feldspars, 360–420  C for metamorphic muscovite,
and 280–330  C for metamorphic biotite, respectively.
The exact “closure” temperatures for which timing infor-
mation is obtained may be more difficult to determine,
as they depend on cooling rate and radii of the crystal
domains over which diffusion occurs. The determination
of the exact closure temperature may be hindered by par-
tial recrystallization or deformation, as under such condi-
tions non-volume diffusion processes may have altered
the age of the mineral.
As a consequence, the K-Ar and 40Ar/39Ar techniques
can typically be used to measure the age of metamorphic
peak temperatures in those areas where the peak tempera-
ture has not exceeded temperatures of 400–550  C for
considerable amounts of time, i.e., at the prehnite–
pumpellyite, greenschist, blueschist, and lower amphibo-
lite facies metamorphic grades (Warren et al., 2012).
In higher-grade metamorphic terranes, the K-Ar and
40
Ar/39Ar methods are useful to constrain the cooling path
of the rocks to temperatures below the closure tempera-
tures of hornblendes, muscovites, biotites, and K-feld-
spars, respectively. The K–Ar and 40Ar/39Ar methods
can be applied to a wide range of K-bearing minerals
and rocks, although hornblendes, micas, and K-feldspar
are the most widely used minerals used for dating in meta-
morphic terranes.
Historical perspectives
The K-Ar method has been in use since the 1950s,
whereas the 40Ar/39Ar method was first developed in the
mid-1960s and has since become the preferred method
for dating by the 40K-40Ar isotopic system. Whereas the
K–Ar technique was quite successful in dating of volcanic
and simple igneous rocks, it became apparent that the
method did not always work well in old, complex Precam-
brian metamorphic basement rocks, where consistently
younger ages were often found in rocks that had demon-
strably complex extended tectonic histories (e.g., Kinny
et al., 1990). Moreover, researchers working in younger,
Alpine metamorphic rocks found that in rocks and min-
erals dated with both the Rb/Sr dating technique and the
544 METAMORPHIC TERRANES (K–Ar/40Ar/39Ar)
Caledonides
North
American
Cordillera
Appalachians
Mauritanides
Andes
40 39
Metamorphic Terranes (K–Ar/ Ar/ Ar), Figure 1 Global distribution of orogenic belts containing metamorphic terranes,
highlighting linear belts from ca. 2.0–1.7 Ga.
K–Ar dating technique, the K-Ar technique often yielded
systematically younger ages (e.g., Armstrong et al., 1966).
This observation was used with advantage, when com-
bined with estimates for metamorphic temperature
deduced from the rocks used for dating, to estimate an
empirical temperature reflecting the opening and closing
of the isotopic system, a so-called “blocking” temperature.
The blocking temperature for a mineral was defined as the
temperature at which the isotopic system was reset during
metamorphic overprinting, yielding for the first time tem-
peratures above which isotopic ages could not be pre-
served (Jäger, 1979). Clearly, under such extreme
conditions, the K–Ar technique may only yield a partial
answer to the question of timing of the tectonic evolution
of the crust.
During this same period, the 40Ar/39Ar step-heating
method was used to derive diffusion parameters for the dif-
fusion of argon through the crystal lattices of various
K-bearing minerals. Results of these first attempts to mea-
sure diffusion rates in hydrous minerals such as biotite
and hornblende were unreliable as the mineral structures
become instable due to dehydration reactions in vacuo
(Mussett, 1969; Hoffmann, 1979). When such minerals
were kept stable at high temperatures under hydrothermally
controlled pressures, the diffusion results were more pre-
dictable (e.g., phlogopite, Giletti, 1974; hornblende,
Urals
Alps
alayas
m
Hi
Time span (billions of years ago)
0.25 to 0 2.5 to 1.7 2.5 to 0.7
0.7 to 0.25 3.8 to 2.5 3.8 to 1.7
1.7 to 0.7
Harrison, 1981). The experimentally determined diffusion
parameters, i.e., the pre-exponential coefficient, Do, and
activation energy, Ea, could thus be used to predict diffusion
loss from the crystal lattice of minerals at elevated tempera-
tures. This approach led to the definition of a closure tem-
perature (Dodson, 1973), formally defined in a system
cooling from high temperature as the temperature of the
mineral at the time determined by its isotopic age. In con-
trast, a blocking temperature refers to an empirically cali-
brated temperature of resetting of isotopic ages in
minerals. Although very powerful for the description of
cooling systems, the use of closure temperatures to describe
the loss of isotopic daughters during reheating is more com-
plex, because in practice when the original mineral remains
stable, the duration of the overprinting pulse, grain size, and
degree of deformation play a role in affecting the age of the
rock or mineral. Several researchers have attempted to
model diffusion numerically using modified heat flow
codes (Lister and Baldwin, 1996; Wheeler, 1996); this
approach is particularly useful to test the effects of various
parameters that may influence resetting by diffusion.
Diffusion is typically a rate-limiting process, so to
be measured accurately a diffusion experiment needs
to be carried out within the mineral’s stability field.
Because the amounts of diffusion loss during experiments
are very low, diffusion experiments present many
 METAMORPHIC TERRANES (K–Ar/40Ar/39Ar) 545

Greenschist
Amphibolite
High P High P, High T Granulite
Low T Decreasing Grade
Blueschist
Eclogite

IV

I III

II

Temperature (°c)
0 100 200 300 400 500 600 700 800 900 1000 1100
0
Diagenesis
Hornfels facies
Prehnite-Pumpelleyite IV
10 facies CONTACT
Greenschist METAMORPHISM
facies (High-temperature,
Pressure (kilobars)

5 I
Depth (kilobars)

Amphibolite low-pressure)
20 facies III
Blueschist Granulite
facies facies
II
30 Eclogite
10 facies
Eclogite facies REGIONAL
40 SUBDUCTION ZONE BURIAL METAMORPHISM
METAMORPHISM
(High-pressure,
low-temperature)
15 50
Lower limit
of metamorphism
Metamorphic Terranes (K–Ar/40Ar/39Ar), Figure 2 Section through a convergent margin and PT diagram showing the PT fields of the
various metamorphic grades. I, prograde path; II, peak metamorphic pressure; III, peak metamorphic temperature; IV, retrograde path.

analytical challenges; the results of several carefully con- Metamorphic Terranes (K–Ar/40Ar/39Ar), Table 1 Closure
trolled experiments are given in Table 1. An important temperature ranges
consideration is that Ar diffusion may result from different Mineral Temperature range ( C)
mechanisms that occur simultaneously in minerals. In
a perfect crystal lattice, diffusion of argon will proceed K-Feldspar ~150–400
by a point defect process (also commonly known as vol- Biotite ~280–330
ume diffusion); the large argon molecule has to push lat- Muscovite ~360–420
tice ions out of the way in order to migrate through the Hornblende ~500–550
lattice. Because the neutral argon molecule cannot substi-
Modified from Johnson and Harley (2012)
tute for the lattice-forming ions, an exchange mechanism
for Ar diffusion is not likely to occur. However, deformed
crystals will contain defects or dislocations of the crystal
lattice. Such defects will also diffuse through a crystal lat- closure temperatures. These considerations have led some
tice at elevated temperature. Dislocation-driven diffusion researchers to suggest that true volume diffusion is so slow
and the point defect diffusion process are likely to have that it is not an effective means of mass transport, even at
different diffusion constants and activation energies geological timescales at temperatures commonly occur-
(Lee, 1995) and thus may lead to varying estimates for ring during metamorphism (Villa, 1998).
546 METAMORPHIC TERRANES (K–Ar/40Ar/39Ar)
Under the assumption that anhydrous minerals
such a K-feldspar do not undergo phase changes during
laboratory heating in vacuum, and hence argon loss
during stepwise heating will follow a volume diffusion-
governed process, it was argued that it is possible to
model complex patterns of apparent age spectra assum-
ing a finite number of sub-crystal sizes with different
radii for diffusion (Lovera et al., 1989). Such a model
could then be used to imply a discrete cooling history at
geologically relevant time scales. Another approach,
using higher-temperature geochronometers and lower-
temperature geochronometers based on other isotopic
decay systems such as U/Pb, Sm/Nd, and Rb/Sr, has been
very useful in the quantification of crustal evolution stud-
ies (Kalt et al., 2000).
General description
The 40Ar/39Ar method can be applied successfully in
metamorphic terranes provided that the temperature his-
tory is well known from thermobarometric studies, thus
linking the measured ages to the PT history of the rock.
In this section, some examples of different time slices of
Earth history will be discussed.
The oldest rocks on Earth dated successfully with
the 40Ar/39Ar method are from the Kaapvaal and Pilbara
cratons, both now situated in the southern hemisphere.
In the Kaapvaal craton, komatiites from the Onverwacht
Group in Barberton were dated at 3.45 Ga by Martinez
et al. (1984), and the ages were interpreted to record the
timing of lower-greenschist facies overprinting of the vol-
canics shortly after deposition. Layer et al. (1998)
obtained an age of 3.18 Ga on the Nelshoogte pluton.
In the Pilbara craton of Western Australia, Wijbrans
and McDougall (1987) published hornblende ages as old
as 3.2 Ga from the Western Shaw greenstone belt and
demonstrated that within the adjacent granitoid domes,
amphibolite facies metamorphism continued until
2.95 Ga. Subsequent work (Davids et al., 1997; Zegers
et al., 1999) showed that in the Pilbara craton, also
3.4 Ga rocks have been preserved in the greenstone belts.
The metamorphic evolution of several Proterozoic
metamorphic terranes has been studied using the
40
Ar/39Ar method. Studies have focused on the metamor-
phic evolution of the early Proterozoic Arunta block in
Australia and the eastern part of the Baltic shield in Scan-
dinavia. Early Proterozoic mobile belts are relatively nar-
row metamorphic terranes in the Archean cratons, for
example, the Limpopo belt in Southern Africa, the Capri-
corn belt in Western Australia, and the Nagsuggtoqidian
belt in Greenland are examples of such terranes that expe-
rienced metamorphic overprinting in the period ca.
2,000–1,700 Ma. The younger Grenville metamorphic
event is most pronounced in North America character-
ized by ages in the range 1,200–800 Ma. Grenville-aged
rocks are widespread and have been found in the Baltic
Shield and in South America, where they can be straight-
forwardly linked to the core Grenville of North America;
Grenville-age rocks also occur in Asia.
Younger Proterozoic metamorphic terranes can be
found in the southern continents that formed part of Gond-
wana at a later stage of their history. For example, Pan-
African metamorphic belts occur as relatively narrow
mobile belts of ca. 500–600 Ma in age, and Pan-African
age mobile belts occur, for example, in Northeast Africa,
Egypt and the Sinai, Southeast Africa, Madagascar, India,
and Western Australia.
Early Paleozoic metamorphic terranes occur in the
Atlantic domain, in the Caledonian belt of western Scandi-
navia, eastern Greenland, and Eastern North America. HP
and UHP rocks of the same age occur in the Urals and the
Central metamorphic belt of China, whereas in southeast-
ern Australia the Ordovician Lachlan fold belt is a low-
grade slate belt.
Very young metamorphic terranes have been described
tectonic systems that are active at present, such as the case
of ongoing continental collision in, e.g., the Karakorum
(Zeitler, 1985), South Island, NZ (Chamberlain et al.,
1995), and the D’Entrecasteaux Islands off the coast of
Papua New Guinea (Baldwin et al., 2004), where Pliocene
UHP rocks are exposed.
Summary and conclusion
Much of the continental crust exposed within the conti-
nental interiors consists of metamorphosed rocks.
These rocks can be conveniently grouped in terms of
the age of metamorphism as recorded by common
40
Ar/39Ar geochronometers such as K-feldspar, biotite,
muscovite, and hornblende. Such a division illustrates
the evolution of the continental crust through time from
the early Archean exposed in the continental interiors to
the youngest metamorphic rocks found in the rapidly
exhumed rocks at the continental margins. Metamor-
phic rocks are important in that they often constitute
the only record of evolution of the continental
crust and bear evidence for tectonic processes in the
geological past.
Bibliography
Armstrong, R. L., Jäger E., and Eberhardt, A., 1966. Comparison of
K/Ar and Rb/Sr ages on Alpine biotites. Earth and Planetary
Science Letters, 1, 13–19.
Baldwin, S. L., Monteleone, B. D., Webb, L. E., Fitzgerald, P. G.,
Grove, M., and Hill, E. J., 2004. Pliocene eclogite exhumation
at plate tectonic rates in eastern Papua New Guinea. Nature,
431, 263–267.
Brown, M., 2009, Metamorphic Patterns in Orogenic Systems and
the Geological Record. London: Geological Society, Special
publications number 318, pp. 37–74.
Chamberlain, C. P., Zeitler, P. K., and Cooper, A. F., 1995. Geochro-
nologic constraints of the uplift and metamorphism along the
Alpine Fault, South Island, New Zealand. New Zealand Journal
of Geology and Geophysics, 38, 515–523.
Davids, C., Wijbrans, J. R., and White, S. H., 1997. 40Ar/39Ar
laserprobe ages of metamorphic hornblendes from the Coongan
Belt, Pilbara, Western Australia. Precambrian Research,
83, 221–242.
 METEORIC 10Be
Dodson, M. H., 1973. Closure temperature in cooling geochrono-
logical and petrological systems. Contributions to Mineralogy
and Petrology, 40, 259–274.
Gilletti, B.J., 1974. Studies in diffusion I: argon in phlogopite mica.
In Hofmann, A. W., Giletti, B. J., Yoder, H. S., Jr., and Yund,
R. A. (eds.), Geochemical Transport and Kinetics. Washington,
DC: Carnegie Institute of Washington. Carnegie Institute of
Washington Publication, Vol. 634, pp. 107–115.
Harrison, T. M., 1981. Diffusion of 40Ar in hornblende. Contribu-
tions to Mineralogy and Petrology, 78, 324–331.
Hofmann, A. W., 1979. Diffusion experiments in geology. In Jäger,
E., and Hunziger, J. C. (eds.), Lectures in Isotope Geology.
Berlin: Springer, pp. 189–193.
Jäger, E., 1979. Introduction to geochronology. In Jäger, E., and
Hunziger, J. C. (eds.), Lectures in Isotope Geology. Berlin:
Springer, pp. 1–12.
Johnson, M. R. W., and Harley, S. L., 2012. Orogenesis, The Mak-
ing of Mountains. Cambridge: Cambridge University Press,
p. 388.
Kalt, A., Corfu, F., and Wijbrans, J. R., 2000. Time calibration of
a P-T path from a Variscan high-temperature low-pressure meta-
morphic complex (Bayerische Wald, Germany), and the detec-
tion of inherited monazite. Contributions to Mineralogy and
Petrology, 138, 143–163.
Kinny, P. D., Wijbrans, J. R., Froude, D. O., Williams, I. S., and
Compston, W., 1990. Age constraints on the geological evolu-
tion of the Western Gneiss Terrain in the vicinity of Mount
Narryer, Yilgarn Block, W.A. Australian Journal of Earth Sci-
ences, 37, 51–70.
Layer, P. W., Lopez-Martinez, M., Kroner, A., York, D., and
McWilliams, M., 1998. Thermochronometry and
palaeomagnetism of the Archaean Nelshoogte Pluton, South
Africa. Geophysical Journal International, 135, 129–145.
Lee, J. K. W., 1995. Multipath diffusion in geochronology. Contri-
butions to Mineralogy and Petrology, 120, 60–82.
Lister, G. S., and Baldwin, S. L., 1996. Modelling the effect of arbi-
trary P-T-t histories on argon diffusion in minerals using the
MacArgon program for the Apple Macintosh. Tectonophysics,
253, 83–109.
Lovera, O. M., Richter, F. M., and Harrison, T. M., 1989. The
40
Ar/39Ar geochronology for slowly cooled samples having
a distribution of diffusion domain sizes. Journal of Geophysical
Research, 94, 17917–17935.
Martinez, M. L., York, D., Hall, C. M., and Hanes, J. A., 1984.
Oldest reliable 40Ar/39Ar ages for terrestrial rocks – Barberton
Mountain Komatiites. Nature, 307, 352–354.
Mussett, A. E., 1969. Diffusion measurements and the potassium
argon method of dating. Geophysical Journal of the Royal Astro-
nomical Society, 18, 257–303.
Villa, I. M., 1998. Isotopic closure. Terra Nova, 10, 42–47.
Warren, C. J., Hanke, F., and Kelley, S. P., 2012. When can musco-
vite 40Ar/39Ar dating constrain the timing of metamorphic exhu-
mation? Chemical Geology, 291, 79–86.
Wheeler, J., 1996. DIFFARG: a program for simulating argon diffu-
sion profiles in minerals. Computers & Geosciences, 22,
919–929.
Wijbrans, J. R., and McDougall, I., 1987. On the metamorphic his-
tory of an Archaean granitoid greenstone terrane, East Pilbara,
Western Australia, using the 40Ar/39Ar age spectrum technique.
Earth and Planetary Science Letters, 84, 226–242.
Zegers, T. E., Wijbrans, J. R., and White, S. H., 1999. 40Ar/39Ar
age constraints on tectonothermal events in the Shaw area of
the eastern Pilbara granite-greenstone terrain (W Australia):
700 Ma of Archean tectonic evolution. Tectonophysics, 311,
45–81.
Zeitler, P. K., 1985. Cooling history of the NW Himalaya, Pakistan.
Tectonics, 4, 127–151.
547
Cross-references
Alpine Terranes (K–Ar/Ar–Ar)
Luminescence, Earthquake and Tectonic Activity
Rb–Sr Dating
Sm–Nd Dating
Uranium–Lead, Metamorphic Rocks
METEORIC 10Be
Jane Willenbring
Department of Earth and Environmental Sciences,
University of Pennsylvania, Philadelphia, PA, USA
Synonyms
Atmospheric 10Be; Fallout 10Be; Garden-variety 10Be
Definition
Most terrestrial beryllium-10 (10Be) is produced by spall-
ation of nitrogen and oxygen nuclei in the atmosphere by
cosmic rays. The relatively high atmospheric production
rate of 10Be (global mean of ~106 atoms cm
2 year
1)
varies regionally by latitude by a factor of ~2 (Lal and
Peters, 1967). Meteoric 10Be is scavenged quickly from
the atmosphere by aerosols and is transported to the
Earth’s surface, mostly by precipitation. When 10Be falls
onto land, it binds tightly to soil particles. Stable soil with
prolonged exposure to rain becomes highly enriched in
10
Be, whereas soil in areas of high erosion accumulates lit-
tle 10Be. When 10Be falls on water, it eventually accumu-
lates on sediments or in precipitates. Movement and
cycling of 10Be in the Earth’s reservoirs has been reviewed
in detail by Morris et al. (2002) and Willenbring and von
Blanckenburg (2010).
In recent years, 10Be has been measured most widely in
continental ice cores to determine snow accumulation
rates and for comparison with cosmogenically produced
radiocarbon to determine variations in solar activity
through time. Meteoric 10Be is used for a growing number
of geomorphic studies including discriminating river sed-
iment sources, understanding soil transport, and quantify-
ing long-term sediment fluxes from watersheds.
Beryllium-10 is radioactive with a half-life of 1.39 million
years and is often used in conjunction with the native,
stable nuclide beryllium-9 (9Be) to account for variable
absorption behavior. The 10Be/9Be ratio is used in deep-
sea sediments for estimating ages via 10Be decay, for calcu-
lating the growth rates of manganese encrustations, and for
understanding the past variability in the Earth’s magnetic
field and chemical weathering fluxes.
Most natural concentrations of 10Be range from ~103
atoms g
1 in rain and ice to ~106–108 atoms g
1 in marine
and terrestrial sediments. These levels are now easily mea-
sured by accelerator mass spectrometry after a chemical
process to purify the Be from other elements and boron –
an isobar and the daughter product of 10Be decay (Granger
548 METEORITES (36Cl)
et al., 2013). Well-known geochronologic techniques
using cosmogenic 10Be called surface exposure dating
typically uses the abundance of 10Be produced in situ
within crystal lattices of minerals rather than the abun-
dance of meteoric 10Be. The larger flux of meteoric 10Be
(on a per-surface-area basis) that is about two to four
orders of magnitude greater than the rate of in situ produc-
tion complicates the chemical purification of the in situ
produced component by contamination of grain surfaces.
Etching is specifically carried out to remove every trace
of atmospheric 10Be adhered to surfaces of grains.
Bibliography
Granger, D., Lifton, N., and Willenbring, J. K., 2013. A cosmic trip:
25 years of cosmogenic nuclides in geology. Geological Society
of America Bulletin, 125, 1379.
Lal, D., and Peters, B., 1967. Cosmic ray produced radioactivity on
the Earth. Handbuch Der Physik, 46, 551.
Morris, J. D., Gosse, J. C., and Brachfeld, S., 2002. Cosmogenic
10
Be and the solid earth: studies in geomagnetism, subduction
zone processes, and active tectonics. Reviews of Mineralogy
and Geochemistry, 50, 40.
Willenbring, J. K., and von Blanckenburg, F., 2010. Meteoric cos-
mogenic beryllium-10 adsorbed to river sediment and soil: appli-
cations for Earth-surface dynamics. Earth-Science Reviews,
98, 105.
METEORITES (36Cl)
Kees Welten
Space Sciences Laboratory, University of California,
Berkeley, CA, USA
Definition
Cosmogenic nuclide: A stable or radioactive isotope that is
produced by the interaction of terrestrial or extraterrestrial
materials with galactic and/or solar cosmic rays.
Meteorites: An extraterrestrial rock or iron that survived
passage through the atmosphere and landed on Earth.
Most meteorites are pieces of collisional debris from aster-
oids or comets, although a small percentage are fragments
of the moon and Mars.
Introduction
Cosmogenic nuclides provide information on
a meteorite’s history, specifically during their travel
through space as meter-sized objects, in which they were
exposed to cosmic rays. More than 50 different cosmo-
genic nuclides have been detected in meteorites, including
the stable isotopes of the noble gases, He, Ne, Ar, Kr, and
Xe, and radioactive nuclides with half-lives ranging from
less than 1 day (24Na) to 1.28 Ga (40K). The measured
concentrations of cosmogenic nuclides in meteorites pro-
vide information on the total exposure time of meteorites
to cosmic rays, also known as the cosmic-ray exposure
(CRE) age. Since cosmic rays penetrate about 1 m deep,
the CRE age represents the time since a meteorite was
ejected as a meter-sized object from its parent body,
although a few meteorites show evidence of CRE on the
surface of their parent body. The most commonly mea-
sured nuclides in meteorites are listed in Table 1, along
with their production mechanism and the analytical
method used to measure each nuclide. Although most cos-
mogenic nuclides are mainly produced by interactions
with galactic cosmic rays (GCR), a few of the nuclides
(including 22Na, 22Ne, 26Al, 54Mn, and 56Co) are also pro-
duced by solar cosmic rays (SCR). However, since the
SCR flux is dominated by particles with energies
<100 MeV, SCR contributions are only expected in the
outer few cm of a meteoroid, which is usually lost during
atmospheric passage. Therefore, SCR effects are rarely
observed in meteorites.
In addition, the cosmogenic radionuclides (such as 14C,
41
Ca, 36Cl, and 81Kr) also provide information on the ter-
restrial age of a meteorite, the time elapsed since it fell
on Earth and was shielded from the high cosmic ray flux
it experienced in space. Terrestrial ages of meteorites are
especially interesting for meteorites from hot deserts
(e.g., Australia, Southwestern USA, Sahara, Atacama,
and the Arabian Peninsula) and from Antarctica. In addi-
tion, the terrestrial age of a meteorite find is needed to cal-
culate the ejection age, i.e., the time a meteorite was
ejected from its parent body, which equals the sum of the
CRE age and the terrestrial age. For most meteorites, the
terrestrial age is small compared to the cosmic-ray expo-
sure age, but for lunar meteorites and certain carbonaceous
chondrites, both ages have to be taken into account.
Analytical methods
Cosmogenic noble gases are measured using a noble gas
mass spectrometer. Samples of 10–500 mg are usually
wrapped in a metal (Al, Ni) foil and stored under vacuum
in one of several storage arms that are connected to the
mass spectrometer. After preheating the samples at
100–120  C to remove absorbed atmospheric gases, the
cosmogenic (as well as radiogenic and trapped) noble
gases are extracted by heating the sample in a furnace
for 15–30 min at 1,750–1,800  C. The noble gases are
then separated and purified, and their abundance and iso-
topic composition are measured in a mass spectrometer
with a Nier-type ion source (Begemann et al., 1976;
Signer et al., 1977). Using known ratios for the cosmo-
genic and trapped noble gases, the individual contribu-
tions of cosmogenic and trapped He, Ne, Ar, Kr, and Xe
are calculated (e.g., Eugster, 1988).
Short-lived cosmogenic radionuclides as well as 26Al are
measured by gamma-ray spectrometry (Evans et al., 1982).
This method requires samples of 5–100 g and is nondestruc-
tive, but generally requires counting times of several days to
several weeks. Long-lived cosmogenic radionuclides
(including 14C, 10Be, 26Al, 36Cl, 41Ca, 53Mn, 59Ni, 60Fe,
and 129I) are measured by accelerator mass spectrometry
(AMS). For this method, samples of 100 mg are
 METEORITES (36Cl) 549

Meteorites (36Cl), Table 1 Cosmogenic nuclides produced in meteorites

Nuclide t1/2 (year) Major target elements Main production mechanism Method of analysis
54
Mn 0.855 Fe, Ni Spallation g-ray spectrometry
22
Na 2.61 Mg, Si Spallation g-ray spectrometry
60 59
Co 5.27 Co Co(n, g) g-ray spectrometry
14
C 5.730 O, Mg, Si Spallation AMS
59
Ni 7.6  104 Fe, Ni Spallation, 58Ni(n, g) AMS
41
Ca 1.04  105 Fe, Ca Spallation, 40Ca(n, g) AMS
81
Kr 2.29  105 Rb, Sr, Y, Zr Spallation Noble gas MS
36
Cl 3.01  105 K, Ca, Ti, Fe, Cl Spallation, 36Cl(n, g) AMS
26
Al 7.05  105 Si, Al, Mg, (Fe) Spallation AMS, g-ray spectrometry
10
Be 1.36  106 O, Mg, Si, (Fe) Spallation AMS
53
Mn 3.7  106 Fe, Ni Spallation AMS, INAA
129
I 1.57  107 Te, Ba, La Spallation, 128Te(n, g) AMS
40
K 1.28  109 Fe, Ni Spallation
3
He Stable O, Mg, Si, Fe Spallation Noble gas MS
20
Ne, 21Ne, 22Ne Stable Mg, Al, Si GCR Noble gas MS
36
Ar, 38Ar Stable Ca, Fe GCR Noble gas MS
78, 80, 82, 83, 84
Kr Stable Rb, Sr, Y, Zr, Nb Spallation, 79Br(n, g)80Kr, 81Br(n, g)82Kr Noble gas MS
126-136
Xe Stable Ba, REE Spallation Noble gas MS

Nth: thermal neutron capture

dissolved in the presence of known carrier amounts of the
stable nuclide of each element of interest. After dissolu-
tion, a small aliquot of the sample is taken for chemical
analysis, and each cosmogenic nuclide is separated by
ion exchange and solvent extraction techniques. The iso-
topic ratio of the radioactive to the stable isotope is mea-
sured by AMS, and the concentration of the radionuclide
(in atoms per gram meteorite) is calculated from the mea-
sured isotope ratio, the amount of carrier added, and the
sample mass. Radionuclide concentrations in meteorites
are generally reported in decay rates, i.e., disintegrations
per minute per kg (dpm/kg). Since most meteorites were
exposed to cosmic rays in space long enough (>107
years) to reach saturation levels for radionuclides with
half-lives up to a few Ma, the measured activities gener-
ally also represent the production rates (in atoms per
minute per kg).
Cosmic-Ray exposure age methods
Many different methods have been used to determine the
CRE ages of meteorites (e.g., Herzog, 2005). Some expo-
sure age methods can only be applied for iron meteorites
(and the metal phase of stony irons), while others only
apply to stone meteorites. Production rates of cosmogenic
nuclides in meteorites depend on their chemical composi-
tion, the size of meteoroid during irradiation in space, and
depth of the samples within the meteoroid. The size and
depth effects are generally lumped together under the term
“shielding,” reflecting the attenuation of primary cosmic
ray particles as well as the buildup of secondary particles
(protons and neutrons). Since meteorites typically lose
50–99 % of their pre-atmospheric mass during atmo-
spheric passage (breakup and ablation processes), the
original size and pre-atmospheric depth of individual
meteorites are not known and can only be estimated from
the measured cosmogenic nuclide concentrations or
cosmic-ray track densities. Many empirical relationships
between different cosmogenic nuclides have been
established that help in calculating absolute production
rates in individual samples and thus to obtain shielding-
corrected CRE ages.
In general, the most reliable CRE age methods are the
ones that combine a stable nuclide (e.g., 36Ar, 21Ne,
83
Kr) and a radionuclide (e.g., 10Be, 26Al, 36Cl, 81Kr). In
these methods, the radionuclide (if it has reached satura-
tion level) serves as an internal shielding parameter, which
can be used to calibrate the production rate of the stable
nuclide. However, many exposure ages of stony meteor-
ites, which generally come from objects of less than 1 m
in diameter, are based on noble gas concentrations only,
in combination with the 22Ne/21Ne ratio, which serves as
a shielding parameter for objects <1 m in diameter.
Noble gas exposure ages of stony meteorites
The majority of CRE ages of stony meteorites are based on
the concentrations of cosmogenic 3He, 21Ne, and 38Ar
(Marti and Graf, 1992; Graf and Marti, 1994, 1995;
Eugster and Michel, 1995; Welten et al., 1997; Scherer
et al., 1998; Scherer and Schultz, 2000; Rai et al., 2003).
The concentration of cosmogenic 21Ne is generally con-
sidered the most reliable of the three nuclides, since 3He
is more susceptible to diffusion losses and 38Ar is more
sensitive to variations in chemical composition (mainly
Ca, Fe). In some achondrites, including howardites and
eucrites, the 38Ar ages are more reliable than 3He and
21
Ne, as the production of 38Ar is higher due to the high
Ca concentrations, while 3He and 21Ne suffer diffusive
losses (mainly from plagioclase).
550 METEORITES (36Cl)
The production rates of 3He, 21Ne, and 38Ar are gener-
ally derived from empirical relationships with the
22
Ne/21Ne ratio, which serves as a shielding indicator
and generally ranges from 1.05 at high shielding to
1.25 at low shielding. The relationship between 21Ne and
the 22Ne/21Ne ratio in chondrites was first established by
Eberhardt et al. (1966) and was later refined by Herzog
and Anders (1971a), Nishiizumi et al. (1980), and Eugster
(1988). The most commonly used equation for the 21Ne
production rate, P(21Ne), expressed in 10
8 cm3 STP/g/
Ma, in chondrites is
 
P 21 Ne ¼ 1:61
h   i
F21= 21:77 22 Ne=21 Ne
19:32 , ð1Þ
in which F21 is the chemical correction factor for different
types of chondrites and the term in the denominator serves
as shielding correction.
Similar empirical relationships were proposed for the
production rates of 3He, 38Ar, 83Kr, and 126Xe in chondrites
(Eugster, 1988) as well as for noble gas production rates in
achondrites (Eugster and Michel, 1995). Model calcula-
tions have shown that the simple relationships between pro-
duction rates and the 22Ne/21Ne ratio are generally valid for
chondritic objects <1 m in diameter, which dominate the
meteoroid flux to Earth. However, these simple relation-
ships fail for larger objects, for which they result in
overestimation of the production rates (Graf et al., 1990b;
Masarik et al., 2001; Welten et al., 2003; Leya and Masarik,
2009). For larger objects, it is recommended to use expo-
sure age methods which are based on the combination of
a stable nuclide and a radionuclide with similar production
mechanisms. We will discuss the most commonly used
methods for stony and iron meteorites.
The 81Kr–Kr method
Cosmogenic Kr in stony meteorites is mainly produced
from the trace elements Rb, Sr, Y, Zr, and Nb. Although
the concentration of cosmogenic Kr is several orders of
magnitude lower than for cosmogenic He, Ne, and Ar,
cosmogenic Kr was first detected in lunar samples
(Marti, 1967) and later in many meteorites, including
achondrites and chondrites (Eugster et al., 1967; Finkel
et al., 1978; Freundel et al., 1986; Lavielle and Marti,
1988; Eugster, 1988). The 81Kr-Kr method is based on
the combination of stable Kr isotopes, mainly 78Kr, 80Kr,
82
Kr, and 83Kr, and the radioactive isotope 81Kr, which
reaches saturation after 1.5 Ma of exposure. To deter-
mine the CRE age of a meteorite, one has to know the
P(81Kr)/P(83Kr) ratio, which is shielding dependent. There
are generally two ways to derive this ratio, i.e., by using
the empirical relationship between 81Kr/83Kr and
78
Kr/83Kr ratio derived from Apollo 12 samples (Eq. 2)
(Marti and Lugmair, 1971) or the theoretical relationship
for Kr production by spallation (Eq. 3), in which the
81
Kr/83Kr production ratio is determined from the
measured 80Kr/83Kr and 82Kr/83Kr ratios assuming an iso-
baric fraction yield of 0.95 (Marti, 1967):
   
P 81 Kr =P 83 Kr ¼ 1:262 78 Kr=83 Kr þ 0:381 ð2Þ
81  83 
P Kr =P Kr ¼ 0:95
 
80 Kr=83 Kr þ82 Kr=83 Kr =2 ð3Þ
The latter approach has the advantage that it is less
dependent on variations in chemical composition of differ-
ent meteorites, but its disadvantage is that in large meteor-
ites, 80Kr and 82Kr are not only produced by spallation
reactions but also by neutron-capture reactions on 79Br
and 81Br, respectively, thus leading to an overestimation
of the P(81Kr)/P(83Kr) values. Therefore, Eq. 2, which
uses the 78Kr/83Kr ratio as shielding indicator, is more
widely applicable and thus has been used for many differ-
ent types of meteorites (e.g., Eugster et al., 1967; Finkel
et al., 1978; Freundel et al., 1986; Lavielle and Marti,
1988; Eugster, 1988; Shukolyukov and Begemann,
1996; Leya et al., 2004). Since the relatively small cosmo-
genic Kr contributions have to be determined from
a mixture of cosmogenic and trapped Kr components,
the 81Kr–Kr method is most reliable for meteorites with
long exposure ages (>10 Ma), high concentrations of tar-
get elements (e.g., howardites, eucrites), and low concen-
trations of trapped Kr (either planetary or terrestrial). The
latter condition is valid for equilibrated chondrites and
achondrites and implies that observed falls are more suit-
able than weathered finds.
The 36Cl–36Ar method
The 36Cl–36Ar technique is generally considered the most
reliable dating method for iron meteorites (e.g., Lavielle
et al., 1999) and the metal phase of stony irons
(Begemann et al., 1976). The 36Cl and 36Ar are produced
by spallation reactions on Fe and Ni (e.g., Begemann
et al., 1976). Since the reaction mechanism of the two
products is almost identical, the 36Cl/36Ar production rate
ratio that is essentially independent from shielding condi-
tions (Schaeffer and Heymann, 1965; Lavielle
et al., 1999). In addition, more than 98 % of the produced
36
Cl, which is in saturation after a few million years of
cosmic-ray exposure, decays to 36Ar. As a result, more
than 80 % of the 36Ar in the metal phase is from decay
of the radionuclide 36Cl (half-life ¼ 3.01  105 year).
Since 36Cl and 36Ar concentrations can be determined
accurately and are relatively insensitive to silicate contam-
ination, typical uncertainties in the 36Cl–36Ar ages are
3–5 % (Lavielle et al., 1999; Graf et al., 2001). For calcu-
lating 36Cl–36Ar exposure ages, T36 (in Ma), the following
equation can be used (Lavielle et al., 1999; Leya and
Masarik, 2009):
     
T36 ¼ 511 P 36 Cl =P 36 Ar 36 Ar = 36 Cl
þ 0:36, ð4Þ
 METEORITES (36Cl)
in which [36Ar] and [36Cl] are the measured concentrations
of cosmogenic 36Ar [in units of 10
8 cm3 STP g
1] and
36
Cl (in dpm kg
1) in metal and P(36Cl)/P(36Ar) ¼
0.835  0.040 atom/atom (Lavielle et al., 1999). The addi-
tional constant (of 0.36 Ma) takes into account that – at the
time of the meteorite fall – some of the cosmogenic 36Cl
has not yet decayed to 36Ar (Leya and Masarik, 2009).
The 36Cl–36Ar method is most reliable if the measured
36
Cl concentration represents the saturation value, i.e., for
meteorites with CRE ages >2 Ma and short terrestrial ages,
although corrections can be made for radioactive decay of
36
Cl during the terrestrial residence time if the terrestrial
age is known (Lavielle et al., 1999). The 36Cl–36Ar method
has also been applied to the metal phases of stony irons
(Begemann et al., 1976; Albrecht et al., 2000; Honda
et al., 2002) and the metal phase of stone meteorites (Graf
et al., 2001; Leya et al., 2001).
The 26Al–21Ne method
Iron meteorites
The energy dependence of 26Al and 21Ne production from
spallation reactions on Fe (and Ni) is very similar, which
makes the 26Al/21Ne ratio in metal insensitive to variations
in shielding and thus suitable for determining CRE ages of
iron meteorites. Lipschutz et al. (1965) calculated CRE
ages of iron meteorites from the measured 26Al/21Ne ratio,
in which 26Al and 21Ne concentrations are expressed in
dpm/kg and 10
8 cm3 STP/g, respectively. Using Eq. 5,
assuming a constant 26Al/21Ne production rate ratio of
0.37 (atom/atom), while later studies used slightly differ-
ent ratios of 0.38 (Hampel and Schaeffer, 1979; Lavielle
et al., 1999) and 0.35 (Honda et al., 2002).
      values of 0.55 (Lavielle et al., 1999) and 0.50 atom/atom
T 26 Al
21 Ne ¼ 511 P 26 Al =P 21 Ne
 
21 Ne = 26 Al ð5Þ
The 26Al–21Ne ages of 17 iron meteorites of different
pre-atmospheric size (Lavielle et al., 1999) and of 8 metal
separates from a wide range of shielding depths in the
large Brenham stony iron meteorite (Honda et al., 2002)
generally yield good agreement (within 10 %) with the
corresponding 36Cl–36Ar ages. Although the 26Al/21Ne
method is reliable for most iron meteorites and the metal
phase of stony irons, it has not been applied to the metal
phase of chondritic meteorites due to large contributions
of 26Al and 21Ne from silicates, which are difficult to
remove from the chondritic metal phase. However,
for chondrites and other stone meteorites, the 26Al/21Ne
method can also be applied to bulk samples, albeit with
a different production ratio.
Stony meteorites
The production mechanisms of 26Al and 21Ne in stony
meteorites are dominated by low-energy reactions on Al +
Si (for 26Al) and Mg, Al, and Si (for 21Ne). One therefore
expects that the production rates of 26Al and 21Ne in stone
meteorites show very similar dependence on shielding
551
conditions (Herzog and Anders, 1971b). The constancy
of the 26Al/21Ne ratio in chondrites has been investigated
by several empirical and semiempirical studies (e.g., Graf
et al., 1990a; Graf et al., 1990b; Nagai et al., 1993; Honda
et al., 2002; Welten et al., 2003) as well as by model calcu-
lations (Leya et al., 2001; Leya and Masarik, 2009).
Although the shielding dependence of the 26Al/21Ne ratio
varies slightly among different studies, a relatively con-
stant 26Al/21Ne production ratio of 0.35–0.42 atom/atom
seems consistent with most studies. This method has been
used to calculate CRE ages for several large chondrites
(e.g., Leya et al., 2001; Welten et al., 2001, 2003), for
which the CRE age cannot be calculated from cosmogenic
Ne data alone.
The 10Be–21Ne method
Iron meteorites
Cross-section data for the production of 10Be and 21Ne
from high-energy proton reactions on Fe suggest that the
production ratio of 10Be/21Ne in irons does not show
a significant depth dependency, which is confirmed by rel-
atively constant 10Be/21Ne ratios measured in samples
from different shielding depths in two iron meteorites
(Graf et al., 1987; Lavielle et al., 1999) and in metal
phases of the Brenham stony iron (Honda et al., 2002).
Therefore, Eq. 6 can be used to calculate the CRE ages
from the measured 21Ne/10Be ratios in iron meteorites
     
T ¼ 511 P 10 Be =P 21 Ne 21 Ne = 10 Be ð6Þ
in which [21Ne] and [10Be] are the measured concentra-
tions of 21Ne [in 10
8 cm3 STP g
1] and of 10Be
(in dpm kg
1) in metal. For the P(10Be)/P(21Ne) ratio,
(Honda et al., 2002) have been used. The latter study
found that the 10Be/21Ne ages of metal phases in the
Brenham pallasite agree with the 36Cl–36Ar ages, while
absolute production rates varied by three orders of magni-
tude. Despite empirical evidence that the 10Be/21Ne ratio
in iron meteorites is remarkably constant, Ammon
et al. (2009) caution against using the 10Be/21Ne ratio for
CRE ages of iron meteorites, since their model calcula-
tions indicate that trace elements (such as P and S) can
affect the 10Be/21Ne production ratio.
Stony meteorites
Depth profiles of 10Be and 21Ne in the large Knyahinya
chondrite indicate that the 10Be/21Ne ratio in a meter-sized
object is relatively constant even though concentrations of
both nuclides increase by more than 20 % from the surface
to the center of the meteorite (Graf et al., 1990a). The
semiempirical model of Graf et al. (1990b) suggests that
the 10Be/21Ne ratio is relatively constant at 0.14 atom/
atom, although it may show small variations (5–10 %)
as a function of shielding. Subsequent model calculations
based on particle fluxes and proton and neutron cross sec-
tions (Leya et al., 2000; Leya and Masarik, 2009) suggest
that the 10Be/21Ne production ratio is more sensitive to
552 METEORITES (36Cl)
shielding than in the semiempirical model of Graf and is
probably lower for high shielding conditions. This conclu-
sion is also confirmed by empirical studies, which suggest
production ratios of 0.10–0.12 atom/atom for high
shielding (Leya et al., 2001; Welten et al., 2004). Although
the exact relationship between the 10Be/21Ne ratio and
shielding is still a topic of ongoing research, and the
10
Be/21Ne ratio thus has an uncertainty of 20 %, the
10
Be/21Ne method is already more reliable for very large
meteorites, such as the Gold Basin meteorite (Welten
et al., 2003), for which the simple relationship between
21
Ne production rate and 22Ne/21Ne ratio fails.
Terrestrial age methods
Cosmogenic radionuclide concentrations also provide
information on the terrestrial ages of meteorite finds. The
method that is most suitable for terrestrial age determina-
tion of a meteorite depends on the metal content of the
meteorite (metal rich vs. metal poor), its find location,
and its expected terrestrial age. We will discuss the most
commonly used methods below.
14
C and 14C/10Be terrestrial ages
The terrestrial ages of stony meteorites that fell less than
30–40 ka ago can best be determined by measuring the
cosmogenic 14C concentration of finds relative to falls.
Since 99 % of cosmogenic 14C in stony meteorites is pro-
duced from oxygen, the 14C production rate is simply
a function of bulk O content, which is generally well
known and does not vary more than 20–30 % among
different types of stone meteorites. Measurements of
14
C in observed falls suggest saturation values of
30–60 dpm/kg for chondrites with radii of <50 cm (Jull
et al., 1989; Wieler et al., 1996), which is valid for more
than 90 % of all stony meteorites. For average-size mete-
orites (<1 m in diameter), a constant production rate,
P(14C), is generally assumed, leading to the following
equation for the terrestrial age, T(14C), in which T is
expressed in ka and P(14C) and A(14C) in dpm/kg:
       Cl/10Be ratio. Although this ratio was initially assumed
T 14 C ¼ 8:27 ln P 14 C =A 14 C ð7Þ
This method was first applied to meteorite finds from
the Southwestern USA (Boeckl, 1972), revealing terres-
trial ages up to 13 ka. With the introduction of 14C
AMS measurements, the required sample size was
reduced to <1 g. The method was applied to meteorites
from other hot deserts (Jull et al., 1990; Bland
et al., 1998) as well as Antarctica (Jull et al., 1989).
However, the assumption of an average production rate
of 45–50 dpm/kg is not valid for very large meteorites, as
was shown for the Gold Basin chondrite shower (Kring
et al., 2001). In very large meteorites, the terrestrial age
is best determined by measuring the ratio of 14C/10Be,
which is relatively insensitive to shielding effects and is
assumed to be constant at 2.5 in chondrite falls. This
method is valid on the condition that the meteorite had
a minimum CRE of 5–7 Ma, i.e., long enough to saturate
both 14C and 10Be, or that the CRE age is well known, so
that corrections for undersaturation of 10Be can be made.
Kring et al. (2001) showed that the measured 14C/10Be
ratio in fragments of the large Gold Basin meteorite is rel-
atively constant, even though absolute 14C and 10Be con-
centrations varied by more than a factor of 10 due to
differences in shielding depth. Comparison of 14C ages
with shielding-corrected 14C–10Be ages for a set of mete-
orites from the Nullarbor desert in Australia shows that the
two ages generally agree within 1–2 ka, which is the typ-
ical uncertainty in the 14C method (Jull et al., 2010). For
meteorites older than 30–40 ka, the 14C method only pro-
vides a lower limit, and measurements of longer-lived
radionuclides, such as 41Ca, 36Cl, 26Al, or 81Kr, are needed
to further constrain the terrestrial age (Nishiizumi
et al., 1989; Welten et al., 2004, 2006, 2008).
36
Cl and 36Cl/10Be terrestrial ages
For Antarctic meteorites, which have typical terrestrial
ages of 10 ka to 1 Ma, the radionuclide 36Cl is most suit-
able for terrestrial age determinations due to its half-life of
0.3 Ma and the relatively constant 36Cl production rate of
20–25 dpm/kg in small irons and the metal phase of most
chondrites (Nishiizumi et al., 1989). The terrestrial age (in
ka) can then be calculated from the following equation:
      
T 36 Cl ¼ 434 ln P 36 Cl =A 36 Cl ð8Þ
in which P(36Cl) is generally assumed to be
22.1  2.8 dpm kg
1 (Nishiizumi, 1995), leading to typi-
cal 2s uncertainties of 55 ka in the terrestrial age.
As with most other cosmogenic nuclide methods for
meteorites, this simple approach is not valid for large iron
meteorites (Chang and Wänke, 1969; Nishiizumi
et al., 1987; Lavielle et al., 1999) or large chondrite
showers, such as Queen Alexandra Range 90201
(Welten et al., 2011). A more reliable terrestrial age
method for iron meteorites, which often show large varia-
tions in shielding and large chondrites, is based on the
36
to be constant at 4.3 (Chang and Wänke, 1969), addi-
tional measurements have shown that the ratio increases
systematically from 3.5 in small irons to 6 in large
irons (Lavielle et al., 1999). The 36Cl/10Be saturation ratio
shows a strong correlation with the 10Be concentration in
the metal phase, yielding the following equation:
h i 
36
Cl=10 Be ¼ 5:97
0:184 10 Be
0:024
10 2
Be ð9Þ
This equation was later updated with the three constants
as 6.01, 0.28, and 0.009, respectively (Welten et al., 2006),
in which the 10Be saturation value ranges from 0 to
7 dpm/kg. The 2s uncertainty in the 36Cl/10Be ratio is
estimated to be 6 %, corresponding to an uncertainty of
35 ka in the shielding-corrected terrestrial age. The terres-
trial age (in ka) is calculated from the measured
 METEORITES (36Cl)
36
Cl/10Be ratio and 10Be concentrations in a few iterative
steps, in which the 36Cl/10Be saturation value is recalculated
in each step from the decay-corrected 10Be concentration:
 
T 36 Cl=10 Be ¼ 557
h    i
ln 36 Cl=10 Be = 36 Cl=10 Be ð10Þ
sat m from independent measurements (such as cosmogenic
This method is straightforward for iron meteorites for
which both 36Cl and 10Be are generally saturated at the
time of fall, since they have CRE ages >50 Ma. For the
metal phase of chondrites, sometimes a small correction
has to be made for undersaturation of the 10Be concentra-
tion if the CRE age is <10 Ma (Welten et al., 2006, 2011).
41 41 36
Ca and Ca/ Cl terrestrial ages
The production mechanism of 41Ca is similar to that of
36
Cl, i.e., it is mainly produced by spallation from
Fe. The average saturation value of 41Ca in small iron
meteorites is 24  1 dpm/kg (Fink et al., 1991), i.e.,
slightly higher than the average 36Cl value of
22.1  1.4 dpm/kg. Although it was initially assumed that
the 41Ca/36Cl ratio is relatively constant at 1.0–1.1 (Fink
et al., 1991; Klein et al., 1991), additional measurements
in larger iron meteorites and the metal phases of stony
irons and chondrites have shown that the ratio exhibits
a small shielding dependence (Nishiizumi and Caffee,
1998). This shielding dependence is also confirmed by
model calculations (Leya and Masarik, 2009). The
shielding dependence of the 41Ca/36Cl ratio is best
described by the following empirical relationship with
the 36Cl concentration (Welten et al., 2006, 2011):
h i  derived from their radiocarbon content. Nature, 236, 25–26.
41
Ca=36 Cl ¼ 1:25 þ 0:003 36 Cl
 2

0:0006 36 Cl ð11Þ
in which the 41Ca and 36Cl concentrations in the metal
fraction are expressed in dpm kg
1.
The terrestrial age (in ka), T(41Ca/36Cl), is then calcu-
lated from the measured 41Ca/36Cl ratio and 36Cl concen-
tration, using equation 12. The 41Ca/36Cl saturation value
is calculated in a few iterative steps using Eq. 11 and the
decay-corrected 36Cl concentration until the terrestrial
age changes less than 0.1 ka:
 
T 41 Ca=36 Cl ¼ 229
h    i
ln 41 Ca=36 Cl = 41 Ca=36 Cl ð12Þ
sat m dence for galactic cosmic ray variations during the period
81
Kr–Kr terrestrial ages
Terrestrial ages >40 ka are more difficult to determine for
achondrites, which usually have very little (if any) metal,
therefore hampering the determination of 36Cl or 41Ca
ages from the metal phase. For meteorites with high
553
plagioclase contents, such as eucrites, the combination of
radioactive 81Kr with stable 83Kr provides a good alterna-
tive for terrestrial age determinations. In this method, the
measured 81Kr/83Kr ratio is both a function of the CRE
age (as described above) and of radioactive decay of
81
Kr during the terrestrial residence time (Freundel
et al., 1986). If the CRE age of the meteorite is well known
38
Ar), then the 81Kr/83Kr ratio provides a measure of the
terrestrial age. This method is most suitable for ages of
100–600 ka and has been applied to Antarctic eucrites
(Freundel et al., 1986) as well as for a few achondrites
from hot deserts (Nishiizumi et al., 2002).
Bibliography
Albrecht, A., Schnabel, C., Vogt, S., Xue, S., Herzog, G. F.,
Begemann, F., Weber, H. W., Middleton, R., Fink, D., and Klein,
J., 2000. Light noble gases and cosmogenic radionuclides in
Estherville, Budulan and other mesosiderites: implications for
exposure histories and production rates. Meteoritics & Planetary
Science, 35, 975–986.
Ammon, K., Masarik, J., and Leya, I., 2009. New model calcula-
tions for the production rates of cosmogenic nuclides in iron
meteorites. Meteoritics & Planetary Science, 44, 485–503.
Begemann, F., Weber, H. W., Vilcsek, E., and Hintenberger, H.,
1976. Rare gases and 36Cl in stony-iron meteorites: cosmogenic
elemental production rates, exposure ages, diffusion losses and
thermal histories. Geochimica et Cosmochimica Acta, 40,
353–368.
Bland, P. A., Sexton, A. S., Jull, A. J. T., Bevan, A. W. R., Berry,
F. J., Thornley, D. M., Astin, T. R., Britt, D. T., and Pillinger,
C. T., 1998. Climate and rock weathering: a study of terrestrial
age dated ordinary chondritic meteorites from hot desert regions.
Geochimica et Cosmochimica Acta, 62, 3169–3184.
Boeckl, R., 1972. Terrestrial age of nineteen stony meteorites
Chang, C., and Wänke, H., 1969. Beryllium-10 in iron meteorites,
their cosmic ray exposure and terrestrial ages. In Millman,
P. M. (ed.), Meteorite Research. Dordrecht: Reidel Publishing
Company.
Eberhardt, P., Eugster, O., Geiss, J., and Marti, K., 1966. Rare gas
measurements in 30 stone meteorites. Zeitschrift für
Naturforschung, 21a, 414–426.
Eugster, O., 1988. Cosmic-ray production rates of 3He, 21Ne, 38Ar,
83
Kr and 126Xe in chondrites based on 81Kr-Kr exposure ages.
Geochimica et Cosmochimica Acta, 52, 1649–1662.
Eugster, O., and Michel, T., 1995. Common asteroid break-up
events of eucrites, diogenites and howardites and cosmic-ray
production rates for noble gases in achondrites. Geochimica et
Cosmochimica Acta, 59, 177–199.
Eugster, O., Eberhardt, P., and Geiss, J., 1967. 81Kr in meteorites
and 81Kr radiation ages. Earth and Planetary Science Letters,
2, 77–82.
Evans, J. C., Reeves, J. H., Rancitelli, L. A., and Bogard, D. D.,
1982. Cosmogenic nuclides in recently fallen meteorites – evi-
1967–1978. Journal of Geophysical Research, 87, 5577–5587.
Fink, D., Klein, J., Middleton, R., Vogt, S., and Herzog, G. F., 1991.
41
Ca in iron falls, Grant and Estherville: production rates and
related exposure age calculations. Earth and Planetary Science
Letters, 107, 115–128.
Finkel, R. C., Kohl, C. P., Marti, K., and Martinek, B., 1978. The
cosmic ray record of the San Juan Capistrano meteorite.
Geochimica et Cosmochimica Acta, 42, 241–250.
554 METEORITES (36Cl)
Freundel, M., Schultz, L., and Reedy, R. C., 1986. Terrestrial 81Kr-
Kr ages of Antarctic meteorites. Geochimica et Cosmochimica
Acta, 50, 2663–2673.
Graf, T., and Marti, K., 1994. Collisional records in LL-chondrites.
Meteoritics, 29, 643–648.
Graf, T., and Marti, K., 1995. Collisional history of H chondrites.
Journal of Geophysical Research, 100, 21247–21263.
Graf, T., Vogt, S., Bonani, G., Herpers, U., Signer, P., Suter, M.,
Wieler, R., and Wölfli, W., 1987. Depth dependence of 10Be
and 26Al production rates in the iron meteorite Grant. Nuclear
Instruments and Methods in Physics Research, B29, 262–265.
Graf, T., Signer, P., Wieler, R., Herpers, U., Sarafin, R., Vogt, S.,
Fieni, C., Pellas, P., Bonani, G., Suter, M., and Wölfli, W.,
1990a. Cosmogenic nuclides and nuclear tracks in the chondrite
Knyahinya. Geochimica et Cosmochimica Acta, 54, 2511–2520.
Graf, T., Baur, H., and Signer, P., 1990b. A model for the production
of cosmogenic nuclides in chondrites. Geochimica et
Cosmochimica Acta, 54, 2521–2534.
Graf, T., Caffee, M. W., Marti, K., Nishiizumi, K., and Ponganis,
K. V., 2001. Dating collisional events: 36Cl–36Ar exposure ages
of H-chondritic metal. Icarus, 150, 181–188.
Hampel, W., and Schaeffer, O. A., 1979. Al-26 in iron meteorites
and the constancy of cosmic ray intensity in the past. Earth
and Planetary Science Letters, 42, 348–358.
Herzog, G. F., 2005. Cosmic-ray exposure ages of meteorites.
In A. M. Davis (ed.), Meteorites, Comets and Planets: Treatise on
Geochemistry, Volume 1 (Exec. Ed. Holland, H. D., and Turekian,
K. K.), Elsevier, Amsterdam, The Netherlands, pp. 347–380.
Herzog, G. F., and Anders, E., 1971a. Radiation age of the Norton
County meteorite. Geochimica et Cosmochimica Acta,
35, 239–244.
Herzog, G. F., and Anders, E., 1971b. Absolute scale for radiation
ages of stony meteorites. Geochimica et Cosmochimica Acta,
35, 605–611.
Honda, M., Caffee, M. W., Miura, Y. N., Nagai, H., Nagao, K., and
Nishiizumi, K., 2002. Cosmogenic nuclides in the Brenham
pallasite. Meteoritics & Planetary Science, 37, 1711–1728.
Jull, A. J. T., Donahue, D. J., and Linick, T. W., 1989. Carbon-14
activities in recently fallen meteorites and Antarctic meteorites.
Geochimica et Cosmochimica Acta, 53, 2095–2100.
Jull, A. J. T., Wlotzka, F., Palme, H., and Donahue, D. J., 1990. Distri-
bution of terrestrial age and petrologic type of meteorites from west-
ern Libya. Geochimica et Cosmochimica Acta, 54, 2895–2898.
Jull, A. J. T., McHargue, L. R., Bland, P. A., Greenwood, R. C.,
Bevan, A. W. R., Kim, K. J., LaMotta, S. E., and Johnson,
J. A., 2010. Terrestrial ages of meteorites from the Nullarbor
region, Australia, based on 14C and 14C–10Be measurements.
Meteoritics & Planetary Science, 45, 1271–1283.
Klein, J., Fink, D., Middleton, R., Nishiizumi, K., and Arnold, J. R.,
1991. Determination of half-life of 41Ca from measurements of
Antarctic meteorites. Earth and Planetary Science Letters,
103, 79–83.
Kring, D. A., Jull, A. J. T., McHargue, L. R., Bland, P. A., Hill,
D. H., and Berry, F. J., 2001. Gold Basin meteorite strewn field,
Mojave Desert: relict of a small Late Pleistocene impact event.
Meteoritics & Planetary Science, 36, 1057–1066.
Lavielle, B., and Marti, K., 1988. Cosmic-ray-produced Kr in
St. Severin core AIII. In Proceedings 18th Lunar and Planetary
Science Conference, Houston, Texas, pp. 565–572.
Lavielle, B., Marti, K., Jeannot, J.-P., Nishiizumi, K., and Caffee,
M. W., 1999. The 36Cl–36Ar–40K–41K records and cosmic-ray
production in iron meteorites. Earth and Planetary Science Let-
ters, 170, 93–104.
Leya, I., and Masarik, J., 2009. Cosmogenic nuclides in stony mete-
orites revisited. Meteoritics & Planetary Science, 44, 1061–1086.
Leya, I., Lange, H.-J., Neumann, S., Wieler, R., and Michel, R.,
2000. The production of cosmogenic nuclides in stony
meteoroids by galactic cosmic ray particles. Meteoritics &
Planetary Science, 35, 259–286.
Leya, I., Wieler, R., Aggrey, K., Herzog, G. F., Schnabel, C.,
Metzler, K., Hildebrand, A. R., Bouchard, M., Jull, A. J. T.,
Andrews, H. R., Wang, M.-S., Ferko, T. E., Lipschutz, M. E.,
Wacker, J. F., Neumann, S., and Michel, R., 2001. Exposure his-
tory of the St-Robert (H5) fall. Meteoritics & Planetary Science,
36, 1479–1494.
Leya, I., Gilabert, E., Lavielle, B., Wiechert, U., and Wieler, R.,
2004. Production rates for cosmogenic krypton and argon iso-
topes in H-chondrites with known 36Cl–36Ar ages. Antarctic
Meteorite Research, 17, 185–199.
Lipschutz, M. E., Signer, P., and Anders, E., 1965. Cosmic ray
exposure ages of iron meteorites by the Ne21/Al26 method. Jour-
nal of Geophysical Research, 6, 1473–1489.
Marti, K., 1967. Mass-spectrometric detection of cosmic-ray pro-
duced 81Kr in meteorites and the possibility of Kr-Kr dating.
Physical Review Letters, 18, 264–266.
Marti, K., and Graf, T., 1992. Cosmic-ray exposure history of ordi-
nary chondrites. Annual Review of Earth and Planetary
Sciences, 20, 221–243.
Marti, K., and Lugmair, G. W., 1971. Kr81-Kr and K-Ar40 ages,
cosmic-ray spallation products and neutron effects in lunar sam-
ples from Oceanus Procellarum. Proceedings of the Lunar
Science Conference, 2, 1591–1605.
Masarik, J., Nishiizumi, K., and Reedy, R. C., 2001. Production
rates of cosmogenic helium-3, neon-21 and neon-22 in ordinary
chondrites and the lunar surface. Meteoritics & Planetary
Science, 36, 643–650.
Nagai, H., Honda, M., Imamura, M., and Kobayashi, K., 1993.
Cosmogenic 10Be and 26Al in metal, carbon and silicate
of meteorites. Geochimica et Cosmochimica Acta, 57, 3705–3723.
Nishiizumi, K., 1995. Terrestrial ages of meteorites from cold and
cold regions. In L. Schultz, J. O. Annexstad, and M. E. Zolensky
(eds.), Workshop on Meteorites from Cold and Hot Deserts.
Houston: Lunar and Planetary Institute. LPI Technical Report
No. 95-02, pp. 53–55.
Nishiizumi, K., and Caffee, M. W., 1998. Measurements of cosmo-
genic calcium-41 and calcium-41/chlorine-36 terrestrial ages
(abstract). Meteoritics & Planetary Science, 33, A117.
Nishiizumi, K., Klein, J., Middleton, R., and Arnold, J. R., 1987.
Long-lived cosmogenic nuclides in the Derrick Peak and
Lazarev iron meteorites. Lunar and Planetary Science Confer-
ence, 18, 724–725.
Nishiizumi, K., Regnier, S., and Marti, K., 1980. Cosmic ray expo-
sure ages of chondrites, pre-irradiation and constancy of cosmic
ray flux in the past. Earth and Planetary Science Letters,
56, 156–170.
Nishiizumi, K., Elmore, D., and Kubik, P. W., 1989. Update on ter-
restrial ages of Antarctic meteorites. Earth and Planetary Sci-
ence Letters, 93, 299–313.
Nishiizumi, K., Okazaki, R., Park, J., Nagao, K., Masarik, J., and
Finkel, R. C., 2002. Exposure and terrestrial histories of Dhofar
019 Martian meteorite. 33rd Lunar and Planetary Science Con-
ference, Houston, Texas, Abstract #1366 (CD-ROM).
Rai, V. K., Murty, S. V. S., and Ott, U., 2003. Noble gases in
ureilites: cosmogenic, radiogenic and trapped components.
Geochimica et Cosmochimica Acta, 67, 4435–4456.
Reedy, R. C., 1985. A model for GCR-particle fluxes in stony mete-
orites and production rates of cosmogenic nuclides. Journal of
Geophysical Research, 90, C722–C728.
Schaeffer, O. A., and Heymann, D., 1965. Comparison of 36Cl–36Ar
and 39Ar–38Ar cosmic-ray exposure ages of dated fall iron mete-
orites. Journal of Geophysical Research, 70, 215–224.
Scherer, P., and Schultz, L., 2000. Noble gas record, collisional his-
tory, and pairing of CV, CO, CK and other carbonaceous chon-
drites. Meteoritics & Planetary Science, 35, 145–153.
 METEORITES (Lu–Hf)
Scherer, P., Herrmann, S., and Schultz, L., 1998. Noble gases in
twenty-one Saharan LL-chondrites: exposure ages and possible
pairings. Meteoritics & Planetary Science, 33, 259–265.
Shukolyukov, A., and Begemann, F., 1996. Cosmogenic and
fissiogenic noble gases and 81Kr-Kr exposure age clusters of
eucrites. Meteoritics & Planetary Science, 31, 60–72.
Signer, P., Baur, H., Derksen, U., Etique, P., Funk, H., Horn, P., and
Wieler, R., 1977. Helium, neon, and argon records of lunar soil
evolution. Proceedings of the Lunar Planetary Science Confer-
ence, 8, 3657–3683.
Welten, K. C., Lindner, L., Van der Borg, K., Loeken, T., Scherer, P.,
and Schultz, L., 1997. Cosmic-ray exposure ages of diogenites
and the recent collisional history of the howardite, eucrite and
diogenite parent body/bodies. Meteoritics & Planetary Science,
32, 891–902.
Welten, K. C., Nishiizumi, K., Masarik, J., Caffee, M. W., Jull,
A. J. T., Klandrud, S. E., and Wieler, R., 2001. Cosmic-ray expo-
sure history of two Frontier Mountain H-chondrite showers from
spallation and neutron-capture products. Meteoritics & Plane-
tary Science, 36, 301–317.
Welten, K. C., Caffee, M. W., Leya, I., Masarik, J., Nishiizumi, K.,
and Wieler, R., 2003. Noble gases and cosmogenic radionuclides
in the Gold Basin L4-chondrite shower: thermal history, expo-
sure history and pre-atmospheric size. Meteoritics & Planetary
Science, 38, 157–173.
Welten, K. C., Nishiizumi, K., Finkel, R. C., Hillegonds, D. J., Jull,
A. J. T., Franke, L., and Schultz, L., 2004. Exposure history and
terrestrial ages of ordinary chondrites from the Dar al Gani
region, Libya. Meteoritics & Planetary Science, 39, 481–498.
Welten, K. C., Nishiizumi, K., Caffee, M. W., Hillegonds, D. J.,
Johnson, J. A., Jull, A. J. T., Wieler, R., and Folco, L., 2006. Ter-
restrial age, pairing and concentration mechanism of Antarctic
chondrites from Frontier Mountain, northern Victoria Land.
Meteoritics and Planetary Science, 41, 1081–1094.
Welten, K. C., Folco, L., Nishiizumi, K., Caffee, M. W., Grimberg,
A., Meier, M. M. M., and Kober, F., 2008. Meteoritic and bed-
rock constraints on the glacial history of Frontier Mountain in
northern Victoria Land, Antarctica. Earth and Planetary Science
Letters, 270, 308–315.
Welten, K. C., Caffee, M. W., Hillegonds, D. J., McCoy, T. J.,
Masarik, J., and Nishiizumi, K., 2011. Cosmogenic radionu-
clides in L5 and LL5 chondrites from Queen Alexandra Range,
Antarctica: identification of a large L/LL5 chondrite shower with
a pre-atmospheric mass of 50 metric tons. Meteoritics & Plan-
etary Science, 46, 177–198.
Wieler, R., Graf, T., Signer, P., Vogt, S., Herzog, G. F., Tuniz, C., Fink,
D., Fifield, L. K., Klein, J., Middleton, R., Jull, A. J. T., Pellas, P.,
Masarik, J., and Dreibus, G., 1996. Exposure history of the Torino
meteorite. Meteoritics & Planetary Science, 31, 265–272.
METEORITES (Lu–Hf)
Audrey Bouvier
Department of Earth Sciences, The University of Western
Ontario, London, ON, Canada
Definition
Radiogenic dating of meteorites and tracing of planetary
differentiation using the radiogenic decay of 176-Lutetium
to 176-Hafnium and variations in their isotopic abundances.
555
Introduction
The 176-Lutetium-176-Hafnium radiogenic isotopic sys-
tem (176Lu half-life ~37 Ga) is a powerful chronometer
and tracer of geochemical processes and planetary evolu-
tion. It was formerly assumed that the Lu–Hf system
would be a valuable complement to the short-lived
(146Sm–142Nd with a ~68 Ma half-life) and long-lived
(147Sm–143Nd with a ~106 Ga half-life) Sm–Nd radio-
genic systems. The Lu–Hf and Sm–Nd systems involve
refractory and lithophile elements, all of which have mostly
incompatible behavior (e.g., Patchett and Tatsumoto, 1980;
Boyet and Carlson, 2005; also see entry on “Sm–Nd Dat-
ing” for further details). The Lu–Hf pair has an advantage
over the Sm–Nd pairs in that it includes two elements with
distinct chemical behaviors: Lu is a heavy rare-earth ele-
ment (REE) and Hf is a high field strength element
(HFSE). Elemental fractionation between Hf and Lu is
consequently greater in geological systems than for two
REEs that have similar chemical behavior.
There has been tremendous progress made over the last
few years for measuring Lu and Hf isotopic abundances
with the development of multi-collector inductively
coupled plasma mass spectrometry techniques
(MC-ICP-MS). Such technical advances have led to the
collection of enormous volumes of high-precision Hf iso-
topic data over the last 15 years. By measuring precisely
small variations in the abundance of 176Hf in various min-
erals and rocks with different Lu/Hf (parent/daughter)
ratios, we can now date, with ~1–10 % precision, geolog-
ical, and extraterrestrial processes as recent as ~15 Ma in
igneous, metamorphic, or sedimentary rocks. Many recent
studies have been focused on understanding the differenti-
ation of the mantle and crust, and the dynamics of these
reservoirs on the Earth, Moon, Mars, and planetesimals
(e.g., Vervoort et al., 1996; Blichert-Toft et al., 1999,
2002). Studies in igneous geochemistry and metamorphic
petrology particularly benefit from minerals that highly
fractionate Lu from Hf during geological processes. Such
minerals include zircon and ilmenite with very low Lu/Hf
to track crustal evolution, and garnet and phosphate with
very high Lu/Hf, to establish the chronology of metamor-
phic events. These accessory minerals are nevertheless
rare in extraterrestrial materials, which sample primitive,
ultramafic, and rarely evolved planetary crust. Also,
because of the low concentration levels of Hf and REE
in meteorites and limited supplies of these precious plane-
tary materials, most of the cosmochemical studies have
been focusing on bulk-rock analysis. Occasionally, sili-
cate separates such as pyroxene, feldspar series, and oliv-
ine (Bouvier et al., 2008a; Thrane et al., 2010) oxides
(e.g., chromite; Lapen et al., 2010), and phosphates
(Amelin, 2005) have been analyzed for chronological
studies of meteorites.
One area of considerable activity in geochemistry is the
isotopic analysis of Hf in parallel with U-Pb zircon stud-
ies, either to constrain early crustal growth and evolution
or as a complement to detrital zircon studies
556 METEORITES (Lu–Hf)

Meteorites (Lu–Hf), Figure 1 Lu–Hf compositions of unmetamorphosed (petrologic types 1–3, red symbols) and metamorphosed
(petrologic types 4–6, blue symbols) bulk chondrites obtained using efficient dissolution methods (Bizzarro et al., 2003; Patchett
et al., 2004; Bouvier et al., 2008b; Dauphas and Pourmand, 2011) with corresponding isochron regression lines. Thermal
metamorphism produces a redistribution of Lu relative to Hf on the chondrite parent bodies and consequently a rotation and larger
spread of the Lu–Hf compositions of the petrological types 4 to 6 bulk chondrites. Proposed initial solar system, and modern CHUR
and BSE compositions are also indicated (Bouvier et al., 2008b); see text for additional details.

(e.g., Harrison et al., 2005). Interpretation of these data,
however, relies on the availability of unambiguous and
well-constrained decay constants for these isotope sys-
tems. At the moment, a major issue for deciphering plane-
tary differentiation processes is that the apparent
ages deduced from the Lu–Hf isochron method in plane-
tary materials are variable by ~5 %. Consequently, there
is an uncertainty for estimates of the initial Hf isotopic
composition of the solar system and consequently an even
larger uncertainty for the Earth’s bulk composition. The
following sections detail the recent determination of the
176
Lu–176Hf parameters for solar system evolution and
potential building blocks of the Earth. A third section dis-
cusses the recent findings of irradiation effects on Hf iso-
topes due to the long-term exposure of planetary materials
to cosmic rays in space and how it affects the interpretation
of Lu–Hf and Sm–Nd systematics in many meteorites and
lunar samples. Finally, the last section will present Lu–Hf
chronological studies for Martian meteorites, which are
not affected by these processes and can therefore be used
to constrain the formation times of these rocks.
Lu–Hf isotopic composition of the solar system
The b
decay of 176-Lutetium (176Lu; 2.56 % of Lu) pro-
duces 176-Hafnium (176Hf; 5.26 % of Hf). Dixon
et al. (1954) reported that 176Lu also decays by electron
capture to176Yb with a branching ratio of ~3 %. Determi-
nations of the 176Lu decay constant (l176Lu) by physical
counting experiments vary widely, leading geochemists
to use the Lu–Hf isochron method in various meteoritic
and geological samples of known U–Pb ages (see
Albarède et al., 2006 for a review). This method deduces
the value of the decay constant l from the slope s of the
isochron of an object of known age t by the relationship
s ¼ elt
1 (Figure 1). An important requirement is that
both systems closed rapidly to diffusion and remained
subsequently undisturbed.
An unsolved discrepancy of ~5 % exists between the
determinations of l176Lu (and consequently Lu–Hf ages)
based on meteorites at ~1.93–1.99  10
11 year
1 (e.g.,
Patchett and Tatsumoto, 1980; Blichert-Toft et al., 2002;
Bizzarro et al., 2003; Thrane et al., 2010) and those based
on terrestrial samples at 1.867  10
11 year
1 (Scherer
et al., 2001; Söderlund et al., 2004). It was suggested that
this difference was caused by an early episode of irradia-
tion in the solar system that generated 176Hf excesses in
chondrite, eucrite, and angrite meteorites (Albarède
et al., 2006; Thrane et al., 2010; Figure 1). This effect,
however, is not observed by the time of formation of phos-
phates from ordinary chondrite (4,551 Ma) and
acapulcoite (4,557 Ma) meteorites (Amelin, 2005). This
discrepancy hampers accurate determination of the initial
Hf isotopic composition of the solar system and the Earth,
leading to an uncertainty of as much as seven epsilon Hf
units for the composition of the modern BSE (denoted
eHf, expressed as the deviations in parts per ten thousand
relative to the reference CHUR parameters) (Figure 1).
 METEORITES (Lu–Hf)
There are now several additional Lu–Hf isotopic
datasets on bulk chondrites that have been gathered using
MC-ICPMS since the pioneering work by Blichert-Toft
and Albarede (1997). When comparing metamorphosed
(petrologic types 4–6) and unmetamorphosed (petrologic
types 1–3) chondrites, it is evident that there is
a disturbance of the Lu–Hf bulk chondrite isochron during
thermal metamorphism on the chondrite parent body.
Metamorphism on chondrite parent bodies is the conse-
quence of accretionary and radiogenic heat that lasted for
at least 10–100 Ma on planetesimals with tens of kilome-
ter diameters. Later planetary impacts may also have con-
tributed to such disturbances at depth in the planetary
bodies. Recently determined internal Lu–Hf isochrons of
individual refractory inclusions (4,568 Ma) and chondrite,
eucrite and angrite meteorites also show variability in the
deduced l176Lu values and initial Hf isotopic composi-
tions, showing that secondary processes such as thermal
metamorphism and impact must have heterogeneously
affected the parent bodies of these meteorites.
Building blocks of the earth
Because the Lu–Hf and Sm–Nd systems involve refrac-
tory and lithophile elements, it has been always assumed
that the average composition of bulk terrestrial planets
should be the same as that of primitive chondrites, and
for this reason, the term Chondritic Uniform Reservoir
was defined (CHUR). Knowing the bulk composition of
the Earth is essential for these isotopic tools to realize their
potential in planetary studies. The starting composition of
terrestrial planets (e.g., Earth, Mars) is commonly
assumed to be chondritic for Lu–Hf and by extension for
Sm–Nd, which are both refractory elements. There are
nevertheless variations in stable isotopic compositions of
many elements that show that the Earth is made of differ-
ent types of chondritic and unknown planetary materials
such as potentially differentiated materials and comets.
The hypothesis of collisional erosion during accretion
and differentiation would lead to fractionated bulk composi-
tion of planets. This mechanism was suggested to explain
the so-called missing reservoir for the Earth or super-
chondritic composition (Boyet and Carlson, 2005; Caro
et al., 2008; O'Neill and Palme, 2008), but such fraction-
ation and missing component is nevertheless not observed
for large ~400 km diameter planetesimals such as 4 Vesta,
the accepted parent body of basaltic eucrite meteorites
which have near-chondritic Lu–Hf compositions
(Blichert-Toft et al., 2002). Evidence from lunar Lu–Hf
and Sm–Nd isotopic compositions also suggests that the
chondritic model holds for these two systems for the Moon
and consequently also for the Earth (Sprung et al., 2013).
Lu–Hf isotopic variations in meteorites
The improvements made on the precision of isotopic mea-
surements allow scientists to generate new insights into
the origin of isotopic variation and the stellar contribution
(e.g., giant stars, supernovae) to the matter present in the
557
solar system (e.g., Carlson et al., 2007). Some isotopes
of interest (e.g.,178Hf,152Sm,146Nd) can also be affected
by neutron capture which is a quite common process in
meteorites and planetary materials (e.g., lunar rocks) due
to their journey into space or long-term exposure on plan-
etary surfaces. If there are large isotopic anomalies due to
cosmic-ray exposure on the normalizing isotopes, it will
ultimately result in incorrect normalizations of all the
ratios used for Lu–Hf dating.
Nucleosynthetic anomalies
It was suggested that Lu and Yb isotopic variations could
potentially explain discrepancies in the 176Lu decay rates
deduced in meteorites if there were detectable irradiation
effects for Lu or a branching decay of 176Lu to 176Yb
(Amelin and Davis, 2005; Scherer et al., 2005).
175
Lu/176Lu ratios have been measured in planetary mate-
rials and do not show any resolvable variations that could
explain such discrepancy (Wimpenny et al., 2013). For Hf,
Sprung et al. (2010) show that there is no resolvable vari-
ation in the stable isotopic composition of bulk chondrites
(Sprung et al., 2010), while small variations in leached
chondrites (Qin et al., 2011) and refractory inclusions
(Akram et al., 2013; Bouvier and Boyet, 2013) were most
recently found and likely caused by nucleosynthetic
anomalies but are too small to affect Lu–Hf dating of
meteorites.
Irradiation effects
The Lu–Hf dating of meteorites can be significantly
affected by variations in stable isotopic abundances for
Lu and Hf isotopic compositions due to irradiation effects
from cosmic-ray exposure. The isotope 149Sm is very use-
ful as it has a large thermal neutron capture cross section
and can be used to monitor irradiation effects on isotopic
compositions (e.g., Lingenfelter et al., 1972). For exam-
ple, it has been used for correcting irradiated Nd isotopic
compositions of lunar samples (Nyquist et al., 1995). Sec-
ondary neutron capture effects on Hf isotopes have been
detected in some achondrites (mesosiderites, aubrites)
and lunar samples (Sprung et al., 2010) and were corrected
using the measured 149Sm/144Sm ratio of the meteorites.
These findings have changed the view of Hf–Nd isotopic
evolution and differentiation of the lunar mantle and crust
and also provided an additional constraint on the validity
of the chondritic model for the Hf–Nd isotopic system
for the Moon and by extension the Earth (Sprung
et al., 2010).
Lu–Hf dating martian meteorites
Martian meteorites are the only meteorites not affected by
nucleosynthetic anomalies and irradiation effects that
affect most of the planetary materials, as described in the
previous sections. There has been extensive geochemical
and chronological research on Martian meteorites as more
than 150 individual specimens have been found so far.
Using Lu–Hf chronometry, the unique orthopyroxenite
558 METEORITES (Lu–Hf)
Allan Hills 84001 (found in 1984 in Antarctica) has been
dated at 4.1 Ga (Lapen et al., 2010) consistent with its
Pb–Pb age (Bouvier et al., 2009), while shergottites show
a broad range of Lu–Hf ages from about 185 Ma for
Zagami (fell in 1962 in Nigeria) (Bouvier et al., 2005) to
580 Ma for Tissint (fell in 2012 in Morocco) (Grosshans
et al., 2013) contrasting with their corresponding Pb–Pb
ages of 4.1 Ga and 4.3 Ga, respectively (Bouvier
et al., 2008a). The explanation for why only the Lu–Hf
ages of shergottites are much younger than their
corresponding Pb–Pb ages is an intensive area of current
research. It is good to note that shergottite meteorites have
been extremely shocked by impact on the surface of Mars
and contain abundant impact melt and high-pressure poly-
morphs of many minerals (maskelynite replacing plagio-
clase, stishovite replacing quartz, ringwoodite replacing
olivine). These observations may explain why Lu–Hf
and other similar parent-daughter radiogenic isotopic
chronometers have been disturbed in these meteorites
(El Goresy et al., 2013).
Summary
The most accurate value for l176Lu deduced from geolog-
ical materials is 1.867  10
11 year
1 with
a corresponding half-life of 37.1 Ga. There is no robust
evidence for an accelerated decay of 176Lu in meteorites.
The calculated initial ratio of 176Hf/177Hf ¼ 0.27979, as
deduced from the modern CHUR parameters of
176
Lu/177Hf ¼ 0.0336 and 176
Hf/177Hf ¼ 0.282785,
remains the best estimate for the initial isotopic composi-
tion of the solar system. The Lu–Hf system is extremely
useful for tracking early planetary processes and chemical
reservoirs formed in the interior of planets. Its chronolog-
ical applications to meteorites are nevertheless limited due
to its lower precision (a few million years to tens of mil-
lions years) and higher susceptibility to disturbance com-
pared to the Pb–Pb chronometer, which is the most
accurate and precise absolute chronometer used to estab-
lish the time scales of formation of meteorites, planets,
and solar system (see entry on “Uranium–Lead, Extrater-
restrial, Planetary Materials”).
Bibliography
Akram, W., Schönbächler, M., Sprung, P., and Vogel, N., 2013.
Zirconium–Hafnium isotope evidence from meteorites for the
decoupled synthesis of light and heavy neutron-rich nuclei. The
Astrophysical Journal, 777, 169.
Albarède, F., Scherer, E. E., Blichert-Toft, J., Rosing, M. T.,
Simionovici, A., and Bizzarro, M., 2006. Gamma-ray irradiation
in the early solar system and the conundrum of the 176Lu decay
constant. Geochimica et Cosmochimica Acta, 70, 1261–1270.
Amelin, Y., 2005. Meteorite phosphates show constant 176Lu decay
rate since 4557 million years ago. Science, 310, 839–841.
Amelin, Y., and Davis, W. J., 2005. Geochemical test for branching
decay of 176Lu. Geochimica et Cosmochimica Acta, 69,
465–473.
Bizzarro, M., Baker, J. A., Haack, H., Ulfbeck, D., and Rosing, M.,
2003. Early history of the Earth’s crust-mantle system inferred
from hafnium isotopes in chondrites. Nature, 421, 931–933.
Blichert-Toft, J., and Albarede, F., 1997. The Lu-Hf isotope geo-
chemistry of chondrites and the evolution of the mantle-crust
system. Earth and Planetary Science Letters, 148, 243–258.
Blichert-Toft, J., Gleason, J. D., Télouk, P., and Albarède, F., 1999.
The Lu-Hf isotope geochemistry of shergottites and the evolu-
tion of the Martian mantle-crust system. Earth and Planetary
Science Letters, 173, 25–39.
Blichert-Toft, J., Boyet, M., Télouk, P., and Albarède, F., 2002.
147
Sm-143Nd and 176Lu-176Hf in eucrites and the differentiation
of the HED parent body. Earth Planetary Science Letters,
204, 167–181.
Bouvier, A., and Boyet, M., 2013. Lu-Hf and Sm-Nd systematics
of the first solids in the Solar System. Mining Magazine, 77, 754.
Bouvier, A., Blichert-Toft, J., Vervoort, J. D., and Albarède, F.,
2005. The age of SNC meteorites and the antiquity of the
Martian surface. Earth and Planetary Science Letters, 240,
221–233.
Bouvier, A., Blichert-Toft, J., Vervoort, J. D., Gillet, P., and
Albarède, F., 2008a. The case for old basaltic shergottites. Earth
and Planetary Science Letters, 266, 105–124.
Bouvier, A., Vervoort, J. D., and Patchett, P. J., 2008b. The Lu-Hf
and Sm-Nd isotopic composition of CHUR: constraints from
unequilibrated chondrites and implications for the bulk composi-
tion of terrestrial planets. Earth and Planetary Science Letters,
273, 48–57.
Bouvier, A., Blichert-Toft, J., and Albarède, F., 2009. Martian mete-
orite chronology and effects of impact metamorphism. EOS
Trans AGU Fall Meet. Suppl. 90: Abstract MR12A-04.
Boyet, M., and Carlson, R. W., 2005. 142Nd evidence for early
(>4.53 Ga) global differentiation of the silicate Earth. Science,
309, 576–581.
Carlson, R. W., Boyet, M., and Horan, M., 2007. Chondrite Barium,
Neodymium, and Samarium isotopic heterogeneity and early
earth differentiation. Science, 316, 1175–1178.
Caro, G., Bourdon, B., Halliday, A. N., and Quitté, G., 2008. Super-
chondritic Sm/Nd ratios in Mars, the Earth and the Moon.
Nature, 452, 336–339.
Dixon, D., McNair, A., and Curran, S. C., 1954. The natural radio-
activity of Lutetium. Philosophical Magazine, 45, 683–694.
El Goresy, A., Gillet, P., Miyahara, M., Ohtani, E., Ozawa, S., Beck,
P., and Montagnac, G., 2013. Shock-induced deformation of Sher-
gottites: shock-pressures and perturbations of magmatic ages on
Mars. Geochimica et Cosmochimica Acta, 101, 233–262.
Grosshans, T. E., Lapen, T. J., Andreasen, R., and Irving, A. J.,
2013. Lu-Hf and Sm-Nd ages and source compositions for
depleted Shergottite Tissint. In 44th Lunar and Planetary
Science Conference, The Woodlands, TX, p. 2872.
Harrison, T. M., Blichert-Toft, J., Müller, W., Albarède, F., Holden,
P., and Mojzsis, S. J., 2005. Heterogeneous Hadean hafnium:
evidence of continental crust at 4.4 to 4.5 Ga. Science,
310, 1947–1950.
Lapen, T. J., Righter, M., Brandon, A. D., Debaille, V., Beard, B. L.,
Shafer, J. T., and Peslier, A. H., 2010. A younger age for
ALH84001 and its geochemical link to shergottite sources in
Mars. Science, 328, 347–351.
Lingenfelter R.E., Canfield E.H., Hampel V.E., 1972. The lunar
neutron flux revisited. Earth and Planetary Science Letters 16,
355–369.
Nyquist, L. E., Wiesman, H., Bansal, B., Shih, C.-Y., Keith, J. E.,
and Harper, C. L., 1995. 146Sm-142Nd formation interval for
the lunar mantle material. Geochimica et Cosmochimica Acta,
59, 2817–2837.
O'Neill, H. S. C., and Palme, H., 2008. Collisional erosion and the
non-chondritic composition of terrestrial planets. Philosophical
Transactions of the Royal Society A, 366, 4205–4238.
Patchett, P. J., and Tatsumoto, M., 1980. Lu-Hf total-rock isochron
for the eucrite meteorites. Nature, 288, 571–574.
 METEORITES (U–Pb)
Qin, L., Carlson, R. W., and Alexander, C. M. O. D., 2011. Corre-
lated nucleosynthetic isotopic variability in Cr, Sr, Ba, Sm, Nd
and Hf in Murchison and QUE 97008. Geochimica et
Cosmochimica Acta, 75, 7806–7828.
Scherer, E., Münker, C., and Mezger, K., 2001. Calibration of the
Lutetium-Hafnium clock. Science, 293, 683–687.
Scherer, E., Münker, C., Kleine, T., Mezger, K., Albarède, F., and
Blichert-Toft, J. (2005). The isotopic composition of Lu in mete-
orites and lunar rocks: implications for the decay constant of
176
Lu. Geophysical Journal Abstracts, 7.
Söderlund, U., Patchett, P. J., Vervoort, J. D., and Isachsen, C. E.,
2004. The176Lu decay constant determined by Lu-Hf and U-Pb
isotope systematics of Precambrian mafic intrusions. Earth and
Planetary Science Letters, 219, 311–324.
Sprung, P., Scherer, E. E., Upadhyay, D., Leya, I., and Mezger, K.,
2010. Non-nucleosynthetic heterogeneity in non-radiogenic sta-
ble Hf isotopes: implications for early solar system chronology.
Earth and Planetary Science Letters, 295, 1–11.
Sprung, P., Kleine, T., and Scherer, E. E., 2013. Isotopic evidence
for chondritic Lu/Hf and Sm/Nd of the Moon. Earth and Plane-
tary Science Letters, 380, 77–87.
Thrane, K., Connelly, J. N., Bizzarro, M., Meyer, B. S., and
The, L.-S., 2010. Origin of excess 176Hf in meteorites. Astro-
physics Journal, 717, 861–867.
Vervoort, J. D., Patchett, P. J., Gehrels, G. E., and Nutman, A. P.,
1996. Constraints on early earth differentiation from hafnium
and neodymium isotopes. Nature, 379, 624–627.
Wimpenny, J., Amelin, Y., and Yin, Q. Z., 2013. Isotopic composi-
tion of Lu in terrestrial samples and Achondrites: implications
for Lu-Hf systematics in the early solar system. Meteoritics
and Planetary Science, 76, 5218.
Cross-references
Lu-Hf Dating: The Lu-Hf Isotope System
Meteorites (U–Pb)
Sedimentary Rocks (Rb-Sr Geochronology)
Sm–Nd Dating
METEORITES (U–Pb)
Yuri Amelin
Research School of Earth Sciences, The Australian
National University, Canberra, Australia
Synonyms
207
Pb/206Pb dating; Pb-isotopic dating; Pb-Pb dating;
U-Pb dating of meteorites
Definition
Determination of ages of meteorites and their components
using decay of radioactive isotopes 238U and 235U
and accumulation of their decay products (daughter iso-
topes) 206Pb and 207Pb.
Introduction
Meteorites contain trace amounts of uranium, and their
ages can be calculated from the quantities of accumulated
radiogenic isotopes of Pb. Two long-lived isotopes of
uranium 235U (0.7205 %, half-life 4.468  109 year)
559
and 238 U(99.2736 %, half-life 0.704  109 year; isotopic
abundances based on the data of Brennecka and Wadhwa,
2012, half-lives from Jaffey et al., 1971) decay to stable iso-
topes of lead 207Pb and 206Pb. The dual nature of the U–Pb
decay scheme has two important implications for geochro-
nology (Ludwig, 1998; Amelin, 2006; Schoene, 2014).
First, consistency of the dates calculated from two parent-
daughter pairs, 238U– 206Pb* and 235U– 207Pb* (the asterisk
indicates radiogenic Pb), can be checked using concordia
diagrams (Wetherill, 1956; Tera and Wasserburg, 1972),
and their consistency provides evidence that the
U–Pb-isotopic system remained closed since formation of
the rock. Second, the two decay schemes can be combined,
and the date can be calculated from the ratio of radiogenic
isotopes 207Pb* and 206Pb*. The 207Pb*/206Pb* dates are
insensitive to recent migration of U or recent loss of radio-
genic Pb, so it may be possible to get accurate ages even
from meteorites affected by recent impacts, terrestrial
weathering, and laboratory treatment that is usually neces-
sary for removal of non-radiogenic Pb. Modern U-Pb chro-
nology of meteorites is based on 207Pb*/206Pb*, whereas
206
Pb*/238U and 207Pb*/235U provide supporting
information.
Concentration and distribution of uranium in
meteorites
Modern “terrestrial” U-Pb geochronology mainly relies on
dating minerals such as zircon, baddeleyite, and monazite
that usually contain 10–1,000 ppm or more U and negligi-
ble quantities of non-radiogenic Pb. In contrast, minerals
with U content greater than several ppm are very rare in
chondrites and their components and in achondrites of aster-
oidal origin. Uranium concentrations in various types of
meteorites are summarized in Table 1. Because of low
U concentration, several milligrams of bulk meteorite, or
a chondrule or CAI material, are required for a single anal-
ysis, compared to microgram or smaller quantities of zircon
and other U-rich minerals commonly analyzed in “terres-
trial” U–Pb geochronology. Some meteorites contain very
rare minerals with 10 ppm or more of uranium: perovskite
in some CAIs in CV chondrites, silica-rich apatite in plu-
tonic angrites, and small grains of zircon on some chon-
drites and eucrites. Despite the advantage provided by
higher U concentrations and the evidence for usually con-
cordant isotopic systems in these minerals, their use in
meteorite U–Pb chronology has been limited because of
their extreme rare occurrence. The mainstream meteorite
U–Pb chronology still relies on representative fractions of
whole meteorites, bulk chondrules, and CAIs and on rock-
forming minerals such as pyroxene, all of which contain
well below 1 ppm of uranium.
From isotopic ratio to the age of a meteorite
in five steps
Obtaining a meaningful Pb-isotopic age of a meteorite
from precise Pb-isotopic measurements involves five
steps (Amelin et al., 2009, including their Electronic
560 METEORITES (U–Pb)

Meteorites (U–Pb), Table 1 Examples of uranium concentration (in parts per million) in meteorite classes, selected meteorites, and
their components

Uranium concentration, ppm,

Meteorite class Meteorite group Component or mineral range, and (median)

Bulk meteorites and their major components

Ordinary chondrites H Bulk meteorite 0.009–0.135 (0.022)
Ordinary chondrites L and LL Bulk meteorite 0.007–0.027 (0.012)
Enstatite chondrites EH Bulk meteorite 0.006–0.008
Carbonaceous chondrites CI, CM, CV, CR, ungrouped Bulk meteorite and chondrules 0.012–0.491 (0.026)
Carbonaceous chondrites CV Bulk Ca-Al-rich refractory inclusions (CAIs) 0.041–0.192 (0.113)
Eucrites Bulk meteorite 0.008–0.305 (0.100)
Angrites Quenched and plutonic Bulk meteorite 0.030–0.302 (0.128)
Rare high-U minerals in meteorites of asteroidal (non-planetary) origin
Carbonaceous chondrites CV Perovskite in CAIs 3–23
Chondrites and achondrites Zircon 30–133
Angrites Plutonic Silica-rich apatite 11–22

Data from Kaltenbach et al. (2012) and Amelin et al. (2009) and references therein

Annex EA-1). First, the Pb-isotopic ratios are corrected for
all sources of analytical biases, such as mass fractionation,
interferences, and matrix effects. Second, laboratory blank
is subtracted, resulting in accurate isotopic composition of
the Pb that was isolated from the sample. Third, radiogenic
207
Pb*/206Pb* ratio is calculated or inferred from blank-
corrected isotopic ratios that may include initial Pb, radio-
genic Pb, and terrestrial contaminant Pb using an isochron
(preferably) or a model date approach. Fourth, the
207
Pb*/206Pb* ratio is converted to a date – a measure of
time, using known values of uranium decay constants
and isotopic composition. Finally, the meaning of the date
as marking the timing of a real, natural event is established,
and the nature of this event is identified. Systematic uncer-
tainties possibly leading to inaccuracy in 207Pb*/206Pb*
ages of meteorites can be introduced at all five steps, and
their sources have to be carefully examined. Below we dis-
cuss two of the most important sources of potential inaccu-
racy in meteorite ages.
Removal of non-radiogenic Pb
Along with radiogenic Pb, meteorites contain several
components of non-radiogenic Pb: initial Pb of primordial
composition, Pb mobilized from the target and projectile
rocks by impacts in the parent asteroid and implanted into
surrounding rocks, and terrestrial Pb introduced by
weathering and handling. In U-Pb dating of meteorites
containing as little as 10–200 ppb of U (and similar con-
centrations of radiogenic 206Pb), we have to work with
larger samples and rely on laboratory treatments, such as
multistep acid leaching, to separate radiogenic Pb from
several components of non-radiogenic Pb. Modern
multistep leaching schemes produce radiogenic Pb with
reasonably high 206Pb/204Pb (up to several thousands in
favorable cases) from most refractory inclusions and igne-
ous meteorites and are usable even if less efficient for
chondrules. However, most radiogenic Pb is lost in the
earlier leaching steps along with non-radiogenic Pb, so
that the final, usually the purest, leaching steps typically
contain only 5–10 % of the total radiogenic Pb. This is
a serious disadvantage in dating materials of limited size
or quantity, such as individual chondrules, or low-U min-
eral phases from CAIs and achondrites. Furthermore, acid
leaching can fractionate radiogenic Pb from its parent U,
thus rendering a concordance test of closed system behav-
ior invalid. A search for alternative treatments, such as
nonacidic selective dissolution of minerals and volatility-
driven separation of Pb components by evaporation under
controlled conditions, can bring about more efficient
methods for extraction of radiogenic Pb from meteorites.
Isotopic composition of uranium
The 238U/235U ratio was considered constant and equal to
137.88 until recently, but several studies published since
2010 show that it is variable. Uranium isotopic variations
among the CAIs, ranging from 137.409  0.039 to
137.885  0.009 (Brennecka et al., 2010), are most prom-
inent. Most recent data from bulk chondrites and achon-
drites suggest uniform 238U/235U ratio (Bouvier
et al., 2011; Brennecka and Wadhwa, 2012; Connelly
et al., 2012), but it is unclear whether variations at
a smaller scale observed in these and ongoing studies are
significant and whether the meteorite and terrestrial
U isotopic compositions are significantly offset from each
other. Considering possible variability, the U isotope
ratios should be determined for every meteorite that is
being precisely dated with the 207Pb/206Pb method. These
determinations are, however, performed at the limit of
analytical sensitivity of modern multicollector inductively
coupled plasma mass spectrometry (MC-ICPMS) and typ-
ically require several times larger quantities of material
than Pb-isotopic analyses by ionization in thermal ioniza-
tion mass spectrometry (TIMS). Development of a more
sensitive method for precise (0.01 % or better)
 METEORITES (U–Pb)
measurement of the 238U/235U ratios is one of the most
desperately needed analytical advancements in U–Pb dat-
ing of meteorites. As the analytical uncertainties of U–Pb
dating decrease, methods of non-radiogenic Pb removal
improve, understanding of chemical and isotopic system-
atics of meteorite progresses, and correction for
variable 238U/235U becomes one of the dominant compo-
nents in the total uncertainty of meteorite U-Pb dates.
Reducing this uncertainty will require thorough under-
standing of U isotopic variation in meteorites, along with
improvements in analytical sensitivity and precision.
Precision of Pb-isotopic dates of meteorites
Fast decay of 235U and its high relative abundance at the
time of the solar system formation make the change of
radiogenic 207Pb*/206Pb* over time strongly nonlinear
and initially very rapid, so that the relative uncertainty of
the 207Pb*/206Pb* dates for the oldest minerals and rocks
(>4,500 Ma) is 3.2 times smaller than the relative uncer-
tainty of the 207Pb*/206Pb* ratio. Together with the mod-
ern advancements in techniques of isotopic analysis, this
makes the 207Pb*/206Pb* ratio an exceptionally precise
chronometer for ancient rocks and minerals. Precision of
0.2–0.5 Ma is now common in dating early meteoritic
solids such as refractory inclusions, chondrules, and dif-
ferentiated asteroidal materials (Amelin, 2008; Wadhwa
et al., 2009; Connelly et al., 2012). This precision is simi-
lar to that obtained with the popular extinct radionuclide
systems 182Hf-182W and 53Mn-53Cr. In the most favorable
cases, the uncertainty of the 207Pb/206Pb can be as low as
0.1 Ma, rivaling precision of the state-of-the-art
26
Al-26Mg dating. Similar precision levels form a strong
basis for building a coherent timescale of the early solar
system processes with a combination of Pb-isotopic and
extinct radionuclide chronometers.
The prospect of further improvement of precision
in Pb-isotopic dating of meteorites
Even the most precise Pb-isotopic dates reported so far
have 30–100 times larger uncertainties than the ultimate
precision level based on counting statistics and the number
of atoms of 207Pb and 206Pb consumed during analysis:
about 0.11 Ma for 1 pg of radiogenic Pb and a mere
3,400 years for 1 ng of radiogenic Pb. The main sources
of additional analytical uncertainties are incomplete con-
version of Pb atoms in the sample to the ions that reach
the detectors and noise and distortions in the ion beam reg-
istration. The efficiency of Pb ionization in TIMS using
the best existing emitters is about 5–10 %, and this alone
increases the counting statistics-related uncertainty by
3–4.5 times compared to the hypothetical 100 % ioniza-
tion and extraction. In MC-ICPMS, another technique
commonly used in meteorite U–Pb chronology, the ioni-
zation efficiency for Pb, is close to 100 %, but the limited
ion extraction efficiency makes the total ion yield of even
the most sensitive instrument configurations (2–3.5 %)
lower than in TIMS. The detector systems also add
561
substantial uncertainties to the isotope ratio measure-
ments. The noise of the Faraday cup arrays with amplifiers
in the current mode, which is used in modern commercial
mass spectrometers, is limited by the thermal noise of the
feedback resistors. This source of noise can be greatly
reduced by using amplifiers with high input impedance
in charge mode. Ion counting with an array of multipliers
is an alternative solution for measuring very low ion
beams, but it also requires many improvements in linear-
ity, cross-calibration, gain stability, and other parameters.
Summary
The 207Pb*/206Pb* form of the U-Pb method is the only
extant radionuclide dating method with sufficient preci-
sion for resolving individual events in accretion of the
solar protoplanetary disk and for cross-calibration with
extinct radionuclide chronometers. It allows dating refrac-
tory inclusions, chondrules, and achondrites with preci-
sion of 0.1–0.5 Ma using samples of several milligrams
to tens of milligrams. Accuracy of the dates is similar, if
we do not consider uncertainty in the decay constants of
uranium isotopes, which is only relevant in comparisons
to other long-lived radionuclide chronometers. Multistep
isotope leaching can efficiently separate radiogenic
Pb. Pb isotope dating must be accompanied by measure-
ments of U isotopic ratios because of their known variabil-
ity. Further substantial improvement of precision and
reduction in sample sizes is likely to be achieved in the
future by improvements of analytical sensitivity and
reduction of all components of instrument noise and labo-
ratory contamination.
Bibliography
Amelin, Y., 2006. The prospect of high-precision Pb isotopic dating
of meteorites. Meteoritics and Planetary Science, 41, 7–17.
Amelin, Y., 2008. U–Pb ages of angrites. Geochimica et
Cosmochimica Acta, 72, 221–232.
Amelin, Y., Connelly, J., Zartman, R. E., Chen, J. H., Göpel, C., and
Neymark, L. A., 2009. Modern U–Pb chronometry of meteor-
ites: advancing to higher time resolution reveals new problems.
Geochimica et Cosmochimica Acta, 73, 5212–5223.
Bouvier, A., Spivak-Birndorf, L. J., Brennecka, G. A., and Wadhwa,
M., 2011. New constraints on early Solar System chronology
from Al–Mg and U–Pb isotope systematics in the unique basal-
tic achondrite Northwest Africa 2976. Geochimica et
Cosmochimica Acta, 75, 5310–5323.
Brennecka, G. A., and Wadhwa, M., 2012. Uranium isotope compo-
sitions of the basaltic angrite meteorites and the chronological
implications for the early solar system. Proceedings of the
National Academy of Sciences, 109, 9299–9303.
Brennecka, G. A., Weyer, S., Wadhwa, M., Janney, P. E., Zipfel, J.,
and Anbar, A. D., 2010. 238U/235U variations in meteorites:
extant 247Cm and implications for Pb–Pb dating. Science,
327, 449–451.
Connelly, J. N., Bizzarro, M., Krot, A. N., Nordlund, Å., Wielandt,
D., and Ivanova, M. A., 2012. The absolute chronology and ther-
mal processing of solids in the solar protoplanetary disk.
Science, 338, 651–655.
Hiess, J., Condon, D. J., McLean, N., and Noble, S. R., 2012.
238 235
U/ U systematics in terrestrial uranium-bearing minerals.
Science, 335, 1610–1614.
562 METEORITES, RUBIDIUM–STRONTIUM, AND SAMARIUM–NEODYMIUM CHRONOLOGY
Kaltenbach, A, Stirling, C.H., and Amelin, Y., 2012. Uranium isoto-
pic composition of carbonaceous chondrites. 43rd Lunar and
Planetary Science Conference, abstract 1691.
Jaffey, A. H., Flynn, K. F., Glendenin, L. E., Bentley, W. C., and
Essling, A. M., 1971. Precision measurement of half-lives and spe-
cific activities of 235U and 238U. Physical Review C,
4, 1889–1906.
Ludwig, K. R., 1998. On the treatment of concordant uranium–lead
ages. Geochimica et Cosmochimica Acta, 62, 665–676.
Schoene, B., 2014. U–Th–Pb geochronology. In Rudnick, R. (ed.),
Treatise on Geochemistry (Second Edition), Vol. 4: The Crust,
Oxford, UK, Elsevier, pp. 341–378.
Tera, F., and Wasserburg, G. J., 1972. U–Th–Pb systematics in three
Apollo 14 basalts and the problem of initial Pb in lunar rocks.
Earth and Planetary Science Letters, 14, 281–304.
Wadhwa, M., Amelin, Y., Bogdanovski, O., Shukolyukov, A.,
Lugmair, G. W., and Janney, P., 2009. Ancient relative and abso-
lute ages for a basaltic meteorite: implications for timescales of
planetesimals accretion and differentiation. Geochimica et
Cosmochimica Acta, 73, 5189–5201.
Wetherill, G. W., 1956. Discordant uranium–lead ages. Transac-
tions of the American Geophysical Union, 37, 320–326.
Cross-references
Mass Spectrometry
Meteorites (Lu–Hf)
Meteorites, Rubidium–Strontium, and Samarium–Neodymium
Chronology
Rhenium–Osmium Dating (Meteorites)
Sedimentary Rocks (Rb-Sr Geochronology)
Terrestrial Cosmogenic Nuclide Dating
Uranium–Lead Dating
Uranium–Lead, Extraterrestrial, Early Solar System
Uranium–Lead, Extraterrestrial, Planetary Materials
METEORITES, RUBIDIUM–STRONTIUM, AND
SAMARIUM–NEODYMIUM CHRONOLOGY
Lars Borg
Chemical Sciences Division, Lawrence Livermore
National Laboratory, Livermore, CA, USA
Rb–Sr chronology of meteorites
Rb–Sr isotopic systematics
The Rb–Sr isotopic system has been used to constrain the
evolution of Solar System since it was first applied to basal-
tic achondrite meteorites by Papanastassiou and
Wasserburg (1969). Although this method was first used
to obtain an age by Hahn et al. (1943), it was not until
Nier-type mass spectrometers (see entry “Mass Spectrome-
try”) were readily available that it came into widespread
use. The principles of the Rb–Sr system and its application
to geologic materials are discussed by several authors (e.g.,
Faure, 1998; Dickin, 2005; see entry “Rb–Sr Dating”) and
are not reiterated here. However, it is important to under-
stand that Rb–Sr ages are based on the isochron method
in which several aliquots, usually separated mineral frac-
tions, of a sample are analyzed for 87Rb/86Sr and 87Sr/86Sr
ratios and then plotted on an isochron diagram. The slope
of a regression line through the data defines the age of the
sample, and the y intercept defines the initial Sr isotopic
composition of the fractions at the time they formed from
a common source. Uncertainty in the ages is based primar-
ily on the scatter of the data points on the isochron diagram.
Ages determined from this method are only valid if the ali-
quots that define the isochron regression were in isotopic
equilibrium when they formed. Thus, the age records the
time elapsed since the aliquots shared a common Sr isotopic
composition. This is usually considered to be when the
minerals crystallized from a magma.
In order to obtain an age from the slope of the isochron,
the 87Rb decay constant must be known. Unfortunately,
the decay of 87Rb to 87Sr is a low-energy transformation
of only 275 keV, involving the emission of
a beta particle and an antineutrino (3787)Rb ! (3887)
Rb + b + v
+ Q, and is therefore difficult to measure
accurately. As a result, meteorite Rb–Sr ages
have been calculated using several decay constants rang-
ing from 1.39  10
11 year
1 to 1.42  10
11 year
1
(Aldrich et al., 1956; Steiger and Jäger, 1977). Ages
reported in the literature must therefore be adjusted to
a common decay constant for comparison. After
a detailed analysis of published 87Rb decay constants,
Begemann et al. (2001) suggested using the value of
1.402  10
11 year
1 determined by Minster et al. (1982)
and Shih et al. (1985). This is the value most commonly
used for meteorite and lunar chronology studies today.
Another factor that must be considered when examining
Rb–Sr data in the literature is the range of 87Sr/86Sr values
obtained on standards. This is complicated by the fact that
early investigations used different strontium isotopic stan-
dards, such as seawater or plagioclase separated from
meteorites. In the mid-1970s, the NBS-987 standard
became widely used, and individual laboratories often
normalized their data to the certified 87Sr/86Sr ratio of
0.71014. More recently, the consensus of the Rb–Sr iso-
tope community is that 87Sr/86Sr value of this standard is
0.71025. In this discussion, all 87Sr/86Sr data are
renormalized to this 87Sr/86Sr value, and Rb–Sr ages are
recalculated using an 87Rb decay constant of
1.402  1011 year
1.
Chondrites, angrites, and eucrites
Ideally the slope of a regression line constructed to pass
through a suite of whole rock and/or mineral fractions,
and typically shown on a Rb–Sr isochron diagram, is used
to constrain the age of a sample. In some cases, ancient
Rb–Sr ages have been obtained for individual basaltic
achondrites, such as Ibitira and Juvinas (Allègre
et al., 1975; Birck and Allègre, 1978), which record the
crystallization age of the sample. However, many meteor-
ite samples either have limited internal variation in
87
Rb/86Sr or demonstrate evidence for significant distur-
bance of the Rb–Sr system by secondary processes occur-
ring on their parent bodies or on Earth. To address these
 METEORITES, RUBIDIUM–STRONTIUM, AND SAMARIUM–NEODYMIUM CHRONOLOGY 563

0.7008
Achondrite Whole Rocks
0.7006 Age = 4.35 ± 0.24 Ga
Stannern
Initial 87Sr/ 86Sr = 0.699099 ± 0.000044
0.7004 MSWD = 1.2

0.7002
Jonzac
0.7000
Nuevo Laredo
87Sr / 86Sr

0.6998

0.6996 Sioux County

Moore Pasamonte
0.6994 County
Juvinas
0.6992
Basaltic Achondrite Best Initial (BABI)
0.6990
Angra dos Reis (ADOR)
0.6988
Allende CAI (ALL)
0.6986
0.000 0.005 0.010 0.015 0.020 0.025 0.030
87Rb/ 86Sr

Meteorites, Rubidium–Strontium, and Samarium–Neodymium Chronology, Figure 1 A Rb-Sr isochron diagram of achondrite
meteorite whole rock analyses replotted from data of Papanastassiou and Wasserburg (1969).

Meteorites, Rubidium–Strontium, and Samarium–Neodymium Chronology, Table 1 Initial Sr isotopic compositions of ancient
samples

Sample Measured initial87Sr/86Sr Normalized initial87Sr/86Sra Reference

Allende CAI D7 (
ALL) 0.69876  0.00004 0.69887  0.00004 Gray et al. (1973)
Allende CAI N17 0.69861  0.00004 0.69872  0.00004 Tatsumoto et al. (1976)
Allende CAI 3529-30 0.69885  0.00002 0.69891  0.00002 Podosek et al. (1991)
Allende CAI Al3S4 0.69894  0.00001 0.69894  0.00001 Marks et al. (2013)
Chondritic meteorites 0.69885  0.00001 0.69895  0.00001 Minster et al. (1982)
Angra dos Reis (
ADOR) 0.69884  0.00004 0.69895  0.00004 Papanastassiou (1970)
Angra dos Reis 0.69897  0.00002 0.69897  0.00002 Nyquist et al. (1994)
LEW 86010 0.69897  0.00001 0.69897  0.00001 Nyquist et al. (1994)
Basaltic Achondrites (
BABI) 0.69899  0.00004 0.69910  0.00004 Papanastassiou and Wasserburg (1969)
Basaltic Achondrites 0.69899  0.00004 0.69909  0.00004 Birck and Allègre (1978)
a
Data normalized to NBS-987 ¼ 0.71025. Initial isotopic compositions calculated from isochron or least radiogenic sample using of
l87Rb ¼ 1.402  1011 year
1. More significant figures are often available from original references

problems, ages of individual samples have been calculated
from an assumed initial 87Sr/86Sr ratio. In one of the first
investigations of meteorite samples, Papanastassiou and
Wasserburg (1969) measured Rb–Sr in a group of basaltic
achondrites obtaining an age of 4.35  0.24 billion years
(Ga) and an initial 87Sr/86Sr of 0.699099  0.000044
(Figure 1). This is a whole rock isochron and is based on
the premise that all the analyzed meteorites are derived
from a common source region at the same time. As such,
the initial Sr isotopic composition determined from the
isochron records the value of the source region. This initial
value is known as BABI (Basaltic Achondrite Best Initial)
and is used as a reference point for model ages calculated
from the Rb–Sr isotopic compositions of other meteoritic
samples. BABI model ages have proven invaluable in
constraining the relative age of the earliest meteorite sam-
ples (e.g., Papanastassiou, 1970) due to the disturbed
nature of their Rb–Sr isotopic systematics.
A summary of initial 87Sr/86Sr values from the most
ancient meteorite samples is presented in Table 1. From
this table, it is apparent that BABI is not the least radio-
genic Sr isotopic composition measured in the Solar Sys-
tem. This reflects the fact that other meteorite samples,
such as calcium-aluminum-rich inclusions (CAIs) from
564 METEORITES, RUBIDIUM–STRONTIUM, AND SAMARIUM–NEODYMIUM CHRONOLOGY
carbonaceous chondritic meteorites and angrite meteor-
ites, are older than the basaltic achondrites. The initial Sr
isotopic compositions measured by Gray et al. (1973) of
these samples were given acronyms as follows: Allende
(ALL) and Angra dos Reis (ADOR). Subsequent mea-
surements on the same sample types are often slightly dif-
ferent than the original values and are listed separately in
Table 1.
Martian meteorites
Whereas obtaining internal Rb–Sr isochron ages for the
most ancient meteorites has often proven to be
a difficult task, defining Rb–Sr ages for young meteorites
from differentiated bodies, such as Mars and the Moon,
has been less difficult. The first unequivocal young Rb–
Sr date was determined for the meteorite Nakhla, which
belongs to the SNC (shergottite-nakhlite-chassigny)
group of meteorites. Ages of 1.30  0.02 Ga and
1.23  0.01 Ga were determined by Papanastassiou and
Wasserburg (1974) and Gale et al. (1975). Young ages
of 163–178 million years (Ma) were later obtained on
the related Shergotty and Zagami meteorites by Nyquist
et al. (1979a) and Shih et al. (1982). Initially the young
ages obtained on the SNC meteorites were interpreted
to reflect isotopic re-equilibrium associated with exten-
sive metamorphism of their parent body. The extent of
metamorphism required to reset the Rb–Sr system led
Nyquist et al. (1979b) to conclude that the parent body
was very large and likely to be the planet Mars. Later
petrologic work (Jones 1986) and isotopic analysis
involving the Sm–Nd chronometer (Borg et al., 1997)
demonstrated that the ages most likely recorded
a crystallization event rather than a metamorphic event.
Numerous Rb–Sr (and Sm–Nd) ages have now been
obtained of the SNC suite of Martian meteorites. The
shergottites define 4 general age groups of ~170 Ma,
~330 Ma, ~475 Ma, and ~575 Ma (see summaries by
Nyquist et al., 2001a; Borg and Drake, 2005). Several
additional nakhlites have also been dated at 1.3 Ga
(e.g., Shih et al., 1999). An age of 2,089  81 Ma has
been reported for NWA7034 meteorite (Agee
et al., 2013). This meteorite is a breccia, and mineral frac-
tions were derived from the bulk sample. Implicit in this
age determination is the assumption that all of the breccia
fragments and matrix minerals crystallized from the same
magma at the same time. Finally a Rb–Sr age of
3.90  0.04 Ga was determined on secondary carbonates
handpicked from the orthopyroxenite ALH84001 (Borg
et al., 1999).
Initial 87Sr/86Sr values have been determined from the
Rb–Sr isochrons and demonstrate a huge range from
~0.701 (e.g., Borg et al., 1997) to 0.723 (e.g., Nyquist
et al., 1979a). The initial 87Sr/86Sr ratios, along with initial
143
Nd/144Nd ratios, indicate that the shergottites are derived
from source regions that have elemental abundances that
range from incompatible-element depleted to
incompatible-element enriched. The shergottites have
consequently been subdivided on this basis (Symes
et al., 2008). A sampling of Rb–Sr ages determined for
the various types of Martian meteorites is presented in
Table 2, with an emphasis on the first age determinations
for each meteorite.
Lunar meteorites
Several relatively young ages have also been determined
for lunar meteorites. For the most part, however, Rb–Sr
ages have been confined to basaltic samples. Like high-
land rocks returned by the Apollo missions, the Rb–Sr iso-
topic systematics of highland clasts present in brecciated
meteorites appear to be disturbed by metamorphic pro-
cesses associated with impact. A good example of such
disturbance is in anorthositic meteorite Y86032 that yields
a Rb–Sr age of 4.62  0.89 Ga (Nyquist et al., 2006).
Metamorphism appears to result in the mobilization, and
loss, of Rb from the mafic mineral fractions. Thus, Rb–Sr
isochrons are generally disturbed to the older side of the
age spectrum.
A few basaltic lunar meteorites have been dated using
the Rb–Sr chronometer. An age of 2,990  18 Ma was
reported for the incompatible-element-enriched LaPaz
02205 meteorite (Rankenburg et al., 2006), whereas low
Ti-basaltic meteorites NWA 032 and NWA 4734 yielded
ages of 2,947  16 Ma (Borg et al., 2009) and
3,083  42 Ma (Elardo et al., 2014), respectively. Many
of the lunar meteorites were found in desert environments
and consequently have Rb–Sr systematics that have been
disturbed by terrestrial weathering. The Rb–Sr systemat-
ics of NWA4734, for example, is based on a two-point
tie line. Luckily this age was confirmed by a more robust
Sm–Nd isochron that was derived from additional mineral
fractions. Ages determined on lunar basaltic meteorites
demonstrate that these samples have crystallization ages
that are younger than any samples returned by the Apollo
and Luna missions. In fact, the meteorites are the youngest
lunar samples known (Borg et al., 2004). This is consistent
with the hypothesis that they are derived from different
localities on the Moon.
Sm–Nd chronology of meteorites
Introduction
Samarium-neodymium ages of meteorites are derived from
two independent parent-daughter isotope pairs. The first
decay reaction is 147 62 Sm ! 60 Nd + a + Q. The parent
143
147
isotope Sm has a well-defined half-life of 106 billion
years corresponding to a 147Sm decay constant of
6.54  10
12 year
1. Like the 87Rb–87Sr decay system
(see entry “Rb–Sr Dating”), this system is based on the iso-
chron method in which separated mineral fractions of an
individual samples are analyzed for147Sm/144Nd and
143
Nd/144Nd and plotted on an isochron diagram (Faure,
1998; Dickin, 2005; see entry “Sm–Nd Dating”). Ages
determined from this method are only valid if the aliquots
that define the isochron regression were in isotopic equilib-
rium. It should be noted that different laboratories have
 METEORITES, RUBIDIUM–STRONTIUM, AND SAMARIUM–NEODYMIUM CHRONOLOGY 565

Meteorites, Rubidium–Strontium, and Samarium–Neodymium Chronology, Table 2 Representative ages and initial Sr isotopic
compositions of Martian meteorites

Sample Rb–Sr age Initial87Sr/86Sr Reference

Zagami (E) 171  8 Ma 0.721566  0.000082 Shih et al. (1982)

EET79001B (I) 172  18 Ma 0.71217  0.00003 Nyquist et al. (1986)
ALH77005 (I) 187  12 Ma 0.71037  0.00005 Shih et al. (1982)
QUE94201 (D) 327  12 Ma 0.701298  0.000014 Borg et al. (1997)
Nakhla 1.30  0.02 Ga 0.70232  0.00006 Papanastassiou and Wasserburg (1974)
Chassigny 1.22  0.01 Ga 0.70253  0.00004 Nakamura et al. (1982)

Letters after shergottite names refer to incompatible-element characteristics of their source regions (E) ¼ enriched, (D) ¼ depleted, and
(I) ¼ intermediate

used different Nd isotopic ratios, including 142Nd/144Nd, Co-analysis of Sm–Nd and Rb–Sr has demonstrated that
146
Nd/144Nd, and 148Nd/144Nd, as well as different isotopic the Sm–Nd system is less easy to disturb by post-
compositions, to correct for instrument mass fractionation. crystallization metamorphism and secondary alteration
Caution must therefore be taken when comparing the initial than the Rb–Sr system. Thus, many of the most reliable
143
Nd/144Nd values reported from different laboratories. ages obtained on meteorites are currently derived from
The second decay reaction is 146 62 Sm ! 60 Nd + a + Q.
142
the Sm–Nd system.
146
The parent isotope, Sm, is extinct so this system can only
be used to date samples older than ~4.3 Ga. As a conse- Primitive achondrites
quence, this system is used almost exclusively to date mete-
Lugmair (1974) first proposed using the 147Sm– 143Nd
orite samples. Model ages are obtained by determining
system as a dating method and demonstrated its feasibility
the initial 146Sm/144Sm ratio of the sample at the time it
by obtaining an age of 4.56  0.08 Ga on the eucrite
formed from the slope of a 144Sm/144Nd-142Nd/144Nd iso-
Juvinas. This was followed by application of the Sm–Nd
chron plot. A model age is defined by the time required to
system to the angrite Angra dos Reis, yielding an age of
produce the initial146Sm/144Sm ratio of the sample starting
4.55  0.04 Ga (Lugmair and Marti, 1977), as well as to
from the initial146Sm/144Sm ratio Solar System at
lunar samples (e.g., Lugmair et al., 1976). Early in the
4,567 Ma. In turn, the initial146Sm/144Sm ratio of the Solar
development of the Sm–Nd chronometer, evidence for
System is calculated from the initial146Sm/144Sm ratio mea-
live146Sm was sought, but the results were equivocal
sured on a reference sample of known age. The reference
(Notsu and Mabuchi, 1973; Lugmair et al., 1975). The
samples used to determine146Sm
142Nd ages have tradi-
first clear evidence for live 146Sm came from analysis of
tionally been angrite meteorites (e.g., Nyquist
Angra dos Reis by Lugmair and Marti (1977). They
et al., 1994), although eucrites (Boyet et al., 2010) and CAIs
reported a 60 ppm difference between phosphate and
(Marks et al., 2014) have been proposed more recently. One
pyroxene mineral fractions and attributed the difference
obstacle to the application of the short-lived Sm–Nd system
to radiogenic decay of 146Sm. The 146Sm
142Nd and
to date meteorite samples has been defining the appropriate 147
Sm
143Nd isotopic systems have been used to date
half-life for 146Sm. Values of 68 Ma (Kinoshita et al., 2012)
many differentiated achondrite meteorites including
and 103 Ma (Friedman et al., 1966; Meissner et al., 1987)
angrites, eucrites, mesosiderites, a brachinite, and an
have both been used to calculate 146Sm–142Nd ages. Either
alcapulcoite. A forthcoming manuscript by Marks
half-life appears to reproduce Pb–Pb ages on the oldest
et al. (2014) provides a summary of some of these results.
samples, because the uncertainties on the 146Sm–142Nd
ages are often larger than the difference in ages calculated
using the 68 Ma and 103 Ma half-lives. The 103 Ma half- Martian and lunar meteorites
life appears to better reproduce the Pb–Pb and 147Sm– The Sm–Nd system has also been applied to the Martian
143
Nd ages of young lunar samples, however (Marks suite of meteorites that includes shergottites (basalts),
et al., 2014). nakhlites (clinopyroxenites), chassignites (dunites), as
The chemical purification procedures developed to sep- well as an orthopyroxenite (ALH84001). Application of
arate Sm and Nd from matrix start with the purification of the Sm–Nd chronometer to these samples is very difficult
Rb–Sr. As a consequence, most Sm–Nd isochrons because they have very low abundances of Sm and Nd,
obtained today on meteorite samples are accompanied by contain phosphate minerals that host the vast majority of
Rb–Sr data from the same fractions. The simultaneous the Sm and Nd in the rock, and often contain copious
analysis of Sm–Nd and Rb–Sr has made the interpretation amounts of impact melt. As a consequence, mineral frac-
of some ages significantly easier, because concordant tions derived from Martian meteorites have some of the
Sm–Nd and Rb–Sr ages typically indicate that these chro- lowest abundances of Nd of any planetary sample. Not
nometers record the timing of igneous crystallization. surprisingly, initial attempts to date Martian meteorites
566 METEORITES, RUBIDIUM–STRONTIUM, AND SAMARIUM–NEODYMIUM CHRONOLOGY
Meteorites, Rubidium–Strontium, and Samarium–Neodymium Chronology, Table 3 Representative ages and initial Nd isotopic
compositions of Martian meteorites
Sample Sm–Nd age
Zagami (E) 166  12 Ma
EET79001B (I) 169  23 Ma
ALH77005 (I) 173  6 Ma
QUE94201 (D) 327  19 Ma
Governador Valadares 1.37  0.02 Ga
Letters after shergottite names refer to incompatible-element characteristics of their source regions (E) ¼ enriched, (D) ¼ depleted, and
(I) ¼ intermediate
yielded highly disturbed Sm–Nd isochrons (e.g., Shih
et al., 1982) that seemed to confirm the original hypothesis
that these meteorites underwent extensive metamorphism.
Thus, young Rb–Sr and U–Th–Pb ages determined from
these meteorites were interpreted to record the timing of
this event (Papanastassiou and Wasserburg, 1974; Nyquist
et al., 1979a; Nyquist et al., 1979b; Chen and Wasserburg,
1986). However, Borg et al. (1997) obtained high purity
mineral fractions from QUE94201 that were devoid of
impact melt glass that yielded young concordant Rb–Sr
and Sm–Nd ages of ~330 Ma and suggested that the
young ages recorded the crystallization of the sample.
Concordant ages for numerous Martian meteorites have
now been obtained using Rb–Sr, Sm–Nd, U–Pb, and
Lu–Hf chronometers that support this contention.
Although most researchers consider these ages to record
igneous crystallization, some still support the older
hypothesis that the young ages reflect secondary resetting
of the isotopic systems by metamorphism (e.g., Bouvier
et al., 2009).
Summaries of Sm–Nd ages for Martian meteorites can
be found in Nyquist et al. (2001a) and Borg and Drake
(2005). Like the Rb–Sr system, the Sm–Nd system can
be used to define 4 general age groups for the shergottites
of ~170 Ma, ~330 Ma, ~475 Ma, and ~575 Ma. In addi-
tion, the initial eNd values (see entry “Model Ages
(Sm–Nd)”) determined from the Sm–Nd isochrons dem-
onstrate a huge range from
7 to +48 (e.g., Borg
et al., 1997, 2005). The initial eNd values, along with ini-
tial 87Sr/86Sr ratios, indicate that the shergottites are
derived from source regions that have elemental abun-
dances that range from incompatible-element depleted to
incompatible-element enriched. Samarium-neodymium
ages of ~1.3 Ga have been determined from the nakhlites
(e.g., Shih et al., 1999), and ages ranging from 4.0 to
4.5 Ga have been obtained for ALH84001 (Nyquist
et al., 1995a; Lapen et al., 2010). A sampling of Sm–Nd
ages determined for the various types of Martian meteor-
ites is presented in Table 3. In contrast to the Rb–Sr ages
presented in the Rb–Sr chronology of meteorites (see
entry “Rb–Sr Geochronology (Igneous Rocks)”), this
table emphasizes the more recent chronology for each
meteorite because the newer data is generally less
disturbed.
Initial eNd Reference

7.23  0.17 Borg et al. (2005)

16.9  1.4 Nyquist et al. (2001b)
+11.1  0.2 Borg et al. (2002)
+47.6  1.7 Borg et al. (1997)
+16.7  0.4 Shih et al. (1999)
Several lunar meteorites have been dated using the
Sm–Nd system. Unlike many Apollo samples, the
Sm–Nd isotopic systematics of most meteorites appear
to be minimally affected by the capture of thermal and
epithermal neutrons that are produced in the lunar subsur-
face by bombardment of galactic cosmic rays. The oldest
sample is an anorthositic clast from Y86032 which yields
an age of 4.43  0.03 Ga (Nyquist et al., 2006), making it
one of the oldest dated rocks from the Moon. Most other
Sm–Nd chronometry has been applied to lunar basaltic
meteorites. Samarium-neodymium ages of
2,931  92 Ma, 2,993  32, 3,106  44 Ma, and
2,922  85 Ma have been determined for NWA032,
NWA773 NWA 4734, and LAP02205 (Borg et al., 2004;
Rankenburg et al., 2006; Borg et al., 2009; Elardo
et al., 2014) that are in good agreement with Rb–Sr ages
determined on the same samples. This suite of basalts is
the youngest group of lunar rocks that have been dated
from the Moon so far and serves to demonstrate that there
is significantly more geologic variation on the Moon than
is represented in the Apollo (or Luna) sample collections.
Finally, the age of differentiation of both Moon and
Mars has been estimated from Sm–Nd whole rock iso-
chrons completed on lunar samples, including meteorites,
and Martian meteorites. The age of lunar differentiation is
determined using the Sm/Nd ratio estimated for basalt
magma sources from measured 143Nd/144Nd ratios and
142
Nd/144Nd ratios measured for the whole rocks. The
model age is calculated from the slope of a line regressed
through whole rock data on a 146Sm–142Nd isochron plot.
The basis of these model ages is the assumption that isoto-
pic variation observed in the basalt source regions reflects
fractionation of Sm from Nd associated with solidification
of the lunar magma ocean. Ages of 4,329+40/
56 Ma
(Nyquist et al., 1995a), 4,352+21/-23 Ma (Rankenburg
et al., 2006), 4,313+25/
30 (Boyet and Carlson, 2007),
4,340+20/-24 (Brandon et al., 2009), and 4,355+31/
39 Ma
(Gaffney and Borg, 2014) have been determined by these
studies. All ages are within uncertainty of the original
model age obtained by Nyquist et al. (1995a) and yield
a weighted average age of 4,341  21 Ma.
The model age of Martian differentiation has been esti-
mated using the slope of shergottite whole
rock142Nd/144Nd –143Nd/144Nd isotopic data to be
 METEORITES, RUBIDIUM–STRONTIUM, AND SAMARIUM–NEODYMIUM CHRONOLOGY
4,513+27/
33 Ma (Borg et al., 2003), 4,525+19/
21 Ma
(Foley et al., 2005), and 4,502  5 Ma (Symes et al., 2014).
It has also been determined from 146Sm–142Nd whole rock
isochron to be 4,527  18 Ma (Caro et al., 2008) and from
model age determined on individual samples to be
4,535  7 Ma (Debaille et al., 2007). All ages are within
uncertainty of the original model age determined on
QUE94201 of 4,525  25 Ma (Borg et al., 1997). The
weighted average of all these ages is 4,505  11 Ma, indi-
cating that the Moon differentiated significantly
after Mars.
Bibliography
Agee, C. B., Wilson, N. V., McCubbin, F., M., Ziegler, K., Polyak,
V. J., Sharp, Z. D., Asmerom, Y., Numm, M. H., Shaheen, R.,
Thiemens, M. H., Steele, A., Fogel, M. L., Bowden, R.,
Glamoclija, M., Zhang, Z., and Elardo, S. M., 2013. Unique
meteorite from early Amazonian Mars: Water-rich basaltic brec-
cias Northwest Africa 7013. Science, 339, 780–785.
Aldrich, L. T., Wetherill, G. W., Tilton, G. R., and Davis, G. L.,
1956. Half-life of Rb87. Physics Review, 103, 1045–1047.
Allègre, C. J., Birck, J. L., Fourcade, S., and Semet, M. P., 1975.
87Rb-87Sr age of Juvinas basaltic achondrite and early igneous
activity in the Solar System. Science, 187, 436–438.
Begemann, F., Ludwig, K. R., Lugmair, G. W., Min, K., Nyquist,
L. E., Patchett, P. J., Renee, P. R., Shih, C.-Y., Villa, I. M., and
Walker, R. J., 2001. Call for an improved set of decay constants
for geochronological use. Geochimica et Cosmochimica Acta,
65, 111–121.
Birck, J. L., and Allègre, C. J., 1978. Chronology and chemical his-
tory of the parent body of basaltic achondrites studied by the
87Rb-87Sr method. Earth and Planetary Science Letters,
39, 37–51.
Borg, L. E., and Drake, M. J., 2005. A Review of Meteorite Evi-
dence for the Timing of Magmatism and of Surface or Near-
Surface Liquid Water on Mars. Journal of Geophysical
Research, 110, E12SO3, doi:10.1029/2005JE002402.
Borg, L. E., Nyquist, L. E., Taylor, L. A., Wiesmann, H., and Shih,
C.-Y., 1997. Constraints on Martian Differentiation Processes
from Rb-Sr and Sm-Nd Isotopic Analyses of the Basaltic Sher-
gottite QUE 94201. Geochimica et Cosmochimica Acta,
61, 4915–4931.
Borg, L. E., Connelly, J. N., Nyquist, L. E., Shih, C.-Y., Wiesmann,
H., and Reese, Y., 1999. The age of carbonates in Martian mete-
orite ALH 84001. Science, 286, 90–94.
Borg, L. E., Nyquist, L. E., Wiesmann, H., and Reese, Y., 2002.
Constraints on the petrogenesis of Martian meteorites from
Rb-Sr and Sm-Nd isotopic systematics of the lherzolitic shergot-
tites ALH77005 and LEW88516. Geochimica et Cosmochimica
Acta, 66, 2037–2053.
Borg, L. E., Nyquist, L. E., Wiesmann, H., and Reese, Y., 2003. The
age of Dar al Gani 476 and the differentiation history of the Mar-
tian meteorites inferred from their radiogenic isotopic systemat-
ics. Geochimica et Cosmochimica Acta, 67, 3519–3536.
Borg, L. E., Shearer, C. K., Asmerom, Y., and Papike, J. J., 2004.
Evidence for prolonged KREEP magmatism on the Moon from
the youngest dated lunar igneous rock. Nature, 432, 209–211.
Borg, L. E., Edmunson, J. E., and Asmerom, Y., 2005. Constraints
on the U-Pb isotopic systematics of Mars inferred from
a combined U-Pb, Rb-Sr, and Sm-Nd isotopic study of the Mar-
tian meteorite Zagami. Geochimica et Cosmochimica Acta,
69, 5819–5830.
Borg, L. E., Gaffney, A. M., Shearer, C. K., DePaolo, D. J., Hutcheon,
I. D., Owens, T. L., Ramon, E., and Brennecka, G., 2009.
567
Mechanisms for incompatible-element enrichment on the
Moon deduced from the lunar basaltic meteorite Northwest Africa
032. Geochimica et Cosmochimica Acta, 73, 3963–3980.
Bouvier, A., Blichert-Toft, J., and Albarède, F., 2009. Martian mete-
orite chronology and the evolution of the interior of Mars. Earth
and Planetary Science Letters, 280, 285–295.
Boyet, M., and Carlson, R. W., 2007. A highly depleted Moon or a -
non-magma origin for the lunar crust? Earth and Planetary Sci-
ence Letters, 262, 505–516.
Boyet, M., Carlson, R. W., and Horan, M., 2010. Old Sm-Nd ages
for cumulate eucrites and redetermination of the solar system ini-
tial 146Sm/144Sm ratio. Earth and Planetary Science Letters,
291, 172–181.
Brandon, A. D., Lapen, T. L., Debaille, V., Beard, B. L.,
Rankenburg, K., and Neal, C. R., 2009. Re-evaluating
142
Nd/144Nd in lunar mare basalts with implications for the early
evolution and bulk Sm/Nd of the Moon. Geochimica et
Cosmochimica Acta, 73, 6421–6445.
Caro, G., Bourdon, B., Halliday, A. N., and Quitte, G., 2008. Super-
chondritic Sm/Nd ratios in Mars, the Earth and the Moon.
Nature, 452, 336–339.
Chen, J. H., and Wasserburg, G. J., 1986. Formation ages and evo-
lution of Shergotty and its parent planet from U-Th-Pb systemat-
ics. Geochimica et Cosmochimica Acta, 50, 955–968.
Debaille, V., Brandon, A. D., Yin, Q. Z., and Jacobsen, B., 2007.
Coupled 142Nd-143Nd evidence for a protracted magma ocean
in Mars. Nature, 450, 525–528.
Dickin, A. P., 2005. Radiogenic Isotope Geology. Cambridge: Cam-
bridge University Press, p. 492.
Elardo, S. M., Shearer, C. K., Fagan, A. L., Borg, L. E., Gaffney,
A. M., Burger, P. V., Neal, C. R., Fernandes, V. A., and
McCubbin, F. M., 2014. The origin of young Mare basalts
inferred from lunar meteorites Northwest Africa 4734,
032, and LaPaz Icefield 02205. Meteoritics & Planetary Science,
49, 261–291.
Faure, G., 1998. Principles and Applications of Geochemistry:
A Comprehensive Textbook for Geology Students. Upper Saddle
River, NJ: Prentice Hall, p. 600.
Foley, C. N., Wadhwa, M., Borg, L. E., Janney, P. E., Hines, R., and
Grove, T. L., 2005. The early differentiation history of Mars
from 182W-142Nd isotope systematics in the SNC meteorites.
Geochimica et Cosmochimica Acta, 69, 4557–4571.
Friedman, A. M., Milsted, J., Metta, D., Henderson, D., Lerner, J.,
Harkness, A. L., and Rokop, D. J., 1966. Alpha decay half-
lives of 148Gd, 150Gd and 146Sm. Radiochimica Acta, 5,
192–194.
Gaffney, A. M., and Borg, L. E., 2014 (in review). A young crystal-
lization age for the lunar magma ocean. Geochimica et
Cosmochimica Acta
Gale, N. H., Arden, J. W., and Hutchison, R., 1975. The chronology
of the Nakhla achondrite meteorite. Earth and Planetary Science
Letters, 26, 195–206.
Gray, C. M., Papanastassiou, D. A., and Wasserburg, G. J., 1973.
The identification of early condensates from the solar nebula.
Icarus, 20, 213–239.
Hahn, O., Strassman, F., Mattauch, J., and Ewald, H., 1943.
Geologische Altersbestimmungen mit der strontiummethode.
Chemiker Zeitung, 67, 55–56.
Jones, J. H., 1986. A Discussion of Isotopic Systematics and Min-
eral Zoning in the Shergottites: Evidence for a 180 Myr Igneous
Crystallization Age. Geochimica et Cosmochimica Acta, 50,
969–977.
Kinoshita, N., Paul, M., Kashiv, Y., Collon, P., Deibel, C. M.,
DiGiovine, B., Greene, J. P., Henderson, D. J., Jiang, C. L.,
Marley, S. T., Nakanishi, T, Pardo, R. C., Rehm, K. E.,
Roberston, D., Scott R., Schmitt C., Tang, X. D., Vondrasek,
R., and Yokoyama, A., 2012. A shorter 146Sm Half-life measured
568 METEORITES, RUBIDIUM–STRONTIUM, AND SAMARIUM–NEODYMIUM CHRONOLOGY
and implications for 146Sm-142Nd chronology in the solar sys-
tem. Science, 335, 1614–1617.
Lapen, T. J., Righter, M., Brandon, A. D., Debaille, V., Beard, B. L.,
Shafer, J. T., and Peslier, A. H., 2010. A younger age for
ALH84001 and its geochemical link to shergottite sources. Sci-
ence, 328, 347–351.
Lugmair, G. W., 1974. Sm-Nd ages: a new dating method
[Abstract]. Meteoritics, 9, 369.
Lugmair, G. W., and Marti, K., 1977. Sm-Nd-Pu timepieces in the
Angra dos reis meteorite. Earth and Planetary Science Letters,
35, 273–284.
Lugmair, G. W., Scheinin, N. B., and Marti, K., 1975. Search for
extinct 146Sm, 1. The isotopic abundance of 142Nd in the
Juvinas meteorite. Earth and Planetary Science Letters, 27,
79–84.
Lugmair, G. W., Marti, K., Kurtz, J. P., and Scheinin N. B., 1976.
History and genesis of lunar troctolite 76535. In Proceedings
of 7th Lunar Planetary Science Conference, Houston, TX,
pp. 2009–2033.
Marks, N. A., Borg, L. E., and Gaffney, A. M., 2013. Evidence for
young anorthositic magmatism on the Moon from Sm-Nd isoto-
pic measurements for ferroan anorthosite clast 3A from breccia
60016. In XLV Lunar and Planetary Science Conference, Hous-
ton, TX. Abstract #1129.
Marks, N., Borg, L. E., Hutcheon, I. D., Jacobsen, B., and Clayton
R. N., 2014 (in review). Samarium-neodymium chronology of
an Allende calcium-aluminum-rich inclusion with implications
for 146Sm isotopic evolution. Earth Planetary Science Letters
Meissner, F., Schmidt-Ott, W.-D., and Ziegeler, L., 1987. Half-life
a-ray energy of 146Sm*. Zeitschrift für Physik A Atomic Nuclei,
327, 171–174.
Minster, J.-F., Birck, J.-L., and Allègre, C. J., 1982. Absolute ages
of chondrites studied by the 87Rb–87Sr method. Nature, 300,
414–419.
Nakamura, N., Komi, H., and Kagami, H., 1982. Rb-Sr Isotopic and
REE Abundances in the Chassigny Meteorite. Meteoritics, 17,
257–258.
Notsu, K., and Mabuchi, H., 1973. Evidence of the extinct nuclide
146Sm in “Juvinas” achondrite. Earth and Planetary Science
Letters, 19, 29–36.
Nyquist, L. E., Wooden, J., Bansal, B., Wiesmann, H., McKay, G.,
and Bogard, D. D., 1979a. Rb-Sr age of the Shergotty Achon-
drite and Implications for Metamorphic Resetting of Isochron
Ages. Geochimica et Cosmochimica Acta, 43, 1057–1074.
Nyquist, L. E., Bogard, D. D., Wooden, J. L., Wiesmann, H., Shih,
C.-Y., Bansal, B. M., and McKay, G., 1979b. Early Differentia-
tion, Late Magmatism, and Recent Bombardment on the Sher-
gottite Parent Planet. Meteoritics, 14, 502.
Nyquist, L. E., Wiesmann, H., Shih, C. -Y., and Bansal, B. M., 1986.
Sr isotopic systematics of EETA79001. In Proceedings of 17th
Lunar Planetary Science Conference, Houston, TX,
pp. 624–625 (abstract).
Nyquist, L. E., Bansal, B., Wiesmann, H., and Shih, C.-Y., 1994.
Neodymium, strontium and chromium isotopic studies of the
LEW86010 and Angra dos Reis meteorites and the chronology
of the angrite parent body. Meteoritics & Planetary Science,
29, 872–885.
Nyquist, L. E., Bansal, B. M., Wiesmann, H., Shih, C. -Y., 1995.
“Martians” young and old: Zagami and ALH84001. In Proceed-
ings of 26th Lunar Planetary Science Conference, Houston, TX,
pp. 1065–1066 (abstract).
Nyquist, L. E., Wiesmann, H., Bansal, B. M., Shih, C.-Y., Keith,
J. E., and Harper, C. L., 1995b.146Sm
142Nd formation interval
for the lunar mantle. Geochimica et Cosmochimica Acta, 59,
2817–2837.
Nyquist, L. E., Bogard, D. D., Shih, C.-Y., Greshake, A.,
Stoffler, D., and Eugster, O., 2001a. Ages and geologic histories
of the Martian meteorites. In Kallenbach, R., Geiss, J., and
Hartmann, W. K. (eds.), Chronology and Evolution of Mars.
Dordrecht: Kluwer. Space Science Reviews, Vol. 96.
Nyquist, L. E., Reese, Y., Wiesmann, H., and Shih, C. -Y., 2001b.
Age of EET79001B and implications for Shergottite origins’.
In Proceedings of 32nd Lunar Planetary Science Conference,
abstract #1407 (CD-ROM), Houston, TX.
Nyquist, L. E., Bogard, D., Yamaguchi, A., Shih, C.-Y., Karouji,
Y., Ebihara, M., Reese, Y. D., Garrison, D., McKay, G., and
Takeda, H., 2006. Feldspathic clasts in Yamato-086032: rem-
nants of the lunar crust with implications for its formation
and impact history. Geochimica et Cosmochimica Acta, 70,
5990–6015.
Papanastassiou, D. A., 1970. The determination of small time differ-
ences in the formation of planetary objects. Thesis, California
Institute of Technology.
Papanastassiou, D. A., and Wasserburg, G. J., 1969. Initial stron-
tium isotope abundances and the resolution of small time differ-
ences in the formation of planetary objects. Earth and Planetary
Science Letters, 5, 361–376.
Papanastassiou, D. A., and Wasserburg, G. J., 1974. Evidence for
late formation and early metamorphism in the achondrite
Nakhlite. Geophysical Research Letters, 1, 23–26.
Podosek, F. A., Zinner, E. K., Macpherson, G. J., Lundberg, L. L.,
Brannon, J. C., and Fahey, A. J., 1991. Correlated study of
initial 87Sr/86Sr and Al-Mg isotopic systematics and petrologic
properties in a suite of refractory inclusions from the
Allende meteorite. Geochimica et Cosmochimica Acta, 55,
1083–1110.
Rankenburg, K., Brandon, A. D., and Neal, C. R., 2006. Neodym-
ium isotope evidence for a chondritic composition of the Moon.
Science, 312, 1369–1372.
Rankenburg, K., Brandon, A. D., and Norman, M. D. A., 2007. Rb–
Sr and Sm–Nd isotope geochronology and trace element study
of lunar meteorite LaPaz Icefield 02205. Geochim. Cosmochim
Acta, 71, 2120–2135.
Shih, C.-Y., Nyquist, L. E., Bogard, D. D., McKay, G. A.,
Wooden, J. L., Bansal, B. M., and Wiesmann, H., 1982.
Chronology and Petrogenesis of Young Achondrites,
Shergotty, Zagami, and ALHA 77005:Late Magmatism on a
Geologically Active Planet. Geochimica et Cosmochimica
Acta, 46, 2323–2344.
Shih, C.-Y., Nyquist, L. E., Bogard, D. D., Wooden, J. L., Bansal,
B. M., and Wiesmann, H., 1985. Chronology and petrogenesis
of a 1.8 g lunar granitic clast: 14321,1062. Geochimica et
Cosmochimica Acta, 49, 411–426.
Shih, C. Y., Nyquist, L. E., and Weismann, H., 1999. Samarium-
neodymium and rubidium-strontium systematics of nakhlite
Governador Valadares. Meteoritics & Planetary Science,
34, 647–655.
Steiger, R. H., and Jäger, E., 1977. Subcommission on geochronol-
ogy: convention on the use of decay constants in geo- and cos-
mochemistry. Earth and Planetary Science Letters,
36, 359–362.
Symes, S. J., Borg, L. E., Shearer, C. K., and Irving, A. J., 2008. The
age of the martian meteorite Northwest Africa 1195 and the dif-
ferentiation history of the shergottites. Geochimica et
Cosmochimica Acta, 72, 1696–1710.
Symes, S. J. K., Borg, L. E., and Brennecka, G. A., 2014. A young
differentiation age of Mars deduced from high-precision142Nd
analyses of Martian meteorites. In Proceedings of Lunar
Planetary Science Conference, abstract #1777 (CD-ROM),
Houston, TX.
Tatsumoto, M., Unru, D. M., and Desborough, G. A., 1976.
U-Th-Pb and Rb-Sr systematics of Allende and U-Th-Pb sys-
tematics of Orgueil. Geochimica et Cosmochimica Acta,
40, 617–634.
 MINERALS (40Ar–39Ar)
MINERALS (40Ar–39Ar)
Simon Kelley1, Clare Warren1 and Camilla Wilkinson2
1 what you’re dating in order to interpret the age. Mineral
Department of Earth and Environmental Sciences, The
Open University, Milton Keynes, UK
2 sections, often combined with elemental composition
Norges Geologiske Undersøkelse (NGU), Trondheim,
Norway
Definition
Dating potassium-bearing minerals using the 40Ar-39Ar
technique is the most popular approach to determining
the rates and timescales of geological processes across a
wide range of geological time and is applied to dating
igneous, sedimentary, and metamorphic rocks including
meteorites and Moon rocks. Dating minerals using the
Ar-Ar dating method is performed on purified mineral
separates, individual mineral grains, and in situ in thick
polished slabs.
Introduction
The earliest Ar-Ar dating studies were analyses of whole
rock samples to investigate the age of meteorites and the
first Moon rocks returned by the Apollo 11 astronauts
(e.g., Turner et al., 1966, 1971). Terrestrial studies
quickly applied the new dating techniques to mineral sep-
arates (e.g., Lanphere and Dalrymple, 1971). Whole rock
dates using the Ar-Ar technique are used predominantly
for volcanic rocks where they can distinguish between
instantaneous eruption ages and those disturbed by alter-
ation of additional excess argon. Whole rock Ar-Ar dates
of basalt eruptions are in fact dominated by potassium in
the glass phase (potassium is incompatible in the main
mineral phases pyroxene, olivine, and plagioclase).
Unfortunately glass is also commonly altered to clay
minerals, some of which yield ages reflecting the later
fluid flow events (e.g., illite) and some do not retain
argon quantitatively (e.g., smectite). Moreover argon is
more soluble in melt than minerals so glass can exhibit
anomalously old ages caused by excess argon. So even
in this scenario, understanding mineral Ar-Ar dates can
be important.
Mineral dates have been measured for both K-Ar and
Ar-Ar dating because of their versatility. For example,
muscovite mineral ages have been used to date high-level
igneous intrusions, metamorphic cooling, and sediment
provenance, and alkali feldspars have been used to date
mega-eruptions and metamorphic thermal histories.
Criteria for successful mineral Ar-Ar dating
The fundamental reason for mineral dating is that the
ages can be interpreted in terms of geological processes
including crystallization and cooling. While some ages
yield simple ages of instantaneous events, others yield
additional information on cooling patterns and fluid flow.
There are several important mineral- and rock-specific
parameters that should be considered in order to interpret
the ages.
569
Mineral characterization and analytical approach
Ensuring mineral separate purity is an essential prerequi-
site for mineral Ar-Ar dating; in short you have to know
identification is normally undertaken optically using thin
measured using electron microprobe. Several techniques
are commonly used to measure mineral Ar-Ar ages includ-
ing stepped heating mineral separates of 10–100 mg and
laser heating of mineral separates commonly less than
1 mg or as single grains (see “Single-Crystal Laser
Fusion”; McDougall and Harrison, 1999; Kelley, 2002a).
In addition, Ar-Ar age variations within individual min-
erals can be measured using laser ablation (pits
25–200 mm diameter) to extract argon from individual
laser pits in mineral grains in rock sections or separately
(see “Potassium–Argon (Argon–Argon), Structural Fab-
rics”; Kelley et al., 1994).
The potassium content and age of minerals are the two
key parameters in determining the analytical approach.
Higher potassium and older ages make analysis of smaller
samples possible. Such samples can usefully be analyzed
using UV laser to extract additional temporal and thermal
information (e.g., Flude et al., 2013). High-potassium
minerals are more commonly dated, but precise and geo-
logically meaningful ages can be measured on minerals
with low potassium content (e.g., Renne et al., 1997). In
contrast Ar-Ar mineral dating of young volcanic rocks
requires multiple grains analyzed as a pure mineral sepa-
rate (e.g., Ton-That et al., 2001). Greater care is essential
in dating all minerals since contamination is more likely
to cause problems of interpretation. For example, a young
volcanic rock can be dated using plagioclase, but it has
become a common practice to briefly etch the grains using
acid such as HF to remove any adhering glass or clay min-
erals because glass is more likely to contain high potas-
sium contents and excess argon (e.g., Esser et al., 1997),
and clay minerals may also contain significant potassium
and younger ages.
What do mineral ages mean – closure and other
models
The key property of mineral is that the ages they yield can
be interpreted as a single geological event or a process
such as cooling. Only in the case of geologically instanta-
neous volcanic eruption and cooling do all minerals in a
rock yield a single age. When rocks cool more slowly or
recrystallize, minerals yield different ages.
The capacity of mineral Ar-Ar dating to date a range of
different types of event is the result of differing argon
retention properties. Argon is a gas and diffuses through
minerals at high temperature, but diffusion rates slow
down at lower temperatures. The relationship between dif-
fusion rate and temperature is exponential, so small
changes in temperature cause rapid changes in diffusion
rate. Considered over geological time, this diffusion rate
change can approach a switch-like behavior and the point
570 MINERALS (40Ar–39Ar)
at which the Ar-Ar clock starts. Dodson (1973) formulated
a mathematical solution to this process which made sim-
plifying assumptions but proved to be very powerful tool
in quantifying link between thermal closure and age that
all later studies followed. The “closure temperature” con-
cept (Dodson, 1973) has been central to the popularity of
mineral Ar-Ar dating because it provides both time and
temperature simultaneously.
The early literature on closure temperatures is charac-
terized by empirical closure temperatures for minerals;
for example, biotite closure was considered to be around
300  C, and most mineral closure temperatures are in
geological useful range from around 150  C to 500  C.
However, reliable experimental argon diffusion data is
now available for all common potassium-bearing min-
erals, so mineral closure should always be calculated
using the mineral-specific diffusion parameters (e.g.,
Baxter, 2010), estimated grain size, and cooling rate.
While the Dodson (1973) formula was a good estimate
for simple cooling, many recent studies use a finite differ-
ence model (e.g., Wheeler, 1996) both for whole-grain
ages and within grain age variation. This type of model
is capable of simulating the coupled loss of argon with
varying temperature history, and coupled radiogenic
production of 40Ar, and has enabled a better understand-
ing of metamorphic ages (e.g., Warren et al., 2012)
including partial argon retention of Ar-Ar ages in high-
pressure micas cycled through rapid orogenic burial and
exhumation.
Another extension of closure temperature is the use of
diffusion data derived from stepped or cycled heating of
complex alkali feldspar mineral grains from granites and
basement metamorphic rocks (e.g., Harrison and Lovera,
2013). This approach has produced thermal histories from
many individual mineral samples, although the interpreta-
tion has sometimes been controversial (e.g., Parsons
et al., 1999).
Not all mineral Ar-Ar ages can be interpreted in terms
of closure due to slowing argon diffusion with tempera-
ture. In conditions close to the closure temperature, argon
loss can also be controlled by deformation, grain-size
reduction, and recrystallization (e.g., see “Potassium–
Argon (Argon–Argon), Structural Fabrics”). Fluids play
a key role in recrystallization and mineral growth and
some metamorphic mineral Ar-Ar ages, particularly
amphiboles (e.g., Villa et al., 2000) and low-temperature
K-feldspar (e.g., Mark et al., 2007).
Argon solubility, excess argon, and mineral Ar-Ar
ages
One of the basic assumptions of Ar-Ar dating is that the
mineral contains negligible amounts of 40Ar at the point
of closure. However, this assumption is “at odds” with
experimental noble gas research which indicates that par-
tition coefficients are not zero (e.g., Chamorro et al.,
2002). Non-zero argon partition coefficients in turn imply
that all minerals that reach equilibrium with silicate melt
or hydrous liquid acquire low concentrations of non-atmo-
spheric argon contamination. This conclusion appears to
be a fundamental challenge to Ar-Ar mineral dating suc-
cess, and yet most measurements yield geologically mean-
ingful ages – so why is this?
There are some geological circumstances where Ar-Ar
dating works less well and minerals yield anomalous
ages caused by high concentrations of radiogenic argon
not related to in situ potassium decay (Note that
40
Ar/36Ar ratios of metamorphic fluids are never the
same as the Earth’s atmosphere and commonly very
high). There are two scenarios in which this can affect
the mineral Ar-Ar age, which can be described as fluid
present or open system and fluid absent or closed system
(Kelley, 2002b). In the fluid present system, high concen-
trations of excess argon in the fluid (melt or hydrous) lead
to partition into the mineral or incorporation in a crystal-
lizing mineral. In the fluid absent case, the lack of any
fluid means that the minerals become the main reservoir
for argon despite the temperatures exceeding those
normally associated with closure (e.g., Kelley and
Wartho, 2000).
Recent work has helped to characterize these scenar-
ios since argon solubility has been measured experimen-
tally in hydrous liquids including sea water (summarized
in Kelley, 2002), and in silicate melts at high temperature
(e.g., Carroll and Stolper, 1993), and is in the range
~10–100 ppm for hydrous liquids and ~0.1–1 ppm for
silicate melts. Experimental data has now revealed a
reproducible pattern of solubility (Brooker et al., 2003;
Heber et al., 2007) and discovered significant variations
between mineral types (e.g., Jackson et al., 2013).
Excess argon should thus be understood in terms of
variable solubility and partition, in the same way as other
trace elements are understood to partition between
minerals.
Minerals used for Ar-Ar dating
Minerals of several classes are used for mineral Ar-Ar dat-
ing, some more commonly than others. Orthosilicates
have rarely been used for Ar-Ar dating and had a reputa-
tion for excess argon, but it seems this was ill founded
and mainly the result of their low potassium contents
(100 s of ppm). There are several records of successful
pyroxene dates in meteorites and recent work has shown
that they can yield useful ages and argon diffusion rates
in terrestrial rocks (Cassata et al., 2011). Framework sili-
cates are very commonly dated, particularly alkali feld-
spars in igneous rocks where sanidine has yielded some
of the youngest reliable Ar-Ar ages including the eruption
of Vesuvius in AD 79 (Renne et al., 1997). Alkali
feldspars have also been used in granites and basement
rocks to date cooling and yield thermal histories (e.g., Har-
rison and Lovera, 2013). Plagioclase is also a very com-
monly dated mineral, although the low potassium
contents (0.05–1.0 wt%) make sample purity essential.
Feldspathoids (up to 14 wt%) and anorthoclase (3–4 wt
 MINERALS (40Ar–39Ar)
%) also yield reproducible ages but are less common.
Chain silicates (amphiboles) are commonly dated,
although they generally have lower potassium contents
(generally 0.05–1.5 wt%). Sheet silicates including bio-
tite, phlogopite, muscovite, and phengite (6–9 wt%) are
all common mineral targets for Ar-Ar dating. Finally, clay
minerals have also been dated using the Ar-Ar technique
and illite is perhaps the most commonly used since it has
similar potassium content to muscovite. Clay mineral
dating using the Ar-Ar technique involves particular
challenges caused by the recoil of 39Ar during the irradia-
tion process, although this has been successfully over-
come by encapsulating pure clay mineral separates (see
“Ar–Ar and K–Ar Dating” and “Clays and Glauconites
(K–Ar/Ar–Ar)”).
Interpretation of mineral Ar-Ar dates
Mineral ages from igneous rocks
Mineral ages from lavas and volcanic deposits (i.e., tuff and
tephra layers) provide chronostratigraphic markers that rep-
resent geologically instantaneous events, which are rou-
tinely used to calibrate the geological timescale (e.g.,
Kuiper et al., 2008; Rivera et al., 2011; Storey
et al., 2012) and date paleomagnetic events, such as polarity
reversals (e.g., Deino et al., 2002). Mineral ages are also
used to assess the impact of explosive volcanic eruptions
(e.g., Ton-That et al., 2001), correlate stratigraphy and bio-
stratigraphy over large areas, and place an age upon and
hominid activities (e.g., Deino and Potts, 1992; Storey
et al., 2012). High-temperature alkali feldspars, in particular
sanidine, appear to be less susceptible to both xenocrystic
contamination and excess argon and are readily used in
Ar-Ar geochronology to determine the timing of volcanic
eruption. Plagioclase is commonly used for dating basaltic
igneous rocks where it is often the only K-bearing phase
available (e.g., Renne et al., 1995), although their low
K contents do mean that the incorporation of inherited
and/or excess 40Ar can cause significant increases to
Ar-Ar apparent age determinations. Plutonic rocks are also
dated using mineral Ar-Ar dates most commonly using
micas or amphiboles since the lower-temperature feldspars
yield cooling rather than crystallization ages, but even these
minerals can yield ages significantly younger than U-Pb
ages on minerals from the same intrusion.
Mineral ages from metamorphic rocks
Mineral Ar-Ar ages from metamorphic rocks provide
key information on both time and temperature during
cooling or crystallization as discussed above. Sequences
of mineral ages such as muscovite, biotite, and alkali
feldspar can yield extended portions of thermal history
in some cases. However mineral Ar-Ar dates can also
be complicated by processes that alter the Ar concentra-
tion both during and after crystallization (see “Metamor-
phic Terranes (K–Ar/40Ar/39Ar)”). Measured mineral
Ar-Ar ages may therefore relate to the timing of mineral
crystallization and recrystallization, cooling following
571
crystallization, and deformation with associated recrys-
tallization. Careful petrographical observations allow
the relative timing of (re)crystallization and/or deforma-
tion and the pressure-temperature history following
crystallization to be estimated. Geochronological data
from other decay systems, such as U-Pb where “age”
and metamorphic “stage” can clearly be linked, are use-
ful for providing a solid framework in which to interpret
the Ar-Ar data. In situ laser ablation methods allow
Ar-Ar ages within grains to be measured to determine
whether ages may be linked to changes in chemical
composition, textural position, grain size, or regions of
grain deformation (see “Potassium–Argon (Argon–
Argon), Structural Fabrics”). Numerical modeling pro-
vides constraints on the expected ages for different grain
sizes at different cooling/decompression rates in cases
where an open grain boundary network may be assumed
(e.g., Warren et al., 2012). In all cases the relative petro-
graphical framework and geochronological constraints
from other systems can be used to determine whether
mineral grains degassed efficiently at high temperatures
or whether they might contain additional or “excess Ar”
sourced from a precursor K-bearing mineral, a neighbor-
ing K-bearing mineral in a closed system where
fluid absence prevented its escape, or alternatively argon
in a concentrated fluid in which the mineral was
crystallizing.
Sedimentary rocks
Mineral Ar-Ar ages provide two important types of infor-
mation, detrital mineral ages from clastic rocks (often
muscovite but also occasionally alkali feldspar and bio-
tite) and ages for newly grown low-temperature minerals
including clays (see “Clays and Glauconites (K–Ar/Ar–
Ar)”) and authigenic alkali feldspar. Ar-Ar ages for detrital
minerals became popular when techniques were devel-
oped for analyzing individual grains (e.g., Kelley and
Bluck, 1992). Bulk ages were only ever useful as general
indicators, whereas dating individual mineral grains
enabled comparison not only of the time lag between
cooling/erosion and deposition but also age variations
within individual samples. This capability has been
exploited to understand mountain building and erosion
processes including recycling (e.g., Sherlock, 2001;
Najman et al., 2001) and determining glacial source
regions (e.g., Hemming and Hajdas, 2003; Reynolds
et al., 2004). Mineral Ar-Ar ages for authigenic minerals
are less common mainly because of the issues of obtaining
pure separates and recoil effects upon the submicron
grains (see “Clays and Glauconites (K–Ar/Ar–Ar)”).
However other authigenic minerals are known to yield
meaningful ages. Alkali feldspar overgrowths have been
successfully dated using UV laser ablation (e.g., Mark
et al., 2007) and can indicate both authigenesis and multi-
ple fluid flow events (e.g., Sherlock et al., 2005). Gangue
minerals such as cryptomelane have also been success-
fully dated (Vasconcelos et al., 1992).
572 MINERALS (40Ar–39Ar)

Bibliography Kelley, S. P., 2002b. Excess argon in K-Ar and Ar-Ar geochronol-
ogy. Chemical Geology, 188, 1–22.
Baxter, E. F., 2010. Diffusion of noble gases in minerals. In Zhang,
Kelley, S. P., and Bluck, B. J., 1992. Laser 40Ar-39Ar ages for indi-
Y., and Cherniak, D. (eds.), Diffusion in Minerals and Melts.
vidual detrital muscovites in the Southern Uplands of Scotland,
Chantilly/Saint Louis: Mineralogical Society of America/Geo-
UK. Chemical Geology (Isotope Geoscience section),
chemical Society. Reviews in Mineralogy & Geochemistry,
101, 143–156.
Vol. 72, pp. 509–557.
Kelley, S. P., and Wartho, J. A., 2000. Rapid Kimberlite ascent and
Brooker, R. A., Du, Z., Blundy, J. D., Kelley, S. P., Allan, N. L.,
the significance of Ar-Ar ages in Xenolith Phlogopites. Science,
Wood, B. J., Chamorro, E. M., Wartho, J. A., and Purton, J. A.,
289, 609–611.
2003. The ‘zero charge’ partitioning behaviour of noble gases
Kelley, S. P., Arnaud, N. O., and Turner, S. P., 1994. High spatial
during mantle melting. Nature, 423, 738–741.
resolution 40Ar-39Ar investigations using an ultra-violet laser
Carroll, M. R., and Stolper, E. M., 1993. Noble gas solubilities in
probe extraction technique. Geochimica et Cosmochimica Acta,
melts and glasses: new experimental results for argon and the
58, 3519–3525.
relationship between solubility and ionic porosity. Geochimica
Kuiper, K. F., Deino, A., Hilgen, F. J., Krijgsman, W., Renne, P. R.,
Et Cosmochimica Acta, 57, 5039–5051.
and Wijbrans, J. R., 2008. Synchronizing rock clocks of earth
Cassata, W. S., Renne, P. R., and Shuster, D. L., 2011. Argon diffu-
history. Science, 320, 500–504.
sion in pyroxenes: implications for thermochronometry and
Lanphere, M. A., and Dalrymple, G. B., 1971. A test of the
mantle degassing. Earth and Planetary Science Letters, 40
Ar/39Ar age spectrum technique on some terrestrial materials.
304, 407–416.
Earth and Planetary Science Letters, 12, 359–372.
Chamorro, E. M., Brooker, R. A., Wartho, J. A., Wood, B. J., Kelley,
Lanphere, M. A., Champion, D. E., Christiansen, R. L.,
S. P., and Blundy, J. D., 2002. Ar and K partitioning between
Izett, G. A., and Obradovich, J. D., 2002. Revised ages for
clinopyroxene and silicate melt to 8 GPa. Geochimica Et
tuffs of the Yellowstone Plateau volcanic field: assignment
Cosmochimica Acta, 66, 507–519.
of the Huckleberry Ridge Tuff to a new geomagnetic
Deino, A., and Potts, R., 1992. Age-probability spectra for examina-
polarity event. Geological Society of America Bulletin,
tion of single-crystal 40Ar/39Ar dating results: examples from
114, 559–568.
Olorgesailie, Southern Kenya Rift. Quaternary International,
Mark, D. F., Parnell, J., Kelley, S. P., and Sherlock, S. C., 2007.
13(14), 47–53.
Resolution of regional fluid flow related to successive orogenic
Deino, A. L., Tauxe, L., Monaghan, M., and Hill, A., 2002.
40 events on the Laurentian margin. Geology, 35, 547–550.
Ar/39Ar geochronology and paleomagnetic stratigraphy of
McDougall, I., and Harrison, T. M., 1999. Geochronology and
the Lukeino and lower Chemeron Formations at Tabarin and
Thermochronology by the 40Ar/39Ar Method. New York: Oxford
Kapcheberek, Tugen Hills, Kenya. Journal of Human Evolution,
University Press.
42, 117–140.
Najman, Y., et al., 2001. Dating of the oldest continental sedi-
Dodson, M. H., 1973. Closure temperature in cooling geochrono-
ments from the Himalayan foreland basin. Nature, 410,
logical and petrological systems. Contributions to Mineralogy
194–197.
and Petrology, 40, 259–274.
Parsons, I., Brown, W. L., and Smith, J. V., 1999. 40Ar/39Ar
Dodson, M. H., 1986. Closure profiles in cooling systems. Mate-
thermochronology using alkali feldspars: real thermal history
rials Science Forum, 7, 145–154.
or mathematical mirage of microtexture? Contributions to Min-
Esser, R. P., McIntosh, W. C., Heizler, M. T., and Kyle, P. R., 1997.
eralogy and Petrology, 136, 92–110.
Excess argon in melt inclusions in zero-age anorthoclase
Pringle, M. S., McWilliams, M., Houghton, B. F., Lanphere, M. A.,
feldspar from Mt Erebus, Antarctica, as revealed by the
and Wilson, C. J. N., 1992. 40Ar/39Ar dating of quaternary feld-
40
Ar/39Ar method. Geochimica et Cosmochimica Acta, 61,
spar: examples from the Taupo Volcanic Zone, New Zealand.
3789–3801.
Geology, 20, 531–534.
Flude, S., Halton, A. M., Kelley, S. P., Sherlock, S. C., Schwanethal,
Renne, P. R., Zhang, Z. C., Richards, M. A., Black, M. T., and Basu,
J., and Wilkinson, C. M., 2013. Observation of centimetre-scale
A. R., 1995. Synchrony and causal relations between Permian-
argon diffusion in alkali feldspars: implications for 40Ar/39Ar
Triassic Boundary crises and Siberian Flood Volcanism. Science,
thermochronology. In advances in 40Ar/39Ar dating: from
269, 1413–1416.
archaeology to planetary sciences. Geological Society Special
Renne, P. R., et al., 1997. Ar-40/Ar-39 dating into the historical
Publication, 378, 265–275.
realm: calibration against Pliny the Younger. Science,
Harrison, T. M., and Lovera, O. M., 2013. The multi-diffusion
277, 1279–1280.
domain model: past, present and future. Geological Society Spe-
Reynolds, B. C., et al., 2004. Radiogenic isotope records of quater-
cial Publication, 378, 91–106.
nary glaciations: changes in the erosional source and weathering
Heber, V. S., et al., 2007. Crystal-melt partitioning of noble gases
processes. Geology, 32(10), 861–864.
(helium, neon, argon, krypton, and xenon) for olivine and
Rivera, T. A., Storey, M., Zeeden, C., Hilgen, F. J., and Kuiper, K.,
clinopyroxene. Geochimica Et Cosmochimica Acta, 71,
2011. A refined astronomically calibrated 40Ar/39Ar age for Fish
1041–1061.
Canyon sanidine. Earth and Planetary Science Letters,
Hemming, S. R., and Hajdas, I., 2003. Ice-rafted detritus evidence
311, 420–426.
from Ar-40/Ar-39 ages of individual hornblende grains for evo-
Semaw, S., Renne, P., Harris, J. W. K., Feibel, C. S., Bernor, R. L.,
lution of the eastern margin of the Laurentide ice sheet since
Fesseha, N., and Mowbray, K., 1997. 2.5-million-year-old stone
43 C-14 ky. Quaternary International, 99, 29–43.
tools from Gona, Ethiopia. Nature, 385, 333–336.
Jackson, C. R. M., et al., 2013. Noble gas transport into the mantle
Sherlock, S. C., 2001. Two-stage erosion and deposition in a conti-
facilitated by high solubility in amphibole. Nature Geoscience,
nental margin setting: a 40Ar/39Ar laserprobe study of offshore
6, 562–565.
detrital white micas in the Norwegian Sea. Journal of the
Kelley, S. P., 2002a. K-Ar and Ar-Ar dating. In Porcelli, D.,
Geological Society of London, 158, 793–800.
Ballentine, C. J., Wieler, R., and Ribbe, P. H. (eds.), Noble Gases
Sherlock, S. C., et al., 2005. A high resolution record of multiple
in Geochemistry and Cosmochemistry. Chantilly/Saint Louis:
diagenetic events: ultraviolet laser microprobe Ar/Ar analysis
Mineralogical Geochemical Society/Geochemical Society.
of zoned K-feldspar overgrowths. Earth and Planetary Science
Reviews in Mineralogy & Geochemistry, Vol. 47, pp. 785–818.
Letters, 238(3–4), 329–341.
 MODEL AGES (Sm–Nd)
Storey, M., Roberts, R. G., and Saidin, M., 2012. Astronomically
calibrated 40Ar/39Ar age of the Toba supereruption and global
synchronization of late quaternary records. Proceedings of the
National Academy of Sciences of the United States of America,
109, 18684–18688.
Ton-That, T., Singer, B. S., and Paterne, M., 2001. 40Ar/39Ar of lat-
est Pleistocene (41 ka) marine tephra in the Mediterranean Sea:
implications for global climate records. Earth and Planetary Sci-
ence Letters, 184, 645–658.
Turner, G., 1971. 40Ar-39Ar ages from the lunar Maria. Earth and
Planetary Science Letters, 11, 169–191.
Turner, G., et al., 1966. The thermal history of the Bruderheim mete-
orite. Earth and Planetary Science Letters, 1, 155–157.
Vasconcelos, P. M., Becker, T. A., Renne, P. R., and Brimhall, G. H.,
1992. Age and duration of weathering By K-40 Ar-40 and Ar-40/
Ar-39 analysis of potassium-manganese oxides. Science,
258, 451–455.
Villa, I. M., Hermann, J., Muntener, O., and Trommsdorff, V., 2000.
Ar-39-Ar-40 dating of multiply zoned amphibole generations
(Malenco, Italian Alps). Contributions to Mineralogy and
Petrology, 140, 363–381.
Warren, C. J., Hanke, F., and Kelley, S. P., 2012. When can musco-
vite Ar-40/Ar-39 dating constrain the timing of metamorphic
exhumation? Chemical Geology, 291, 79–86.
Wheeler, J., 1996. A program for simulating argon diffusion profiles
in minerals. Computers and Geosciences, 28, 919–929.
Cross-references
Ar–Ar and K–Ar Dating
Clays and Glauconites (K–Ar/Ar–Ar)
Metamorphic Terranes (K–Ar/40Ar/39Ar)
Model Ages (Sm–Nd)
Potassium–Argon (Argon–Argon), Structural Fabrics
Single-Crystal Laser Fusion
MODEL AGES (Sm–Nd)
Alan P. Dickin
ð143
School of Geography and Earth Sciences, McMaster
Nd Nd
University, Hamilton, ON, Canada


ð144 NdÞsample ð144 NdÞCHUR
1 þ 0 0
Definition
ð147 SmÞ ð147 SmÞ
A neodymium model age is an estimate of the age of
ð144 Nd Þsample ð144 NdÞCHUR
“crustal formation” based on the assumption of a model
composition for the mantle source. It is most commonly
used to date the ages of crustal terranes within Precam-
brian orogenic belts or to estimate the provenance age of
clastic sedimentary rocks.
Introduction
Nd model age dating is probably the most misunderstood
of the major geological dating methods. As the name indi-
cates, Nd model ages are dependant on a mantle model to
determine ages, and their accuracy is therefore dependant
on the suitability of the mantle model. Nd model ages
also have relatively large analytical uncertainties of
around +/
20 Myr. However, they are uniquely powerful
in determining crustal formation ages, representing the
time when crustal blocks were first created by mantle-
573
derived magmatism, most commonly above subduction
zones. In contrast, the U-Pb dating method provides much
more precise and accurate ages for igneous crystallization,
but can only provide a minimum age for crustal formation,
since it cannot date the time of crustal separation from the
mantle.
Historical development
Nd model ages were conceived by DePaolo and
Wasserburg (1976b), based on the observation by
DePaolo and Wasserburg (1976a) that the initial ratios of
terrestrial igneous rocks appeared to lie on the same Nd
isotope growth curve as chondritic meteorites. Hence,
they inferred that the mantle source of terrestrial igneous
rocks had a uniform “Bulk Earth” signature equivalent to
a Chondritic Uniform Reservoir (CHUR).
Since Nd and Sm are rare earth elements (REE) with
very similar chemistry, it was inferred that within mag-
matic systems, these elements only undergo significant
relative fractionation at low percentages of melting, such
as when magmas are first extracted from the mantle. Since
Nd is the more incompatible of the two elements, the pro-
cess of “crustal extraction” or “crustal formation” leads to
a markedly lower Sm/Nd ratio in crustal rocks relative to
the mantle. In contrast, subsequent intra-crustal differenti-
ation processes are not thought to cause significant further
Sm/Nd fractionation. This leads to the scenario shown in
Figure 1.
The determination of the Nd isotope ratio and Sm/Nd
ratio of a single crustal rock, presumed to have been an
effectively closed system since crustal extraction, allows
its Nd isotope growth line to be extrapolated back to its
intersection with the CHUR evolution line, so that the
point of intersection yields a TCHUR model age for the ana-
lyzed sample. Algebraically, the age is defined as follows:
Þ
0
ð143
Þ
0
T CHUR ¼ ln
l
The non-deviation of the crustal growth line through
metamorphic and sedimentation events in Figure 1 is
a schematic representation of the closed-system concept.
Empirical evidence suggests that given an adequate-sized
whole-rock sample of suitable chemistry, Sm-Nd system-
atics are resistant to high-grade metamorphism, even
under severe deformation (e.g., Barovich and Patchett,
1992). The Sm-Nd system may also remain unfractionated
through erosion-sedimentation events, although the integ-
rity of the system depends on the type and conditions of
sedimentation. Clastic sediments have been shown to give
good results, but chemical sediments have not (e.g., Bock
et al., 1994).
In their pioneering demonstration of the Nd model age
method, McCulloch and Wasserburg (1978) applied the
574 MODEL AGES (Sm–Nd)
Model Ages (Sm–Nd), Figure 1 Plot of Nd isotope evolution
against time, showing the concept of Nd model ages based on
the intersection of the isotopic growth line of the sample with
that of the Bulk Earth, based on a chondritic model. McCulloch
and Wasserburg (1978) and Dickin (2005).
model age dating technique to composite samples of the
Canadian Shield, in order to achieve maximum coverage
from a limited number of samples. They showed that
Archean crust is preserved in several Proterozoic orogenic
belts for which other dating systems (e.g., Rb-Sr and
K-Ar) had been reset.
Problems with the chondritic earth model appeared
when young oceanic volcanics were shown to deviate
from the Bulk Earth composition, reflecting sources in
the depleted upper mantle. It was realized that such
a depleted mantle was an inevitable result of crustal
extraction over earth history, leaving an incompatible-
depleted residue with a higher Sm/Nd ratio than Bulk
Earth (see vector in Figure 1).
To solve this problem, DePaolo (1981) derived an
empirical model for a “depleted mantle” reservoir, which
he believed would be typical of the asthenospheric “man-
tle wedge” that provides the source for subduction-related
magmatism. He fitted a quadratic equation to three points
to model this DM source: a chondritic source at 4.57 Ga,
the juvenile Paleoproterozoic Idaho Springs Formation,
and modern island arcs (Figure 2). Although the model
was ad hoc, the resulting TDM ages have often yielded sur-
prising accurate estimates of the crustal formation ages of
major crustal blocks, when compared with other geochro-
nological or geological evidence.
Controversies
Apart from the assumption of the mantle model and the
closed-system assumption, the other major assumption
Model Ages (Sm–Nd), Figure 2 Plot of Nd isotope ratio against
time using the epsilon notation, which normalizes Nd isotope
ratios relative to CHUR (epsilon Nd ¼ part per 10,000 deviation
from the CHUR line). The plot shows DePaolo’s quadratic curve
for depleted mantle evolution and the effects of subsequent
intra-crustal melting to form granite batholiths such as Boulder
Creek and Pikes Peak (After DePaolo (1981) and Dickin (2005)).
of Nd model ages is that the crustal sample in question
was extracted from the mantle during a single event. This
is a good assumption for the formation of geological prov-
inces such as the Superior Province of the Canadian
Shield, which largely consists of a collage of accreted oce-
anic arcs (Hoffman, 1989). However, continental margin
arcs such as the Andes (ensialic arcs) are notorious for
complex magmatic mixing of components with different
ages (e.g., Mamani et al., 2010). If such samples are used
to calculate Nd model ages, the result can be meaningless
mixed ages (Figure 3).
A warning against such mixed ages was given by Arndt
and Goldstein (1987), in their notorious paper “The use
and abuse of Nd crust formation ages.” The warning was
perhaps necessary, because in their haste to characterize
vast areas of continental crust with a very few Nd model
ages, authors such as Nelson and DePaolo (1985) had
cut corners and reached some erroneous conclusions. For
example, they characterized the crustal formation age of
the Penokean event as 2.3 Ga, based on a chance cluster-
ing of three Nd model ages with this value from Michigan.
Unfortunately, the geological community became so
obsessed by this warning that later estimates of crustal for-
mation age would be questioned (e.g., Hanmer
et al., 2000), even if based on much larger Nd data sets
yielding consistent results.
Some ground rules for future work
In their critique, Arndt and Goldstein (1987) argued that
Nd model ages should only be interpreted as true crustal
formation ages “if they coincide with U-Pb zircon ages
 MODEL AGES (Sm–Nd)
Model Ages (Sm–Nd), Figure 3 Nd isotope evolution diagram
using the epsilon notation, schematically showing the effects of
Nd mixing to generate a composite sample whose Nd model
age may not correspond with any real geological event (After
Arndt and Goldstein (1987) and Dickin (2005)).
or other independent evidence of an orogenic event.” This
is actually a misleading claim, which has led to additional
confusion. An orogenic event is most often a collisional
event and may not be a significant crust-forming event.
A good example is the Grenville Orogeny, during which
essentially no new crust was extracted from the mantle
(Dickin et al., 2010). In such circumstances, large num-
bers of U-Pb ages may be generated that significantly
post-date crustal formation (e.g., McNutt and Dickin,
2012).
Ideally, if Nd model ages are used to determine the
crustal formation age of accreted oceanic arc fragments
(perhaps in a Precambrian crustal terrane), we may hope
that U-Pb ages will be recovered that date igneous crystal-
lization within the original arc. In that case, Nd model ages
may coincide with U-Pb ages. However, even this sce-
nario is oversimplified. The typical granitoid rocks that
make up the bulk of the continental crust (tonalites and
granodiorites) cannot be direct products of the mantle.
When Nd model ages are determined on such material,
they are probably dating their mafic crustal precursors,
which may be millions of years older than the crystalliza-
tion age of the plutons. Since it may take 50 Myr to create
a mature island arc, we should expect that the Nd model
ages for juvenile terranes may be 50 Myr older that the
oldest U-Pb ages, which therefore provide only
a minimum age for crustal formation. Therefore, if TDM
model ages fall within 100 Myr or so of U-Pb zircon ages,
this provides good evidence that we have a juvenile arc
terrane (e.g., Daly and McLelland, 1991).
If Nd model ages are more than 100 Myr older than the
oldest U-Pb age, this may mean that an original accreted
arc terrane was subsequently involved in a continental
margin arc. Such arcs may generate pervasive magmatic
reworking that almost completely masks the older crust-
575
Model Ages (Sm–Nd), Figure 4 Nd isotope evolution diagrams
using the epsilon notation, showing data for the Lewisian
gneisses of NW Scotland. (a) initial ratios for three isochron sets,
(b) growth lines for individual samples comprising the tonalitic
gneiss isochron (After Whitehouse (1988) and Dickin (2005)).
forming event. A good example is the 1.65 Ga Labrador-
ian Orogeny of eastern Canada, which was partly
established on an older 1.9 Ga Makkovikian accreted mar-
gin. For many years, the only evidence for the older event
was the observation of ca. 2 Ga TDM model ages (Dickin,
2000). However, pre-Labradorian U-Pb ages eventually
confirmed the accuracy of the TDM ages as accurate esti-
mates of the crustal formation age of the crust (Gower
et al., 2008).
Nd model ages and isochron ages
Contrary to popular belief, Sm-Nd isochrons are another
valid way of testing Nd model ages. When a system is dis-
turbed or mixed, it is unlikely that model ages and iso-
chron ages will be affected in the same way. An
illustration of these effects during high-grade metamor-
phism is shown in Figure 4 (Whitehouse, 1988). In this
case, the Sm-Nd isochron age of Lewisian intermediate
and granitic gneisses from NW Scotland was reset by
high-grade metamorphism, causing the initial ratio of the
isochron to lie off the DM line (Figure 4a). However, the
model ages of these rocks still approximately indicate
576 MOLECULAR CLOCK CALIBRATION
their crustal formation age of 2.9 Ga (Figure 4b), consis-
tent with isochron ages for associated mafic gneisses.
An example where Nd model ages and isochron ages
are consistent is provided by Mesoproterozoic gray
gneisses from the Grenville Province of Quebec. A suite
of over 70 such gneisses from an area of
nearly 100,000 km2 gave a Sm-Nd isochron age of
1.51 +/
0.05 Ga and an average model age of 1.57 Ga
(Dickin et al., 2010). These two ages are essentially within
error and therefore validate the DM mantle model,
suggesting that the suite represents a juvenile arc terrane
extracted from the mantle over a few tens of millions of
years around 1.5 Ga. This is confirmed by published
U-Pb data within this terrane, yielding ages as old as
1.45 Ga (see references in Dickin, 2000).
Conclusions
Nd model ages are uniquely powerful in providing an esti-
mate of the formation age of Precambrian crustal terranes.
They rely on geological assumptions and can therefore
lead to erroneous ages if not correctly applied. However,
given adequate sample sizes to test for effects such as
mixing and disturbance, they can yield geologically accu-
rate estimates for the time of crustal extraction from the
mantle.
Bibliography
Arndt, N. T., and Goldstein, S. L., 1987. Use and abuse of crust-
formation ages. Geology, 15, 893–895.
Barovich, K. M., and Patchett, P. J., 1992. Behaviour of isotopic
systematics during deformation and metamorphism: a Hf, Nd
and Sr isotopic study of mylonitized granite. Contributions to
Mineralogy and Petrology, 109, 386–393.
Bock, B., McLennan, S. M., and Hanson, G. N., 1994. Rare earth
element redistribution and its effects on the neodymium isotope
system in the Austin Glen Member of the Normanskill Forma-
tion, New York, USA. Geochimica et Cosmochimica Acta,
58, 5245–5253.
Daly, J. S., and McLelland, J. M., 1991. Juvenile Middle Protero-
zoic crust in the Adirondack Highlands, Grenville province,
northeastern North America. Geology, 19, 119–122.
DePaolo, D. J., 1981. Neodymium isotopes in the Colorado Front
Range and implications for crust formation and mantle evolution
in the Proterozoic. Nature, 291, 193–197.
DePaolo, D. J., and Wasserburg, G. J., 1976a. Nd isotopic variations
and petrogenetic models. Geophysical Research Letters,
3, 249–252.
DePaolo, D. J., and Wasserburg, G. J., 1976b. Inferences about
magma sources and mantle structure from variations of
143
Nd/144Nd. Geophysical Research Letters, 3, 743–746.
Dickin, A. P., 2000. Crustal formation in the Grenville Province:
Nd-isotope evidence. Canadian Journal of Earth Sciences,
37, 165–181.
Dickin, A. P., 2005. Radiogenic Isotope Geology. Cambridge: Cam-
bridge University Press.
Dickin, A. P., McNutt, R. H., Martin, C., and Guo, A., 2010. The
extent of juvenile crust in the Grenville Province: Nd isotope evi-
dence. Geological Society of America Bulletin, 122, 870–883.
Gower, C. F., Kamo, S., Kwok, K., and Krogh, T. E., 2008. Protero-
zoic southward accretion and Grenville orogenesis in the interior
Grenville Province in eastern Labrador: evidence from U-Pb
geochronological investigations. Precambrian Research,
165, 61–95.
Hanmer, S., Corrigan, D., Pehrsson, S., and Nadeau, L., 2000. SW
Grenville Province, Canada: the case against post-1.4 Ga accre-
tionary tectonics. Tectonophysics, 319, 33–51.
Hoffman, P. F., 1989. Precambrian geology and tectonic history of
North America. In Bally, A. W., and Palmer, A. R. (eds.), The
Geology of North America. Boulder: Geological Society of
America, Vol. A, 447–512.
Mamani, M., Worner, G. W., and Sempere, T., 2010. Geochemical
variations in igneous rocks of the Central Andean orocline
(13S to 18S): tracing crustal thickening and magma generation
through space and time. Geological Society of America Bulletin,
122, 162–182.
McCulloch, M. T., and Wasserburg, G. J., 1978. Sm-Nd and Rb-Sr
chronology of continental crust formation. Science,
200, 1003–1011.
McNutt, R. H., and Dickin, A. P., 2012. A comparison of Nd model
ages and U-Pb zircon ages of Grenville granitoids: constraints on
the evolution of the Laurentian margin from 1.5 to 1.0 Ga. Terra
Nova, 24, 7–15.
Nelson, B. K., and DePaolo, D. J., 1985. Rapid production of con-
tinental crust 1.7 to 1.9 b.y. ago: Nd isotopic evidence from the
basement of the North American mid-continent. Geological
Society of America Bulletin, 96, 746–754.
Whitehouse, M. J., 1988. Granulite facies Nd-isotopic homogenisa-
tion in the Lewisian complex of northwest Scotland. Nature,
331, 705–707.
MOLECULAR CLOCK CALIBRATION
Rachel Warnock
School of Earth Sciences, University of Bristol,
Bristol, UK
Synonyms
Calibration prior; Fossil calibration; Molecular clock
constraint; Time prior
Definition
Molecular clock calibration. The temporal information
used to calibrate the molecular clock and estimate the
molecular substitution rate.
The molecular clock
Establishing an evolutionary timeline for the tree of life
(TOL) is essential for understanding the history of biodi-
versity. The incompleteness of the rock and fossil record
means this cannot be achieved using paleontological evi-
dence alone. The molecular clock provides the only viable
means of obtaining precise estimates of evolutionary time.
The mutations that arise along a gene sequence represent
copy errors that accumulate randomly over time. The dif-
ferences observed between the genes of two living species
are therefore a function of the time since they last shared
a common ancestor. If the age of the last common ancestor
can be determined from the fossil record for a pair of living
species, the clock can be calibrated and the rate at which
 MOLECULAR CLOCK CALIBRATION
mutations have accumulated can be calculated. The
molecular clock can then be used to infer the time of diver-
gence of other species pairs. In practice, different branches
of the TOL accumulate genetic changes at different rates,
and precise calibrations cannot be obtained from the fossil
record. Over the past 50 years, molecular clock methods
have evolved to become increasingly sophisticated.
Relaxed clock models can be used to accommodate rate
variation, the process of molecular evolution can now be
modeled more realistically, and these methods continue
to be developed (Sanderson, 1997; Thorne et al., 1998;
Sanderson, 2002; Drummond et al., 2006; Rannala and
Yang, 2007; Drummond and Suchard, 2010). There have
been parallel developments in the way in which the clock
is calibrated (Sanderson, 1997; Thorne et al., 1998; Drum-
mond et al., 2006; Yang and Rannala, 2006; Inoue et al.,
2010). Deriving paleontologically based calibrations,
and quantifying the associated uncertainty, requires
a careful evaluation of the way in which temporal informa-
tion can be extrapolated from the fossil record (Reisz and
Müller, 2004; Benton and Donoghue, 2007; Donoghue
and Benton, 2007; Parham et al., 2012).
Uncertainty in the fossil record
The fossil record is incomplete. Only a tiny fraction of all
life that has ever lived has been preserved in the rock
record, and some branches of the TOL are entirely absent
(Raup, 1972). However, the process of preservation and
discovery is far from random, and the quality of the fossil
record is immensely variable over geological time. There
are many variables which predicate the probability that
an organism will ever be preserved in the fossil record,
and further obstacles that determine the probability that
it will later be sampled and identified correctly by
paleontologists.
Fossils are the remains of once living organisms that
have been mineralized and preserved in the rock record.
Different anatomical and skeletal features have variable
fossilization potential, contingent on their chemical struc-
ture and the environment in which they degrade and
become deposited. An organism’s physiology, habitat
preferences, and life history all contribute to the likelihood
of preservation, and the probability of becoming fossilized
can alter over time, for example, by the acquisition or loss
of hard parts.
The rock record, in which the fossil record is embed-
ded, is also spatially and temporally biased (Smith,
2007), and there exists a strong, unresolved correlation
between the two (Smith and McGowan, 2007; Peters,
2005; Benton et al., 2011). To some extent the fossil
record must be subject to the artifacts of the rock record,
because the appearance of an organism in the fossil record
is contingent on the availability of rock units suitable for
preservation and subsequent sampling (Peters and Foote,
2001). Widespread heterogeneity in the deposition of sed-
iments means the fossil record is heterogeneous both tem-
porally and regionally. Older rocks are also more likely to
577
have been eroded or altered by temperature and pressure.
The way paleontologists sample the rock record is also
influenced by the exposure and access to fossil-bearing
rocks, and variable interest in different groups of fauna
and formations (Dunhill, 2012).
Most importantly for telling evolutionary time, the first
appearance of a species in the fossil record is not coinci-
dent with the time of origin, because the probability of
sampling the earliest member of a lineage is virtually zero
and because it will be indistinguishable from its direct
ancestor. The large number of variables that contribute to
the probability that an organism will be preserved and later
sampled creates huge variability in the extent of the gaps
in the fossil record across the TOL. This means different
fossils approximate divergence times to different degrees.
Understanding the incompleteness of the fossil record
forms an active and important branch of paleontology.
It is important to recognize the factors that both prohibit
and facilitate the process of fossilization as part of any
attempt to extrapolate information from the fossil record.
How we calibrate the molecular clock from such an
incomplete record is not obvious.
Minimum and maximum constraints
The fossil record can never provide a precise time of origin
for any species, but this is not equivalent to saying that the
fossil record cannot provide accurate estimates. The order
in which organisms first appear in the fossil record is
roughly congruent with the order in which the branches
of the TOL diverged genetically, meaning that the fossil
record captures reality on some level (Benton et al.,
2000). Using a critical approach, paleontological evidence
can instead be used to constrain divergence events by pro-
viding minimum (or maximum) ages that can be incorpo-
rated into the calibration of the molecular clock (Figure 1).
First appearances provide an obvious source for defining
minimum constraints, because they must postdate the time
of origin of all evolutionary lineages.
An individual fossil specimen can never provide more
than a minimum point of divergence, but maximum con-
straints are also essential for calibrating the molecular
clock (Sanderson, 1997; Kishino et al., 2001; Thorne
and Kishino, 2005; Yang and Rannala, 2006). The maxi-
mum has to define the point before which evidence sug-
gests the divergence event has not yet occurred. How far
back beyond the minimum constraint is the evolutionary
history of a clade likely to extend? The variable length
of gaps in the fossil record makes this extremely difficult
to estimate.
Maximum constraints can be established by taking into
account a wider suite of paleontological evidence. One
strategy is to use the fossil record of closely related evolu-
tionary branches (phylogenetic bracketing) (Reisz and
Müller, 2004; Müller and Reisz, 2005). The first appear-
ance of the descendants of the proceeding divergence
event may be useful for defining the maximum. This
requires an assessment of the quality of the records on
578 MOLECULAR CLOCK CALIBRATION

Molecular Clock Calibration, Figure 1 Tree illustrating the minimum and maximum divergence times among several model animal
species based on fossil evidence. As an example, the divergence event between human (Actinopterygii) and zebra fish (Sarcopterygii)
is highlighted. The minimum constraint (highlighted in white) is represented by one of the earliest known sarcopterygians, the
lungfish Diabolepis. Minimum constraints for all divergence events are shown in bold; maximum dates are shown below. These
examples are taken from Benton and Donoghue (2007).

either side of the divergence event, because this approach
relies on the assumption that there is perfect congruence
between the order of branching events and the sequence
in which lineages first appear in the fossil record. We
know that this assumption will be violated for some spe-
cies due to nonrandom preservation. An alternative, more
conservative approach can be adopted to define the maxi-
mum, by tracing the record back to a time period during
which conditions conducive to preservation are met, but
nothing resembling the group of interest has ever been dis-
covered, taking into account a species’ ecology, biogeog-
raphy, preservation potential, and sampling (Benton and
Donoghue, 2007; Parham et al., 2012).
distinguish from the ancestral members, and in some cases
the most distinguishing features of a group may be among
the first to be degraded prior to fossilization (Donoghue
and Purnell, 2009; Sansom et al., 2010). There can also
be a dramatic conflict between the relationships suggested
by independent pools of phylogenetic evidence (e.g.,
molecular versus morphological data). If the relationship
between a fossil specimen and other species (relative to
the divergence event) cannot be established, that specimen
cannot be used reliably to inform the minimum constraint.
It is necessary to examine as much evidence as possible
and use robust tree building programs to establish the rela-
tionships among fossils and living species.

The relationship between fossils and living species
For a fossil to be useful for defining the minimum con-
straint of a divergence event, it must be certain that the fos-
sil belongs to one of the descendant branches (the crown
group), and not to some extinct ancestral branch that pre-
dates the point of divergence (the stem group) (Figure 2).
Understanding the evolutionary – or phylogenetic – rela-
tionships among living species is greatly facilitated by
molecular sequence data; however, for organisms known
only from the fossil record, relationships can only be
inferred from morphological characters. This can be diffi-
cult because many extinct species are known only from
very fragmentary fossil remains, or come from rocks,
which have been highly eroded over time. The earliest
members of a group will also be the most difficult to
Establishing the absolute age of a calibration
We need to know the age of the source rock yielding the
fossil specimens used to inform calibrations. It is rarely
possible to derive the age directly from the specimens,
so the age of the locality from where the fossil was sam-
pled must be established. Firstly, accurate locality infor-
mation must be obtained. This information varies in
precision among specimens; some records will refer to
a well-constrained level in a stratigraphic section, while
others only indicate the relevant geological formation or
group. Secondly, the age of the locality must be
established. If the locality has not been directly dated
(absolute dating), or the available radiometric date is not
stratigraphically relevant for defining the minimum or
maximum age limits, it will usually be necessary to
 MOLECULAR CLOCK CALIBRATION
Molecular Clock Calibration, Figure 2 Definitions of crown and stem groups. The crown group (shaded in grey) represents all living
members of a group, along with their common ancestor, and all extinct relatives in between. The stem group (blue) represents all the
extinct members of a group that are more closely related to their crown group than to any other.
establish the age through stratigraphic correlation (relative
dating). Once the age of the source rock is known relative
to the geological timescale, the absolute age can be
defined with reference to the international chronostra-
tigraphic chart. A conservative interpretation is necessary
to maintain accuracy. Minimum (or maximum) constraints
should be based on the youngest (or oldest) possible inter-
pretation of the fossil specimen. The age should corre-
spond to the explicitly defined top (or base) of
a geological time period and should include the associated
radiometric dating error (Benton and Donoghue, 2007;
Parham et al., 2012).
Tectonic calibrations
Tectonic calibrations provide an alternative to fossil cali-
brations, based on the tectonic events that lead to the initial
separation of two evolutionary lineages, such as the frag-
mentation of supercontinents, the opening of oceans, or
the formation of volcanic islands. The tectonic movements
that were the key to stimulating divergence among species
may be manifested in the modern biogeographic ranges of
the independent lineages. The use of tectonic calibrations
circumvents many of the problems associated with fossil-
based constraints, such as the necessity to define evolu-
tionary relationships among extinct and living species
a priori. But like fossil-based calibrations, the uncertainty
associated with the age of tectonic events must be
established, and in some cases this uncertainty can be
extremely large.
Firstly, it must be established whether a given tectonic
event was actually causal to the speciation event
(Kodandaramaiah, 2011). This requires establishing
whether the modern species ranges reflect the ancestral
biogeography. Tectonic events will also have variable
impacts on the ecology and biogeography of different spe-
cies (Goswami and Upchurch, 2010). For example, an
579
oceanic barrier might limit the dispersal of a freshwater
amphibian species, but will not pose the same limitations
for some bird species. The precise point at which
a tectonic event becomes a complete dispersal barrier will
be hard to define. Secondly, the age of the event must be
established. Many tectonic events can be protracted over
long episodes (e.g., tens of millions of years), and it can
be difficult to pinpoint rock sources that constrain the tem-
poral episode. The dating errors associated with the rock
ages must also be considered, as above.
For groups with a very sparse fossil record, tectonic cal-
ibrations provide an excellent source, independent from
the fossil record, for establishing constraints (Bromham
and Penny, 2003). However, fossils (perhaps of other spe-
cies) will remain integral for verifying the relationship of
tectonic events on modern biogeography. Tectonic cali-
brations should be subject to the same scrutiny with which
fossil calibrations are established (Goswami and
Upchurch, 2010), but similarly the associated uncertainty
can be quantified and modeled in molecular clock
analyses.
The development of different approaches to
implementing calibrations
Traditionally calibrations were implemented as fixed
points (Zuckerkandl and Pauling, 1962), making an
assumption that there was no uncertainty associated with
the calibrations and that ages obtained from the fossil
record were known perfectly. However, there is often
a great deal of uncertainty associated with the calibrations.
Computational progress has enabled molecular clock
methods to become increasingly sophisticated. Maximum
likelihood approaches allowed the implementation of
minimum and maximum constraints that could reflect
a range of calibration ages, instead of fixed points
(Sanderson, 1997, 2002). There has now been a shift into
580 MOLECULAR CLOCK CALIBRATION

Molecular Clock Calibration, Figure 3 Example calibration priors used in Bayesian molecular clock analyses. The top panel shows
a hypothetical divergence event (shaded in gray) and fossil occurrences (squares). The oldest fossil that can be unambiguously
assigned to the crown group and the minimum constraint are highlighted in blue. The maximum constraint is represented by the
dashed red line. The bottom panel shows the prior probability functions. The y-axis represents the relative probability of divergence. In
a and b the calibration prior is defined using the lognormal distribution based on the minimum constraint only (1 Ma) (a: m ¼ 1;
s ¼ 0.5; b: m ¼ 1.5; s ¼ 0.5). c is an example using the skew-t distribution. The parameters were selected such that the 97.5 % interval
matches the maximum constraint (10 Ma) (x ¼ 1, o ¼ 2, a ¼ 100, n ¼ 2.73). d is an example using a uniform distribution between the
minimum and maximum constraints.

the Bayesian framework (Thorne et al., 1998; Drummond
et al., 2006; Yang and Rannala, 2006). The posterior prob-
ability for divergence times
Pr½TimesjSequence data ¼ Pr½Sequence data j Times
Pr½Times=Pr½Sequence Data
combines the prior probability of divergence times
(Pr [Times]) with the likelihood (Pr [Sequence data |
Times]). The prior probability is used to express our prior
knowledge of divergence times and contains the calibra-
tion information.
In the original Bayesian implementation, the calibra-
tions could be implemented as minimum and maximum
constraints (Thorne et al., 1998), but further development
has led to increased flexibility. Each variable that
describes the probability of divergence times can now be
modeled independently, and all unknown parameters can
be assigned an independent prior distribution (Drummond
and Rambaut, 2007). The calibrations can be expressed
using any of a huge variety of functions that best describe
the relationship between the uncertainty inherent in the
calibrations and the true divergence times (Yang and
Rannala, 2006; Drummond et al., 2006; Inoue et al.,
2010) (Figure 3). The uncertainty associated with all
unknown parameters is considered in the estimation of
evolutionary rates and times, and the output (the posterior
probability) will be a distribution of values that reflects the
underlying uncertainty in the estimates.
The Bayesian approach to estimating divergence times
is appealing not only because it offers a great deal of flex-
ibility, but it is also the only method that will allow the
uncertainty associated with the calibrations to be reflected
in the results (Thorne and Kishino, 2005). Because this
 MOLECULAR CLOCK CALIBRATION
error can be large, this is extremely important. The way in
which calibration uncertainty is represented in molecular
clock analysis can have a pejorative impact on the results
(Ho and Phillips, 2009; Inoue et al., 2010; Clarke et al.,
2011; Warnock et al., 2012). It is therefore imperative
for the accurate estimation of rates and times that calibra-
tion uncertainty be accurately quantified and represented.
Characterizing the shape of uncertainty
from the fossil record
Beyond establishing minimum and maximum constraints,
perhaps the most challenging element of establishing
molecular clock calibrations is characterizing the temporal
relationship between the fossil evidence and time of ori-
gin. The function used to express the calibration prior
has to reflect the probability of divergence relative to the
first appearance in the fossil record. A huge variety of
functions, such as the lognormal, exponential, and uni-
form (Figure 3), have been integrated into Bayesian
molecular clock software for this purpose. The distribu-
tion used for the calibration prior should depend on the
available fossil evidence.
The minimum constraint represents the youngest possi-
ble time of divergence and can be used to define the lowest
limits of the calibration prior. The peak in prior probabil-
ity, representing the most likely time of divergence, occurs
at some time before the first appearance in the fossil
record, with a diminishing probability that divergence
has occurred as you go further back in time. For an organ-
ism with a good, dense fossil record, the first appearance
might provide a very good approximation of the true
divergence time, so the peak in prior probability might
occur close to the minimum constraint (Figure 3a). Con-
versely, for a group with a very sparse record, the first
appearance may provide a very poor approximation of
the time of origin, and a more diffuse prior distribution
might be more appropriate (Figure 3b). In reality, it is
extremely difficult to infer this relationship, and limited
protocols exist for choosing the shape of the prior proba-
bility function and choosing the distribution parameters.
These parameters should not be chosen arbitrarily:
a meaningful maximum constraint is also required for the
estimation of the substitution rate (Sanderson, 1997; Kishino
et al., 2001; Thorne and Kishino, 2005; Yang and Rannala,
2006). A soft maximum constraint provides a useful guide
for the selection of the distribution parameters, by defining
the upper confidence limits of the distribution (e.g., the
95 % or 99 % limits) (Figure 3c) (Ho and Phillips, 2009;
Warnock et al., 2012). However, this does not eliminate
the requirement to specify the location of the peak in prior
probability. In the absence of any alternative justification,
a uniform distribution with an equal probability between
the hard minimum and soft maximum constraints may
provide the best characterization of the available evidence
(Figure 3d) (Benton and Donoghue, 2007).
More comprehensive approaches are available for
constructing probabilistic calibration priors based on
additional phylogenetic and paleontological information.
581
The most likely divergence time can be calculated based
on the first occurrences of closely related species
(Hedman, 2010) or modeled using total group fossil
occurrence data (Strauss and Sadler, 1989; Marshall,
1997). The most complex models can incorporate esti-
mates of speciation and extinction rates over time and
nonrandom sampling of the fossil record (Foote et al.,
1999; Tavaré et al., 2002; Wilkinson et al., 2010). These
approaches present an excellent opportunity to integrate
more paleontological evidence into the molecular estima-
tion of divergence times, but the required comprehensive
and well-curated fossil datasets are often unavailable.
Like any scientific variable based on independent
sources of evidence, calibrations are dynamic values. The
understanding of relationships among different branches
of the TOL is continually updated, new fossil finds can
update minimum or maximum constraints, and the geolog-
ical timescale is perpetually refined and revised. Calibra-
tions should be updated in light of evolving evidence.
The inclusion of extinct species in the estimation
of divergence times
Phylogenetic tree building methods also enable the simulta-
neous estimation of evolutionary relationships and diver-
gence times (Drummond et al., 2006; Ronquist et al.,
2012a). However, the popular approach of incorporating
calibrations in the form of prior constraints on divergence
events necessitates that the relationships between the fossils
and living species be known a priori. An alternative
approach is to include the fossils of known age as branches
in the phylogeny (Pyron, 2011; Ronquist et al., 2012b).
This allows the relationships of extinct and extant species,
as well as the evolutionary times to be co-estimated (this
is how viral trees are typically dated, using the molecular
sequences of archived strains). Although molecular data
can rarely be obtained from extinct or fossil specimens,
morphological data can be obtained from fossils, and
because living (or extant) organisms possess both sources
of data, the information can be combined. The major advan-
tage of this approach is that it eliminates the difficult quest
of assigning a probability of divergence time relative to the
paleontological evidence. However, this approach is
accompanied by its own caveats. It requires the fossil record
of the group to be relatively comprehensive, and appropri-
ate morphological datasets are rarely available. The
fragmented or eroded nature of many fossils, and the
absence of any molecular data to accompany the morpho-
logical data for fossil specimens, introduces a large amount
of missing data and additional uncertainty into the analysis.
It also requires the assumption that a clock-like model of
evolution can be applied to morphological data.
Current and ongoing controversies
Initially there were huge discrepancies between the evolu-
tionary times implied by paleontological evidence and the
times inferred from the molecular clock. This is partially
because traditional molecular clock methods demanded
582 MOLECULAR CLOCK CALIBRATION
the use of point (or fixed) calibrations. These were typi-
cally based on inaccurate interpretations of first appear-
ances, and early molecular clock estimates for many
groups hinted at much more ancient origins than those
inferred from the fossil record alone. The timing of origin
of the organisms present during and at the end of the Cam-
brian explosion is one notable example (which represents
the first appearance of all the major animal phyla in the
fossil record, around 541 Ma). Early molecular clock esti-
mates suggested an age of more than twice that implicated
by the fossil record for the origin of most animal phyla
(Hedges and Kumar, 2003). The large deviations between
the timescales obtained from the molecular data were ini-
tially used to denigrate the veracity of the fossil record as
an accurate document of evolutionary history.
Parallel improvements in the way in which rate varia-
tion and calibration uncertainty can be modeled in molec-
ular clock analyses have resulted in the reducing of the gap
between molecular clock estimates and the fossil record.
Recent estimates place the origins of the Cambrian animal
fauna at 619–546 Ma (Rota-Stabelli et al., 2013). A degree
of discrepancy is inevitable given the paucity of the fossil
record, but the questions have now become more intellec-
tual. How much of the fossil record can be anticipated to
be missing? Could the morphology of the common ances-
tor of animals explain its absence from the record?
A concrete collaboration between geologists and molecu-
lar biologists is essential, both for tackling these questions
and for establishing a more precise timescale for the evo-
lution of life.
In reality, we can never know the correct time of diver-
gence relative to appearances in the fossil record, so we
can never assess the accuracy of molecular clock estimates
using empirical datasets. The solution is to use artificial, or
simulated, datasets where the relationship between the fossil
evidence and genetic divergence is known. Fossil occur-
rences can be simulated using a simple model of nonuniform
preservation that describes the stratigraphic distribution of
fossils observed in the fossil record (Holland, 1995, 2000).
Models of speciation, molecular sequence evolution, and
fossil preservation can be combined in simulations to test
different approaches to establishing calibrations and, ulti-
mately, the performance of the molecular clock.
Summary
The incompleteness of the rock and fossil records means
they cannot be used to establish a precise timeline for
evolutionary history. The molecular clock provides
a useful alternative, but the temporal calibrations required
to estimate the substitution rate must be derived from
paleontological evidence. The nonrandom nature of fossil
preservation and recovery makes this extremely difficult.
Although precise calibrations cannot be derived from
the fossil record, they can instead be used to accurately
define minimum and maximum constraints on diver-
gence ages. The next challenge is to characterize the rela-
tionship between the constraints and divergence times.
Approaches to do so have become increasingly complex
and continue to progress to allow more information to be
captured in the estimation of evolutionary times. Molecu-
lar estimates of times and rates will only ever be as accu-
rate as the calibrations on which they are based, making
it essential to embrace and understand the uncertainties
inherent in the fossil record. Adopting a holistic approach
through the integration of paleontological and biological
expertise can make the molecular clock one of the most
powerful tools in evolutionary biology.
Bibliography
Benton, M. J., and Donoghue, P. C., 2007. Paleontological evidence
to date the tree of life. Molecular Biology and Evolution,
24, 26–53.
Benton, M. J., Wills, M. A., and Hitchin, R., 2000. Quality of the
fossil record through time. Nature, 403, 534–537.
Benton, M. J., Dunhill, A. M., Lloyd, G. T., and Marx, F. G., 2011.
Assessing the quality of the fossil record: insights from verte-
brates. Geological Society, London, Special Publications,
358, 63–94.
Bromham, L., and Penny, D., 2003. The modern molecular clock.
Nature Reviews Genetics, 4, 216–224.
Clarke, J. T., Warnock, R. C. M., and Donoghue, P. C., 2011.
Establishing a time-scale for plant evolution. New Phytologist,
192, 266–301.
Donoghue, P. C., and Benton, M. J., 2007. Rocks and clocks: cali-
brating the Tree of Life using fossils and molecules. Trends in
Ecology and Evolution, 22, 424–431.
Donoghue, P. C., and Purnell, M. A., 2009. Distinguishing heat
from light in debate over controversial fossils. Bioessays,
31, 178–189.
Drummond, A. J., and Rambaut, A., 2007. BEAST: Bayesian evo-
lutionary analysis by sampling trees. BMC Evolutionary Biol-
ogy, 7, 214.
Drummond, A. J., and Suchard, M. A., 2010. Bayesian random
local clocks, or one rate to rule them all. BMC Biology, 8, 114.
Drummond, A. J., Ho, S. Y. W., Phillips, M. J., and Rambaut, A.,
2006. Relaxed phylogenetics and dating with confidence. PLoS
Biology, 4, e88.
Dunhill, A. M., 2012. Problems with using rock outcrop area as
a paleontological sampling proxy: rock outcrop and exposure
area compared to coastal proximity, topography, land use and
lithology. Paleobiology, 38, 840–857.
Foote, M., Hunter, J. P., Janis, C. M., and Sepkoski, J. J., 1999. Evo-
lutionary and preservational constraints on origins of biologic
groups: divergence times of eutherian mammals. Science,
283, 1310–1314.
Goswami, A., and Upchurch, P., 2010. The dating game: a reply to
Heads (2010). Zoologica Scripta, 39, 406–409.
Hedges, S. B., and Kumar, S., 2003. Genomic clocks and evolution-
ary timescales. Trends in Genetics, 19, 200–206.
Hedman, M. M., 2010. Constraints on clade ages from fossil
outgroups. Paleobiology, 36, 16–31.
Ho, S. Y. W., and Phillips, M. J., 2009. Accounting for calibration
uncertainty in phylogenetic estimation of evolutionary diver-
gence times. Systematic Biology, 58, 367–380.
Holland, S. M., 1995. The stratigraphic distribution of fossils.
Paleobiology, 21, 92–109.
Holland, S. M., 2000. The quality of the fossil record: a sequence
stratigraphic perspective. Paleobiology, 26, 148–168.
Inoue, J. G., Donoghue, P. C. J., and Yang, Z., 2010. The impact of
the representation of fossil calibrations on Bayesian estimation
of species divergence times. Systematic Biology, 59, 74–89.
 MOLECULAR CLOCKS
Kishino, H., Thorne, J. L., and Bruno, W. J., 2001. Performance of
a divergence time estimation method under a probabilistic model
of rate evolution. Molecular Biology and Evolution, 18, 352–361.
Kodandaramaiah, U., 2011. Tectonic calibrations in molecular dat-
ing. Current Zoology, 57, 116–124.
Marshall, C. R., 1997. Confidence intervals on stratigraphic ranges
with nonrandom distributions of fossil horizons. Paleobiology,
23, 165–173.
Müller, J., and Reisz, R. R., 2005. Four well-constrained calibration
points from the vertebrate fossil record for molecular clock esti-
mates. Bioessays, 27, 1069–1075.
Parham, J. F., Donoghue, P. C. J., Bell, C. J., et al., 2012. Best practices
for justifying fossil calibrations. Systematic Biology, 61, 346–359.
Peters, S. E., 2005. Geologic constraints on the macroevolutionary
history of marine animals. Proceedings of the National Academy
of Sciences, 102, 12326–12331.
Peters, S. E., and Foote, M., 2001. Biodiversity in the Phanerozoic:
a reinterpretation. Paleobiology, 27, 583–601.
Pyron, R. A., 2011. Divergence time estimation using fossils as ter-
minal taxa and the origins of Lissamphibia. Systematic Biology,
60, 466–481.
Rannala, B., and Yang, Z., 2007. Inferring speciation times under an
episodic molecular clock. Systematic Biology, 56, 453–466.
Raup, D. M., 1972. Taxonomic diversity during the
Phanerozoic. Science, 177, 1065–1071.
Reisz, R. R., and Müller, J., 2004. Molecular timescales and the fos-
sil record: a paleontological perspective. Trends in Genetics,
20, 237–241.
Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D. L., Darling,
A., Höhna, S., Larget, B., Liu, L., Suchard, M. A., and
Huelsenbeck, J. P., 2012a. MrBayes 3.2: efficient Bayesian phy-
logenetic inference and model choice across a large model space.
Systematic Biology, 61, 539–542.
Ronquist, F., Klopfstein, S., Vilhelmsen, L., Schulmeister, S., Mur-
ray, D. L., and Rasnitsyn, A. P., 2012b. A total-evidence
approach to dating with fossils, applied to the early radiation of
the hymenoptera. Systematic Biology, 61, 973–999.
Rota-Stabelli, O., Daley, A. C., and Pisani, D., 2013. Molecular
timetrees reveal a cambrian colonization of land and a new sce-
nario for ecdysozoan evolution. Current Biology, 23, 392–398.
Sanderson, M. J., 1997. A nonparametric approach to estimating
divergence times in the absence of rate constancy. Molecular
Biology and Evolution, 14, 1218–1231.
Sanderson, M. J., 2002. Estimating absolute rates of molecular evo-
lution and divergence times: a penalized likelihood approach.
Molecular Biology and Evolution, 19, 101–109.
Sansom, R., Gabbott, S., and Purnell, M., 2010. Non-random decay
of chordate characters causes bias in fossil interpretation. Nature,
463, 797–800.
Smith, A. B., 2007. The shape of the Phanerozoic marine
palaeodiversity curve: how much can be predicted from the sed-
imentary rock record of Western Europe? Palaeontology,
50, 765–774.
Smith, A. B., and McGowan, A. J., 2007. The shape of the Phaner-
ozoic marine palaeodiversity curve: how much can be predicted
from the sedimentary rock record of Western Europe?
Palaeontology, 50, 765–774.
Strauss, D., and Sadler, P. M., 1989. Classical confidence intervals
and Bayesian probability estimates for ends of local taxon
ranges. Mathematical Geology, 21, 411–427.
Tavaré, S., Marshall, C. R., Will, O., Soligo, C., and Martin, R. D.,
2002. Using the fossil record to estimate the age of the last com-
mon ancestor of extant primates. Nature, 416, 726–729.
Thorne, J. L., and Kishino, H., 2005. Estimation of divergence times
from molecular sequence data. In Nielsen, R. (ed.), Statistical
Methods in Molecular Evolution. New York: Springer,
pp. 233–256.
583
Thorne, J. L., Kishino, H., and Painter, I. S., 1998. Estimating the
rate of evolution of the rate of molecular evolution. Molecular
Biology and Evolution, 15, 1647–1657.
Warnock, R. C. M., Yang, Z., Donoghue, P. C. J., 2012. Exploring
uncertainty in the calibration of the molecular clock. Biology
Letters, 8, 156–159.
Wilkinson, R. D., Steiper, M. E., Soligo, C., Martin, R. D., Yang, Z.,
and Tavaré, S., 2011. Dating primate divergences through an
integrated analysis of palaeontological and molecular data. Sys-
tematic Biology, 60, 16–31.
Yang, Z., and Rannala, B., 2006. Bayesian estimation of species
divergence times under a molecular clock using multiple fossil
calibrations with soft bounds. Molecular Biology and Evolution,
23, 212–226.
Zuckerkandl, E., and Pauling, L., 1962. Molecular disease, evolution
and genetic heterogeneity. In Kasha, M., and Pullman, B. (eds.),
Horizons in Biochemistry. New York: Academic, pp. 189–225.
MOLECULAR CLOCKS
Simon Y. W. Ho
School of Biological Sciences, University of Sydney,
Sydney, NSW, Australia
Definition
Molecular clock. A hypothesis that predicts a constant
rate of molecular evolution among species. It is also
a method of genetic analysis that can be used to estimate
evolutionary rates and timescales using data from DNA
or proteins.
Introduction
The tempo and mode of evolution are central themes of
biological research. This places importance on the estima-
tion of evolutionary timescales, which provide the back-
drop for our interpretations of evolutionary patterns and
processes. Traditionally, such inferences were made from
the fossil record, coupled with radiometric dating. Fossils
can provide an estimate of when different lineages first
appeared and when species diverged from each other. In
many cases, however, such data are unavailable, forcing
us to look elsewhere for a source of temporal information.
The “molecular clock,” proposed in the 1960s
(Zuckerkandl and Pauling, 1962, 1965), allows evolution-
ary timescales to be estimated using genetic data. Molecu-
lar clocks have provided insights into significant
evolutionary events such as human evolution and the radi-
ations of birds, mammals, and flowering plants. Molecu-
lar-clock methods continue to undergo improvement,
which is crucial if we are to take advantage of the growing
wealth of genomic data (Kumar, 2005).
The molecular-clock hypothesis posits that rates of
molecular evolution, as reflected in changes in DNA or
protein sequences through time, are constant among line-
ages (but not across different regions of the genome).
Nucleotide mutations cause DNA sequences to change
584 MOLECULAR CLOCKS
over time, whereas mutations of amino acids lead to evo-
lution in proteins. These mutations occur with a constant
probability rather than at a constant frequency – that is,
the molecular clock is stochastic rather than metronomic
(Zuckerkandl and Pauling, 1965). The occurrence of
mutations in a DNA sequence is often modelled using
a Poisson process.
A corollary of the molecular-clock hypothesis is that
the genetic difference between any two species is propor-
tional to the time since they last shared a common ances-
tor. This is a useful relationship because it allows
evolutionary timescales to be estimated from genetic data,
provided that the rate of molecular evolution is known.
The molecular clock represents the only method for study-
ing the evolutionary timescales of organisms that have
failed to leave any trace in the fossil record, including
many groups of invertebrates, bacteria, and viruses. It also
allows us to estimate the timing of events that are too
recent to be resolved by fossil evidence, such as diver-
gences among conspecific populations (Arbogast et al.,
2002).
History of the molecular clock
The molecular-clock hypothesis was put forward by Emile
Zuckerkandl and Linus Pauling (1962), who assumed
a constant evolutionary rate in their analysis of globin pro-
teins from vertebrates. They observed about 18 differ-
ences in amino acids between horse and human and
estimated the mutation rate by assuming that the diver-
gence between these two species occurred 100–160 Ma
ago. Upon extrapolating this rate, Zuckerkandl and Pau-
ling (1962) estimated that humans diverged from gorillas
about 11 Ma ago. They also estimated that different copies
of globin genes first diverged from each other in the late
Precambrian. When reporting these estimates,
Zuckerkandl and Pauling (1962) warned about the possi-
ble confounding effects of a number of factors, including
natural selection, variations in population size, and muta-
tional saturation.
In the following years, further studies produced evi-
dence of clocklike evolution in other proteins. Doolittle
and Blomback (1964) found a simple relationship
between sequence identity and time since divergence in
mammalian fibrinopeptides (Figure 1). A year later,
Zuckerkandl and Pauling (1965) coined the term “molec-
ular evolutionary clock.” It promised to be a useful tool
in biological research, as demonstrated shortly afterward
by Sarich and Wilson (1967a, b) in their pioneering stud-
ies of the evolutionary timescale of hominids and other
primates.
At the time, the idea of a constant substitution rate was
controversial (Morgan, 1998). The molecular clock was
received unenthusiastically by George Gaylord Simpson
(1964), among others. Criticisms of the molecular clock
were partly motivated by the apparent lack of uniformity
in the pace of morphological evolution, which was pre-
sumably linked to molecular evolution. There was no
Molecular Clocks, Figure 1 Plot of amino acid identity of
fibrinopeptides from various pairs of mammals, plotted against
time since divergence. The divergence times are based on
estimates from the fossil record (Data are from Doolittle and
Blomback (1964)).
evidence to suggest that adaptive change occurred at
a uniform rate and it seemed implausible that the complex
process of molecular evolution could be described by such
a simple statistical model.
Meanwhile, there was growing evidence of high evolu-
tionary rates in various proteins, suggesting that a large
proportion of the changes in amino acids must have
a negligible impact on evolutionary fitness. As
a response, Motoo Kimura (1968) proposed the neutral
theory of molecular evolution, which states that many
mutations have such a small effect on the fitness of an
organism that they can be considered as “neutral.” This
can be explained by the fact that many amino acids in
a protein can be exchanged for other amino acids with
similar biochemical properties, with negligible impact on
the overall function or structure of the protein. At the level
of DNA, many mutations in protein-coding genes are
“synonymous” because they do not result in any changes
in the amino acid sequence of the protein. The fate of these
“neutral” mutations is determined stochastically by
a process known as genetic drift. The neutral theory was
proposed independently by Jack King and Thomas Jukes
(1969), who cited a range of evidence to support their
hypothesis.
One of the predictions of the neutral theory is that rates
of molecular evolution are constant among lineages. How-
ever, this prediction refers specifically to the rate of
genetic change per generation. As a consequence, we
expect to see a generation-time effect, whereby species
with shorter generations tend to evolve more quickly per
unit of time. For example, a higher rate would be observed
 MOLECULAR CLOCKS
in rodents than in whales. This is based on the assumption
that most mutations occur during the replication of
germline DNA. Such a generation-time effect was
observed in the analyses of non-coding DNA (Laird
et al., 1969; Kohne, 1970). This was in contrast with the
rate of protein evolution, which appeared to be indepen-
dent of generation time.
In response to the shortcomings of the neutral theory,
Tomoko Ohta (1972, 1973) proposed the nearly neutral
theory of molecular evolution. In this framework, there
is a large class of “nearly neutral” mutations that have
small effects on an organism’s fitness. In contrast with
the results of the neutral theory, the nearly neutral theory
states that the population sizes of species have
a significant influence on the molecular evolutionary pro-
cess. Many mutations are slightly harmful and are gradu-
ally removed from the population. In large populations,
natural selection is effective at removing mutations that
are detrimental to the organism’s fitness. In small
populations, natural selection tends to be ineffective and
most genetic change is driven by genetic drift. Generally,
drift acts more quickly in small populations than does
selection in large populations. However, a greater number
of mutations appear each generation in large populations,
simply because there are more individuals that can experi-
ence mutations. These two effects offset each other, lead-
ing to a relatively constant evolutionary rate among
species per unit of time.
Through the ensuing decades, the molecular clock was
used to study the evolutionary timescales of a range of
organisms. Some of the most prominent studies involved
analyses of the metazoan evolutionary timescale, with
a focus on the divergences among animal phyla. In consid-
erable conflict with the “Cambrian explosion” scenario
supported by the fossil record, estimates from the molecu-
lar clock suggested a protracted Precambrian timescale for
basal metazoan divergences (e.g., Dickerson, 1971;
Runnegar, 1982; Doolittle et al., 1996). Similarly, molec-
ular evidence pointed to much deeper diversifications of
avian and mammalian orders than those suggested by
paleontological evidence (e.g., Cooper and Penny, 1997;
Springer, 1997). These discrepancies remain some of the
key sources of debate about the accuracy of the molecular
clock (Bromham and Penny, 2003).
Meanwhile, there was a growing body of evidence that
pointed to rate variation among lineages. This came
partly from various statistical tests that had been devel-
oped to test for clocklike evolution in genetic data
(Wu and Li, 1985; Tajima, 1993). Departures from the
molecular clock provided a challenge for studies of evo-
lutionary timescales. It was not until the late 1990s,
however, when molecular-clock methods were devel-
oped that were able to account for variation in evolution-
ary rates (Sanderson, 1997; Thorne et al., 1998). Known
as “relaxed” molecular clocks because they relax the
assumption of rate constancy among lineages, these have
become the standard techniques in molecular evolution-
ary analysis.
585
Rates of molecular evolution
Absolute rates of molecular evolution vary among regions
of the genome because of differing constraints. Function-
ally important genes often evolve at an extremely slow
pace, because many new mutations are detrimental to the
organism and are rapidly removed by natural selection.
As a consequence, such genes are often conserved even
among distantly related organisms, making them useful
genetic markers for studying deep evolutionary relation-
ships. For example, histones, which are proteins that play
an important role in binding and packaging DNA, have
a very low substitution rate. In contrast, non-coding
DNA has fewer functional constraints and can evolve at
a higher rate than protein-coding DNA.
The mitochondrial genome evolves rapidly in animals
and experiences about 10
8 substitutions per nucleotide
per year, which is an order of magnitude higher than the
average rate in the nuclear genome. This is in contrast with
the pattern of molecular evolution in plants, where nuclear
genomes evolve more quickly than either chloroplast or
mitochondrial genomes. Rates of change in the genomes
of viruses are higher by several orders of magnitude. In
rapidly evolving RNA viruses, such as influenza virus
and human immunodeficiency virus (HIV), mutation rates
can exceed 10
3 mutations per nucleotide per year, and
measurable genetic change can occur over a matter of
weeks.
There is substantial variation in rates of molecular
evolution across the tree of life. This variation is possibly
driven by biological factors such as generation time, pop-
ulation size, longevity, and body temperature, as well as
abiotic factors such as ultraviolet radiation. The relative
importance of each of these factors is still poorly under-
stood (Bromham, 2009). Studies have found evidence
that molecular evolutionary rates are associated with lon-
gevity in mammals (Welch et al., 2008), with generation
time in invertebrates (Thomas et al., 2010), and with
height in plants (Lanfear et al., 2013). Research into the
factors governing rate variation among organisms is
ongoing.
Molecular clocks and phylogenetic analysis
Molecular clocks are typically used in phylogenetic ana-
lyses, which aim to reconstruct evolutionary trees that
show the relationships among species of interest
(Figure 2). Internal nodes in the tree represent evolution-
ary divergence events. The timing of these events can be
estimated using molecular clocks. A number of statistical
methods are available for testing the molecular-clock
hypothesis for a given set of DNA or protein sequences.
When the molecular clock is rejected for a data set, one
can use a statistical model to account for rate variation
when estimating evolutionary timescales (Welch and
Bromham, 2005).
When estimating evolutionary timescales in
a phylogenetic analysis, the molecular clock needs to be
“calibrated.” This can be done by assigning an absolute
586 MOLECULAR CLOCKS
Molecular Clocks, Figure 2 Phylogenetic tree showing the
relationships of species from three genera, A, B, and C. Genera
A and B are found on the mainland, whereas species of Genus C
are found on an offshore island that was formed by volcanic
activity. The tree is drawn to an unspecified timescale, so that
the branch lengths are proportional to time. The tree illustrates
the use of two molecular-clock calibrations. First, a fossil taxon
has been assigned to the lineage leading to Genus A. This places
a minimum age constraint on the divergence of A from B and C.
Second, a geological calibration is placed on the divergence
between B and C. The split between these genera is assumed to
coincide with the formation of the offshore island on which
Genus C is found. The age of the island is estimated using
radiometric dating. The uncertainty in this date estimate is
incorporated into the calibration, as indicated by the black bar.
Gray error bars represent the uncertainty in the molecular-clock
estimates of lineage divergence times.
age to one or more nodes in the phylogeny, which can then
act as a reference point for estimating the ages of the
remaining nodes. Calibrations are often based on the fossil
record, which can provide date estimates for the diver-
gence events in the phylogeny (Figure 2). If a fossil taxon
can be reliably assigned to one of the lineages in the tree,
then the divergence of that lineage from its sister lineage
must be older than the age of the fossil. Calibrations can
also be based on geological events, including the separa-
tion of continents or the emergence of islands, if they are
linked to evolutionary divergences. An example of this
might involve two sister genera, one of which is endemic
to the mainland and the other endemic to an offshore
island that formed as a result of volcanic activity
(Figure 2). The timing of the divergence between the
two genera can be estimated by the age of the island
which, in turn, can be estimated using radiometric dating.
In some cases, previous genetic estimates of dates are
employed as calibrations. These are known as “second-
ary” calibrations and are generally used when other cali-
brating information is unavailable.
Once the phylogenetic tree is calibrated by fixing or
constraining the ages of one or more nodes in the
phylogeny, the ages of the remaining nodes can be esti-
mated using a molecular-clock analysis of the genetic
data. This process is referred to as “molecular dating” or
“divergence-time estimation.” There is a large range of
methods and models that have been developed for this
purpose, implemented in various statistical frameworks.
These methods are available in a range of computer pro-
grams, including most standard phylogenetic software.
Universal molecular clocks
There is abundant evidence of evolutionary rate variation
among species, dispelling any hope of a universal molec-
ular clock across the tree of life. However, some
researchers entertain the idea of homogeneous rates of
mitochondrial evolution within certain groups of organ-
isms, such as birds, mammals, and arthropods. This is
a convenient assumption because it allows evolutionary
rates to be applied in molecular-clock analyses when fossil
or geological calibrations are otherwise unavailable.
Many studies of birds assume that the mitochondrial
genome evolves at about 1 % per million years, a value
that was initially based on a study of five species of geese
(Shields and Wilson, 1987). Various analyses of avian
mitochondrial DNA have supported this estimate, includ-
ing an extensive study involving 74 calibrations and
genetic data from 12 orders of birds (Weir and Schluter,
2008). The avian mitochondrial clock has been used to
investigate key questions in the evolution of birds, includ-
ing the impact of Pleistocene glaciation on the diversifica-
tion of passerines. Some have questioned the reliability of
this clock, citing evidence of significant rate variation
among lineages and across timescales (García-Moreno,
2004; Ho, 2007). Despite this criticism, the avian mito-
chondrial clock is still widely used.
In a similar fashion, a universal mitochondrial clock of
1 % per million years is often assumed for mammals. This
rate was initially based on a mitochondrial study of pri-
mates and rodents (Brown et al., 1979), but it soon found
support from estimates derived from other organisms
(Wilson et al., 1985). Comprehensive analyses of mam-
malian DNA have demonstrated that there is substantial
variation in rates among species, partly driven by differ-
ences in longevity (Nabholz et al., 2008; Welch et al.,
2008).
Studies of evolutionary timescales in arthropods have
often employed a rate of 1.15 % per million years, which
is based on an analysis of insects and crustaceans (Brower,
1994). This arthropod mitochondrial clock is widely used,
partly because of the relative paucity of reliable fossil cal-
ibrations for many invertebrate taxa. However, this clock
was based on a very small number of data points, leading
to its validity being questioned (Papadopoulou et al.,
2010; Ho and Lo, 2013). Moreover, there is strong evi-
dence of mitochondrial and nuclear rate variation among
invertebrate species (Thomas et al., 2010).
The employment of “standard” mitochondrial clocks,
such as those described above, has often been criticized.
 MOLECULAR CLOCKS
This is because the mitochondrial genome is subject to dif-
fering degrees of natural selection among species, which
can lead to heterogeneity in evolutionary rates. Similar
reasoning applies to evolutionary rates in the nuclear
genome. Given that rates can show substantial variation
even among closely related species, the use of standard
clocks has the potential to yield date estimates that are
highly misleading.
Controversies
The molecular clock has had a long history of contro-
versy, beginning with the criticisms of the assumption
of rate constancy and continuing with the debate over
the relative merits of the neutral theory compared with
natural selection. Current debates include the use of cali-
brations in genetic dating analyses and the constancy of
evolutionary rates across timescales (Pulquério and
Nichols, 2007).
The choice of calibrations has a substantial influence on
the outcome of a molecular-clock analysis. Inadequate
modelling of calibrations can lead to highly misleading
estimates of evolutionary timescales. Accordingly, recent
work has focused on accounting for the uncertainty in fos-
sil and geological calibrations (Ho and Phillips, 2009).
There is also ongoing development of methods for model-
ling and quantifying the uncertainty in fossil calibrations
(e.g., Marshall, 2008; Wilkinson et al., 2011). Some
aspects of calibration methodology are poorly understood,
such as the impact of using different methods for model-
ling the uncertainty in calibrations. One of the key points
of agreement is that best practice involves the use of mul-
tiple calibrations throughout the phylogenetic tree.
The use of secondary calibrations in molecular dating
has been the target of strong criticism (Graur and Martin,
2004). Although the use of secondary calibrations is
sometimes unavoidable, good practice involves taking
into account the uncertainty in previous molecular-clock
estimates. Ignoring this uncertainty can lead to artificially
precise estimates of evolutionary timescales. Instead, the
uncertainty should be incorporated into the dating analysis
so that it can be included in the resulting estimates. This
can readily be done in a Bayesian framework, in which
the user can choose prior distributions that reflect the
degree of uncertainty in the parameters and node times.
There is growing evidence that estimates of rates
depend on the timescale of observation. Low rates of evo-
lution are seen across long evolutionary time frames, such
as those analyzed in phylogenetic analyses of distant spe-
cies. In contrast, high evolutionary rates have been esti-
mated in studies of populations and pedigrees (Howell
et al., 2003). The exact causes of this disparity remain
unclear, although they are likely to include the effects of
natural selection and inaccurate modelling of the molecu-
lar evolutionary process (Ho et al., 2011). If the time
dependence of molecular rates is not taken into account,
evolutionary timescales can be under- or overestimated
by an order of magnitude.
587
Conclusions
The molecular clock has undergone considerable evolu-
tion during its long history. It is useful as a hypothesis in
molecular evolution and as a tool for estimating evolution-
ary rates and timescales. New molecular-clock methods
are being developed in order to take advantage of the large
amounts of genomic data that are being generated. With
further refinement and development, the molecular clock
will continue to play an important role in understanding
the evolution of life on Earth.
Bibliography
Arbogast, B. S., Edwards, S. V., Wakeley, J., Beerli, P., and
Slowinski, J. B., 2002. Estimating divergence times from molec-
ular data on phylogenetic and population genetic timescales.
Annual Review of Ecology, Evolution, and Systematics,
33, 707–740.
Bromham, L., 2009. Why do species vary in their rate of molecular
evolution? Biology Letters, 5, 401–404.
Bromham, L., and Penny, D., 2003. The modern molecular clock.
Nature Reviews. Genetics, 4, 216–224.
Brower, A. V. Z., 1994. Rapid morphological radiation and conver-
gence among races of the butterfly Heliconius erato inferred
from patterns of mitochondrial DNA evolution. Proceedings of
the National Academy of Sciences of the United States of Amer-
ica, 91, 6491–6495.
Brown, W. M., George, M., and Wilson, A. C., 1979. Rapid evolu-
tion of animal mitochondrial DNA. Proceedings of the National
Academy of Sciences of the United States of America,
76, 1967–1971.
Cooper, A., and Penny, D., 1997. Mass survival of birds across the
Cretaceous-Tertiary boundary: molecular evidence. Science,
275, 1109–1113.
Dickerson, R. E., 1971. The structure of cytochrome c and the rates of
molecular evolution. Journal of Molecular Evolution, 1, 26–45.
Doolittle, R. F., and Blomback, B., 1964. Amino-acid sequence
investigations of fibrinopeptides from various mammals: evolu-
tionary implications. Nature, 202, 147–152.
Doolittle, R. F., Feng, D.-F., Tsang, S., Cho, G., and Little, E., 1996.
Determining divergence times of the major kingdoms of living
organisms with a protein clock. Science, 271, 470–477.
García-Moreno, J., 2004. Is there a universal mtDNA clock for
birds? Journal of Avian Biology, 35, 465–468.
Graur, D., and Martin, W., 2004. Reading the entrails of chickens:
molecular timescales of evolution and the illusion of precision.
Trends in Genetics, 20, 80–86.
Ho, S. Y. W., 2007. Calibrating molecular estimates of substitution
rates and divergence times in birds. Journal of Avian Biology,
38, 409–414.
Ho, S. Y. W., and Lo, N., 2013. The insect molecular clock. Austra-
lian Journal of Entomology, 52, 101–105.
Ho, S. Y. W., and Phillips, M. J., 2009. Accounting for calibration
uncertainty in phylogenetic estimation of evolutionary diver-
gence times. Systematic Biology, 58, 367–380.
Ho, S. Y. W., Lanfear, R., Bromham, L., Phillips, M. J., Soubrier, J.,
Rodrigo, A. G., and Cooper, A., 2011. Time-dependent rates of
molecular evolution. Molecular Ecology, 20, 3087–3101.
Howell, N., Bogolin Smejkal, C., Mackey, D. A., Chinnery, P. F.,
Turnbull, D. M., and Herrnstadt, C., 2003. The pedigree rate of
sequence divergence in the human mitochondrial genome: there
is a difference between phylogenetic and pedigree rates. Ameri-
can Journal of Human Genetics, 72, 659–670.
Kimura, M., 1968. Evolutionary rate at the molecular level. Nature,
217, 624–626.
588 MOLECULAR CLOCKS, HUMAN EVOLUTION
King, J. L., and Jukes, T. H., 1969. Non-Darwinian evolution. Sci-
ence, 164, 788–798.
Kohne, D. E., 1970. Evolution of higher-organism DNA. Quarterly
Reviews of Biophysics, 3, 327–375.
Kumar, S., 2005. Molecular clocks: four decades of evolution.
Nature Reviews. Genetics, 6, 654–662.
Laird, C. D., McConaughy, B. L., and McCarthy, B. J., 1969. Rate
of fixation of nucleotide substitutions in evolution. Nature,
224, 149–154.
Lanfear, R., Ho, S. Y. W., Davies, T. J., Moles, A. T., Aarssen, L.,
Swenson, N. G., Warman, L., Zanne, A. E., and Allen, A. P.,
2013. Taller plants have lower rates of molecular evolution.
Nature Communications, 224, 1879.
Marshall, C. R., 2008. A simple method for bracketing absolute
divergence times on molecular phylogenies using multiple fossil
calibration points. American Naturalist, 171, 726–742.
Morgan, G. J., 1998. Emile Zuckerkandl, Linus Pauling, and the
molecular evolutionary clock, 1959–1965. Journal of the His-
tory of Biology, 31, 155–178.
Nabholz, B., Glémin, S., and Galtier, N., 2008. Strong variation of
mitochondrial mutation rate across mammals – the longevity
hypothesis. Molecular Biology and Evolution, 25, 120–130.
Ohta, T., 1972. Evolutionary rate of cistrons and DNA divergence.
Journal of Molecular Evolution, 1, 150–157.
Ohta, T., 1973. Slightly deleterious mutant substitutions in evolu-
tion. Nature, 246, 96–98.
Papadopoulou, A., Anastasiou, I., and Vogler, A. P., 2010.
Revisiting the insect mitochondrial molecular clock: the mid-
Aegean trench calibration. Molecular Biology and Evolution,
27, 1659–1672.
Pulquério, M. J. F., and Nichols, R. A., 2007. Dates from the molec-
ular clock: how wrong can we be? Trends in Ecology & Evolu-
tion, 22, 180–184.
Runnegar, B., 1982. A molecular-clock date for the origin of the ani-
mal phyla. Lethaia, 15, 199–205.
Sanderson, M. J., 1997. A nonparametric approach to estimating
divergence times in the absence of rate constancy. Molecular
Biology and Evolution, 14, 1218–1231.
Sarich, V. M., and Wilson, A. C., 1967a. Rates of albumin evolution
in primates. Proceedings of the National Academy of Sciences of
the United States of America, 58, 142–148.
Sarich, V. M., and Wilson, A. C., 1967b. Immunological time scale
for hominid evolution. Science, 158, 1200–1203.
Shields, G. F., and Wilson, A. C., 1987. Calibration of mitochon-
drial DNA evolution in geese. Journal of Molecular Evolution,
24, 212–217.
Simpson, G. G., 1964. Organisms and molecules in evolution. Sci-
ence, 146, 1535–1538.
Springer, M. S., 1997. Molecular clocks and the timing of the pla-
cental and marsupial radiations in relation to the Cretaceous-
Tertiary boundary. Journal of Mammalian Evolution, 4,
285–302.
Tajima, F., 1993. Simple methods for testing the molecular evolu-
tionary clock hypothesis. Genetics, 135, 599–607.
Thomas, J. A., Welch, J. J., Lanfear, R., and Bromham, L., 2010.
A generation time effect on the rate of molecular evolution in
invertebrates. Molecular Biology and Evolution, 27, 1173–1180.
Thorne, J. L., Kishino, H., and Painter, I. S., 1998. Estimating the
rate of evolution of the rate of molecular evolution. Molecular
Biology and Evolution, 15, 1647–1657.
Weir, J. T., and Schluter, D., 2008. Calibrating the avian molecular
clock. Molecular Ecology, 17, 2321–2328.
Welch, J. J., and Bromham, L., 2005. Molecular dating when rates
vary. Trends in Ecology & Evolution, 20, 320–327.
Welch, J. J., Bininda-Emonds, O. R. P., and Bromham, L., 2008.
Correlates of substitution rate variation in mammalian protein-
coding sequences. BMC Evolutionary Biology, 8, 53.
Wilkinson, R. D., Steiper, M. E., Soligo, C., Martin, R. D., Yang, Z.,
and Tavaré, S., 2011. Dating primate divergences through an
integrated analysis of palaeontological and molecular data. Sys-
tematic Biology, 60, 16–31.
Wilson, A. C., Cann, R. L., Carr, S. M., George, M., Gyllensten,
U. B., Helm-Bychowski, K. M., Higuchi, R. G., Palumbi,
S. R., Prager, E. M., Sage, R. D., and Stoneking, M., 1985. Mito-
chondrial DNA and two perspectives on evolutionary genetics.
Biological Journal of the Linnean Society, 26, 375–400.
Wu, C.-I., and Li, W. H., 1985. Evidence for higher rates of nucleo-
tide substitutions in rodents than in man. Proceedings of the
National Academy of Sciences of the United States of America,
82, 1741–1745.
Zuckerkandl, E., and Pauling, L., 1962. Molecular disease, evolu-
tion and genetic heterogeneity. In Kasha, M., and Pullman, B.
(eds.), Horizons in Biochemistry. New York: Academic,
pp. 189–225.
Zuckerkandl, E., and Pauling, L., 1965. Evolutionary divergence
and convergence in proteins. In Bryson, V., and Vogel, H. J.
(eds.), Evolving Genes and Proteins. New York: Academic,
pp. 97–166.
Cross-references
Gene Sequencing
Molecular Clock Calibration
Molecular Clocks, Human Evolution
Molecular Clocks, Relaxed Variant
Molecular Dating of Evolutionary Events
Molecular Rate Variation (Molecular Clocks)
Polymerase Chain Reaction DNA Amplification
MOLECULAR CLOCKS, HUMAN EVOLUTION
Simon Y. W. Ho1 and Phillip Endicott2
1
School of Biological Sciences, University of Sydney,
Sydney, NSW, Australia
2
Départment Hommes, Natures, Sociétés, Musée de
l’Homme, Paris, France
Definition
Human evolution (molecular clocks). The timescale of
human evolution and migration can be estimated from
genetic data that have been sampled from living, histori-
cal, and ancient humans. This can be done using statistical
methods based on the molecular clock hypothesis.
Introduction
The timescale of human evolutionary origins and migra-
tion across the globe has been a long-standing focus of sci-
entific research. Genetic studies of the human
evolutionary timescale were first performed in the 1960s,
with the divergence time between humans and chimpan-
zees being estimated using data from the albumin protein
(Sarich and Wilson, 1967). The complete sequence of
the human mitochondrial genome, consisting of about
16.5 thousand nucleotides, was published in 1981
(Anderson et al., 1981). Two decades later, the Interna-
tional Human Genome Consortium (2001) released
 MOLECULAR CLOCKS, HUMAN EVOLUTION
a complete draft sequence of the much larger nuclear
genome, comprising 3.2 billion nucleotides. DNA
sequences of mitochondrial and nuclear genomes have
now been obtained from a range of individuals throughout
the world. The availability of genetic data from ancient
samples has also been growing steadily (see “Timescale
of Human Evolution and Migration” below).
Evolutionary timescales can be estimated from DNA
sequence data using methods based on the molecular
clock. The molecular clock hypothesis states that the rate
of molecular evolution is constant among lineages
(Zuckerkandl and Pauling, 1962, 1965). If this assumption
holds, then the difference between the DNA sequences of
two individuals is proportional to the time since they last
shared a common ancestor. If not, then there are various
statistical methods that can account for rate variation
among lineages and along the genome (Welch and
Bromham, 2005).
Until recently, genetic analyses of human prehistory
were primarily based on mitochondrial DNA. The mito-
chondrial genome evolves rapidly and displays useful var-
iation even over the short timescales associated with
the prehistory of modern humans (Torroni et al., 2006).
Substantial developments in DNA sequencing technology
are enabling a growing focus on the nuclear genome,
greatly increasing the data available for analysis. In addi-
tion, improving molecular techniques have made it possi-
ble to deal with contamination in ancient hominin
samples. This has created opportunities to access genomic
data from ancient modern humans, together with extinct
species of hominins such as Neanderthals (Shapiro and
Hofreiter, 2010). These new sources of data have led to
some significant changes to our understanding of hominin
evolution. For example, DNA sequences of ancient
genomes have shed light on interbreeding between Nean-
derthals and modern humans (Prüfer et al., 2014), our rela-
tionships with ancient hominins (Reich et al., 2010), and
the timescale and patterns of human dispersals across the
globe (Reich et al., 2011).
The human molecular clock
For several decades, the mitochondrial genome was the
dominant source of information for estimating molecular
rates of evolution in humans and other hominins. In ana-
lyses of human mitochondrial genomes, a common
approach is to employ an estimate of the evolutionary
rate from a previous study, with a small number of
“standard” rates dominating the literature (Endicott
et al., 2009). Most of these rates were estimated by
assuming that humans and chimpanzees diverged from
each other about 6.5 million years ago, based on fossil
evidence.
Pedigree studies, which involve the analysis of DNA
sequences from multiple members of a family, have
yielded very high estimates of mitochondrial mutation
rates (Howell et al., 2003). There is a substantial discrep-
ancy between these estimates and those based on
589
comparisons between humans and chimpanzees. This pat-
tern can be explained by a number of biological and statis-
tical factors (Ho et al., 2011), including the effects of
natural selection. The presence of transient deleterious
mutations at the pedigree level, which are removed over
time by negative selection, causes an apparent short-term
elevation of evolutionary rates (Soares et al., 2009).
In analyses of mitochondrial genomes, the evolutionary
rate is typically assumed to be constant among lineages.
However, there is considerable evidence for rate variation
among species and even among human populations (Henn
et al., 2009). In addition, there are limits to the interpreta-
tion of analyses based on mitochondrial DNA, owing to
the fact that it is only a single genetic marker (Balloux,
2010). This is in contrast with the nuclear genome, which
contains thousands of independent markers that can be
used to trace evolutionary patterns and timescales.
In the post-genomic era, sequences from the nuclear
genome have become readily available. This important
source of data has provided a broad range of insights into
human evolution. A prominent feature of genomic evolu-
tion in primates is a decline in evolutionary rates within
hominoids compared with their relatives (Goodman,
1985; Steiper et al., 2004). The hominoid slowdown can
be partly explained by differences in generation times. In
species with longer generations, the genome experiences
fewer replications per unit of time. During replication,
copying errors lead to mutations in the genome, and these
can be inherited by descendants.
More recently, studies at the pedigree level have
yielded estimates of the de novo mutation rate in the
human nuclear genome. These have produced an unex-
pected result, with the de novo mutation rate being about
half of the rate estimated using a human-chimpanzee com-
parison (e.g., Roach et al., 2010). A possible explanation
for this pattern is that humans have longer generation
times than chimpanzees, which would lead to a lower rate
of evolution. This lower rate has the effect of increasing
date estimates for key phases in human prehistory, such
as the most recent common ancestor in Africa (Scally
and Durbin, 2012). The archaeological evidence for
detecting population-level events in human evolution is
rather scant, and it is not yet clear whether such revised
dates are supported.
Timescale of human evolution and migration
Estimates of the human evolutionary timescale have var-
ied considerably, depending on the choice of genomic
marker, the breadth and depth of sampling, the calibra-
tions used for the molecular clock, and the method of anal-
ysis. In some cases, different date estimates support
conflicting interpretations of human evolution and
prehistory.
Studies of mitochondrial DNA have yielded date esti-
mates that have been broadly consistent with the time-
scales inferred using archaeological and paleontological
approaches. The lineage leading to modern humans
590 MOLECULAR CLOCKS, HUMAN EVOLUTION
Molecular Clocks, Human Evolution, Figure 1 Phylogenetic
tree showing the evolutionary timescale of modern humans,
Neanderthal, Denisovan, and chimpanzee. Dashed lines indicate
genetic contributions made from Neanderthals and Denisovans
to populations of modern humans. Date estimates are from
Prüfer et al. (2014). The tree is not drawn to scale.
diverged from Neanderthals about 400,000 years ago
(Endicott et al., 2010). Living modern humans share
a common mitochondrial ancestor, sometimes referred to
as “mitochondrial Eve,” who lived in Africa over
100,000 years ago. Humans then migrated to Eurasia
and Australasia about 50,000–60,000 years ago, reaching
the Americas about 15,000 years ago.
DNA sequences from ancient specimens have had
a major impact on our understanding of human prehistory.
Significant advances in molecular technology have
enabled entire mitochondrial and nuclear genomes to be
sequenced from samples that are tens to hundreds of thou-
sands of years old, provided that they have been preserved
under certain conditions. Estimates from the nuclear
genome indicate that modern humans and Neanderthals
diverged more than 600,000 years ago (Figure 1; Prüfer
et al., 2014), although estimates range from 400,000 to
800,000 years. Neanderthals and another group of
hominins known as Denisovans split from each other
about 430,000 years ago. After this event, Neanderthals
contributed genetic material to non-African modern
humans, whereas significant amounts of Denisovan
DNA appear only in modern humans from Oceania
(Figure 1).
Future directions
Our understanding of the timescale of hominid and human
evolution is constantly being refined. The genomic era has
brought a wealth of data, offering the possibility of resolv-
ing human prehistory at an unprecedented level of detail
and precision. The data being produced can yield large
numbers of neutral markers for use in large-scale analyses.
Additional benefits will come from further sampling of
global genetic diversity, particularly in regions that are
poorly represented. Genetic estimates of the human evolu-
tionary timescale will continue to improve with progress
in identifying reliable calibrations for the human molecu-
lar clock, such as dates based on Paleolithic archaeology
or biogeography. Genomic analysis of ancient human
specimens represents a particularly promising area of
research and has the potential to transform our understand-
ing of the evolutionary origins of our species.
Bibliography
Anderson, S., Bankier, A. T., Barrell, B. G., de Bruijn, M. H. L.,
Coulson, A. R., Drouin, J., Eperon, I. C., Nierlich, D. P., Roe,
B. A., Sanger, F., Schreier, P. H., Smith, A. J. H., Staden, R.,
and Young, I. G., 1981. Sequence and organization of the human
mitochondrial genome. Nature, 290, 457–465.
Balloux, F., 2010. The worm in the fruit of the mitochondrial DNA
tree. Heredity, 104, 419–420.
Endicott, P., Ho, S. Y. W., Metspalu, M., and Stringer, C., 2009.
Evaluating the mitochondrial timescale of human evolution.
Trends in Ecology and Evolution, 24, 515–521.
Endicott, P., Ho, S. Y. W., and Stringer, C., 2010. Using genetic evi-
dence to evaluate four palaeoanthropological hypotheses for the
timing of Neanderthal and modern human origins. Journal of
Human Evolution, 59, 87–95.
Goodman, M., 1985. Rates of molecular evolution: the hominoid
slowdown. BioEssays, 3, 9–14.
Henn, B. M., Gignoux, C. R., Feldman, M. W., and Mountain, J. L.,
2009. Characterizing the time dependency of human mitochon-
drial DNA mutation rate estimates. Molecular Biology and Evo-
lution, 26, 217–230.
Ho, S. Y. W., Lanfear, R., Bromham, L., Phillips, M. J., Soubrier, J.,
Rodrigo, A. G., and Cooper, A., 2011. Time-dependent rates of
molecular evolution. Molecular Ecology, 20, 3087–3101.
Howell, N., Smejkal, C. B., Mackey, D. A., Chinnery, P. F.,
Turnbull, D. M., and Herrnstadt, C., 2003. The pedigree rate of
sequence divergence in the human mitochondrial genome: there
is a difference between phylogenetic and pedigree rates. Ameri-
can Journal of Human Genetics, 72, 659–670.
International Human Genome Consortium, 2001. Initial sequencing
and analysis of the human genome. Nature, 409, 860–921.
Prüfer, K., Racimo, F., Patterson, N., Jay, F., Sankararaman, S.,
Sawyer, S., Heinze, A., Renaud, G., Sudmant, P. H., de Filippo,
C., Li, H., Mallick, S., Dannemann, M., Fu, Q., Kircher, M.,
Kuhlwilm, M., Lachmann, M., Meyer, M., Ongyerth, M.,
Siebauer, M., Theunert, C., Tandon, A., Moorjani, P., Pickrell,
J., Mullikin, J. C., Vohr, S. H., Green, R. E., Hellmann, I., John-
son, P. L., Blanche, H., Cann, H., Kitzman, J. O., Shendure, J.,
Eichler, E. E., Lein, E. S., Bakken, T. E., Golovanova, L. V.,
Doronichev, V. B., Shunkov, M. V., Derevianko, A. P., Viola,
B., Slatkin, M., Reich, D., Kelso, J., and Pääbo, S., 2014. The
complete genome sequence of a Neanderthal from the Altai
Mountains. Nature, 505, 43–49.
Reich, D., Green, R. E., Kircher, M., Krause, J., Patterson, N.,
Durand, E. Y., Viola, B., Briggs, A. W., Stenzel, U., Johnson,
P. L. F., Maricic, T., Good, J. M., Marques-Bonet, T., Alkan,
C., Fu, Q., Mallick, S., Li, H., Meyer, M., Eichler, E. E.,
Stoneking, M., Richards, M., Talamo, S., Shunkov, M. V.,
Derevianko, A. P., Hublin, J.-J., Kelso, J., Slatkin, M., and
Pääbo, S., 2010. Genetic history of an archaic hominin group
from Denisova Cave in Siberia. Nature, 468, 1053–1060.
 MOLECULAR CLOCKS, RELAXED VARIANT
Reich, D., Patterson, N., Kircher, M., Delfin, F., Nandineni, M. R.,
Pugach, I., Ko, A. M.-S., Ko, Y.-C., Jinam, T. A., Phipps, M. E.,
Saitou, N., Wollstein, A., Kayser, M., Pääbo, S., and Stoneking,
M., 2011. Denisova admixture and the first modern human dis-
persals into Southeast Asia and Oceania. American Journal of
Human Genetics, 89, 516–528.
Roach, J. C., Glusman, G., Smith, A. F. A., Huff, C. D., Hubley, R.,
Shannon, P. T., Rowen, L., Pant, K. P., Goodman, N., Bamshad,
M., Shendure, J., Drmanac, R., Jorde, L. B., Hood, L., and Galas,
D. J., 2010. Analysis of genetic inheritance in a family quartet by
whole-genome sequencing. Science, 328, 636–639.
Sarich, V. M., and Wilson, A. C., 1967. Immunological time scale
for hominid evolution. Science, 158, 1200–1203.
Scally, A., and Durbin, R., 2012. Revising the human mutation rate:
implications for understanding human evolution. Nature
Reviews Genetics, 13, 745–753.
Shapiro, B., and Hofreiter, M., 2010. Analysis of ancient human
genomes. BioEssays, 32, 388–391.
Soares, P., Ermini, L., Thomson, N., Mormina, M., Rito, T., Röhl,
A., Salas, A., Oppenheimer, S., Macaulay, V., and Richards,
M. B., 2009. Correcting for purifying selection: an improved
human mitochondrial molecular clock. American Journal of
Human Genetics, 84, 740–759.
Steiper, M. E., Young, N. M., and Sukrarna, T. Y., 2004. Genomic
data support the hominoid slowdown and an early Oligocene
estimate for the hominoid–cercopithecoid divergence. Proceed-
ings of the National Academy of Sciences of the USA,
101, 17021–17026.
Torroni, A., Achilli, A., Macaulay, V., Richards, M., and Bandelt,
H.-J., 2006. Harvesting the fruit of the human mtDNA tree.
Trends in Genetics, 22, 339–345.
Welch, J. J., and Bromham, L., 2005. Molecular dating when rates
vary. Trends in Ecology and Evolution, 20, 320–327.
Zuckerkandl, E., and Pauling, L., 1962. Molecular disease, evolu-
tion and genic heterogeneity. In Kasha, M., and Pullman,
B. (eds.), Horizons in Biochemistry. New York: Academic Press,
pp. 189–225.
Zuckerkandl, E., and Pauling, L., 1965. Evolutionary divergence
and convergence in proteins. In Bryson, V., and Vogel, H. J.
(eds.), Evolving Genes and Proteins. New York: Academic
Press, pp. 97–166.
Cross-references
Gene Sequencing
Molecular Clocks
Molecular Clock Calibration
Molecular Clocks, Relaxed Variant
Molecular Dating of Evolutionary Events
Molecular Rate Variation (Molecular Clocks)
Polymerase Chain Reaction DNA Amplification
MOLECULAR CLOCKS, RELAXED VARIANT
Simon Y. W. Ho
School of Biological Sciences, University of Sydney,
Sydney, NSW, Australia
Definition
Relaxed molecular clock. A statistical model of molecular
evolution that allows the evolutionary rate to vary among
591
organisms. Relaxed molecular clocks can be used to esti-
mate rates and timescales of evolution using data from
DNA or proteins.
Introduction
Evolutionary timescales can be estimated from genetic
data using molecular clocks. The molecular clock hypoth-
esis, developed in the early 1960s (Zuckerkandl and
Pauling, 1962, 1965), predicts a constant rate of evolution-
ary change among organisms. A natural consequence of
this is a linear accumulation of genetic change over time,
meaning that the genetic difference between two species
is proportional to the time since they diverged from their
most recent common ancestor. In some cases, the age of
this divergence event can be inferred from independent
data (such as fossil evidence), making it possible to esti-
mate the rate of molecular evolution. If this rate is assumed
to be constant among lineages (i.e., a “strict” or “global”
molecular clock), we can use genetic data to estimate the
divergence times for other species. Such analyses are nor-
mally done in a phylogenetic framework, which takes into
account the evolutionary relationships among the organ-
isms being studied. Most genetic dating methods assume
that a reliable estimate of these evolutionary relationships
is available.
Evidence of departures from clocklike behavior
became apparent soon after the molecular clock was pro-
posed (e.g., Laird et al., 1969). It is now generally
accepted that genetic data rarely conform to a strict-clock
model. A variety of statistical tests have revealed signifi-
cant rate heterogeneity among lineages, even among
closely related species. From a biological perspective, it
is unsurprising to see such variation in molecular rates.
Different species vary in body size, longevity, body tem-
perature, generation time, and other heritable traits. Some
of these factors have been shown to affect rates of molec-
ular evolution (Bromham, 2009). Even within species,
such variation can be seen among populations. For exam-
ple, different human populations have experienced differ-
ing demographic histories and are subject to
heterogeneous environmental stresses, such that mutation
rates vary among human lineages (Henn et al., 2009).
For several decades, the strict clock provided the only
statistical framework for estimating evolutionary rates
and timescales using genetic data. One of the first
approaches for dealing with rate variation among lineages
was to remove any data that violated the assumption of
rate constancy (Li and Tanimura, 1987; Takezaki et al.,
1995). However, it is preferable to avoid discarding poten-
tially valuable data and to model rate variation explicitly,
particularly if it is an integral property of the molecular
evolutionary process (Gillespie and Langley, 1979). Vari-
ous methods have been developed to account for variation
in molecular rates. They can be broadly divided into local
molecular clocks and relaxed molecular clocks, which dif-
fer in how they assign rates to branches in the phylogeny
(Figure 1). Genetic dating methods that can accommodate
592 MOLECULAR CLOCKS, RELAXED VARIANT

Molecular Clocks, Relaxed Variant, Figure 1 Three different classes of molecular clocks are available. The thickness of each line
reflects the magnitude of the evolutionary rate along a branch. Strict or global clocks assume a constant rate throughout the tree.
Local clocks allow a limited number of rates throughout the tree. In this example, the left side of the tree has a different evolutionary
rate from the right side of the tree. Relaxed clocks allow each branch in the tree to have a distinct evolutionary rate.

such rate variation have been the subject of several
reviews (Renner, 2005; Welch and Bromham, 2005;
Rutschmann, 2006).
Local molecular clocks
Local molecular clocks assume that evolutionary rates vary
among branches in the phylogeny but that this variation can
be described using only a few discrete rates. In many cases,
this would be considered a reasonable assumption because
we expect closely related species to evolve at similar rates.
As a result of common ancestry, species tend to share many
biological features with their relatives, including heritable
traits that influence rates of evolution. Local-clock methods
permit different molecular clocks to exist in different groups
of organisms. For example, one might distinguish between
rates in baleen whales (mysticetes) and toothed whales
(odontocetes), because the latter tend to evolve much more
rapidly (Dornburg et al., 2012).
Local-clock methods retain some of the advantages of
a strict molecular clock, without the restrictive assumption
of a single rate of evolution across all lineages. However,
there are significant difficulties in implementing local-
clock methods, particularly in choosing the number of
local clocks and determining how they are assigned
among the species being studied. Recently, methods have
been developed to conduct some of these steps objectively
(Drummond and Suchard, 2010; Paradis, 2013), but their
performance has not yet been examined comprehensively.
Rate-smoothing methods
Given that the rate of evolution is likely to be influenced
by heritable traits, we can also treat the rate as
a biological character that evolves over time (Gillespie,
1991). Consequently, we would expect rates to be similar,
but not necessarily identical, between neighboring
branches in the phylogeny. One way of putting this
assumption into practice is to penalize large changes in
rates between neighboring branches. A number of such
“rate-smoothing” methods have been developed, includ-
ing nonparametric rate-smoothing (Sanderson, 1997) and
penalized likelihood (Sanderson, 2002).
The nonparametric rate-smoothing method operates by
minimizing an objective function based on the sum of rate
changes between branches in the phylogeny (Sanderson,
1997). In penalized likelihood, rate changes between
branches incur a penalty (Sanderson, 2002). The weight
of this penalty is governed by a smoothing parameter,
the value of which can be optimized objectively. Both of
these methods assume that the phylogeny is known or
has been estimated reliably.
Bayesian relaxed clocks
Some relaxed-clock methods use an explicit model of rate
variation among branches. These models are statistical
rather than mechanistic, in that they model rate variation
without making any statements about the causes of this
variation. Relaxed-clock models tend to be parameter rich
and have been implemented in Bayesian methods for phy-
logenetic analysis. The models of rate change can be
broadly divided into two classes, based on whether evolu-
tionary rates are assumed to be correlated between related
lineages (Ho, 2009).
In autocorrelated models of rate change, the rate is
assumed to “evolve” gradually along branches of the tree
(Thorne et al., 1998; Aris-Brosou and Yang, 2002). The rate
along a descendent branch is drawn from a distribution cen-
tered on the rate in its ancestral branch. As a consequence,
neighboring branches are likely to have similar evolution-
ary rates. Various parametric distributions have been used
in autocorrelated models of rate change, including lognor-
mal and exponential distributions. The statistical properties
of some of these models have been questioned (Welch et al.,
2005; Lepage et al., 2006).
In uncorrelated models of rate change, the rate along
each branch in the phylogeny bears no a priori similarity
 MOLECULAR DATING OF EVOLUTIONARY EVENTS
to the rates along its neighboring branches (Drummond
et al., 2006). However, the rates across the tree are drawn
from a single underlying distribution, such as a lognormal,
exponential, or gamma distribution.
Models of autocorrelated and uncorrelated rates are
probably applicable to different data sets (Ho, 2009).
When comparing closely related species, much of the rate
variation among lineages can probably be explained by
differences in biological characteristics. When the analy-
sis involves distantly related organisms, however, it is pos-
sible that traces of rate autocorrelation break down.
Various statistical methods can be used to choose the
best-fitting model of rate variation among lineages for
a given data set (Baele et al., 2013; Paradis, 2013).
Conclusions
Relaxed molecular clocks make it possible to estimate
evolutionary rates and timescales from genetic data even
when rates vary among lineages. Various relaxed-clock
methods have been developed, with statistical models
designed to capture and account for heterogeneity in evo-
lutionary rates. In view of the growing understanding of
the factors driving variation in molecular rates, the next
step is to incorporate biological knowledge into models
of rate variation.
Bibliography
Aris-Brosou, S., and Yang, Z., 2002. Effects of models of rate evo-
lution on estimation of divergence dates with special reference to
the metazoan 18S ribosomal RNA phylogeny. Systematic Biol-
ogy, 51, 703–714.
Baele, G., Li, W. L. S., Drummond, A. J., Suchard, M. A., and
Lemey, P., 2013. Accurate model selection of relaxed molecular
clocks in Bayesian phylogenetics. Molecular Biology and Evo-
lution, 30, 239–243.
Bromham, L., 2009. Why do species vary in their rate of molecular
evolution? Biology Letters, 5, 401–404.
Dornburg, A., Brandley, M. C., McGowen, M. R., and Near, T. J.,
2012. Relaxed clocks and inferences of heterogeneous patterns
of nucleotide substitution and divergence time estimates across
whales and dolphins (Mammalia: Cetacea). Molecular Biology
and Evolution, 29, 721–736.
Drummond, A. J., and Suchard, M. A., 2010. Bayesian random
local clocks, or one rate to rule them all. BMC Biology, 8, 114.
Drummond, A. J., Ho, S. Y. W., Phillips, M. J., and Rambaut, A.,
2006. Relaxed phylogenetics and dating with confidence. PLOS
Biology, 4, e88.
Gillespie, J. H., 1991. The Causes of Molecular Evolution. New
York: Oxford University Press.
Gillespie, J. H., and Langley, C. H., 1979. Are evolutionary rates
really variable? Journal of Molecular Evolution, 13, 27–34.
Henn, B. M., Gignoux, C. R., Feldman, M. W., and Mountain, J. L.,
2009. Characterizing the time dependency of human mitochon-
drial DNA mutation rate estimates. Molecular Biology and Evo-
lution, 26, 217–230.
Ho, S. Y. W., 2009. An examination of phylogenetic models of sub-
stitution rate variation among lineages. Biology Letters,
5, 421–424.
Laird, C. D., McConaughy, B. L., and McCarthy, B. J., 1969. Rate
of fixation of nucleotide substitutions in evolution. Nature,
224, 149–154.
593
Lepage, T., Lawi, S., Tupper, P., and Bryant, D., 2006. Continuous
and tractable models for the variation of evolutionary rates.
Mathematical Biosciences, 199, 216–233.
Li, W. H., and Tanimura, M., 1987. The molecular clock runs more
slowly in man than in apes and monkeys. Nature, 326, 93–96.
Paradis, E., 2013. Molecular dating of phylogenies by likelihood
methods: a comparison of models and a new information crite-
rion. Molecular Phylogenetics and Evolution (67 436–444).
Renner, S., 2005. Relaxed molecular clocks for dating historical
plant dispersal events. Trends in Plant Science, 10, 550–558.
Rutschmann, F., 2006. Molecular dating of phylogenetic trees:
a brief review of current methods that estimate divergence times.
Diversity and Distributions, 12, 35–48.
Sanderson, M. J., 1997. A nonparametric approach to estimating
divergence times in the absence of rate constancy. Molecular
Biology and Evolution, 14, 1218–1231.
Sanderson, M. J., 2002. Estimating absolute rates of molecular evo-
lution and divergence times: a penalized likelihood approach.
Molecular Biology and Evolution, 19, 101–109.
Takezaki, N., Rzhetsky, A., and Nei, M., 1995. Phylogenetic test of
the molecular clock and linearized trees. Molecular Biology and
Evolution, 12, 823–833.
Thorne, J. L., Kishino, H., and Painter, I. S., 1998. Estimating the
rate of evolution of the rate of molecular evolution. Molecular
Biology and Evolution, 15, 1647–1657.
Welch, J. J., and Bromham, L., 2005. Molecular dating when rates
vary. Trends in Ecology & Evolution, 20, 320–327.
Welch, J. J., Fontanillas, E., and Bromham, L., 2005. Molecular
dates for the “Cambrian Explosion”: the influence of prior
assumptions. Systematic Biology, 54, 672–678.
Zuckerkandl, E., and Pauling, L., 1962. Molecular disease, evolu-
tion and genetic heterogeneity. In Kasha, M., and Pullman, B.
(eds.), Horizons in Biochemistry. New York: Academic,
pp. 189–225.
Zuckerkandl, E., and Pauling, L., 1965. Evolutionary divergence
and convergence in proteins. In Bryson, V., and Vogel, H. J.
(eds.), Evolving Genes and Proteins. New York: Academic,
pp. 97–166.
Cross-references
Amino Acid
Gene Sequencing
Molecular Clocks
Molecular Clock Calibration
Molecular Dating of Evolutionary Events
Molecular Rate Variation (Molecular Clocks)
Polymerase Chain Reaction DNA Amplification
MOLECULAR DATING OF EVOLUTIONARY EVENTS
David Duchene and Lindell Bromham
Centre for Macroevolution and Macroecology, Division
of Evolution, Ecology, and Genetics, Research School
of Biology, Australian National University, Canberra,
ACT, Australia
Definition
Inference of the age of evolutionary events using statisti-
cal analysis of rates of change of DNA or amino acid
sequences.
594 MOLECULAR DATING OF EVOLUTIONARY EVENTS
Introduction
Molecular dating is used in the biological sciences to esti-
mate the age of evolutionary events. Changes to DNA and
amino acid sequences accumulate continuously in the
genome over time, so comparing DNA sequences between
lineages allows us to estimate the time since they last
shared a common ancestor. However, the rate of change
varies across the genome and among species. So in order
to use molecular data to date evolutionary events, we need
a way of estimating the rate of change in genetic
sequences over time for any given dataset. Molecular dat-
ing requires a set of homologous genetic sequences (all
related by descent from a single ancestral sequence),
a method for inferring the number of changes that have
occurred during the evolution of these sequences, and cal-
ibrating information to estimate their rate of change.
An increasing variety of analytical tools are available
for molecular dating of evolutionary events. All assume
that the rate of genetic sequence evolution has some
degree of predictability, so that the age of lineage diver-
gence can be estimated from molecular data using
a model of evolutionary change and some information
about timing of evolutionary events (typically derived
from fossils, biogeography, pedigree, or ancestral genetic
sequences).
Methods for dating evolutionary events
Molecular dating methods begin by estimating the amount
of genetic difference that has occurred between sequences.
Then, assumptions about the rate of change are used to
infer the amount of time needed to explain that amount
of genetic difference. The difference between genetic
sequences is estimated using a model of the frequency of
each type of substitution. Models of molecular evolution
form the basis of methods to estimate the relatedness of
multiple sequences and the timing since their evolutionary
divergence (for more detail on the models, see Yang,
2006). These models can be simple and give an equal
probability to every type of substitution (Jukes and Can-
tor, 1969), or be complex and account for the attributes
of specific genetic datasets (e.g., Yang, 1994). For
instance, the mitochondrial ND6 genetic region in marine
mammals has virtually no transversions (changes from the
purines, A and G, to pyrimidines, C and T, or vice versa),
so it is appropriate to use parameter settings that model
a large number of transitions (changes from purine to
another purine, or pyrimidine to pyrimidine) (Duchêne
et al., 2011).
The simplest way to use genetic difference estimates to
infer the timing of evolutionary divergences is to select
“clocklike” data, where the amount of genetic difference
accumulates at the same rate in all lineages. There are
methods to detect significant variation in rates (Langley
and Fitch, 1973; Takezaki et al., 1995). For example, Like-
lihood Ratio Test can be used to ask whether data can be
best described by a single uniform rate, or a multiple rate
model (Brown and Yang, 2011). Importantly, while some
datasets may approximate uniform rates, clocklike behav-
ior cannot be generalized to a particular gene, lineage, or
taxonomic level.
Another way to account for rate variation is to have
multiple rate categories in the same analysis. For example,
sections of the dataset may have their own “local clock”
(Drummond and Suchard, 2010; Rambaut and Bromham,
1998; Yoder and Yang, 2000). Other methods account for
rate variation by “relaxing” the molecular clock constraint
on all branches, so every branch in a phylogeny has
a different rate (e.g., Drummond et al., 2006; Yang,
2007); for more on this topic, see Springer Reference arti-
cle Relaxed Molecular Clocks; for reviews on dating
methods, (see Magallón, 2004; Welch and Bromham,
2005; Rutschmann, 2006).
Calibrating the rate of substitutions
Some molecular dating analyses use an assumed rate of
substitutions to estimate dates. However, given that rates
of molecular evolution vary across the genome and
between lineages, it is preferable to use independent infor-
mation to calibrate rates of molecular evolution for each
dataset analyzed. The most common way of calibrating
rates is to use a known date of one or more divergence
events in the phylogeny.
The calibration must correspond to the earliest possible
age of a node in the phylogeny. The dates of divergence
events are rarely known with certainty, and the confidence
limits on calibrations can be very large. In the case of fossil
evidence, part of the uncertainty comes from the determina-
tion of the age of the fossil itself, drawn from stratigraphy
and isotopic composition of the fossil (Benton and
Donoghue, 2007). However, the exact relationship between
the fossil taxon and the divergence in the phylogeny is also
typically unknown (Sauquet et al., 2012). A fossil taxon
is unlikely to be identifiable as a member of a particular
lineage until some time after the origin of the lineage, when
key diagnostic characters have had time to evolve. So fossil
dates typically represent minimum ages of lineages: they
provide evidence that a lineage must have originated some
time before that date (Bromham et al., 1999).
One way to account for calibration uncertainty is to have
a probability distribution of node ages. A hard-bound can
define the latest or the earliest point where a divergence
may have occurred (also called the minimum age and the
maximum age, respectively). While a minimum age can
be defined by the age of a fossil known to occur on the lin-
eage in question, because the lineages must have originated
before the occurrence of the first fossil, special caution is
required when placing a maximum constraint because
it assumes with total confidence that a lineage was absent
before a given time (Benton and Donoghue, 2007;
Hug and Roger, 2007; Ho and Phillips, 2009).
To overcome the strict assumptions of hard-bounded
calibrations, an alternative is to use soft-bounded calibra-
tions, so a nonzero probability can be assigned to all ages
(Yang and Rannala, 2006). To define a soft-bounded
 MOLECULAR DATING OF EVOLUTIONARY EVENTS
calibration, the user must specify a probability distribution
for the date of the calibrating divergence. Given that the
confidence in particular date estimates is specific to cur-
rent knowledge, expert advice is often required to define
these calibrations. Broadly, a normally distributed calibra-
tion is conservative for fossil data that is imprecise or
equivocal. A normally distributed calibration can also be
used for biogeographic data that provides reasonable evi-
dence for causing a divergence event, or for calibrations
based on previous date estimation analyses (also called
secondary calibrations, Ho and Phillips, 2009). Lognor-
mal distributions are often appropriate for fossil data as
they provide a hard minimum bound, which is the age of
the fossil. They also allow a higher probability for the
divergence of a node at a point slightly older than the
age of the fossil. These distributions are also applied when
using previous date estimates from Bayesian analyses that
produced lognormal distributions.
Exponentially distributed calibrations can be used if
there is little evidence that an event occurred at any point
before the fossil, or if the fossil is likely to be very close
to the divergence event (Ho and Phillips, 2009). To choose
the values of the parameters of these distributions, the user
must be well informed about the sources of error in the cal-
ibration. Poor calibrations may be useful if their uncertainty
can be quantified, but if only poor calibrations are used, then
the uncertainty in date estimates will be significant.
Summary and conclusions
The field of molecular dating of evolutionary events can
be divided by the methods to deal with variation in the rate
of molecular substitutions. To date evolutionary events
using molecular data, it is necessary to evaluate and
choose a method to model molecular evolution. It is also
necessary to find information to calibrate rates of substitu-
tions, and all analyses must account for calibration uncer-
tainty. Although modern software for this purpose can be
deceptively simple, there is a plethora of considerations
required before performing molecular dating analyses. In
particular, Bayesian methods for molecular dating can
involve complex assumptions about the model of molecu-
lar evolution and arguably are not entirely understood.
Methods in place to compare and choose the most appro-
priate dating schemes are computationally demanding
and may be severely biased (Lartillot and Philippe,
2006; Xie et al., 2011; Baele et al., 2012). However, with
appropriate information and a clear understanding of the
uncertainty inherent in molecular date estimates, molecu-
lar dating is a powerful tool that can be extended to study
the realms of evolution, ecology, biogeography, and past
population processes.
Bibliography
Baele, G., Lemey, P., Bedford, T., et al., 2012. Improving the accu-
racy of demographic and molecular clock model comparison
while accommodating phylogenetic uncertainty. Molecular Biol-
ogy and Evolution, 29, 2157–2167, doi:10.1093/molbev/
mss084.
595
Benton, M. J., and Donoghue, P. C. J., 2007. Paleontological evi-
dence to date the tree of life. Molecular Biology and Evolution,
24, 26–53, doi:10.1093/molbev/msl150.
Bromham, L., Phillips, M., and Penny, D., 1999. Growing up with
dinosaurs: molecular dates and the mammalian radiation. Trends
in Ecology & Evolution, 14, 113–118.
Brown, R. P., and Yang, Z., 2011. Rate variation and estimation of
divergence times using strict and relaxed clocks. BMC Evolu-
tionary Biology, 11, 271, doi:10.1186/1471-2148-11-271.
Drummond, A. J., and Suchard, M. A., 2010. Bayesian random
local clocks, or one rate to rule them all. BMC Evolutionary Biol-
ogy, 8, 114, doi:10.1186/1741-7007-8-114.
Drummond, A. J., Ho, S. Y. W., Phillips, M. J., and Rambaut, A.,
2006. Relaxed phylogenetics and dating with confidence. PLoS
Biology, 4, e88, doi:10.1371/journal.pbio.0040088.
Duchêne, S., Archer, F. I., Vilstrup, J., et al., 2011. Mitogenome
phylogenetics: the impact of using single regions and
partitioning schemes on topology, substitution rate and diver-
gence time estimation. PloS One, 6, e27138, doi:10.1371/jour-
nal.pone.0027138.
Ho, S. Y. W., and Phillips, M. J., 2009. Accounting for calibration
uncertainty in phylogenetic estimation of evolutionary diver-
gence times. Systematic Biology, 58, 367–380, doi:10.1093/
sysbio/syp035.
Ho, S. Y. W., Phillips, M. J., Drummond, A. J., and Cooper, A.,
2005. Accuracy of rate estimation using relaxed-clock models
with a critical focus on the early metazoan radiation. Molecular
Biology and Evolution, 22, 1355–1363, doi:10.1093/molbev/
msi125.
Hug, L. A., and Roger, A. J., 2007. The impact of fossils and taxon
sampling on ancient molecular dating analyses. Molecular Biol-
ogy and Evolution, 24, 1889–1897, doi:10.1093/molbev/
msm115.
Jukes, T. H., and Cantor, C. R., 1969. Evolution of protein mole-
cules. In Munro, H. (ed.), Mammalian Protein Metabolism.
New York: Academic, pp. 21–132.
Langley, C. H., and Fitch, W. M., 1973. The Constancy of Evolu-
tion: A Statistical Analysis of a and b Haemoglobins, Cyto-
chrome c, and Fibrinopeptide A. Genetic Structure of
Populations. Honolulu, HI: University of Hawaii Press,
pp. 246–262.
Lartillot, N., and Philippe, H., 2006. Computing Bayes factors using
thermodynamic integration. Systematic Biology, 55, 195–207,
doi:10.1080/10635150500433722.
Magallón, S., 2004. Dating lineages: molecular and paleontological
approaches to the temporal framework of clades. International
Journal of Plant Sciences, 165, S7–S21.
Rambaut, A., and Bromham, L., 1998. Estimating divergence dates
from molecular sequences. Molecular Biology and Evolution,
15, 442–448.
Rutschmann, F., 2006. Molecular dating of phylogenetic trees:
a brief review of current methods that estimate divergence times.
Diversity and Distributions, 12, 35–48, doi:10.1111/j.1366-
9516.2006.00210.x.
Sauquet, H., Ho, S. Y. W., Gandolfo, M. A., et al., 2012. Testing the
impact of calibration on molecular divergence times using
a fossil-rich group: the case of Nothofagus (Fagales). Systematic
Biology, 61, 289–313, doi:10.1093/sysbio/syr116.
Takezaki, N., Rzhetsky, A., and Nei, M., 1995. Phylogenetic test of
the molecular clock and linearized trees. Molecular Biology and
Evolution, 12, 823–833.
Welch, J. J., and Bromham, L., 2005. Molecular dating when rates
vary. Trends in Ecology & Evolution, 20, 320–327,
doi:10.1016/j.tree.2005.02.007.
Xie, W., Lewis, P. O., Fan, Y., et al., 2011. Improving marginal like-
lihood estimation for Bayesian phylogenetic model selection.
Systematic Biology, 60, 150–160, doi:10.1093/sysbio/syq085.
596 MOLECULAR RATE VARIATION (MOLECULAR CLOCKS)
Yang, Z., 1994. Estimating the pattern of nucleotide substitution.
Journal of Molecular Evolution, 39, 105–111.
Yang, Z., 2006. Computational Molecular Evolution. Oxford:
Oxford University Press.
Yang, Z., 2007. PAML 4: phylogenetic analysis by maximum like-
lihood. Molecular Biology and Evolution, 24, 1586–1591,
doi:10.1093/molbev/msm088.
Yang, Z., and Rannala, B., 2006. Bayesian estimation of species
divergence times under a molecular clock using multiple fossil
calibrations with soft bounds. Molecular Biology and Evolution,
23, 212–226, doi:10.1093/molbev/msj024.
Yoder, A. D., and Yang, Z., 2000. Estimation of primate speciation
dates using local molecular clocks. Molecular Biology and Evo-
lution, 17, 1081–1090.
MOLECULAR RATE VARIATION (MOLECULAR
CLOCKS)
Robert Lanfear
Ecology, Evolution, and Genetics, The Australian
National University, Canberra, ACT, Australia
Definition
The rate of molecular evolution is the rate at which substi-
tutions accumulate in an organism’s genome. Rates of
molecular evolution can vary dramatically, for example,
the rate of molecular evolution in some viruses is around
one substitution per base pair per 1,000 years (Pagn
et al., 2010), while in mammals the rate is around one sub-
stitution per base pair per 1,000 million years (Nabholz
et al., 2008). Even closely related plants and animals can
have rates of molecular evolution that vary by more than
an order of magnitude (Thomas et al., 2006; Smith and
Donoghue, 2008; Welch et al., 2008; Lanfear et al.,
2013). Accounting for molecular rate variation is impor-
tant in methods that use molecular sequence data to date
the divergences between species (molecular dating).
There has been a great deal of research into the causes
of variation in rates of molecular evolution. This research
has identified a range of factors which may cause variation
in the rate of molecular evolution (Bromham, 2011), and
these factors can be broken up into those that affect the
mutation rate and those that affect the rate at which new
mutations are fixed in a population (the fixation rate).
An increase in either the mutation rate or the fixation rate
can potentially increase the overall rate of molecular evo-
lution, although there are some situations in which these
increases will leave the rate of molecular evolution
unchanged (de Visser et al., 1999).
Many factors have been suggested to increase mutation
rates in DNA sequences, including temperature, UV radi-
ation, oxygen radicals, recombination, genome copying,
and natural selection (Martin and Palumbi, 1993; Davies
et al., 2004; Wright et al., 2006; Lanfear et al., 2007;
Thomas et al., 2010; Hodgkinson and Eyre-Walker,
2011; Wright et al., 2011; Lourenço et al., 2013; Lanfear
et al., 2013). In principle, variation in any of these factors
could cause variation in the rates of molecular evolution
between species, although the evidence for this is weak
for UV radiation and oxygen radicals in particular (Davies
et al., 2004; Lanfear et al., 2007; Joyner-Matos et al.,
2011; Lanfear et al., 2013). However, there is good evi-
dence that the many mutations occur during genome-
copying events (meiosis and mitosis) and that variation
in long-term rates of genome copying may explain much
of the variation in the rates of molecular evolution between
species (Smith and Donoghue, 2008; Welch et al., 2008;
Thomas et al., 2010; Lanfear et al., 2013). Intriguingly, it
has also been suggested that natural selection may play
a role in determining mutation rates and rates of molecular
evolution, either through selection for increased longevity
(Nabholz et al., 2008; Welch et al., 2008; Galtier et al.,
2009) or through the reduced efficacy of natural selection
in small populations (Lynch, 2010, 2011).
Anything that increases the fixation rate of new muta-
tions can also increase the substitution rate. Two factors
combine to determine the fixation rate of a new mutation:
the effect of that mutation on the carrier’s fitness and the
effective population size of the host population. As fitness
effects become more beneficial and effective population
sizes get larger, fixation rates increase (Woolfit, 2009).
This is because a population’s effective population size
determines the balance of power between natural selection
and genetic drift. The larger the effective population size,
the stronger natural selection is relative to genetic drift,
and the faster beneficial mutations are fixed in the popula-
tion, and deleterious mutations removed. Because we
expect most mutations to be deleterious, it is generally
thought that organisms with smaller effective population
size should have faster rates of molecular evolution. Some
studies have supported this idea and have shown that
organisms with smaller effective population sizes tend to
evolve more quickly, at least by some measure of the rate
of molecular evolution (Moran, 1996; Woolfit and
Bromham, 2003, 2005). However, other results suggest
that the relationship between population size and the rate
of molecular evolution can be much more complex than
this simple model suggests (Charlesworth and Eyre-
Walker, 2007; Wright et al., 2009).
We have a fairly good understanding of the most impor-
tant causes of variation in rates of molecular evolution.
But despite this, we can often explain less than half of
the variation that we can measure, suggesting that we still
have a lot to learn. A better understanding of molecular
rate variation is important because it can help improve
molecular dating methods. Perhaps the best example of
this is the development of new methods which leverage
what we know about the causes and correlates of molecu-
lar rates to directly improve our estimates of divergence
dates from DNA sequence data (Lartillot and Poujol,
2011; Lartillot and Delsuc, 2012). As our understanding
of molecular rate variation continues to increase, we will
be able to continue to improve the accuracy of molecular
dating methods.
 MOLLUSCS, FORAMINIFERA, AND OTHER CARBONATE FOSSILS 597

Bibliography Thomas, J. A., Welch, J. J., Woolfit, M., and Bromham, L., 2006.
There is no universal molecular clock for invertebrates, but rate
Bromham, L., 2011. The genome as a life-history character: why
variation does not scale with body size. Proceedings of the
rate of molecular evolution varies between mammal species.
National Academy of Sciences of the United States of America,
Philosophical Transactions of the Royal Society, B: Biological
103, 7366–7371.
Sciences, 366, 2503–2513.
Thomas, J. A., Welch, J. J., Lanfear, R., and Bromham, L., 2010.
Charlesworth, J., and Eyre-Walker, A., 2007. The other side of the
A generation time effect on the rate of molecular evolution in
nearly neutral theory, evidence of slightly advantageous back-
invertebrates. Molecular Biology and Evolution, 27, 1173–1180.
mutations. Proceedings of the National Academy of Sciences of
Welch, J. J., Bininda-Emonds, O. R. P., and Bromham, L., 2008.
the United States of America, 104, 16992–16997.
Correlates of substitution rate variation in mammalian protein-
Davies, T. J., Savolainen, V., Chase, M. W., Moat, J., and
coding sequences. BMC Evolutionary Biology, 8, 53.
Barraclough, T. G., 2004. Environmental energy and evolution-
Woolfit, M., 2009. Effective population size and the rate and pattern
ary rates in flowering plants. Proceedings of the Biological Sci-
of nucleotide substitutions. Biology Letters, 5, 417–420.
ences, 271, 2195–2200.
Woolfit, M., and Bromham, L., 2003. Increased rates of sequence
de Visser, J. A. G. M., Zeyl, C. W., Gerrish, P. J., Blanchard, J. L.,
evolution in endosymbiotic bacteria and fungi with small effec-
and Lenski, R. E., 1999. Diminishing returns from mutation sup-
tive population sizes. Molecular Biology and Evolution,
ply rate in asexual populations. Science, 283, 404–406.
20, 1545–1555.
Galtier, N., Blier, P. U., and Nabholz, B., 2009. Inverse relationship
Woolfit, M., and Bromham, L., 2005. Population size and molecular
between longevity and evolutionary rate of mitochondrial pro-
evolution on islands. Proceedings of the Biological Sciences,
teins in mammals and birds. Mitochondrion, 9, 51–57.
272, 2277–2282.
Hodgkinson, A., and Eyre-Walker, A., 2011. Variation in the muta-
Wright, S., Keeling, J., and Gillman, L., 2006. The road from Santa
tion rate across mammalian genomes. Nature Reviews. Genetics,
Rosalia: a faster tempo of evolution in tropical climates. Pro-
12, 756–766.
ceedings of the National Academy of Sciences of the United
Joyner-Matos, J., Bean, L. C., Richardson, H. L., Sammeli, T., and
States of America, 103, 7718–7722.
Baer, C. F., 2011. No evidence of elevated germline mutation
Wright, S. D., Gillman, L. N., Ross, H. A., and Keeling, D. J., 2009.
accumulation under oxidative stress in Caenorhabditis elegans.
Slower tempo of microevolution in island birds: implications for
Genetics, 189, 1439–1447.
conservation biology. Evolution, 63, 2275–2287.
Lanfear, R., Thomas, J. A., Welch, J. J., Brey, T., and Bromham, L.,
Wright, S. D., Ross, H. A., Keeling, D. J., McBride, P., and Gillman,
2007. Metabolic rate does not calibrate the molecular clock. Pro-
L. N., 2011. Thermal energy and the rate of genetic evolution in
ceedings of the National Academy of Sciences of the United
marine fishes. Evolutionary Ecology, 25, 525–530.
States of America, 104, 15388–15393.
Lanfear, R., Ho, S. Y. W., Jonathan Davies, T., Moles, A. T.,
Aarssen, L., Swenson, N. G., Warman, L., Zanne, A. E., and
Allen, A. P., 2013. Taller plants have lower rates of molecular
evolution. Nature Communications, 4, 1879.
Lartillot, N., and Delsuc, F., 2012. Joint reconstruction of diver- MOLLUSCS, FORAMINIFERA, AND OTHER
gence times and life-history evolution in placental mammals
using a phylogenetic covariance model. Evolution, 66,
CARBONATE FOSSILS
1773–1787.
Lartillot, N., and Poujol, R., 2011. A phylogenetic model for inves- Bonnie A. B. Blackwell
tigating correlated evolution of substitution rates and continuous Department of Chemistry, Williams College,
phenotypic characters. Molecular Biology and Evolution, Williamstown, MA, USA
28, 729–744.
Lourenço, J. M., Glémin, S., Chiari, Y., Galtier, N., 2013. The deter-
minants of the molecular substitution process in turtles. Journal Definitions
of Evolutionary Biology, 26(1), 38–50. Mollusc: any member of the Phylum Mollusca. The
Lynch, M., 2010. Evolution of the mutation rate. Trends in Genetics, Mollusca form a diverse group of  85,000 extant species,
26, 345–352. including Bivalvia (clams, oysters, mussels, scallops,
Lynch, M., 2011. The lower bound to the evolution of mutation
rates. Genome Biology and Evolution, 3, 1107–1118. etc.), Gastropoda (snails, slugs), Cephalopoda (octopus,
Martin, A. P., and Palumbi, S. R., 1993. Body size, metabolic rate, squid, cuttlefish, nautilus, ammonoids, etc.),
generation time, and the molecular clock. Proceedings of the Polyplacophora (chitons), Sacophopoda (dentilids),
National Academy of Sciences of the United States of America, Rostroconchia (all extinct), Monoplacophora, and
90, 4087–4091. Aplacophora. In the shell, calcium carbonate, as aragonite,
Moran, N. A., 1996. Accelerated evolution and Muller’s rachet in low or high magnesium calcite, reinforces a conchiolin
endosymbiotic bacteria. Proceedings of the National Academy
of Sciences of the United States of America, 93, 2873–2878. (protein) and chitin matrix.
Nabholz, B., Glémin, S., and Galtier, N., 2008. Strong variations of Barnacle: any of several species of Arthropoda belonging
mitochondrial mutation rate across mammals – the longevity to the Subphylum Crustacea, Infraclass Cirripedia.
hypothesis. Molecular Biology and Evolution, 25, 120–130. Among the 1,220 or so species, most are encrusting filter-
Pagn, I., Firth, C., and Holmes, E. C., 2010. Phylogenetic analysis or suspension-feeding organisms. Although some live as
reveals rapid evolutionary dynamics in the plant RNA virus deep as 600 m, about 75 % live at depths < 100 m in
genus Tobamovirus. Journal of Molecular Evolution,
71, 298–307. marine ecosystems, and 25 % inhabit the intertidal zone.
Smith, S. A., and Donoghue, M. J., 2008. Rates of molecular evolu- Within the intertidal zone, species inhabit a very restricted
tion are linked to life history in flowering plants. Science, range of water depth, allowing them to be used to mark
322, 86–89. paleosealevels with some accuracy. Infamous for fouling
598 MOLLUSCS, FORAMINIFERA, AND OTHER CARBONATE FOSSILS
ships, docks, and intertidal rocks, barnacles have plates
that allow them to close off the internal shell to prevent
desiccation.
Foraminifera (abbreviation: foram): any member of the
Phylum Foraminifera, a group of marine amoeboid
Protista. The Forminifera comprise both extinct and extant
species that all have thin pseudopoda that form an external
net for gathering food. Many species are planktonic, but
several benthic species live on the deep ocean floor, and
a few species can tolerate brackish water.
Carbonate compensation depth (CCD): the depth at
which all carbonate fossils have dissolved from deep
marine sediment. The aragonite compensation depth
varies from that for calcite. Currently, the depths vary
from about 4,200 to 5,000 m, depending on the tempera-
ture, pressure, and the amount of dissolved CO2. In the
past, the depths have varied dramatically, especially
during the Quaternary.
Aragonite: a calcium carbonate (CaCO3) mineral found in
speleothems, travertines, carbonate cements in sedimen-
tary rocks, and live and fossilized shelled organisms, like
molluscs and barnacles. Aragonite often occurs as acicular
crystals in speleothems, travertine, and carbonate
cements, but in fossils, it may take other crystal forms.
Calcite: a calcium carbonate (CaCO3) mineral found in
some speleothems, travertines, carbonate cements in sedi-
mentary rocks, and live and fossilized shelled organisms,
like some molluscs. Calcite can occur as rhombohedral
crystals if allowed to grow uninhibited, but in fossils, it
may take other crystal forms.
Irradiation ramping: a technique used for very small sam-
ples in which only three to eight aliquots are used to build
a growth curve. After reading the spectra following the
first set of irradiations, about half of the aliquots are
reirradiated to higher doses and the spectra for all the ali-
quots are reanalyzed, using the aliquots that have not been
reirradiated to calibrate the spectra for all the newly irradi-
ated samples. Normally, the 0 dose and at least one other
aliquot are not reirradiated. The ramping can be repeated
as many times as necessary to build the growth curve. Nor-
mally, at least 10 different added doses are needed to build
a precise growth curve, but as many as 15–20 doses are
commonly used in ramping.
Ramped box model: a model for deriving time-averaged
cosmic and sedimentary doses for electron spin resonance
(ESR) dating first suggested by Deely et al. (2011). The
model uses a series of time-limited boxes whose end
points have fixed dose rates derived from geological, pale-
ontological, or biological criteria that define the sediment
cover thickness and water depth. Within the intervening
time period, the dose rates are assumed to change at a
constant rate.
Introduction
In marine and lacustrine sedimentary units, open-air
spring deposits, karst fissure fills, and caves and dating
molluscs, foraminifera, barnacles, or other carbonate
fossils found in the sediment (Table 1) can provide diverse
information for Quaternary archaeological, paleontologi-
cal, or geological studies. Electron spin resonance (ESR)
dates for molluscs and planktonic foraminifera have pro-
vided vital geological and paleontological information.
More recently, several researchers have used terrestrial
molluscs to develop sealevel curves, monitor tectonic
uplift, monitor water availability in the desert, and date
archaeological finds. New developments in the field
include applications for benthic foraminifera and
barnacles.
ESR dating issues
Although molluscs and some other carbonate fossils can
act as moderately open systems for U uptake, the minor
discordance between the measured 230Th/234U and
231
Pa/235U ratios suggests that most U uptake accom-
panies sedimentation early in their depositional histories.
Moreover, given the amount of U they normally absorb,
the difference between the EU (early uptake) and LU
(linear uptake) calculated ages for most molluscs and fora-
minifera is usually insignificant compared to their associ-
ated uncertainties.
Most ESR analyses on molluscs, foraminifera, barna-
cles, and other marine shells use the X band, but K band
has been applied to molluscan aragonite for dating
(Kinoshita et al., 2002). The techniques have also been
expanded to recognize irradiated shell for retrospective
accident dosimetry and dosimetric monitoring for the food
industry (e.g., Ikeya et al., 1984; Nakajima et al., 1993;
Raffi et al., 1996).
Aragonitic shells normally show five ESR peaks
(Figure 1; see also Blackwell, 1995, Table 2), but calcitic
shells have more complex spectra. For the peaks at
g ¼ 2.0018, 1.9976, and 2.0007, trap density is related
to Mg/Ca ratios (Mudelsee et al., 1992), which can change
with diagenesis, secondary mineralization, and fossiliza-
tion, making them unsuitable for dating some species.
Generally, either the peaks at g ¼ 2.0014 and 2.0007 in
calcitic shells or the peak at g ¼ 2.0007 in aragonitic shells
are the most reliable, but that must be tested for each spe-
cies individually, because complex peaks do occur and
peaks other than that at g ¼ 2.0007 may be light sensitive
in some species (Bartoll et al., 2000). Secondary mineral-
ization can cause interference that affects the accumulated
dose (AS) measurement and age calculation. Signal
lifetimes vary significantly depending on the peak and
species (e.g., see references in Blackwell, 1995, Table 2).
Some species show inflection points in their growth
curves, such as the peak at g ¼ 2.0018 in Anadara,
Katylesia, Dorax, and Phalium, making it difficult to
select an appropriate set of added doses for measuring
AS (e.g., Brumby and Yoshida, 1994; Shih et al., 2002).
Schellmann and Radtke (2001) advocated using a plateau
technique with 40–60 irradiation steps to maximize accu-
racy in the growth curves. They successfully applied that
technique to dating coastal marine units in Patagonia
Molluscs, Foraminifera, and Other Carbonate Fossils, Table 1 Carbonate samples datable by ESR
Minimum Effects from diagenesis, secondary
sample for mineralization, or cementation
standard
Zeroing Isochrons Ramping Species Best type or ESRc,d (mg/ Signal Inaccurate Incomplete Grinding
Sample type Minerals1 req’d? possible?a possible?b effects?a species? subsample) intensity Interference? ages? zeroing? effects?

Molluscs cct, argt No Unknown Up to 6 Yes Terrestrial or 50–100e,f May increase Likely Possibly N/a Interference
rampings marine
e,f
Coral, argt No Theoret Up to 6 Yes Densest, 50–100 May increase Likely Possibly N/a Interference
echinoderms rampings uneroded
e,f
Foraminifera, cct, argt No Unknown Up to 10 Unknown Planktonic 100–200 May increase Likely Possibly N/a Interference
both rampings or
planktonic & benthonic
benthonic
Ostracodes cct, argt No Unknown Unknown Unknown Unknown 100–200e,f May increase Likely Possibly N/a Interference
Cuttlefish argt Unknown Unknown Unknown Unknown Unknown 100–200e,f Unknown Unknown Unknown Unknown Interference
cuttlebone
Ratite egg cct No Theoret Unknown Unknown Unknown 50–100 May increase Likely Possibly N/a Interference
shells
g,h
Travertine, cct, argt No Yes Unknown N/a Densest 50–100 May increase Likely Possibly N/a Interference
speleothem
g,h
Calcrete, cct No Unknown Unknown N/a Densest 50–100 May increase Likely Possibly Yes, if Interference
caliche, clasts
stromatolites CO3 also
Authigenic cct, argt No Unknown Unknown N/a Densest 100 May increase Likely Possibly N/a Interference
cement
a
Abbreviations: cct calcite, CO3 carbonate, argt aragonite, theoret theoretically
b
Irradiation ramping technique involves reirradiating some aliquots, but can take up to 2 years to complete for 8–10 rampings. The number indicates the maximum number of
rampings that could be required for small samples
c
MOLLUSCS, FORAMINIFERA, AND OTHER CARBONATE FOSSILS

Sizes assume little or no diagenesis is present. For diagenetically altered samples, larger samples are needed
d
For isochron analysis, the sample size must be increased by a factor of 5–8
e
For species that have not been tested for ESR applicability, another 100–200 g is necessary for stability and other background tests
f
Smaller species may require special techniques or mixing multiple individuals into one subsample
g
Large sample sizes ensure sufficient pristine mineral for analysis after mineral separation and for XRD or petrographic analysis to check for recrystallization
h
For samples from new study sites or sample types not yet tested for ESR applicability, another 100–200 g may be necessary
599
600 MOLLUSCS, FORAMINIFERA, AND OTHER CARBONATE FOSSILS
Molluscs, Foraminifera, and Other Carbonate Fossils,
Figure 1 ESR spectra in aragonitic mollusc shells. Three signals
commonly can be used for dating in aragonitic mollusc shells
(adapted from Blackwell, 2006): (a) The signal at g ¼ 2.0058
before and after irradiation measured at room temperature. (b)
The signal at g ¼ 2.0036 measured at room temperature (293 K)
and at 145 K. (c) The signal at g ¼ 2.0007 before and after
irradiation measured at room temperature.
(Schellmann et al., 2008). Dului (2000) showed that the
aragonite cuttlebone in cuttlefish preserved measureable
signals.
Petrographic, X-ray diffraction (XRD), or geochemical
analyses should accompany any ESR date on carbonate
fossils or travertines to avoid remineralized and
recrystallized samples. Contamination from Mn peaks
often requires overmodulation to discriminate the dating
peaks. Due to U uptake, modeling may be required for
uraniferous samples that cannot be analyzed by coupled
ESR-230Th/234U dating. In some fresh and hypersaline
systems, the (234U/238U)o ratio may also need to be mea-
sured or modeled. For each species and signal, the a effi-
ciency factor, Ka, must be measured. Long-term signal
fading may also need to be considered, depending on the
peak selected and its thermal stability (see Blackwell,
1995, Table 2).
Molluscs and barnacles found in life position give the
most reliable dating results, although that does not guaran-
tee that reworking has not affected the unit. Larger species
are preferred so that each subsample represents a single
individual (Table 1), but several shells can be combined
from smaller species, assuming that none have been
reworked. Fragmentary samples still need to be speciated.
Since species effects do occur, analyzing two or three dif-
ferent species from each unit can increase the dating preci-
sion and accuracy. Good agreement between ESR, TL,
OSL, 14C, and AAR (amino acid racemization) ages has
occurred in studies with Hendersonia and Allogona using
g ¼ 2.0007 (Skinner and Mirecki, 1993) and in Lymnaea
baltica and Cerastoderma glaucum using g ¼ 2.0012
(Molod'kov, 1996). Thermal stabilities in Monauha
caucaicala significantly exceeded those in many marine
molluscs. For untested species,  100 g of pristine shell
are needed to perform the necessary signal stability and
calibration tests.
For barnacles, diagenesis or signal interference may
cause some samples to be unsuitable. For example, in bar-
nacles from Norridgewok, ME, Mn interference prevented
the peak heights from being read (Figure 2; Blackwell
et al., 2010). Similar interference problems can occur in
molluscs or foraminifera. Iron contamination can also pro-
duce a very broad Fe3+ signal interference that tilts the
baseline, thus, potentially adding a systematic error to
any peak height measurements (Figure 3; Blackwell
et al., 2010). Both benthic and planktonic foraminifera
can suffer diagenesis due to carbonate dissolution, if they
sit below the carbonate compensation depth (e.g., Tadia
et al., 2002). Since the planktonic foraminifera tests are
larger, however, they persist longer in deep sediment and
are considerably easier to pick from sediment cores. None-
theless, both planktonic and benthonic foraminifera are
datable (Sato, 1981; Blackwell et al., 2013), although both
may need irradiation ramping to measure enough doses to
build the growth curves for these typically tiny samples. In
molluscs, heating to temperatures slightly above 100  C
can start to zero the dating signals, but do not affect the
Fe or Mn signals, allowing the pure interference signals
to be scanned and then removed by signal subtraction
from the dating sample, a technique that drops the preci-
sion, but raises accuracy for the peak height measurements
in contaminated samples. This correction is likely applica-
ble to other contaminated sample types, but it remains to
be tested.
Exactly how deep in time molluscs and foraminifera
can provide reliable ESR dates remains something of an
open question. Traditionally, most experts agreed that
the range for molluscs tended to be about from
 5–10 ka to as old as  500 ka (e.g., see references in
Blackwell, 1995, 2001, 2006), although Barabas
et al. (1988) showed that the signal at g ¼ 2.0036 in fora-
minifera exceeded 1 Ma. Dating samples on the young end
 MOLLUSCS, FORAMINIFERA, AND OTHER CARBONATE FOSSILS 601

Molluscs, Foraminifera, and Other Carbonate Fossils, Table 2 Tests for reworked snails from Medauwara, Kharga Oasis

Sample Description [Umol]a (ppm) Accumulated dose, AS (gray) Agea,b LU (ka)

a. Big Snail Gully

RM47 Grab sample 0.66 67.8 104.1
KAR17  0.02 5.3 9.2
RM51 Grab sample 0.76 61.0 94.1
KAR17c  0.02 2.2 5.2
RM54 Grab sample 0.75 58.6 90.1
KAR17b  0.02 2.9 5.9
RM53 Grab sample 0.67 62.3 95.9
KAR17a  0.02 3.3 6.5
RM52 Grab sample 0.81 68.7 105.5
KAR17d  0.02 4.0 7.6
b. Railway 3
FM85 Unweathered 7.46 29.5 24.0
KAR105e  0.02 1.6 3.0
FM84 Light, rugose 25.97 56.5 31.8
KAR105d Unweathered  0.02 3.1 2.8
FM82 Dark, rugose 15.34 42.8 34.4
KAR105b Unweathered  0.02 4.2 5.1
FM83 Light, smooth 26.09 75.2 40.1
KAR105c Weathered  0.02 3.4 5.3
FM81 Dark, smooth 11.52 57.3 50.2
KAR105a Weathered  0.02 3.3 5.8
c. Parking Lot Basin
FM77 Small, light, rugose 1.16 40.5 63.6
KAR102a Unweathered  0.02 4.0 8.2
FM80 Medium, red-stained, 0.93 54.7 87.2
KAR102f Rugose, unweathered  0.02 4.7 10.5
FM79 Medium, grey, smooth 1.06 42.4 67.1
KAR102e Weathered  0.02 2.1 6.5
FM78 Medium, light, smooth 1.26 54.8 84.3
KAR102d Weathered  0.02 4.6 10.0
a
Abbreviations: LU assuming linear (continuous) U uptake, p = 0
[Umol] = U concentration in the molluscs
b
Ages calculated using a/g factor, ka ¼ 0.10  0.01
initial U activity ratio, (234U/238U)0 ¼ 1.2  0.20
aragonite density, rmol ¼2.95  0.01 g/cm3
radon loss from the shells, Rnmol ¼ 0.0  0.0 vol%
carbonate sediment density, rcal ¼ 2.96  0.01 g/cm3
quartz sediment density, rqtz ¼ 2.66  0.01 g/cm3
quartz sediment density, rqtz ¼ 2.66  0.01 g/cm3
cosmic dose rate, Dcos(t)¼ 0.293  0.010 mGy/y
sedimentary water concentration, Wsed ¼ 10.0  5.0 wt%
All errors are 1s.
Data from Blackwell et al. (2012)

is hampered by inability to recognize the dating signal in
very young molluscs, while the old end is limited by dia-
genetic alteration and signal stability. Periodically, how-
ever, reports surface hinting that the range could be
somewhat longer. For example, Vichaidid et al. (2007)
reported ESR ages as old as 13 Ma for marine molluscs
in Thailand. The reliability for ages this old must await
tests for reproducibility or corroborating ages from
another chronometer, particularly since overmodulating
some aragonitic signals can produce apparently reliable
growth curves without inflection points that hide multiple
signals that each grow at different dose absorption rates,
producing peaks with different lifetimes (stabilities), that
can fade or anneal at different rates under different condi-
tions. Blackwell et al. (2013) did report an age of 2.29 
0.38 Ma for molluscs found encased from Kharga Basin,
Egypt, with which Oldowan tools were associated, which
suggested that the ESR age might be reliable.
Fossils can be easily reworked by sedimentary pro-
cesses, such as erosion/redeposition or bioturbation, from
older into younger depositional units (e.g., Blackwell,
1994; Blackwell et al., 2012). Therefore, any sampling
program should collect at least 8–10 samples from each
stratigraphic unit or sequence of units to increase the
602 MOLLUSCS, FORAMINIFERA, AND OTHER CARBONATE FOSSILS

Molluscs, Foraminifera, and Other Carbonate Fossils, Figure 2 Mn2+ interference peaks in barnacles. In RM37B, from
Norridgewock, the Mn2+ signal completely obscured the unirradiated (0 krad) signal, making it impossible to measure peak heights at
the lower added doses (adapted from Blackwell et al., 2010): (a) The Mn2+ interference peaks. In the natural sample, the Mn2+ signal
exhibited its typically complex spectrum centered at g ¼ 2.0118. Normally, Mn2+ has seven multiplets, but the seventh is off the
spectrum to the right. (b) The dating signal at g ¼ 2.0053. At higher radiation doses, such as with 112 krad, as shown here, the dating
peak’s signal height can be measured, but destructive interference may have reduced the height for the A peak, making the peak
height measurement inaccurate.

Molluscs, Foraminifera, and Other Carbonate Fossils, Figure 3 Fe interference in barnacles. Barnacles have a radiation-sensitive peak
at g ¼ 2.0053. In RM32A from northern BC, shown here, the unirradiated (0 krad) signal grew regularly with irradiation, producing an
identical irradiated signal. Interference from the Fe3+ peak caused the spectrum to have a consistently “tilted” baseline. Nonetheless,
measuring the peak height from the top of the A peak to the base of the B peak did allow an age to be calculated, although the high
signal-to-noise ratio in the lowest dose aliquots made measuring the peak heights difficult (after Blackwell et al., 2010).

chance that the samples analyzed provide dates related to
the event of interest. Nominally, if 4–5 independent grab
analyses from one bag of small or thin-walled molluscs
or foraminifera all give the same age within errors, their
age can probably be considered to date the unit’s deposi-
tion, as in Table 2a, for grab samples of Melanoides
tuberculata. Tables 2b and 2c demonstrate that the
weathering state or diagenetic features provides no clue
to the ESR ages calculated (Blackwell et al., 2012). For
larger shells, 4–5 independent ESR ages that agree well
usually provide a reliable date for a given unit.
Although the required sample weight varies depending
on the auxiliary analyses necessary (Table 2), the ESR
analysis itself, and the associated NAA or geochemical
analyses to measure the internal dose rate, require 1–2 g
of pristine carbonate per standard ESR subsample. Sec-
ondary mineralization or cementation checks require
larger samples, as do samples in which diagenesis may
have occurred or samples needing to be separated into dis-
crete mineral phases. Table 1 lists the minimum sample
sizes normally needed for various sample types. For very
small samples (100–200 mg), the irradiation ramping
technique, in which several aliquots are reirradiated sev-
eral times, can be used to get enough different doses
to build a reliable growth curve. Irradiation ramping,
however, does lengthen the total dating analysis time by
up to 1–2 years, due to the special handling necessary.
Since isochron analysis for all samples, except teeth, is
still experimental, either associated sediment samples
should be collected for sedimentary geochemical analyses
 MOLLUSCS, FORAMINIFERA, AND OTHER CARBONATE FOSSILS 603

Molluscs, Foraminifera, and Other Carbonate Fossils, Figure 4 Modeling the time-averaged cosmic dose rates, D cos ðt Þ. To account
for sedimentation and denudation, time-averaged cosmic dose rates, D cos ðtÞ, must be modeled (adapted from Deely et al., 2011): (a)
A ramped box model for the mollusc RM79, San Salvador: Instantaneous Dcos(t) were integrated over time to calculate the time-
averaged dose rate, D cos ðt Þ: This model assumed that RM79 was deposited or lived in 10 cm of sediment under 5 m of water at 125 ka
giving Dcos(t) ¼ 127 mGy/y. By 115 ka, another 50 cm of sediment had reduced Dcos(t) to 98 mGy/y. Then, sealevel was assumed to have
dropped 5 m in a geologically instantaneous time, which raised Dcos(t) to 194 mGy/y. As the platform was denuded, thinning the
sediment cover to 10 cm today, Dcos(t) rose to 252 mGy/y. (b) The effect of D cos ðtÞ on the coral RC42, San Salvador, assuming that its
possible depth during growth ranged from 1 to 10 m, and its maximum sediment cover did not exceed 2 m more than its current
cover of 1 m yielded ages that ranged from 121 to 128  8 ka. All these ages correlated well with OIS 5e, the known age for the
Cockburn Town Reef.

or the sampling locations need to be tested with in situ g or
TL dosimetry to ensure accurate dates. For TL dosimetry,
the dosimetry site (i.e., the spot from which the samples
were or will be removed) needs to be unaffected by further
excavation or deflation for 6–12 months while the dosim-
eters are in place. Due to the high probability of equipment
tampering or theft in lacustrine or coastal marine sites,
where security can be impossible to maintain over long
periods, either g dosimetry or sedimentary geochemical
analyses are preferred over TL dosimetry.
Water, rock, and sediment attenuate the cosmic dose
reaching any sample below. Since many marine shells
and some lacustrine molluscs are normally deposited in
water, the water depth above the sample presents a poten-
tial source of systematic or random error in the analyses.
For many samples, at some point, more sediment covers
the fossil, while the water could disappear due to tectonic
uplift, the lake drying, or sealevel changes, making it nec-
essary to model the cosmic dose rate changes with time in
order to derive a time-averaged cosmic dose rate. For ter-
restrial molluscs, the speed at which sediment covers the
sample must also be modeled to estimate the time-
averaged sedimentary, as well as the cosmic, dose rates.
In some areas, later denudation may remove the sediment
cover, which again will affect both the sedimentary
and cosmic dose rates. Several ways to model such
changes exist. Deely et al. (2011) used a ramped box
model (e.g., Figure 4a) in which the end points of the
ramped sequences were tied to modern analogue species’
preferences for sediment or water depth, and to
paleodepths and sediment cover thicknesses as estimated
from geological observations. When tested, their model-
ing produced ESR ages that agreed well with known ages
for corals from an MIS 5e reef (Figure 4b). While more
mathematically intricate models could be constructed,
they likely provide little improvement in accuracy given
the uncertainties in the paleowater or sediment cover
depths and the time input required to do the modeling.
After all, such models are only as accurate as the certainty
about the geological conditions under which the sample
was originally deposited, which can be sketchy at best
for species in coastal plain, lacustrine, or terrestrial set-
tings. For most situations, using time slices of 1 ky for such
modeling exercises provides reasonable accuracy, but
water and sediment cover depths need to be refined to
within 0.1–0.5 m for molluscs deposited in shallow water
and to within 1–2 m for species from deeper water up to
200 m (Deely et al., 2011). For species deposited
under > 200 m of water or > 100 m of carbonate sediment,
the cosmic dose rate is negligible.
Applications to quaternary studies
After Ikeya (1980) first described ESR signals in
molluscs, Ikeya and Ohmura (1981), Radtke et al. (1982),
604 MOLLUSCS, FORAMINIFERA, AND OTHER CARBONATE FOSSILS
Baffa and Mascarenhas (1985), Katzenberger and Grün
(1985), Hütt et al. (1983), and Skinner (1986) were among
the first researchers to report ESR ages for marine mol-
luscs. They and their colleagues all later applied these
methods to problems such as dating (e.g., Dash
et al., 2002; Kinoshita et al., 2002) and then correlating
coastal marine units (e.g., Mirecki et al., 1995;
Schellmann et al., 2008) measuring coastal uplift in tec-
tonically active areas (e.g., Hantoro et al., 1994; Doğan
et al., 2011; Tari et al., 2013) or determining the magnitude
of sealevel change in order to build paleosealevel curves in
tectonically stable areas (e.g., Molod'kov and
Bolikhovskaya, 2002; Deely et al., 2011). These methods
are now well established, although its application in each
new study area requires that the tests for signal stability
in new species be completed before the dates can be
assumed to be reliable. Such ESR studies are often
coupled with other applicable dating methods, although
mid Quaternary deposits in many areas may lack suitable
samples for most methods other than ESR (e.g., Takada
et al., 2003; Tari et al., 2013). For more detailed discus-
sions of molluscan dating applications, see Blackwell
(1995, 2001, 2006).
Both terrestrial and freshwater molluscs can give rea-
sonable ESR ages, depending on the species. Molod'kov
(2001) reported ages of 393  27 ka for Layer 5b, for ter-
restrial molluscs preserved in the Lower Paleolithic site at
Treugol'naya Cave, Russia. These dates agreed well with
ESR dates on herbivore teeth from the same sequence
(Blackwell et al., 2005). Given the inherent differences
in the signals in these two chronometers, this suggests that
these terrestrial molluscs do give reliable ESR ages.
In late Pleistocene dune sequences on San Salvador
Island, Deely et al. (2011) showed that the terrestrial mol-
lusc Cerion gave ESR dates that agreed with those
obtained from corals in MIS 5 reefs. They also used ESR
ages on Codakia, a marine mollusc that inhabits a very
restricted tidal and sediment depth range, to pinpoint
sealevel positions during Marine Isotope Stage 5.
After Sato (1981) first used ESR to date planktonic
foraminifera, applications in pelagic marine sediment
have become routine (e.g., Hoffmann et al., 2003). Using
ESR, a provenance study recognized foraminifera in
Cypriote-style limestone statuettes (Polikreti
et al., 2004). Blackwell et al. (2013a) used ESR to date
benthic late Quaternary foraminifera from the Cearà Rise,
in the central Atlantic. Otamendi et al. (2006) suggested
that Fe and Mn2+ signals in Cretaceous to Eocene forami-
nifera could be used for stratigraphy and paleoenvir-
onmental reconstruction.
Summary
ESR dating for molluscs and foraminifera will likely con-
tinue to provide valuable geological and paleontological
dates that can help to answer questions regarding plate tec-
tonic events, geohazard potentials, volcanic histories,
groundwater hydrology, Quaternary climatic and sealevel
fluctuations, Quaternary sedimentation and erosion, as
well as dating archaeological sites and helping to estimate
species’ first and last appearances. The future will likely
see even more applications to groups of organisms hith-
erto untested as well as innovative solutions to identify
and correct interference problems seen in some species.
Bibliography
Baffa, O., and Mascarenhas, S., 1985. ESR dating of shells from
Sambaquis (Brazilian shell mounds). In Ikeya, M., and Miki,
T. (eds.), ESR Dating and Dosimetry. Tokyo: Ionics,
pp. 139–143.
Barabas, M., Bach, A., and Mangini, A., 1988. An analytical model
for the growth of ESR signals. Nuclear Tracks & Radiation Mea-
surements, 14, 231–235.
Bartoll, J., Stößer, R., and Nofz, M., 2000. Generation and conver-
sion of electronic defects in calcium carbonates by UV/Vis light.
Applied Radiation & Isotopes, 52, 1099–1105.
Blackwell, B. A., 1994. Problems associated with reworked teeth in
electron spin resonance dating. Quaternary Science Reviews
(Quaternary Geochronology), 13, 651–660.
Blackwell, B. A., 1995. Electron spin resonance dating. In Rutter,
N. W., and Catto, N. R. (eds.), Dating Methods for Quaternary
Deposits. St. John’s: Geological Association of Canada.
Geotext, 2, 209–251.
Blackwell, B. A. B., 2001. Electron spin resonance (ESR) dating in
lacustrine environments. In Last, W. M., and Smol, J. P. (eds.),
Tracking Environmental Change Using Lake Sediments 1: Basin
Analysis, Coring, and Chronological Techniques. Dordrecht:
Kluwer, pp. 283–368.
Blackwell, B. A. B., 2006. Electron spin resonance (ESR) dating in
karst environments. Acta carsologica, 35, 123–153.
Blackwell, B. A. B., Liang, S. S., Lei, C. Y. Q. Q., Golovanova,
L. V., Doronichev, V. B., Skinner, A. R., and Blickstein,
J. I. B., 2005. ESR at Treugol'naya Cave, Northern Caucasus
Mt., Russia: dating Russia’s oldest archaeological site and paleo-
climatic change in Oxygen Isotope Stage 11. Applied Radiation
& Isotopes, 62, 237–245.
Blackwell, B. A. B., Gong, J. J. J., Skinner, A. R., Blais-Stevens, A.,
Nelson, R. E., and Blickstein, J. I. B., 2010. ESR dating Pleisto-
cene barnacles from BC and Maine: a new method for tracking
sealevel change. Health Physics, 98, 417–426.
Blackwell, B. A. B., Skinner, A. R., Mashriqi, F., Deely, A. E.,
Long, R. A., Gong, J. J. J., Kleindienst, M. R., and Smith,
J. R., 2012. Challenges in constraining pluvial events and
hominin activity: examples of ESR dating molluscs from the
Western Desert, Egypt. Quaternary Geochronology, 10,
430–435.
Blackwell, B. A. B., Skinner, A. R., Blickstein, J. I. B., Montoya,
A. C., Florentin, J. A., Baboumian, S. M., Ahmed, I. J., and
Deely, A. E. (submitted), 2013. ESR dating in the 21st century.
Earth Science Reviews.
Brumby, S., and Yoshida, H., 1994. ESR dating of molluscs shell:
investigations with modern shell of four species. Quaternary
Science Reviews, 13, 157–162.
Brumby, S., and Yoshida, H., 1995. The annealing kinetics of ESR
signals due to paramagnetic centers in mollusk (sic) shell. Radi-
ation Measurements, 24, 255–263.
Dash, J. K., Balakrishnan, S., and Rao, P. S., 2002. A preliminary
study on application of EPR in dating marine shells along the
Pondicherry Coast, India. In Whitehead, N. E., and Ikeya,
M. (eds.), Proceedings of the International Symposium on New
Prospects in ESR Dosimetry and Dating. Osaka: Society of
ESR Applied Metrology. Advances in ESR Applications,
18, 41–46.
 MOLLUSCS, FORAMINIFERA, AND OTHER CARBONATE FOSSILS
Deely, A. E., Blackwell, B. A. B., Mylroie, J. E., Carew, J. L.,
Blickstein, J. I. B., and Skinner, A. R., 2011. Testing cosmic dose
rate models for ESR: dating corals and molluscs on San Salva-
dor, Bahamas. Radiation Measurements, 46, 853–859.
Doğan, U., Koçyiğit, A., Varol, B., Özer, İ., Molodkov, A., and
Zöhra, E., 2011. MIS 5a and MIS 3 relatively high sea-level
stands on the Hatay-Samandaği coast, Eastern Mediterranean,
Turkey. Quaternary International, 262, 65–79.
Duliu, O. G., 2000. Electron paramagnetic resonance identification
of irradiated cuttlefish (Sepia officinalis L.). Applied Radiation
& Isotopes, 52, 1385–1390.
Hantoro, W. S., Pirazzoli, P. A., Jouannic, C., Faure, H., Hoang,
C. T., Radtke, U., Causse, C., Best, M. B., Lafont, R., Bieda,
S., and Lambeck, K., 1994. Quaternary uplifted coral reef ter-
races on Alor Island, East Indonesia. Coral Reef, 13, 215–223.
Hoffmann, D., Woda, C., and Mangini, A., 2003. Equivalent dose
determination in foraminifera, Analytical description of the
CO22signal dose response curve. Radiation Measurements,
37, 95–101.
Hütt, G., Molod'kov, A. N., Punning, J. M., and Pung, L., 1983. The
first experience in ESR dating of fossil shells in Tallinn. PACT,
9, 433–446.
Ikeya, M., 1980. Paramagnetic alanine molecular radicals in fossil
shells and bones. Naturwissenschaften, 67, 474.
Ikeya, M., and Ohmura, K., 1981. Dating of fossil shells with elec-
tron spin resonance. Journal of Geology, 89, 247–251.
Ikeya, M., Miyajima, J., and Okajima, S., 1984. ESR dosimetry for
atomic bomb survivors using shell buttons and tooth enamel.
Japanese Journal of Applied Physics II, 23, L699–L701.
Katzenberger, O., and Grün, R., 1985. ESR dating of circumarctic
molluscs. Nuclear Tracks & Radiation Measurements, 10, 885–890.
Kinoshita, A., Brunetti, A., Avelar, W. E. P., Simões, M. G.,
Fransozo, A., and Baffa, O., 2002. Dating of a subfossil shell
by ESR and K-band spectrum for paramagnetic species assign-
ment. In Whitehead, N. E.¸ and Ikeya, M. (eds), Proceedings of
the International Symposium on New Prospects in ESR Dosime-
try and Dating. Osaka: Society of ESR Applied Metrology.
Advances in ESR Applications, 18, 27–29.
Mirecki, J., Wehmiller, J. F., and Skinner, A. R., 1995. Geochronol-
ogy of Quaternary coast plain deposits, southeastern Virginia,
USA. Journal of Coastal Research, 11, 1135–1144.
Molod'kov, A. N., 1996. ESR dating of Lymnaea baltica and
Cerastoderma glaucum from Low Ancylus Level and Transgress
Littorina Sea deposits. Applied Radiation & Isotopes, 47,
1427–1432.
Molod'kov, A. N., 2001. ESR dating for early man at a Lower
Palaeolithic cave-site (sic) in the Northern Caucasus as derived
from terrestrial mollusc shells. Quaternary Science Reviews,
20, 1051–1055.
Molod'kov, A. N., and Bolikhovskaya, N. S., 2002. Eustatic
sea-level and climate changes over the last 600 ka as derived
from mollusc-based ESR chronostratigraphy and pollen evi-
dence in Northern Eurasia. Sedimentary Geology, 150, 185–201.
Mudelsee, M., Barabas, M., and Mangini, A., 1992. ESR dating of
the Quaternary deep sea sediment core RC17-177. Quaternary
Science Reviews, 11, 181–189.
Nakajima, T., Osuki, T., and Hara, H., 1993. Radiation accident
dosimetry using shells. Applied Radiation & Isotopes, 44,
91–94.
Otamendi, A. M., Díaz, M., Costanzo-Álvarez, V., Aldana, M., and
Pilloud, A., 2006. EPR stratigraphy applied to the study of two
marine sedimentary sections in southwestern Venezuela. Physics
of the Earth & Planetary Interiors, 154, 243–254.
Polikreti, K., Maniatis, Y., Bassiakos, Y., Kourou, N., and
Karageoghis, V., 2004. Provenance of archaeological limestone
with EPR spectroscopy: the case of the Cypriote type statuettes.
Journal of Archaeological Science, 31, 1015–1028.
605
Radtke, U., Hennig, G. J., and Mangini, A., 1982. Untersuchungen
zur Chronostratigraphie mariner Terrassen in Mittelitalien
230
Th/234U und ESR-Datierungen an fossilen Mollusken.
Eiszeitalter und Gegenwart, 32, 49–55.
Raffi, J., Hasbany, C., Lesgards, G., and Ochin, D., 1996. ESR
detection of irradiated shells. Applied Radiation & Isotopes,
47, 1633–1636.
Sato, T., 1981. Electron spin resonance dating of calcareous micro-
fossils in deep-sea sediment. Rock Magnetics Paleogeophysics,
8, 85–88.
Schellmann, G., and Radtke, U., 2001. Progress in ESR dating of
Pleistocene corals: a new approach for DE determination. Qua-
ternary Science Reviews, 20, 1015–1020.
Schellmann, G., Beerten, K., and Radtke, U., 2008. Electron spin
resonance (ESR) dating in Quaternary materials. Eiszeitalter
und Gegenwart, 57, 150–178.
Shih, T. S., Sato, H., Ikeya, M., Liew, P. M., and Chien, S. H., 2002.
Conditions and new extrapolation method for ESR dating of ara-
gonitic mollusk (sic) shells. In Whitehead, N. E., and Ikeya,
M. (eds.), Proceedings of the International Symposium on New
Prospects in ESR Dosimetry and Dating. Osaka: Society of
ESR Applied Metrology. Advances in ESR Applications, 18,
31–39.
Skinner, A. R., 1986. Electron spin resonance dating of aragonitic
materials. In Olin, J. S., and Blackmun, M. J. (eds.), Proceedings
of the 24th International Archaeometry Symposium. Washing-
ton, DC: Smithsonian Institution Press, pp. 477–480.
Skinner, A. R., and Mirecki, J., 1993. ESR dating of molluscs, Is it
only a shell game? Applied Radiation & Isotopes, 44, 139–143.
Tadia, O., Soh, W., and Taira, A., 2002. Estimation of past carbonate
dissolution in deep-sea sediments using the ESR method. In
Whitehead, N. E., and Ikeya, M. (eds.), Proceedings of the Inter-
national Symposium on New Prospects in ESR Dosimetry and
Dating. Osaka: Society of ESR Applied Metrology. Advances
in ESR Applications, 18, 51–55.
Takada, M., Tani, A., Miura, H., Moriwaki, K., and Nagatomo, T.,
2003. ESR dating of fossil shells in the Lützow-Holm Bay
region, East Antarctica. Quaternary Science Reviews, 22,
1323–1328.
Tari, U., Tuysuz, O., Genç, S. C., Đmren, C., Blackwell, B. A. B.,
Lom, N., Tekesin, Ö., Üsküplü, S., Erel, L., Itıok, S., Beyhan,
M., (in press), 2013. Geology and morphology of the Antakya
Graben between the Amik Triple Junction and the Cyprus
Arc. Geodynamica Acta.
Vichaidid, T., Youngchuay, U., and Limsuwan, P., 2007. Dating of
aragonite fossil shell by ESR for paramagnetic species assign-
ment of Mae Moh Basin. Nuclear Instruments & Methods in
Physics Research Section, B62, 323–328.
Cross-references
Carbonates, Lacustrine (U-Series)
Carbonates, Marine Carbonates (U-Series)
Electron Spin Resonance (ESR) Dating of Coral
Electron Spin Resonance (ESR) Dating of Fossil Tooth
Enamel
Electron Spin Resonance (ESR) Dating, General Princi-
ples
Luminescence, Biogenic Carbonates
Marine Isotope Stratigraphy
Radiocarbon Dating of Marine Carbonates
Radiocarbon Dating of Terrestrial Carbonates
Sediment Mixing Rate, 210Pb and 234Th
U–Th/He Dating
 N
NEUTRON ACTIVATION ANALYSIS
Ron Hancock
Toronto, ON, Canada
Definition
Neutron activation analysis is a chemical analysis tech-
nique that may be used to determine the average (bulk)
concentrations of major, minor, and trace elements in
a sample. It may be used to measure the concentration of
a single element or, more often, to measure the concentra-
tions of many elements, as part of a chemical characteriza-
tion of a set of samples to determine chemical similarities
or differences. Samples may be liquid (aqueous or
organic) or solid (e.g., animal tissues, ceramics, clays,
glasses, paper, plants, plastics, lithics, metals).
The most common form of neutron activation analysis
is instrumental neutron activation analysis. In this,
a suitably encapsulated, weighed amount of the sample
to be analyzed is sent, pneumatically or manually, into
an irradiation site of a research nuclear reactor, and is
bombarded by neutrons with a broad range of energies that
the reactor produces.
Once the sample has been irradiated for an appropriate
time, it is removed from the irradiation site and the
induced radioactivity measured by counting the sample
using a gamma-ray spectrometer. The result of counting
produces a gamma-ray spectrum in which specific
gamma-ray energies produce peaks, the areas of which
are proportional to the amount of the element that is
responsible for the gamma-rays.
J.W. Rink, J.W. Thompson (eds.), Encyclopedia of Scientific Dating Methods, DOI 10.1007/978-94-007-6304-3,
© Springer Science+Business Media Dordrecht 2015
To determine the actual amount of an element in
a sample, peak areas are compared with calibrations taken
from a stored library of information or from a similarly run
substandard of the element.
There are three types of neutron activation reactions:
(n, g), a neutron is absorbed by an atomic nucleus and
a gamma-ray is emitted; (n, p), a neutron is absorbed by
an atomic nucleus and a proton is emitted; and (n, a),
a neutron absorbed by an atomic nucleus and an alpha par-
ticle is emitted. The first nuclear reaction is generated by
low-energy (<0.1 eV or thermal) neutrons, and the other
two are produced mainly by epithermal and fast
(>1 MeV) neutrons. Although thermal neutron reactions
are favored highly, some activation products may arise
from elements of different atomic numbers. For
example, 28Al is produced preferentially by the
thermal neutron reaction 27Al(n, g)28Al, by the epithermal
neutron reaction 28Si(n, p)28Al, and by the fast neutron
reaction 31P(n, a)28Al. This means that while it is
potentially impossible to analyze for Si, P, and Al in
a matrix including large amounts of each, it is possible
to use the activation product 28Al to analyze for Al in alu-
mina-silicate materials; for Si in silica-rich, or doped,
materials; and for P in bones.
Only a minute fraction of the neutrons that hit a sample
actually interact with nuclei of atoms in the sample to form
artificial radioactivity. Elements form radioisotopes that
emit one or more gamma-rays with discrete energies and
that decay with characteristic half-lives, labeled short,
medium, or long. For an analysis, knowledge of expected
half-lives helps determine the irradiation time, the delay
time before counting, and the counting time.
608 NOBLE GAS MASS SPECTROMETER
A more detailed view of the principles of neutron acti-
vation analysis (NAA) can be found in Kruger (1971),
Neff (2000), or Pollard and Heron (1996).
Bibliography
Kruger, P., 1971. Principles of Activation Analysis. New York:
Wiley-Interscience.
Neff, H., 2000. Neutron activation analysis for provenance determi-
nation in archaeology. In Ciliberto, E., and Spoto, G. (eds.),
Modern Analytical Methods in Art and Archaeology. New York:
Wiley-Interscience.
Pollard, A. M., and Heron, C., 1996. Archaeological Chemistry.
Cambridge, UK: The Royal Society of Chemistry, 375 pp.
NOBLE GAS MASS SPECTROMETER
Leah E. Morgan
Scottish Universities Environmental Research Centre,
East Kilbride, UK
Definition
Noble gas mass spectrometry is a technique that allows for
the measurement of isotopic ratios of small quantities of
noble gases. Noble gases are typically extracted from sam-
ples in an extraction line, which connects directly to a
mass spectrometer. Measurement of small quantities of
gas is enabled by static-mode analysis, whereby gas is
trapped in the mass spectrometer volume for the entirety
of the measurement. Static mass spectrometric analyses
were first considered by Aldrich and Nier (1948) and Nier
(1981) and fully developed by Reynolds (1956).
Gas extraction and purification
Noble gas extraction and cleaning occur in an “extraction
line” which has been pumped to ultrahigh vacuum (UHV)
pressures of <1  10 9 mbar. Gas extraction from sam-
ples is typically achieved by furnace or laser heating.
Gases thus released expand into the extraction line where
they are purified to prevent isobaric interferences and
source effects. Purification uses chemical traps (“getters”)
made of metals that trap reactive gases and cryogenic
traps, where regions of the line are held at temperatures
below the vapor point of gases to be trapped. In some
cases, different noble gases can be trapped and then later
released for analysis, by varying the temperature of the
cryogenic traps. Isotopic fractionation during extraction
and purification is prevented by providing sufficient time
for equilibration during gas expansions.
Mass spectrometry
Purified noble gases are expanded into a mass spectrome-
ter volume, which has typically been pumped to UHV
until just prior to the inlet time. After allowing for isotopic
equilibrium, the inlet valve is closed and analysis is initi-
ated. Noble gas mass spectrometers contain a Nier-type
electron impact ion source, which relies on a hot filament
to create a beam of electrons that collide with and ionize
gas atoms in the source. Ionized atoms are accelerated
and analyzed as described in the entry on mass spectrom-
etry elsewhere in this volume.
The isotopic composition of gas in the mass spectrom-
eter evolves over the course of a measurement, due both to
gas consumption and to outgassing from the walls of the
mass spectrometer. Thus, each isotope is measured over
a period of time, and the resultant data evolutions are
fit with a line or curve that is extrapolated back to the
time when gas was first expanded into the mass spectrom-
eter. These “time-zero” fits for each isotope are assumed
to represent the relative isotopic abundances of the
samples.
Data must be corrected for background contamination
and isotopic fractionation. Background corrections are
made by replicating run conditions without heating the
sample. Often blanks are run interspersed with samples,
and a long-term fit of blank values is used to correct data.
Isotopic fractionation, or discrimination, is corrected
via the measurement of a gas with a known isotopic
ratio; for Ar this is most commonly the atmospheric
40
Ar/36Ar value.
Geochronometers that rely on noble gas mass spec-
trometry include 40Ar/39Ar, (U–Th)/He, 4He/3He, tritium
(by 3He ingrowth), and cosmogenic radionuclides.
Bibliography
Aldrich, L. T., and Nier, A. O., 1948. Argon 40 in potassium min-
erals. Physical Review, 74, 876–877.
Nier, A. O., 1981. Some reminiscences of isotopes, geochronology,
and mass spectrometry. Annual Review of Earth and Planetary
Sciences, 9, 1–18.
Reynolds, J. H., 1956. High sensitivity mass spectrometer for noble
gas analysis. Review of Scientific Instruments, 27, 928–934.
Cross-references
Ar–Ar and K–Ar Dating
Mass Spectrometry
Minerals (40Ar–39Ar)
Terrestrial Cosmogenic Nuclide Dating
U–Th/He Dating
 O
OBSIDIAN HYDRATION DATING
Ioannis Liritzis
Laboratory of Archaeometry, Department of
Mediterranean Studies, University of the Aegean,
Rhodes, Greece
Definition
Obsidian: Obsidian is an aluminosilicate, or rhyolitic,
glass, formed by rapid cooling of volcanic magma under
the proper geologic conditions. As any other glass, it is
not a crystal, and thus it lacks the lattice structure typical
of crystals at the atomic level. However, glasses do pos-
sess some degree of spatial order. Thus, it is an amorphous
natural glass that contains pristine water 2 and sparse crys-
tals of variable sizes of a few microns. The surface is
weathered in the atmosphere and the environmental
context. Composition: 70–75 % SiO2, 10–15 % Al2O3,
3–5 % Na2O, 2–5 % K2O, 1–5 % FeO3 + FeO
(in contrast to iron-bearing glass and silica-enriched
leached rinds on obsidian glass recently discovered on
Mars that are representative of global processes of explo-
sive volcanism potentially implying widespread acidic
leaching on Mars due to oxidizing and acidic solutions).
Obsidian rocks were used by early peoples for the making
of their tools and implements.
Obsidian hydration
In its most basic aspect, hydration simply describes the
process by which water is absorbed by obsidian and
involves physicochemical changes in the glass
(Doremus, 2002; Anovitz et al., 2008; Liritzis and
Stevenson, 2012). The six steps of the process involve:
1. Environmental water molecules adsorb on the surface
which exhibits roughness at the nanoscale creating a
large surface concentration.
J.W. Rink, J.W. Thompson (eds.), Encyclopedia of Scientific Dating Methods, DOI 10.1007/978-94-007-6304-3,
© Springer Science+Business Media Dordrecht 2015
2. Absorption into the glass and diffusion into the inter-
stices in the glass matrix occurs. The diffusion process
seems to be driven by two properties of the water mol-
ecules: a concentration gradient and osmotic pressure
or chemical potential. Although it has been suggested
that chemical reactions play a role (Doremus, 2002,
108 ff.), it is at present a matter of speculation.
3. A water saturation layer is then formed within a short
period of time after the start of the diffusion process.
Following this period, the saturated layer increases
with depth as time progresses. The saturation condi-
tions on the surface represent the average conditions
of the physical surroundings for the diffusion system
during the elapsed time in a particular archaeological
context. It depends on factors that include the kinetics
of the diffusion mechanism for the water molecules,
the specific chemical structure of obsidian, as well as
the external conditions affecting diffusion
(temperature, relative humidity, pressure) (Smith and
Van Hess, 1987; Liritzis and Diakostamatiou, 2002;
Liritzis, 2006).
4. The diffusing molecules stretch the glass matrix, caus-
ing an increase in volume in the hydrated region and a
stress between unhydrated and hydrated regions.
5. Increased water concentration progresses into the
glass, in time, its rate being a function of the initial
openness of the glass, temperature, and the dynamics
of the process itself.
6. When the hydrated layer becomes thick enough, typi-
cally greater than 20 mm, the accumulated stresses
cause the layer to spall off as perlite.
Three general classes of methods have been proposed
for measuring obsidian hydration: (a) measurement
of water mass uptake or loss versus time by infrared
(IR) (Stolper, 1982; Ebert et al., 1991; Stevenson and
Novak, 2011), (b) direct measurement of water profiles
versus depth by secondary ion mass spectrometry
610 OBSIDIAN HYDRATION DATING

Obsidian Hydration Dating, Figure 1 (a) Schematic diagram of obsidian hydration of diffused water, (b) obsidian hydration rim seen
through optical microscopy.

Obsidian Hydration Dating, Figure 2 SIMS H+ profiles for three samples from Greece, Japan, and Slovakia.

(SIMS) (Anovitz et al., 1999; Liritzis and Diakostamatiou,
2002; Riciputi et al., 2002; Liritzis et al., 2004; Stevenson
et al., 2004; Anovitz et al., 2004, 2008), and
(c) observation of the leading edge of the stress zone by
optical microscopy (e.g., Friedman and Smith, 1960;
Friedman and Long, 1976; Stevenson et al., 1996).
The environmental water molecules entering obsidian
glass depend mainly on (a) environmental temperature,
(b) chemical composition, (c) mineralogical structure,
and (d) intrinsic water content (see, Lanford et al., 1979;
Silver et al., 1990). The deeper the water goes, the greater
the time. The higher the environmental temperature in
burial sites, the higher the diffusion rate, while for average
environmental conditions it is
1 mm/1,000 years
(Figures 1a, b and 2a, b). Attention is exerted on the
precise determination of “hydration front,” i.e., the
 OBSIDIAN HYDRATION DATING
interface layer between unhydrated and hydrated glass.
It is a zone of optical contrast when seen under polarized
light, due to the phenomenon of “stress birefringence”
(Born and Emil, 1980, pp. 703–705). Measurement of this
layer can be determined to within 0.2 mm using digital
imagery at higher magnification (800) (Ambrose and
Novak, 2012). A significant uncertainty can be introduced
into the process as a result of poorly defined image quality
stemming from improper sample preparation and weakly
defined diffusion fronts. It has also been documented that
diffused molecular water extends beyond the limits of the
optical diffusion front (Stevenson et al., 2001; Tokoyama
et al., 2008). Thus, the optical approach does not ade-
quately document the full extent of the diffusion process.
Obsidian hydration dating (OHD)
The technique was initiated by Friedman and Smith
(1960), where they noted that (a) the exposed surfaces of
ancient obsidian artifacts had absorbed water, (b) the
hydration birefringent rims were visible under high-power
magnification (approximately 500), and (c) its width
was dependent on time, chemical composition (i.e., obsid-
ian quarry), and the temperature. They suggested that
hydration process converts a hydration layer of measured
thickness to an absolute date with an established rate
for the inward diffusion of molecular water using the
empirical Eq. (1):
X ¼ ðktÞ1=2 ð1Þ
where X is the hydration rim width in microns (mm), k
is the hydration rate at a particular temperature/relative
humidity, and t is time.
The thickness (X) and the k, which is dependent on the
specific archaeological site conditions, are required for
age determination.
The hydration rim was initially measured by optical
microscopy (Friedman and Smith, 1960; Michels, et al.,
1983; Stevenson et al., 1989a, b), but inaccuracies in the
determination of the end point of the water front
(Anovitz et al., 1999; Hull, 2001; Stevenson et al., 2002)
lead to measurements with other nonoptical techniques,
like nuclear reaction analysis, infrared photoacoustic spec-
troscopy (IR-PAS), and secondary ion mass spectrometry
(SIMS) (Lee et al., 1974; Tsong et al., 1978; Duerden
et al., 1982; Anovitz et al., 1999; Stevenson et al., 2000,
Stevenson et al., 2002; Liritzis and Diakostamatiou, 2002).
The diffusion coefficient was estimated at high temper-
ature (160  C) and extrapolation to ambient site conditions
using the Arrhenius Eq. (3) (Friedman and Long, 1976;
Michels et al., 1983; Stevenson et al., 2002).
Research subsequent to Friedman and Smith’s original
presentation has refined the method into two distinct tech-
niques. The simplest referred to as empirical rate dating
(Meighan et al., 1968; Meighan, 1976; Kimberlin, 1976;
Findlow and Bennett, 1978) requires correlating the width
to optically measured rims to independent chronometric
611
data, such as 14C (e.g., Ambrose, 1976). The second more
complex and widely applied known as intrinsic rate dat-
ing requires experimentally determined rate constants
and a measure of site temperature (Friedman and Long,
1976; Michels et al., 1983; Stevenson and Scheetz,
1989; Hull, 2001).
The mathematical model utilized in the intrinsic rate
OHD was
X2 ¼ A t eE=RT ð2Þ
where X is the depth, A is a source-specific constant, the
diffusion coefficient (mm2/day) is a pre-exponential con-
stant, t is time, and E is the activation energy (calories or
Joules per mole), respectively, R is the gas constant
(calories per degree per mole, i.e., 1.987), and T is the
absolute temperature (Kelvin), in fact the effective hydra-
tion temperature (EHT).
Along with this method, the importance of connate
water, the compositional indices, and density, in control-
ling the hydration rate of obsidians, has been demon-
strated by several authors (e.g., Ambrose and Stevenson,
2004).
Nevertheless, the assumptions made by Eq. (2) are
questioned from the sigmoid hydrogen profiles measured
by SIMS (Figure 2).
However, in spite of over 50 years of development and
application, of the classical OHD versions has produced
consistently reliable results, with the exception of some
sites of young age and/or well calibrated to 14C and
typology. In some cases, ages have so contradicted other
well-established chronometric data that the utility of the
obsidian as an independent chronometer had been
questioned. The problems with OHD are due to the use
of inappropriate analytical techniques, which involved
nonsystematic errors arising there from the inherent
imprecision of optical measurements, the experimental
conditions of estimation of hydration rate, the need for cal-
ibration issues, and an improper (empirical) model of the
hydration process. The scientific background has illus-
trated the complexity of the hydration process, and revised
working assumptions for OHD have been proposed
(Friedman and Long, 1976; Ambrose, 1976; Braswell,
1992; Rindings, 1996; Anovitz et al., 1999; Stevenson
et al., 2000; Hull, 2001; Liritzis et al., 2004; Riciputi
et al., 2002; Liritzis and Laskaris, 2012).
An advancement of OHD with a completely new
approach – the monitoring of hydrogen profile in obsidian
surfaces by SIMS (see below) – suggests that refinement
of the OHD technique is possible in a manner which
improves both its accuracy and precision and potentially
expands the utility by generating chronological as well
as palaeoenvironmental and usage data (Liritzis and
Laskaris 2011). An earlier attempt of using nuclear reac-
tion to monitor H+ by depth has not been followed up
(Tsong et al., 1981).
Another recent example of progress with classical OHD
is based on accurate measurement by IR-PAS and
612 OBSIDIAN HYDRATION DATING
Obsidian Hydration Dating, Figure 3 Overlapping of SIMS H+ profile in atoms/cc versus depth, on the hydration layer (rim), into the
obsidian glass (Liritzis and Laskaris 2011).
calculation of diffusion rates. The rate of environmental
molecular water diffusion depends on the total structural
water content of the obsidian found in the bulk glass. As
the process of mass uptake is a function of temperature,
pressure, and openness of the glass matrix as measured
by intrinsic water concentration, mass gain or loss pro-
ceeds proportional to tn, where t is time and n is an expo-
nent lying between approximately 0.5 and 0.6 shown by
IR-PAS analysis in a series of 90  C isothermal laboratory
hydration experiments (Stevenson and Novak, 2011).
Later on, Stevenson et al. (2013) have applied this
developed calibration to estimate hydration rates for
obsidian based upon the structural water content of
the glass as determined by IR-PAS to 53 Rapa Nui,
Easter Island, Chile. Here, the Arrhenius constants
(i.e., pre-exponential, activation energy) may be estimated
from H2Ot (see below).
Accuracy of hydration rim measurements
In a satisfactory laboratory technique, accuracy is not lim-
ited by resolution of the microscope to ~0.25 mm, but
probably lies in the range of 0.05–0.1 mm, which is consis-
tent with data reported. In a practical sense, measurement
accuracy is more likely limited by the material properties
of the obsidian. In any case, the accuracy of rim measure-
ment is not a large contributor to the error in computed age
in OHD (Rogers, 2010), who points that computing a
hydration rate is not a regression problem, however, but
a problem of parameter optimization.
However, the apparent rim accuracy thus achieved is
not the main factor in age determination. Each data point
is a combination of valid data and experimental error, with
major errors arising from site formation processes, associ-
ation with radiocarbon or luminescence dating or the
archaeological typology and obsidian, uncertainties in
temperature, and fluctuations in the hydration rate which
are not accounted for.
OHD based on SIMS and IR-PAS methodological
approaches
Initiated by Tsong et al. (1980, 1981), OHD based on
SIMS has been followed by dozens of articles on the appli-
cation of SIMS on obsidian artifacts but most on the deter-
mination of hydration layer, the diffusion of cations, and
the dating. The direct measurement of the water profile
measurement is generally performed by SIMS or the
IR-PAS. The principle is to measure the concentration of
H+ ions, as a proxy for water, as a function of depth. The
depth of penetration is typically measured by the end point
(tail) where the water concentration has fallen to the pris-
tine water concentration value at the surface.
Two teams at Oak Ridge (USA) and Rhodes (Greece)
relied on the modeling of the H2O concentration profile
as a function of hydration depth, but following them in dif-
ferent ways. As a result, the ODDSIMS and the SIMS-SS
coined versions were produced, respectively (Anovitz
et al., 1999; Liritzis and Diakostamatiou, 2002; Anovitz
et al., 2004; Liritzis et al., 2004; Liritzis and Laskaris,
2009, 2011).
In using SIMS, an ion beam (usually Cs) bombards the
obsidian surface. (The dating is based on the concentration
of H as a proxy for H2O versus depth) (Figure 3).
 OBSIDIAN HYDRATION DATING
The use of SIMS on obsidian surface investigations,
although relatively infrequent, has produced great pro-
gress in OHD dating. In essence, it is a technique with a
large resolution on a plethora of chemical elements and
molecular structures in an essentially nondestructive man-
ner (Liritzis and Laskaris, 2011).
Anovitz et al. (1999) presented a model which relied
solely on compositionally dependent diffusion, following
numerical solutions (finite difference (FD) or finite ele-
ment) elaborating on the H+ profile acquired by SIMS. A
test of the model followed using results from Mount
65, Chalco in Mexico, by Riciputi et al. (2002). This tech-
nique used numerical calculation to model the formation
of the entire diffusion profile as a function of time and
fitted the derived curve to the hydrogen profile. The FD
equations are based on a number of assumptions about
the behavior of water as it diffused into the glass and char-
acteristic points of the SIMS H+ diffusion profile. In fact,
the depth of the half-amplitude point of S-like H+ profile
is found to be proportional to tn, where t is time and n is
an exponent lying between approximately 0.6 and 0.7
(Anovitz et al., 1999, 2004; Stevenson and Novak, 2011).
The half-amplitude measurement is again a function of tem-
perature and openness of the matrix; however, since the
measurement is made to a half-amplitude point, i.e., a rela-
tive measurement, the technique is not sensitive to the total
amount of water present and should not be affected by
pressure.
In Rhodes, Greece, one dating approach is based on
modeling the hydrogen profile, following Fick’s diffusion
law, and an understanding of the surface saturation layer
(surface saturation, SS) and surface topography (Liritzis
et al., 2004; Liritzis, 2006; Liritzis and Laskaris, 2009).
The saturation conditions on the surface represent the
average conditions of the physical surroundings for the
diffusion system during the elapsed time in a particular
archaeological context.
It has been shown (Liritzis, 2006) that all of the H+
profiles are broadly similar in form; however, this is their
only shared property. Despite the fact that they are
derived from the same type of geological material, close
inspection of the shape of the profiles suggests that defi-
nite differences are present. These which are derived
from the significant environmental impact as well as
intrinsic effects may alter the form of the SIMS hydrogen
profile. It is these slight variations in the profile that are
case specific, and this is the important observation that
the calculated age by SIMS-SS method relies upon.
Therefore, each age calculation is based on this diffusion
profile, which is unique for every obsidian sample. There
are three advantages to this fully intrinsic procedure:
(1) the final shape of the hydrogen profile, properly
modeled, incorporates two of the principal external and
highly variable environmental parameters, those of tem-
perature and humidity; (2) the use of SIMS instrumenta-
tion to measure surface hydration layers results in high
precision thickness (depth) values with an error of
0.02–0.05 mm depending upon the degree of surface
613
roughness; and (3) absolute age estimates may be calcu-
lated. The currently used two methods, one based on
SIMS (namely, SIMS-SS, SS for surface saturation) and
the other on IR-PAS, are briefly described below.
The SIMS-SS method
The water diffusion coefficient at any particular moment is
expressed by the first derivative of the hydrogen profile.
Its inverse ratio is the apparent hydration rate. The average
depth of surface saturation layer Xs and corresponding sat-
uration concentration Cs obtained from the determination
of the SS layer gives the overall error attached to the
SIMS-SS ages.
Using the end product of diffusion, a phenomenologi-
cal model has been developed, based on certain initial
and boundary conditions and appropriate physicochemi-
cal mechanisms that express the H2O concentration ver-
sus depth profile as a diffusion/time equation. The
modeling of this diffusion process is a one-dimensional
phenomenon, where the H2O molecules invade a semi-
infinite medium in a perpendicular direction to the sur-
face. The model is based on the idea that in the SS layer
located near the sample surface, that is, in the first half
of the sigmoid curve, the C is assumed as constant along
a very short distance. Thereafter, C gradually decreases
following the trend of the sigmoid. In brief, the three
principles used for dating are (a) the comparison of a
nondimensional plot with a family of curves of known
exponential diffusion coefficients; (b) the correlation
between the rate of transfer (diffusion) from the surface
with the diffusion duration, the saturation concentration
Cs, the intrinsic (pristine) water concentration Ci, the dif-
fusion coefficient Ds (defined the flux/gradient, where
gradient or tangent ¼ dC/dx), and the following
Boltzmann’s transformation and auxiliary variables;
and (c) the modeled curve of diffusion profile
(concentration versus depth) of Eq. (3) (see Liritzis
et al., 2004; Liritzis and Laskaris, 2012).
C ¼ eaþbxþcx þdx
2 3
ð3Þ
The dating in Eq. (4) that has been proposed incorpo-
rates all the abovementioned parameters:
 2
ðCi  Cs Þ2 1:1280:177 kC
1 Cs i
T¼
2 ð4Þ
dx jx0
4 Ds, eff dC
where Ci is the intrinsic concentration of water, Cs the sat-
uration concentration, Ds ¼ dC/dx the diffusion coeffi-
cient for depth equal to zero, Ds,eff the effective diffusion
coefficient empirically derived from a set of well-known
ages and Eq. (5) as the effective value of the diffusion
coefficient Ds for C ¼ Cs, and k is derived from the family
of Crank’s curve. It is:


Ds, eff ¼ a  Ds þ b= 1022  Ds ð5Þ
614 OBSIDIAN HYDRATION DATING
AFM Topographic Images
AFM image
Height
10
9
300
8
7 250
6
200
mm
nm
5 250
mm
150
4
3 100
100
2 10
50
1 8
0 0
0 1 2 3 4 5 6 7 8 9 10
mm
Crack
Cracks, Voids
Troughs & Peaks
Sample: Fikirtepe-7/
Turkey
Obsidian Hydration Dating, Figure 4 Atomic force microscopy on obsidian surface indicates surface roughness, most significant
parameter for accurate SIMS H+ profiling (Liritzis and Laskaris, 2009).
where Ds ¼ (1/(dC/dx)) * 1011 assuming a constant flux
and taken as unity (Brodkey and Liritzis, 2004; Liritzis,
2006; Liritzis and Laskaris, 2009). Equation (5) and
assumption of unity are a matter of further investigation.
Often, the obsidian surface at the micro- and nanoscale
is not smooth, and this influences the diffusion of water
and, as a result, the SIMS profile. Polarized light micros-
copy (PLM) and atomic force microscopy (AFM)
(Figure 4) have been used to examine the obsidian surface
and investigate any correlation between the surface rough-
ness and the SIMS measurements (Liritzis et al., 2007). As
proposed by Liritzis et al. (2008a, b), a linear regression fit
in the diffused region defines the dispersion which reflects
the degree of surface roughness. The smoother the surface,
the less pronounced the dispersion of the data points to
either side of the trend line and the smoother the SIMS dif-
fusion profile. The surface roughness measured by the
AFM instrument is linearly correlated with the standard
deviation of the residuals between the data points (H+
values of g/mol or atoms/cc) and the linear fit in the dif-
fused region of SIMS. This proportionality aids the selec-
tion of appropriate obsidian samples/surfaces for dating
and also for the proper choice of the spot where the SIMS
analysis is performed. Last, the smooth variation of other
monitored cation profiles (K, Fe, Si, Mg, F, Al) and map-
ping of ion rastered surface areas act as criteria for accept-
able and safe further numerical evaluation of H+ profiles.
The validity of these new procedures is tested through a
comparison with independently derived age determina-
tion. A suite of samples from Easter Island (Chile),
3D representation
nm
300nm
300
200
150
50
Trough
300nm
250nm
6
4
100nm
2
8 10
0 4 6
0 2
mm
X&Y axis: µmxµm
Z axis: nm
Mexico, Greece, Japan, Asia Minor, and Hungary, ranging
from few hundreds to 30,000 years ago, have been used,
derived from archaeological contexts dated by the radio-
carbon method or ceramic associations. Indeed, the con-
vergence between the two dating methods is high
(Figures 5). The SIMS-SS age estimates fall within the
expected age ranges for the archaeological contexts for
all samples (Liritzis and Laskaris, 2009; Liritzis, 2010).
Table 1 gives an example of two dates by SIMS-SS.
Obsidian hydration dating by IR-PAS
The application of IR-PAS (infrared photoacoustic spec-
troscopy) to the dating of obsidian was proposed as a solu-
tion for two central methodological problems associated
with OHD using optical microscopy. First, optical mea-
surements of hydration layer thickness from prepared thin
sections rely upon a visual assessment of hydration layer
strain birefringence to establish a depth of penetration.
This strain is a secondary effect created by a volume
expansion associated with molecular water diffusion that
slows the passage of light through the hydrated layer. Sec-
ond, it has been documented in numerous instances that
the rate of water diffusion is significantly impacted by
the amount of total structural water found within the glass
matrix (Karsten and Delaney, 1981; Tomozawa, 1985;
Zhang and Behrens, 2000). Calibrations have been devel-
oped that correlate the concentration of structural water
with laboratory hydration rates at lower temperatures
(90  C) (Stevenson et al., 1998; Stevenson and Novak,
2011).
 OBSIDIAN HYDRATION DATING 615

Obsidian Hydration Dating, Figure 5 SIMS-SS versus 14C/archaeological ages for obsidians all over the world (Liritzis and Laskaris,
2011).

Obsidian Hydration Dating, Table 1 The SIMS-SS dates for two samples from Ikaria Island, Greece (Kerame site), without any other
chronological control (From Liritzis and Laskaris, 2012)

Sampling Ds,eff Age SIMS-SS

No. Lab. No./site details Xs (cm) Cs (grmol/cc) Ci (grmol/cc) ek (cm2/year) (years BP)

1 RHO-891 Kerame, T,C, Spit 4 3.1486e-5  0.00129197  0.000106014  360 1.1365e-12 11,085  3,282
Ikaria Sq. 5 + 6 1.87e-6 1.79e-5 6.24e-6
2 RHO-892 Kerame, Trench D 4.90433e-5  0.0007287  9.8763e-5  80 3.42e-13 10,152  1,643
Ikaria Spit 4 2.24e-5 2.24e-5 4.44018e-6

With the IR-PAS method, an absorbance spectrum
associated with the vibrational modes of glass functional
groups (Si-OH, Al-OH) and water peaks (OH, H2O) is
produced; at 3,570 cm1, it represents a combination
band for OH/H2O, and the band at 1,630 cm1 reflects
the molecular concentration of H2O whether structural
molecular or diffused molecular water (Newman
et al., 1986), as well as the structural water content of
unhydrated region. The method is partially destructive
since the artifact or flake must be cut to fit within the sam-
ple chamber and the IR-PAS peaks can be impacted by
micro-cracks or voids within the glass that trap and retain
molecular water. Stevenson et al. (2001) developed a cal-
ibration between IR-PAS intensity at 1,630 cm1 and
percent molecular water content (%H2Om), using
polished and parallel-sided obsidian samples with a
range in total water content from 0.09 % to 1.57 %. The
rate constants are then calculated for an archaeological
specimen. The 90  C diffusion coefficient (D) and activa-
tion energy value (E) are estimated from the obsidian
total structural water content (%H2Ot) in weight percent
concentration. The quantity k in age equation is esti-
mated from the Arrhenius equation:
k ¼ D expðE=RTÞ ð6Þ
where D is the diffusion coefficient for H2Om at 90  C, E is
the activation energy in J/mol, R is the gas constant (8.314 J/
mol K), and T is the hydration temperature (Kelvin). The
archaeological site temperature is determined for this appli-
cation by direct measurement of soil temperature using cells
monitoring T for 1 year. The resulting rate was then used to
estimate ages for obsidian artifacts.
Recently, 52 obsidian specimens were measured at
two sites of Rapa Nui Easter Island, Chile. Average mea-
surements per parameter and deduced age for the sample
616 OBSIDIAN HYDRATION DATING
23 of site 15–233, test unit 2 were as follows: tempera-
ture as a function of specimen depth (not really an
effective hydration temperature), 22.1–23.4  C; PAS
hydrated 1,630 cm1, 0.1025; %H2Ome (environmental
diffused molecular water) in hydration layer, 0.10 %;
3,570 cm1 PAS polished, 0.1313; structural %H2Ot,
0.89; D at 90  C as %H2Ome^0.5/day, 0.000000118
(exponent 0.5 as in Eq. (1); E (J/mol), 84.450; diffusion
rate estimated (%H2Ome^0.5/year), 0.000027000; and
dates, A.D. 1,419  30 to A.D. 1,842  30 years. Ages
resulted in an acceptable degree of agreement with radio-
carbon ages that those parameters can be used for other
obsidian specimens excavated at the island (Stevenson
et al., 2013).
However, at Rapa Nui, three problems come into play,
which may be applicable to all methods. (i) Handling dif-
ficulties and the lack of artifact clarity call for different
sampling methods to obtain structural H2Ot and diffused
H2Ome concentration values.
The amount of damage to an artifact may be substan-
tial when the thick section is removed, and in the worst
of cases, small obsidian flakes may be completely
destroyed or too small to be processed at all. In addition,
opaque obsidians, which constitute a large proportion of
the world’s natural glasses, may not achieve clarity even
after thinning glass thick sections to 0.5 mm. These same
issues apply to processing a second subsample in order to
measure the hydrated surface. (ii) Laboratory applica-
tions of obsidian hydration dating include multiple
sources of error that impact the precision of a final age
determination. The precision errors cannot simply be
added to estimate a total error because the variables are
independent of each other with the first associated with
a concentration value and the second to a rate coefficient.
The latter however seems to coincidentally provide a cor-
rect value from hydration experiments at 90  C, while the
used annual results of ambient T need further clarifica-
tion. (iii) The EHT is a problem that in the above example
at Easter Island seems to be overcome due to the short
time archaeological intervals involved, since present
T measurements are similar for the last say 2–5 hundred
of years, but cannot be as such for a few thousand of
years BP.
Diffusion mechanisms in obsidian
Considerable complexity of phase states has been indi-
cated by a wide range of variability of properties from
obsidian sources. The exact mechanism by which water
diffusion in such amorphous glasses takes place, by
either mechanical transport or molecular interconver-
sion, is still subject to research (Doremus, 1969; Crank,
1975; Zhang et al., 1991; Nowak and Behrens, 1997;
Doremus, 2000; Zhang and Behrens, 2000; Doremus,
2002). In either case, diffusion reaction is described
approximately by Eq. (7):
@W=@t ¼ @ ðDe@W=@xÞ=@x; ð7Þ
where W is concentration of total water, H2O + OH
(Zhang and Behrens, 2000; Doremus, 2002, pp. 109–110),
and De is the effective diffusion coefficient given by Eq. (8):
De ¼ 2WD=b; ð8Þ
where D is a constant diffusion coefficient characteristic
of molecular H2O and b is a proportionality constant
related to the Ostwald solubility of water in obsidian
(Doremus, 2002, pp. 78, 110). The general solution of
Eq. (7) requires numerical techniques and has been a sub-
ject of various considerations (Crank, 1975), while an
elaboration based on Fick’s law has led to new age
approaches by SIMS (Anovitz et al., 1999; Liritzis and
Laskaris, 2011; Liritzis et al., 2004).
Accelerated hydration experiments have shown that the
diffusion of water is a complex and dynamic process
(Anovitz et al., 2004; Stevenson and Novak, 2011). The
very early stages of hydration exhibit a high surface water
concentration, a declining diffusion coefficient, and an
advancing diffusion front. This change may be due to
glass surface relaxation as the stress that built up in the
near surface region is released. However, there may be a
difference in the mechanism behind the early stage of
water diffusion that differs from the longer-term develop-
mental process.
Following the work of Doremus (2000, 2002) and
Liritzis (2006), Stevenson and Novak (2011) hypothe-
sized that H2Ot creates a more open glass structure which
allows water molecules to pass through “gateways” in the
silica tetrahedra. It was their observation that the greater
the concentration of H2Ot, the faster the diffusion rate of
ambient water (H2Ome) into the obsidian surface.
Water enters obsidian glass at environmental tempera-
tures as molecular H2O and as hydroxyl OH formed by
reaction with the glass silicate structure. t is believed that
the occurrence of greater amounts of Si-OH in the net-
work (silanols, referred to the intrinsic water molecules
which may be present as OH- with the siliceous network)
may weaken the neighboring Si-O-Si bonds, causing
them to easily break, and therefore lower the overall acti-
vation energy (Brückner, 1970; Ericson et al., 2004).
However, the latter view may be revised as silanols seem
to follow hydrogen sigmoid profile due to diffused water
(see, Fig. 9, Liritzis and Laskaris, 2011).
At any rate, as water enters the glass network, the
structure is depolymerized and hydrogenated and silanol
ions are formed (Si-OH) and allow additional water
to enter the glass at a faster rate, while in the course
of time, chemical reaction follows Fick’s law. This
changing diffusion coefficient results in the formation
of the characteristic S-shaped concentration-depth
profile (Figure 2).
According to another model, H2O molecules choose
various diffusion paths in a nonhomogeneous way, as
the size-dependent mechanism assumes that water mole-
cules of radius Rw ¼ 0.15 nm occupy interstitial sites of the
obsidian and pass through “doorways” of radius RD
 OBSIDIAN HYDRATION DATING
jumping from one doorway to another. The RD ¼ 0.1 nm
derives from the equation regarding activation energy E,
which is recognized as the elastic energy necessary to
dilate a spherical cavity from radius RD to R (Doremus,
2002; Liritzis, 2006). The rate at which this process occurs
depends upon the intermolecular distance of the host
material matrix compared to the size of the diffusing
species.
This diffusion process forms an external layer as water
leaves the atmosphere and humid sediment and diffuses
into the glass structure. In particular, the surface saturated
layer (SS layer) is formed in the first 1–3 mm of the obsid-
ian surface through two diffusion mechanisms. The first
mechanism transfers the ambient water from the environ-
ment into the exterior surface of obsidian, and the second
mechanism is responsible for the diffusion process into the
interior of the artifact (Liritzis and Diakostamatiou, 2002).
The mechanism of water diffusion in obsidian and the
exponent of time in Eq. (1) are still a matter of
investigation.
Discussion of some issues related to
conventional OHD
A variety of strategies have been developed over the years
to calibrate the movement of ambient water into a glass, to
take into account the composition of the glass, and to
model the environmental history of the artifact context
(e.g., temperature, humidity). In essence, the rate of hydra-
tion is dependent on the obsidian chemistry, the intrinsic
water content of the glass, the temperature and relative
humidity, and the chemistry of the diffusing water
(Friedman and Long, 1976; Ambrose, 1984; Mazer et al.,
1991; Friedman et al., 1994; Stevenson et al., 1998;
Stevenson et al., 2002; Ambrose and Stevenson, 2004;
Rogers, 2008, 2012).
Presently, the exponent of Eq. (1) is being questioned as
non-applicable to all environments, and the determination
of the diffusion coefficient from extrapolation of labora-
tory hydrated obsidians down to ambient conditions from
the Arrhenius plot is considered as insecure (Anovitz
et al., 1999; Stevenson and Novak, 2011; Liritzis and
Laskaris, 2011). Several decades of research illustrated
the complexity of the hydration process, and revised work-
ing assumptions or disillusions for OHD were proposed
(Mazer et al., 1991; Ridings, 1996; Stevenson et al.,
1998, 2000; Rogers, 2007). In spite of many years of
development and application, mainly from research
groups in the USA and Australia, it appears that neither
approach has produced consistently reliable results. In
some cases, ages have so contradicted other well-
established chronometric and/or typological data that the
utility of the obsidian as a chronometer has been
questioned (Ridings, 1996; Anovitz et al., 1999; Liritzis,
2006; Rogers, 2008; Eerkens et al., 2008).
Although OHD seemed to give good relative ages in
some case studies as in Easter Islands (Stevenson
et al., 2000; Stevenson et al., 2013), practitioners of the
617
measurement pointed to significant errors in other cases
(Ridings, 1996). The problem of determining accurately
the hydration thickness is solved with new techniques
such as infrared spectroscopy (IR), SIMS. Infrared spec-
troscopy has been used to measure the bulk water content
of obsidian (Newman et al., 1986; Zhang et al., 1996) with
the photoacoustic method (IR-PAS) used to quantify the
depth of surface water diffusion (Stevenson et al., 2001,
2002; Stevenson and Novak, 2011). The latter approach
to hydration layer depth assessment has been calibrated
relatively to SIMS and shown to have a higher precision
and accuracy than obtainable by optical microscopy
(Ambrose and Novak, 2012).
In certain cases, extensive lab experiments using the
same obsidian source, diffusion rate, and average environ-
mental hydration temperature (effective hydration temper-
ature) using temperature cells yielded satisfactory ages.
These were controlled by available 14C ages and ceramic
typology, e.g., for south coast of Peru, using petrographic
microscopic rim measurements (Eerkens et al., 2008), and
Hopewell culture (Stevenson et al., 1998, 2004), using rim
measurements by SIMS.
To summarize, the early problems with OHD were
partly due to inappropriate analytical techniques involving
(i) nonsystematic errors arising from the inherent impreci-
sion for optical measurements, (ii) the experimental condi-
tions and data of estimation of hydration rate, and (iii) on a
theoretical plan an improper model of the hydration
process.
Laboratory-induced hydration does not produce com-
mensurable results with the in situ-induced hydration
and/or silica dilution (Bartholomew et al., 1980;
Morganstein et al., 1999). This was the case even when
using a long-term artificial aging experiment with the
obsidian samples from three separate volcanic provinces
in Papua New Guinea exposed in normal air and water
vapor pressure to controlled temperatures of 10  C,
20  C, 30  C, and 40  C, for up to 26 years (Ambrose
and Novak, 2012). Nevertheless, the latter confirm the
value of the strategy by providing basic hydration rate
constants for obsidian specimens from each of the three
major Papua New Guinea sources. But in an archaeolog-
ical context, these data still require calibration for an opti-
cal microscopy hydration measurement. SIMS provides
an enhanced measurement system and is combined
here with experimentally determined rate constants to
independently cross-check radiocarbon dates and site
temperature calculation at a ~2,100 BP archaeological
site from Papua New Guinea. In fact, Ambrose and
Novak (2012) noted that the error figure in age indicates
the wide divergence from hydration reading error alone,
while additional temperature-based error is even greater
than the reading error by a few hundreds of years and
underlines the critical influence of temperature in
attempting to derive absolute age values from obsidian
hydration dating.
At any rate, the involved age in Eq. (1) is still empirical,
surely far apart of the more satisfactory phenomenological
618 OBSIDIAN HYDRATION DATING
approach and certainly not well substantiated by a scien-
tific approach (Brodkey and Liritzis, 2004).
Sources of uncertainty in OHD include hydrous and
compositional chemistry, humidity, rim measurement,
temperature history, diffusion rate variability due to intrin-
sic water and determination by extrapolation from higher
temperature experiments, site formation processes, and
age errors from comparative methods (with 14C, archaeo-
logical typology) which affect the validity of the age-rim
association.
Nevertheless, in spite of rigorous statistical modeling
and error propagation, the evaluations of all sources
of error impart uncertainties in each variant factor such
that the span of uncertainties may sway between extreme
percentages of the order of some 10s of a percent. The
issue of considering any individual specimen and burial
location as a unique case is a rule. Along this direction,
calibration models per excavated site are always useful
but are cumbersome and timing. A brief evaluation of
main factors related and affecting the traditional OHD is
described below.
Temperature
Perhaps the biggest problem is temperature, since the
hydration process is strongly temperature dependent
and the temperature history of an artifact can never be
completely known. Past historical temperatures vary
over time, both annually and diurnally, and hydration rate
of obsidian is temperature dependent. The effect of the
varying temperature is coined by the concept of effective
hydration temperature (EHT), defined as the constant
temperature which yields same hydration results as the
actual time-varying temperature over the same period of
time and can be computed from the time average of D
(Rogers, 2007). Because of the mathematical form of
the dependence of hydration rate on temperature,
EHT  T mean. In fact, few degrees in error of EHT in
the Arrhenius equation (embedded in Eq. (2)) produce
errors of several centuries. For an activation energy
(E) of 85 KJ/mol, the error in age for 1  C is
11–13 %, for 2  C is 25 %, and for 3  C is around
37 %. The well-known modifiers of obsidian hydration
rate with temperature (the others being chemical compo-
sition and water vapor pressure) may be accounted for,
but none is sufficient for precise prediction of hydration
rates. For this reason, obsidians still need to be indepen-
dently assessed for their own specific rate determination.
On the other hand, another reported temperature
approach using a set of calculated temperature parame-
ters for one site based on a regional temperature scaling
analysis contains uncertainties too and can further be
corrected for first-order changes in paleotemperature if
proxy data are available (West et al., 2007; Rogers,
2010, Rogers and Yohe, 2011; Rogers, 2012). Even so,
it is unlikely that EHT can be corrected much better than
1.0  C (Rogers, 2007). EHT includes the effects of
temperature history in one parameter; furthermore, a
change in EHT (Dte; te is effective hydration tempera-
ture) produces a change in rim value (Dr):


Dr=r ¼  1⁄2 E=RTe 2 DTe ð9Þ
For E ¼ 20 KJ/mol and 90 KJ/mol, Eq. (9) gives 0.0140
and 0.0630, respectively, with a median E value of 0.0007
per KJ/mol. The former leads to a change in rim of about
1–6 %/ C, and the higher holds for nominal conditions.
A cautionary note is that the meteorological data are
for air temperatures, while obsidian usually is found at a
variable depth in the ground where T differs. Alterna-
tively, sensors (e.g., HOBO H8 Family loggers, e.g., Ste-
venson et al., 2013) are put over a year in situ that record
a meaningful annual context meteorological average.
Rim
Typically, the mean and standard deviation of an aggre-
gate of about six individual readings is taken for each
hydration rim measurement reported by a laboratory.
Standard deviations are nearly always <0.1 mm, more
often 0.05–0.01 mm. A previous analysis (Scheetz and
Stevenson, 1988) concluded that the accuracy is
constrained by the resolution of the microscope system
to ~0.25 mm or worse; however, the issue here is not reso-
lution but accuracy of coincidence measurement, which is
well known to be about two orders of magnitude better
than resolution (Jacobs 1943, pp. 86–88), so the error
values quoted by obsidian laboratories are reasonable but
the precision weak as recent methods have shown.
A comparison of optical readings with hydrogen profiles
collected by SIMS (Anovitz et al., 1999; Stevenson
et al., 2001) and infrared profiling (Yokoyama
et al., 2008) show that the depth of water diffusion extends
well beyond the optically defined diffusion front. As a
result, optical techniques cannot accurately track the dif-
fused species. SIMS has been used to significantly
increase precision to 0.05–0.1 mm, but the cost per sam-
ple is prohibitive for routine application.
Higher precision of the rim reading is achieved by
SIMS profiles or IR-PAS peak integration (Anovitz
et al., 1999; Stevenson et al., 2001, 2002).
Diffusion rate, D
Simulation experiments of hydration rate by laboratory
methods (sometimes known as “induced hydration”)
have an equivocal history in archaeology, and since
it depends on temperature, attempts have been made
in the past to measure hydration rate in the laboratory
(e.g., Friedman and Long, 1976; Stevenson et al.,
1998). However, rates measured in the laboratory often
have not agreed well with archaeological data (see, e.g.,
the pointed observations in Hall and Jackson, 1989,
p. 32), while in other cases agree (e.g., for Rapa Nui Eas-
ter Island, Stevenson et al. (2013) and in Topaz Mountain
obsidian from Utah, Rogers and Duke (2011)). The
hydration rate of obsidian, which is strongly temperature
dependent, is allowed to proceed at elevated laboratory
 OBSIDIAN HYDRATION DATING
temperatures, when measurable hydration rims develop
in days rather than years. The temperature dependence
of the hydration rate has a well-attested mathematical
form, so the principle of laboratory hydration is to
hydrate a set of obsidian samples at elevated tempera-
tures, determine the activation energy and diffusion con-
stant, and then compute the hydration rate for
temperatures of archaeological interest. Initially, the dif-
fusion rate D was calculated from hydration that was
experimentally induced under controlled laboratory
experiments for use in the Arrhenius plot, in a high tem-
perature range (90–250  C) (Stevenson et al., 1998).
First, this did not secure safe extrapolation to much lower
temperatures, and second, the simulated hydration does
not produce the anticipated behavior with obsidian of
similar source. Making use of nano-SIMS and simulation
experiments, rates are produced at environmental condi-
tions. Such an aging experiment for Papua New Guinea
obsidians has shown differing diffusion rates for the same
temperature and different storage temperatures of four
different sources. There, the SIMS profile depths and
regression results for the four hydrated obsidians,
namely, Igwageta, Mt Bao, Wekwok, and Umleang, pro-
duced diffusion rates that vary by ~ 25 for a fourfold
temperature increase, between 10  C (0.002–0.004) and
40  C (0.06–0.10) for the four different sources, which,
however, have no marked chemical composition varia-
tion (Ambrose and Novak, 2012). From room tempera-
ture to 75  C, the diffusion coefficient changes by two
to three orders of magnitude confirming the well-known
sensitivity of the intrinsic dating technique to assumed
local temperature (Anovitz et al., 2004).
This had resulted in high measurement error stemming
from the use of optical methods; thus, the regression con-
stants also had a large inherent error. But, data have
shown a relationship between structural water concentra-
tion and D, and an increase in precision is now achieved
through the analysis using infrared measurement of
experimentally developed hydration layers (Stevenson
et al., 2001).
The errors and their affects have been exhaustively
analyzed (Rogers, 2010). Hydration rates can be reliably
estimated by least squares techniques applied to
obsidian-radiocarbon data with errors of about 4–5 %,
while rates estimated by least squares techniques applied
to time-sensitive artifacts are somewhat worse, with about
8–10 % errors. Rates determined from laboratory tech-
niques (induced hydration) are somewhat more accurate,
with about 2–3 % error.
Analytical errors concern (a) the mathematical proce-
dure of a least squares best fit to the logarithmic Arrhenius
equation (see Stevenson and Scheetz, 1989; Rogers and
Duke, 2011), especially if data-point weighting is not
included (otherwise the activation energy and diffusion
constant will be in error); (b) the procedural error that
involves laboratory techniques (e.g., erosion rates and
rim growth, use of proper hydration environment and tem-
perature); (c) the “scaling” problem, i.e., hydration at high
619
lab temperatures and archaeological temperatures (the
scaling is actually a complex function of the both hydra-
tion rates (and diffusion constant) and their activation
energies and requires knowledge of both for each obsidian
source in question); and last, but not least, (d) the role of
phenocrysts presence (size and abundance) and obsidian
transparency should also be reexamined for they obviate
ideal diffusion process.
The dependence of time
The dependence of time on the square exponent of x has
been questioned (Anovitz et al., 1999; Liritzis, 2006;
Anovitz et al., 2004). That is why most reported OHD
ages are compared with other dating means (14C, lumines-
cence, ceramic typology), an expected practice for the ini-
tial development of the method, though in several cases
the square root is valid.
The surface loss
The surface loss is identified as an important factor in
glass hydration dating (Ambrose, 1976, 1994; Morganstein
et al., 1999; Liritzis et al., 2008a; Liritzis et al., 2008b).
Once present on an artifact with fissures, the surface disso-
lution rate of an obsidian artifact can be calculated.
Although OHD theoretically is capable of yielding accurate
dates in the range 1,000,000 to 200 years B.P. (Michels,
1986), the obsidian surface is subjected to environmental
flaking, and this limits its range; on the other hand, the
nano-SIMS technology at least provides an accurate mini-
mal rim reading that can deal with this issue.
Humidity
Humidity effects are relatively small and are typically
ignored in analysis. However, the effect of accelerated
hydrations with 100 % Rh on diffusion kinetics has not
been adequately studied.
Intrinsic (pristine or connate) water
The remaining variable hydration associated with original
fast cooling of lava is an extremely important factor in the
diffusion rate. The hydration rate in obsidian at any given
temperature is a function of the concentration of water in
the glass (Karsten and Delaney, 1981; Karsten et al.,
1982; Zhang et al., 1991; Stevenson et al., 1998; Zhang
and Behrens, 2000). Intra-source variability in intrinsic
water content has been observed in Coso obsidians from
eastern California, USA (Stevenson et al., 1993), and is
likely present in others; thus, measurement of intrinsic
water content per individual specimen is required as a pre-
requisite (Stevenson et al., 2001, 2002).
Some early and recent OHD applications and
development
Friedman et al. (1997) provided a substantive review and
summarized the problems linked to OHD and referred to
case studies including archaeological specimens from
New Mexico, central Mexico, Belize, Honduras, and Ohio
620 OBSIDIAN HYDRATION DATING
(the latter imported from Wyoming). Studies of Mexican,
Guatemalan, and Sardinian obsidians by Michels and his
colleagues however are critiqued therein. The authors
cross-dated OHD results with radiocarbon dating,
archaeomagnetism, and ceramic phasing. Particular points
to take care of included the factors related to hydration
rates (as chemical composition of obsidians, their trace
elements content, relative humidity, and regional rates),
laboratory techniques (preparation of thin sections, optical
measurement of hydration rim/rind thickness), and the
conversion of hydration rate data to age. In their review,
OHD was however promoted as a relatively inexpensive
and “simple” scientific technique for determining absolute
chronologies, so that the current masterful treatments
presented replaced the earlier general assessment.
A series of OHD dates was reported consecutively by
the UCLA laboratory (see Meighan and Scalise, 1988).
Braswell (1992) pointed out problems concerning lab-
oratory determinations of diffusion rate constants and the
estimation of EHT in the field, from a study of the Coner
Phase at Copas (Honduras). He concluded that the
laboratory-induced rate constants used to determine
these dates were of questionable validity and needed to
be independently assessed. He showed how an error of
but a few Kelvin in estimated EHT can lead to dates that
are in error by several centuries. In view of the likelihood
of large errors in the Copán obsidian dates, the assertion
that the Late Classic Coner phase should be extended
beyond 900 A.D. was considered therefore premature
(Webster and Freter, 1990; Braswell, 1992).
Further application in OHD includes the dating of New
Zealand obsidians by Stevenson et al. (1996), with a
recent age span from 13th to 18th A.D., where hydration
rates were calculated on the basis of obsidian water con-
tent as estimated from artifact density, while effective
hydration temperatures were established from published
temperatures derived from thermal cells. The deduced
hydration dates have been successfully compared with
radiocarbon dates from six archaeological sites in the
North Island of New Zealand.
Ambrose (1994) gave a provisional age-depth model
for hydration thicknesses for the Pamwak site, Manus
Island, Papua New Guinea, spanning from 4 to 13 ka BP,
based on internal crack surfaces and compared satisfacto-
rily with 14C age control.
Obsidian hydration dating has served as one of the
chronological indicators for the Hopewell Culture earth-
works (ca. 200 B.C.–A.D. 500) in central Ohio
(Stevenson et al., 2004). This work presented new obsid-
ian hydration dates developed from high precision hydra-
tion layer depth profiling using SIMS, and the data
suggested that long-distance exchange in obsidian
occurred throughout the Hopewell period. From Eq. (2)
parameters, diffusion coefficient A and activation energy
E were estimated by the density versus OH plot;
EHT from cell pairs put in the site and Eq. (2) resulted
in OHD age estimates that range between 258 and
119 B.C. and A.D. 607–94, and these incorporate
the entire Hopewell period, as determined by absolute
C-14 dates of the same site.
Based upon the water diffusion in natural glass, Steven-
son et al. (2007) introduced the conventional OHD tech-
nique to dating high-calcium manufactured glasses with
first preliminary results on seventeenth-century North
American glass wine bottles manufactured by European
settlers, concordant with historical data.
In many cases, intact internal fissures as concealed
cracks can be used for OHD, and the rate of surface loss
can be calculated from paired readings of these inner sur-
faces and the outer weathered (depleted) ones (Ambrose,
1998). When secondary rate factors for these fissures were
compared to radiocarbon-based rates, a discrepancy was
found due to dissolution.
At any rate, since OHD is dependent on processes
which have occurred since an artifact was created, it can
never be as accurate and reliable as radiocarbon or dendro-
chronology, and it should be always a secondary source of
chronological data. Thus, in this frame, using a radiocar-
bon assay remains the basis for relative hydration dating;
but this raises again the question, for a given site, of the
true association of the artifact and the 14C date and, for
nearby sites, the remaining uncertainty of the validity of
the constancy of environmental characteristics.
The ongoing Easter Island OHD by Stevenson
et al. (2011) with more than 800 obsidian specimens ana-
lyzed has been advanced with the IR-PAS measurements
of D and water, and ages are comparable to 14C ones in
the time span of fourteenth to nineteenth century A.D.
SIMS-SS has been applied to obsidians from several
parts of the world (e.g., Greece, Chile, Turkey, Japan,
Hungary, the USA) with ages from a few hundred to
30,000 years BP (Liritzis and Laskaris, 2009, 2012).
Analysis of errors shows that it is unrealistic to expect
age errors much less than 15–20 % in obsidian hydration
dating (Rogers, 2010). With those uncertainties in the
computed ages, great caution must be exerted in inferring
occupation duration of a site used based on OHD.
Conclusion
The process of hydration in glasses is very complex,
involving chemistry, physics, and mathematical modeling.
Obsidian hydration has certain weaknesses in terms of a
chronometric tool like any other dating method. At any
rate, two fundamental equations involved need further
experimentation or prior calibration: an age equation of
classical OHD and a temperature dependence equation,
which is usually of the Arrhenius form and the SIMS-
SS. Since the transport parameters being measured differ,
the age equations differ slightly as well, in particular the
exponent of the time (age) variable. It is likely that the
parameters of the Arrhenius equation are slightly different
as well, although investigations of this point have not yet
been reported. However, it would be unwise to apply,
say, an activation energy determined from water mass
uptake to the optical (classical OHD) case without careful
 OBSIDIAN HYDRATION DATING
study or EHT and hydration experiments at ambient
conditions.
Despite the fact that the classical OHD is furthest
removed from the basic physics of hydration (criticism
by advocates of other techniques, e.g., Anovitz
et al., 1999; Liritzis and Laskaris, 2012), it is still being
used since it seems to satisfy coarse chronologies espe-
cially of recent centuries, whereas other chronometric data
are frequently lacking (Stevenson et al., 2013). The alter-
native SIMS-SS with the mathematical techniques
discussed also offers satisfactory dates for the past
30,000 years (Liritzis and Laskaris, 2009, 2011, 2012).
Probably, the greatest needs at present are for (1) a more
scientific approach of the age equation; (2) more dates on a
“join forces” basis, for the two current approaches, for
specimens throughout the world, and comparison with
luminescence and C-14; and (3) development of not
expensive versions of OHD.
Bibliography
Ambrose, W. R., 1976. Intrinsic hydration rate dating of obsidian. In
Taylor, R. E. (ed.), Advances in Obsidian Glass Studies: Archae-
ological and Geochemical Perspectives. Park Ridge, NJ: Noyes
Press, pp. 81–1054.
Ambrose, W. R., 1984. Soil temperature monitoring at Lake
Mungo; implications for racemisation dating. Australian
Archaeology, 19, 64–74.
Ambrose, W. R., 1994. Obsidian hydration dating of a Pleistocene
age site from the Manus Island, Papua New Guinea. Quaternary
Geochronology (Quaternary Science Reviews), 13, 137–142.
Ambrose, W. R., 1998. Obsidian hydration dating of a recent age
obsidian mining site in Papua New Guinea. In Shackley, M. S.
(ed.), Method and Theory in Archaeological Obsidian Studies.
New York: Plenum Press, pp. 205–222.
Ambrose, W. R., and Novak, S. W., 2012. Obsidian hydration chro-
nometrics using SIMS and optical methods from 26-year-
temperature controlled exposures. In Liritzis, I., and Stevenson,
C. M. (eds.), Obsidian and Ancient Manufactured Glasses.
Albuquerque: New Mexico Press, pp. 15–25, Chap. 2.
Ambrose, W., and Stevenson, C. M., 2004. Obsidian density, con-
nate water and hydration dating. Journal of Mediterranean
Archaeology and Archaeometry, 4, 1–10.
Anovitz, L. M., Elam, J. M., Riciputi, L. R., and Cole, D. R., 1999.
The failure of obsidian hydration dating: sources, implications
and new directions. Journal of Archaeological Science,
26, 735–752.
Anovitz, L. M., Elam, J. M., Riciputi, L. R., and Cole, D. R., 2004.
Isothermal time-series determination of the rate of diffusion of
water in Pachuca obsidian. Archaeometry, 46(2), 301–326.
Anovitz, L. M., Cole, D., and Fayek, M., 2008. Mechanisms of rhy-
olitic glass hydration below the glass transition. American Min-
eralogist, 93, 1166–1178.
Bartholomew, R., Tick, P., and Stookey, S., 1980. Water/glass reac-
tions at elevated temperatures and pressures. Journal of
Non-Crystalline Solids, 38/39, 637–642.
Born, M., and Emil, W., 1980. Principles of Optics, 6th edn. New
York: Pergamon Press.
Braswell, G. E., 1992. Obsidian-hydration dating, the coner phase,
and revisionist chronology at Copan. Honduras Latin American
Antiquity, 3(2), 130–147.
Brodkey, S. R., and Liritzis, I., 2004. The dating of Obsidian: a pos-
sible application for transport phenomena (a tutorial). Mediterra-
nean Archaeology and Archaeometry, 4(2), 67–82.
621
Brückner, R., 1970. Properties and structure of vitreous silica
I. Journal of Non-Crystalline Solids, 5(2), 123–175.
Crank, J., 1975. The Mathematics of Diffusion. Oxford: Oxford
University Press.
Doremus, R.H (1969) The diffusion of water in fused silica. In
Mitchell, J.W, Devries, R.C, Roberts, R.W, Cannon, P (eds),
Reactivity in solids, Wiley, New York, 667–673
Doremus, R. H., 2000. Diffusion of water in rhyolite glass: diffu-
sion–reaction model. Journal of Non-Crystalline Solids, 261,
101–107.
Doremus, R. H., 2002. Diffusion of Reactive Molecules in Solids
and Melts. New York: Wiley-Interscience.
Duerden, P., Cohen, D., and Ambrose, W., 1982. The measurement
of hydration profiles in obsidian in archaeometry: An Austral-
asian prospective In Ambrose, W., and Duerden, P. (eds.),
Archaeometry: An Australian Prospective. Canberra: Australian
National University, pp. 236–242.
Ebert, W. L., Hofburg, R. F., and Bates, J. K., 1991. The sorption of
water on obsidian and a nuclear waste glass. Physics and Chem-
istry of Glasses, 32(4), 133–137.
Eerkens, J. W., Vaughn, K. J., Carpenter, T. R., Conlee, C. A.,
Grados, M. L., and Schreiber, K., 2008. Obsidian hydration dat-
ing on the south Coast of Peru. Journal of Archaeological Sci-
ence, 35(8), 2231–2239.
Ericson, J. E., Dersch, O., and Rauch, F., 2004. Quartz hydration
dating. Journal of Archaeological Science, 31, 883–902.
Findlow, F. J., and Bennett, V. C., 1978. In Meighan, C. W., and
Vandehoeven, P. I. (eds.), Obsidian Dates II. Los Angeles, CA:
UCLA Institute of Archaeology Monograph, Vol. VI,
pp. 127–164.
Friedman, I., and Long, W., 1976. Hydration rate of obsidian.
Science, 159, 347–352.
Friedman, I., and Smith, R. L., 1960. A new dating method using
obsidian: the development of the method. American Antiquity,
25, 476–522.
Friedman, I., Trembour, F. W., Smith, F. L., and Smith, G. I., 1994.
Is obsidian hydration affected by relative humidity? Quaternary
Research, 41, 185–190.
Friedman, I., Trembour, F. W., and Hughes, R. E., 1997. Obsidian
hydration dating. In Taylor, R. E., and Aitken, M. J. (eds.), Chro-
nometric Dating in Archaeology. New York: Plenum Press.
Chapter 10.
Hull, K. L., 2001. Re-asserting the utility of obsidian hydration dat-
ing: a temperature-dependent empirical approach to practical
temporal resolution with archaeological obsidians. Journal of
Archaeological Science, 28, 1025–1040.
Hall, K. L., and Jackson, R. J., 1989. Obsidian hydration rates in
California. In Hughes, R. E. (ed.), Current Directions in Califor-
nia Obsidian Studies. Berkeley: Contributions of the University
of California Archaeological Research Facility No.48,
pp. 31–58.
Jacobs, D. H., 1943. Fundamentals of Optical Engineering.
New York: McGraw-Hill.
Karsten, J. R., and Delaney, J. L., 1981. Ion microprobe studies of
water in silicate melts: concentration-dependent water diffusion
in obsidian. Earth and Planetary Science Letters, 52, 191–202.
Karsten, J. L., Holloway, J. R., and Delaney, J. L., 1982. Ion micro-
probe studies of water in silicate melts: temperature-dependent
water diffusion in obsidian. Earth and Planetary Science Letters,
59, 420–428.
Kimberlin, J., 1976. Obsidian hydration rate determinations on
chemically characterized samples. In Taylor, R. E. (ed.),
Advances in Obsidian Glass Studies. New Jersey: Noyes Press,
pp. 63–79.
Lanford, W. A., Davis, K., Lamarche, P., Laursen, T., and Groleau,
R., Doremus R.H. 1979. Hydration of soda-lime glass. Journal
of Non-Crystalline Solids, 33, 249–266.
622 OBSIDIAN HYDRATION DATING
Lee, R. R., Leich, D. A., Tombrello, T. A., Ericson, J. E., and Fried-
man, I., 1974. Obsidian hydration profile measurements using a
nuclear reaction technique. Nature, 250, 44–47.
Liritzis, I., 2006. SIMS-SS A new obsidian hydration dating
method: analysis and theoretical principles. Archaeometry,
48(3), 533–547.
Liritzis, I., 2010. Strofilas (Andros Island, Greece): new evidence
for the Cycladic final Neolithic period through novel dating
methods using luminescence and obsidian hydration. Journal
of Archaeological Science, 37(6), 1367–1377.
Liritzis, ., and Diakostamatiou, ., 2002. Towards a new method of
obsidian hydration dating with secondary ion mass spectrometry
via a surface saturation layer approach. Mediterranean Archae-
ology and Archaeometry, 2(1), 3–20.
Liritzis, I., and Laskaris, N., 2009. From micron to nano scale in
rhyolites: SEM, PLM and AFM in obsidian hydration dating
by SIMS-SS. In Aegean Nanoscience and Nanotechnology
Workshop. 2nd International Conference from “Nanoparticles
& nanomaterials to nanodevices & nanosystems”, Rhodes
28 June–3 July 2009. Book of Abstracts University of Texas at
Abington, (Meletis et al, organizers), p. 107.
Liritzis, I., and Laskaris, N., 2011. Fifty years of obsidian hydration
dating in archaeology. Journal of Non-Crystalline Solids, 357,
2011–2023.
Liritzis, I., and Laskaris, N., 2012. The SIMS-SS obsidian hydration
dating method. In Ioannis, L., and Christopher, S. (eds.),
Obsidian and Ancient Manufactured Glasses. Albuquerque:
University of New Mexico Press, pp. 26–45.
Liritzis, I., and Stevenson, C. M. (eds.), 2012. Obsidian and Ancient
Manufactured Glasses. Albuquerque, NM: University of New
Mexico Press.
Liritzis, I., Diakostamatiou, M., Stevenson, C. M., Novak, S. W.,
and Abdelrehim, I., 2004. Dating of hydrated obsidian surfaces
by SIMS-SS. Journal of Radioanalytical and Nuclear Chemis-
try, 261(i), 51–60.
Liritzis, I., Stevenson, C., Novak, S., Abdelrehim, I., Perdikatsis, V.,
and Bonini, M., 2007. New prospects in obsidian hydration dat-
ing: an integrated approach. In Proceedings of the Hellenic
Archaeometry Society. Athens, British Archaeological Reports
(BAR) International Series, pp. 9–22.
Liritzis, I., Bonini, M., and Laskaris, N., 2008a. Obsidian hydration
dating by SIMS-SS: surface suitability criteria from atomic force
microscopy. Surface and Interface Analysis, 40, 458–463.
Liritzis, I., Laskaris, N., and Bonini, M., 2008b. Nano- and micro-
scale resolution in ancient obsidian artefact surfaces: the impact
of AFM on the obsidian hydration dating by SIMS-SS. Physica
Status Solidi, 5(12), 3704–3707.
Mazer, J. J., Stevenson, C. M., Ebert, W. L., and Bates, J. K.,
1991. The experimental hydration of obsidian as a function of
relative humidity and temperature. American Antiquity, 56,
504–513.
Meighan, C.W., 1976. In Taylor, R. E. Empirical Determination of
Obsidian Hydration Rates from Archaeological Evidence
Advances in Obsidian Glass Studies. Noyes Press: Park Ridge,
New Jersey, pp. 106–119, ISBN 978-0-8155-5050-1.
Meighan, C. W., and Scalise, J. L., 1988. A Compendium of the
Obsidian Hydration Determinations Made at the UCLA Obsid-
ian Hydration Laboratory: Obsidian Dates. California, LA:
Institute of Archaeology, University of California, Monograph
XXIX, Vol. IV, pp. 473–511.
Meighan, C. W., Foote, L. J., and Aiello, P. V., 1968. Obsidian dat-
ing in West Mexican archaeology. Science, 160, 1069–1075.
Michels, J. W., 1986. Obsidian hydration dating. Endeavour, 10(2),
97–100.
Michels, J. W., Tsong, I. S. T., and Smith, G. A., 1983. Experimen-
tally derived hydration rates in obsidian dating. Archaeometry,
25(2), 107–117.
Morganstein, M. E., Wicket, C. L., and Barkatt, A., 1999. Consider-
ations of hydration-rind dating of glass artifacts: alteration
morphologies and experimental evidence of hydrogeochemical
soil-zone pore water control. Journal of Archaeological Science,
26, 1193–1210.
Newman, S., Stolper, E., and Epstein, S., 1986. Measurement of
water in rhyolitic glasses: calibration of an infrared spectroscopic
technique. American Mineralogist, 71, 15271541.
Nowak, M., and Behrens, H., 1997. An experimental investigation
of diffusion of water in haplogranitic melts. Contributions to
Mineralogy and Petrology, 126, 365–376.
Riciputi, L. R., Elam, M. J., Anovitz, L. M., and Cole, D. R., 2002.
Obsidian diffusion dating by secondary ion mass spectrometry: a
test using results from Mound 65, Chalco, Mexico. Journal of
Archaeological Science, 29, 1055–1075.
Ridings, R., 1996. Where in the world does obsidian hydration
dating work? American Antiquity, 61, 136–148.
Rogers, A. K., 2007. Effective hydration temperature of obsidian: a
diffusion-theory analysis of time-dependent hydration rates.
Journal of Archaeological Science, 34, 656–665.
Rogers, A. K., 2008. Obsidian hydration dating: accuracy and reso-
lution limitations imposed by intrinsic water variability. Journal
of Archaeological Science, 35, 2009–2016.
Rogers, A. K., 2010. Accuracy of obsidian hydration dating based
on obsidian-radiocarbon association and optical microscopy.
Journal of Archaeological Science, 37, 3239–3246.
Rogers, A. K., 2012. Temperature correction for obsidian hydration
dating. In Liritzis, I., and Stevenson, C. (eds.), Obsidian and
Ancient Manufactured Glasses. Albuquerque: University of
New Mexico Press, pp. 46–56.
Rogers, A. K., and Duke, D., 2011. An archaeologically validated
protocol for computing obsidian hydration rates from laboratory
data. Journal of Archaeological Science, 38, 1340–1345.
Rogers, A. K., and Yohe II R. M., 2011. An improved equation for
coso obsidian hydration dating, based on obsidian-radiocarbon
association. Proceedings of the Society for California Archaeol-
ogy, Vol. 25, 1–15. (http://scahome.org/sca-publications/arti-
cles-of-the-sca-proceedings/).
Scheetz, B. E., and Stevenson, C. M., 1988. The role of resolution
and exponential ground temperature and depth of artefact recov-
ery. Journal of Filed Archaeology, 18, 77–85.
Silver, L., Ihinger, P., and Stolper, E., 1990. The influence of bulk
composition on the speciation of water in silicate glasses. Contri-
butions to Mineralogy and Petrology, 104, 142–162.
Smith, J. M., and Van Hess, H. C., 1987. Introduction to
Chemical Engineering Thermodynamics, 4th edn. New York:
McGraw-Hill.
Stevenson, C. M., and Novak, S. W., 2011. Obsidian hydration dat-
ing by infrared spectroscopy: method and calibration. Journal of
Archaeological Science, 38, 1716–1726.
Stevenson, C. M., and Scheetz, B. E., 1989. Induced hydration rate
development for obsidians from the Coso Volcanic Field: a com-
parison of experimental procedures. In Current Directions in
California Obsidian Studies, Contributions of the University of
California Archaeological Research Facility No. 48. Berkeley:
University of California, pp. 23–30.
Stevenson, C. M., Carpenter, J., and Scheetz, B., 1989a. Recent
advances in the experimental determination and applications of
obsidian hydration rates. Archaeometry, 31, 193–206.
Stevenson, C. M., Dinsmoor, D., and Scheetz, B., 1989b. An inter-
laboratory comparison of hydration rim measurements.
Internationales Association of Obsidian Studies Newsletter,
1, 7–13.
Stevenson, C. M., Knauss, E., Mazer, J. J., and Bates, J. K., 1993.
The homogeneity of water content in obsidian from the Coso
volcanic field: implications for obsidian hydration dating.
Geoarchaeology, 8(5), 371–384.
 OBSIDIAN HYDRATION DATING
Stevenson, C. M., Sheppard, P. J., and Sutton, D. G., 1996.
Advances in the hydration dating of New Zealand Obsidian.
Journal of Archaeological Science, 23(2), 233–242.
Stevenson, C. M., Mazer, J. J., and Scheetz, B. E., 1998. Laboratory
obsidian hydration rates: theory, method and application.
In Shackley, M. S. (ed.), Archaeological Obsidian Studies:
Method and Theory. New York: Plenum Press. Advances in
Archaeological and museum Science, Vol. 3, pp. 181–204.
Stevenson, C. M., Gottesman, M., and Macko, M., 2000.
Redefining the working assumptions of obsidian hydration
dating. Journal of California and Great Basin Archaeology,
22(2), 223–236.
Stevenson, C. M., Abdelrehim, I., and Novak, S. W., 2001. Infrared
photoacoustic and secondary ion mass spectrometry measure-
ments of obsidian hydration rims. Journal of Archaeological
Science, 28, 109–115.
Stevenson, C. M., Liritzis, I., Diakostamatiou, M., and Novak,
S. W., 2002. Investigations towards the hydration dating of
aegean obsidian. Mediterranean Archaeology and
Archaeometry, 2(1), 93–109.
Stevenson, C. M., Abdelrehim, I., and Novak, S. W., 2004. High
precision dating of obsidian from the Hopewll Site and Mound
City Group earthworks using secondary ion mass spectrometry.
American Antiquity, 69, 555–568.
Stevenson, C.M., Ladefoged, T.N., Haoa, S., 2007. An upland
agricultural residence on Rapa Nui: occupation of a hare oka
(18-473G) in the Vaitea region. Archaeology in Oceania 42,
72–78.
Stevenson, C. M., Ladefoged, T., and Novak, S. W., 2013. Prehis-
toric settlement chronology on RapaNui, Chile: obsidian hydra-
tion dating using infrared photoacoustic spectroscopy. Journal of
Archaeological Science, 40, 3021–3030.
Stolper, E. M., 1982. Water in silicate glasses: An infrared spectro-
scopic study. Contrib. Mineral. Petrol, 81, p. 1–17.
Tokoyama, T., Okumura, S., and Nakashima, S., 2008. Hydration of
rhyolitic glass during weathering as characterized by IR microspec-
troscopy. Geochimica et Cosmochimica Acta, 72, 117–125.
Tomozawa, M., 1985. Concentration dependence of the diffusion
coefficient of water in SiO2 glass. Journal of the American
Ceramic Society, 68, C251–C252.
Tsong, I. S. T., Houser, C. A., Yusuf, N. A., Messier, R. F.,
White, W. B., and Michels, J. W., 1978. Obsidian hydration
623
profiles measured by sputter-induced optical emission. Science,
201, 339–341.
Tsong, I. S. T., Houser, C. A., and Tsong, S. S. C., 1980. Depth pro-
files of interdiffusing species in hydrated glasses. Physic and
Chemistry of Glasses, 21, 197–198.
Tsong, I. S. T., Smith, G. A., Michels, J. W., Wintenberg, A. L.,
Miller, P. D., and Moak, C. D., 1981. Dating of obsidian artifacts
by depth-profiling of artificially hydrated surface layers. Nuclear
Instruments & Methods, 191, 403–407.
Webster, D., and Freter, A., 1990. Settlement history and the classic
Collapse at Copan: a redefined chronological perspective. Latin
American Antiquity, 1(1), 66–85.
West, G. J., Woolfenden, W., Wanket, J. A., and Anderson, R. S.,
2007. Late pleistocene and holocene environments. In Jones,
T. L., and Klar, K. A. (eds.), California Prehistory: Coloniza-
tion, Culture, and Complexity. Walnut Creek: Altamira Press,
pp. 11–34.
Yokoyama, T., Okumura, S., and Nakashima, S., 2008. Hydration of
rhyolitic glass during weathering as characterized by IR
microspectroscopy. Geochimica et Cosmochimica Acta,
72, 117–125.
Zhang, Y., and Behrens, H., 2000. H2O diffusion in rhyolitic melts
and glasses. Chemical Geology, 169, 243–262.
Zhang, Y., Stolper, E. M., and Wasserburg, G. J., 1991. Diffusion of
water in rhyolytic glasses. Geochimica et Cosmochimica Acta,
55, 441–456.
Zhang, Y., Belcher, R., Ihinger, P. D., Wang, L., Xu, Z., and
Newman, S., 1996. New calibration of infrared measurement
of dissolved water in rhyolitic glass. Geochimica et
Cosmochimica Acta, 61, 3089–3100.
URL
http://www.obsidianlab.com/info_oh.html
http://www.rhodes.aegean.gr/tms/sims-ss/index.html
Cross-references
14
C in Plant Macrofossils
Luminescence Dating
Luminescence, Flints and Stones
 P
PALEOSOL
Gary E. Stinchcomb
Department of Geosciences, Pennsylvania State
University, University Park, PA, USA
Definition
Paleosol is defined as a soil that formed on a landscape of
the past (Yaalon et al., 1971; Retallack, 2001). The defini-
tion of paleosol is not straightforward as it can vary by
geologic discipline. Quaternary geologists often define
paleosols as having been buried, exhumed, and/or at the
surface as relict soils. In this case, a paleosol must contain
evidence of a nonextant soil-forming environment, e.g.,
glacial climate (Birkeland, 1999). Geologists who primar-
ily work in the pre-Quaternary often define paleosols as
lithified, fossil soils that document a past landscape
(Retallack, 2001).
Like modern soils, paleosols record evidence of the
physical, chemical, or biological alteration of
a preexisting material to a more stable form, i.e.,
weathering. These weathering features formed partly as
a result of five primary environmental factors: (i) climate,
(ii) organisms, (iii) relief, and (iv) parent material operat-
ing through (v) time (Jenny, 1941). Paleosols are therefore
valued for their record of paleoenvironment(s) and
paleoclimate(s) in both Quaternary and pre-Quaternary
records. Because paleosols record past landscape condi-
tions through an interval of time, they are often subject
to a variety of dating methods. Generally speaking, dating
methods applied to soils and paleosols are used to assess
the duration of soil formation and/or provide a numerical
age estimate during which the soil formed.
J.W. Rink, J.W. Thompson (eds.), Encyclopedia of Scientific Dating Methods, DOI 10.1007/978-94-007-6304-3,
© Springer Science+Business Media Dordrecht 2015
The duration of formation may be assessed qualita-
tively, e.g., “young” or “old” soil, or quantitatively, e.g.,
a soil formed for 2,000 years. Soil geomorphologists study
how modern soils vary as a function of time when all other
properties are held constant, i.e., a soil chronosequence
(Jenny, 1941; Birkeland, 1999; Holliday, 2004 – see
Table 7.2). A number of profile properties have been
shown to vary as a function of duration of soil formation,
young to old. Age-related properties derived from
chronosequences have been used in geomorphic and
geoarchaeological research of paleosols; however, there
are a number of potential pitfalls and caution should be
exercised when trying to obtain a numerical age estimate
using this approach (Holliday, 2004).
Absolute dating methods are applied to paleosols and
buried soils to derive numerical estimates. Common
approaches include radiocarbon (14C), U-series, lumines-
cence techniques, and cosmogenic-nuclide exposure dat-
ing (Holliday, 2004; Balco and Rovey, 2008).
Radiocarbon is perhaps the most frequently method used
to derive a numerical age estimate for late Quaternary
paleosols (Holliday, 2004). Radiocarbon of soil organic
matter from a buried surface horizon can be dated to derive
an apparent mean residence time (Holliday, 2004 and ref-
erences therein). Both radiocarbon and luminescence dat-
ing have been applied to eolian and alluvial parent
materials on which the paleosol formed to derive
a maximum numerical estimate (age-maxima) of soil for-
mation (Holliday, 2004).
U-series dating has been applied to pedogenic carbon-
ates to assess the age of a paleosol or buried soil (Holliday,
2004). Recent advances in cosmogenic-nuclide exposure
dating have refined the burial-age estimates for paleosols
using an isochron approach (Balco and Rovey, 2008).
 210
626 Pb DATING

The slope of 10Be and 26Al from concentration profiles in already been developed. Krishanswami et al. (1971)
paleosols has been used to derive a numerical estimate of extended the use of this technique to lacustrine deposits
burial age. to reconstruct depositional histories of lake sediment
and, maybe more importantly, contaminant inputs and
Bibliography burial. Thus, the powerful tool for dating recent (up to
Balco, G., and Rovey, C. W., 2008. An isochron method for cosmo-
about one century old) sediment deposits was established
genic-nuclide dating of buries soils and sediments. American and soon widely adopted. Today almost all oceanographic
Journal of Science, 308, 1083 (1880). or limnologic studies that address recent depositional
Birkeland, P. W., 1999. Soils and Geomorphology. Oxford: Oxford reconstructions employ 210Pb as one of several possible
University Press. geochronometers (Andrews et al., 2009; Gale, 2009;
Holliday, V. T., 2004. Soils in Archaeological Research. Oxford: Baskaran, 2011; Persson and Helms, 2011). This entry
Oxford University Press. presents a short overview of the principles of 210Pb dating
Jenny, H., 1941. Factors of Soil Formation: A System of Quantita-
tive Pedology. New York: McGraw-Hill. and provides a few examples that illustrate the utility of
Retallack, G. J., 2001. Soils of the Past: An Introduction to this tracer in contrasting depositional systems. Potential
Paleopedology, 2nd edn. Malden, MA, Oxford: Blackwell Science. caveats and uncertainties (Appleby et al., 1986; Binford,
Yaalon, D. H., International Society of Soil Science, and Interna- 1990; Binford et al., 1993; Smith, 2001; Hancock
tional Union for Quaternary Research, 1971. Paleopedology: et al., 2002) inherent to the use and interpretation of
Origin, Nature, and Dating of Paleosols; Papers. Jerusalem: 210
Pb-derived age models are also introduced. Recom-
International Society of Soil Science.
mendations as to best practices for most reliable uses and
reporting are presented in the summary.
Cross-references
Luminescence Dating 210
Radiocarbon Dating
Pb dating principles
Terrestrial Cosmogenic Nuclide Dating Pb-210 is a naturally occurring radionuclide of the 238U
U-Series Dating radioactive decay chain and has a half-life (t1/2) of
22.23 years (Figure 1). In most environmental systems
that have remained “closed” for more than about 120 years
210
Pb DATING
238 - Uranium series
Peter W. Swarzenski
US Geological Survey, Pacific Coastal and Marine
Science Center, Santa Cruz, CA, USA 238 234
Uranium U U
(4.5E9 a) (2.5E5 a)
Synonyms 234
Proactinium Pa
Mass accumulation rates; Recent sediment dating; Sedi- (6.7 hr)
mentation rates 234 230
Thorium Th Th
(24.1 d) (7.5E4 a)
Definition
Method of dating recent sediment deposition and accumu-
lation using down-core profiles of short-lived radioactive 226
Radionuclide
210
Pb. Radium Ra (half-life)
(1.6E3 a) Alpha
Decay
History Beta
Roughly 50 years ago, a small group of scientists from Radon
222
Rn Decay
Belgium and the USA, trying to better constrain ice sheet (3.8 d)

accumulation rates, attempted to apply what was then

known about environmental lead as a potential 210
geochronometer. Thus Goldberg (1963) developed the Polonium
218
Po
214
Po Po
(3.04 m) (164 µs) (138.4 d)
first principles of the 210Pb dating method, which was 210
soon followed by a paper by Crozaz et al. (1964), who Bismuth
214
Bi Bi
(19.7m) (5.01d)
examined accumulation history of Antarctic snow using
210 214 210 206
Pb. Shortly thereafter, Koide et al. (1972, 1973) Lead Pb Pb Pb
(26.9 m) (22.23 y) stable
adapted this technique to unravel sediment deposition
and accumulation records in deep-sea environments. Ser- 210
Pb Dating, Figure 1 The 238U decay chain showing 226Ra,
endipitously, they chose to work in a deep basin off Cali- 222
Rn, 210Pb, and 210Po. Half-lives are shown in parentheses
fornia, where an independent and robust age model had where a years, d days, hr hours, m minutes, and ms microseconds.
 210
Pb DATING 627

e 210 Pb radio
activ activ
radio e
y deca
222 deca y
Rn 222Rn

gas evasion gas evasion

wet/dry deposition
226
Ra
222
erosion Rn
runoff
226
Ra
226 groundwater
Ra 222
Rn 226
Ra
groundwater 222
flow
Rn 222
Rn
flow

238 226 222 210

U Ra Rn Pb
radioactive decay

210
Pb Dating, Figure 2 Conceptual illustration of the dominant sources and transport pathways for 210Pb.

(five times the half-life of 210Pb), 210Pb is derived from its
parent radionuclide, 226Ra (t1/2 ¼ 1,600 years), that is pre-
sent in the material will reach a state of radioactive secular
equilibrium. In such settings, 210Pb is not a viable
geochronometer. However, in more recently deposited
sediment, 210Pb may not be in equilibrium with parent iso-
tope 226Ra and may have an additional source unrelated to
226
Ra. First, a fraction of the observed 210Pb is continu-
ously produced by decay of 226Ra and represents
“supported” 210Pb, which is considered time independent.
The second fraction originates from the decay of atmo-
spheric 222Rn into 210Pb and is called “excess” 210Pb.
After wet/dry deposition, rapid scavenging processes
remove this radionuclide from the water column
(Chanton et al., 1983; Crusius and Anderson, 1995) via
adsorption onto suspended particulates which may then
become incorporated in sediment (Figure 2). Bottom sed-
iment thus contains a mixture of both supported and
excess 210Pb. This excess 210Pb predictably decays with
its 22.23 year half-life and can be used to establish recent
sediment geochronologies (Goldberg and Bruland, 1974;
Appleby and Oldfield, 1983) under ideal depositional
conditions.
210
If both sediment accumulation and the flux of excess
Pb to a sediment surface are constant over time, and
there are no postdepositional processes that redistribute
excess 210Pb present in the sediment, then a down-core
profile of excess 210Pb will follow a simple exponential
curve that tracks its decay. Although these two assump-
tions are not always observed in natural settings, this sim-
ple premise forms the underlying principle for use of 210Pb
as a geochronometer. More intricate models have been
developed that address nuances in the flux of sediment
and 210Pb (Sanchez-Cabeza and Ruiz-Fernandez, 2012),
as well as the effects of postdepositional mixing
(Pourchet et al., 1989), including bioturbation
(Benninger et al., 1979; Nittrouer et al., 1983).
Goldberg (1963) first proposed the constant flux model,
where the 210Pb flux to sediment is assumed to be constant
over time, while the sedimentation rate may vary. This
model subsequently became known as the constant rate
of supply (CRS) model (Appleby and Oldfield, 1978;
Robbins, 1978; Benoit and Rozan, 2001), which is still
one of the most often used 210Pb dating models
(Sanchez-Cabeza et al., 2000; Persson and Helms 2011).
This model assumes that the down-core excess 210Pb
activity, vertically integrated to a depth, x, or a cumulative
dry mass, m, will equal the flux integrated over the
corresponding time interval. Integrating to either x or m,
Aðx, mÞ ¼ A0 ðltÞ ð1Þ
where A(x, m) is the cumulative residual unsupported
or excess 210Pb activity beneath sediment of depth x,
 210
628 Pb DATING
or mass m, A0 is the total unsupported 210Pb activity in the
sediment column, l is the 210Pb decay constant,
0.03114 year1, and t represents time. The age of sediment
at depth x, m is then described by:

   requires only minor (nondestructive) laboratory work,
1 A0
t¼ ln ð2Þ
l Aðx, mÞ
Unlike the CRS model, in the constant initial concen-
tration (CIC) model, the 210Pb supply varies directly in
proportion to the sedimentation rate (Shukla and Joshi,
1989). Here the age (t) of a sediment layer is related to
the depth (m) by
t ¼ m=s ð3Þ
2
where m denotes the mass depth in the core (g cm ) and s
denotes the sedimentation rate (g cm2 year1). The
unsupported 210Pb activity (A(m)) will thus vary with
depth in accordance with the formula:
AðmÞ ¼ Að0Þelm=t ð4Þ
where l is the decay constant and A(0) is the unsupported
activity at the surface of the core. When plotted on a loga-
rithmic scale, the resulting 210Pb activity versus depth pro-
file will appear linear, and the mean sedimentation rate, s,
can then be determined using a least squares fit procedure.
With these simple equations, 210Pb often can be used as a
reliable geochronometer under suitable depositional con-
ditions, although volumes have been written on more elab-
orate models and their limitations.
Many processes influence the delivery of excess 210Pb
to the bottom sediment of a lake or ocean. For example,
the atmospheric flux of 210Pb varies by latitude, altitude,
and season (Garcia-Orellana et al., 2006; Baskaran and
Swarzenski, 2007). Catchment size and geology impact
scavenging efficiencies and ensuing transport rates of
210
Pb (Nittrouer et al., 1983). Water column depth and res-
idence times and suspended particle composition may also
affect 210Pb cycling (Turner and Delorme, 1996). Based
on Appleby (2008), the CRS model will likely produce
valid geochronologies if sediment transport rates from
the catchment are reasonably constant and produce a
minor component of supported 210Pb compared to atmo-
spheric flux rates contributing to excess 210Pb.
210
Pb measurements
Traditionally, the analytical determination of 210Pb was
based on either radiochemical separation of 210Pb and
measurement of its short-lived beta-emitting daughter,
210
Bi (t1/2 ¼ 5.01 day), or on alpha spectrometry of its
indirect decay product 210Po (t1/2 ¼ 138.4 day) that was
assumed to be in secular equilibrium with 210Pb. While
both of these methods are laborious, today’s alpha-
spectrometric systems often consist of sophisticated, inte-
grated sets of multiple detectors so one sediment core can
theoretically be counted in just a few days. Advanced
planar- or well-type semiconductor detectors enable the
gamma-spectrometric analysis of the 46.5 keV decay
energy of 210Pb with high relative efficiency (Zaborska
et al., 2007). Because gamma analysis of sediment
the use of gamma detectors has considerably reduced the
effort for most routine 210Pb determinations. In addition,
the gamma-spectrometric method provides a concurrent
determination of 226Ra, which allows the activities of
supported 210Pb to be estimated. Ra-226 activity is deter-
mined by quantifying intermediate daughter radionuclides
214
Pb (at 295 and 352 keV) and 214Bi (at 609 keV) after
establishing their radioactive equilibrium with 222Rn
(Kirchner and Ehlers, 1998; Swarzenski et al., 2006).
Importantly, the gamma-spectrometric method also pro-
vides for the determination of 137Cs (t1/2 ¼ 30.17 year;
661 keV), an independent, nuclear bomb-produced
geochronometer.
Based on our own laboratory experience, detection
limits for 210Pb are typically less than 10 mBq g1 for
well-type HPGe gamma detectors and are likely some-
what lower for planar detectors that can accommodate
much larger sediment sample sizes (~20–50 g). Thus,
detection limits for gamma detectors may be considerably
higher than those achievable with alpha-spectrometric
counting of radiochemically separated 210Po. As a conse-
quence, gamma spectrometry may be of limited use in
areas showing small atmospheric deposition rates of
210
Pb.
Prior knowledge of the analytical detection limits of the
instrumentation as well as the sample size and geometry
can affect the decision on how to best section a sediment
core for 210Pb analyses (Kirchner, 2011). Because of the
inverse relationship between particle surface area and par-
ticle size, muddy sediment usually contains the highest
activities of adsorbed 210Pb (Smith and Walton, 1980).
Thinner sediment layers expectedly provide a better time
resolution of sedimentation rates but may compromise
counting statistics if there is not enough excess 210Pb pre-
sent within a sediment layer. For a given flux of excess
210
Pb into the sediment, its activity concentrations are
inversely related to the sedimentation rate (Alexander
and Lee, 2009). Thus, some prior estimate of the expected
sedimentation rate is also desirable to guide the core sec-
tioning. An independent estimate of sedimentation rate
may be available, for example, from lithological analyses
of the core (i.e., varves). Otherwise, the core should be
sectioned into thin, consecutive layers and, starting with
the youngest sediment, measured layer by layer until
excess 210Pb activities approach parent-supported activi-
ties. If 210Pb activities approach instrument detection
limits, the sediment layers can be combined prior to
analysis.
After sectioning, each sediment layer is homogenized
and weighed prior to and after drying at 105 C for at least
24 h (Swarzenski et al., 2006). The wet and dry weights
provide water content information. The porosity (f) of
each sample is calculated as follows:
 210
Pb DATING 629

xs210Pb (dpm g–1) 137

Cs (dpm g–1)
0.01 0.1 1 10 100 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
0 0
a Line 2
5 b
b = –0.158601
5 r2 = 0.98
5
10
corr 137Cs
Line 3
10 10
b = –0.26232
Depth (cm)

r2 = 0.83
15
Sedimentation rate: Sedimentation rate:
15 Line 2 15
0.085 cm yr –1 0.09 cm yr –1
r 2 = 0.98
20
20 Line 3 20
0.52 cm yr –1
r 2 = 0.83
25 25 25
San Pedro Basin SPB San Pedro Basin SPB

Sub-core SPB-B1-GC Sub-core SPB-B1-GC

210
Pb Dating, Figure 3 (a) Excess down-core 210Pb (xs210Pb) and (b) 137Cs profiles (in disintegrations per minute, dpm) and derived
sedimentation rates in a deep submarine basin: San Pedro Basin, California. 60 dpm ¼ 1 Bq ¼ 27.027 pCi. In (a) black and gray
symbols distinguish unique linear sediment rates. Summary parameters of the linear regression are represented as the slope (b)
and r-squared (r2) value.

xs210Pb (dpm g–1) 137Cs (dpm g–1)

1 10 100 0 5 10 15 20 25
0 0
a Line 1 b
b = –0.0320292
r2 = 0.93

5 Line 2 5
b = –0.167993
r2 = 0.97
Depth (cm)

10 10 corr 137Cs
Sedimentation rate:
Line 1
Line 3 0.42 cm yr –1
b = –0.06475 r 2 = 0.93
r2 = 0.99
15 Line 2 15 Sedimentation rate:
0.08 cm yr –1
r 2 = 0.97 0.07 cm yr –1

Line 3
20 0.21 cm yr –1 20
r 2 = 0.99

Copper Lake Copper Lake

25 25

210
Pb Dating, Figure 4 Excess 210Pb and 137Cs down-core profiles and derived sedimentation rates in an alpine lake: Copper Lake,
Washington (Redrawn from Sheibley et al., 2012). In (a) white, black, and gray symbols distinguish unique linear sediment rates.
Summary parameters of the linear regression are represented as the slope (b) and r-squared (r2) value.
 210
630 Pb DATING
f ¼ fw =½fw þ ð1  fw Þrw =rs  ð5Þ
where fw is the fraction of water in the wet sediment (¼1 –
dry wt./wet wt.), rw is the density of pore water (assumed
to be 1.0 g cm3), and rs is the density of dry sediment
particles (assumed to be 2.5 g cm3). The cumulative
mass depth (M) is calculated as follows:
  X  Line 1 (triangles)
M mg cm2 ¼ 1  fi rs xdx ð6Þ
where fi is the porosity at depth “i” and dx is the thickness
of the layer (i.e., 1 cm). The dried sediment layers are pul-
verized using an agate mortar and pestle and prepared for
subsequent analyses.
Three examples of 210Pb profiles in diverse
depositional settings
Submarine basin – The San Pedro Basin is located in the
Southern California Bight, adjacent to Los Angeles, Cali-
fornia, USA. This marine basin acts as an efficient trap for
both natural and anthropogenic materials. Due to the
bathymetry of this deep basin, water circulation is
restricted and, as a consequence, water residence times
are long. Such a depositional setting is ideal for
excess-210Pb-derived geochronologies. Previous work on
sedimentation rates in the San Pedro Basin (Huh
et al., 1990; Alexander and Lee, 2009) corroborate the
excess 210Pb and 137Cs geochronology shown in Figure 3.
Both down-core profiles of 137Cs and excess 210Pb indi-
cate that the sediment has been accumulating at a rate
close to 0.09 cm day1; although finer scale sectioning
would likely show a decline in the sedimentation rates
around the turn of the twentieth century, coincident with
intensified sediment discharge controls on land.
Alpine lake – Copper Lake is located in the North Cas-
cades National Park, Washington State, less than 10 km
from the USA-Canada border (Sheibley et al., 2012).
The lake is at an elevation of about 1,600 m above mean
sea level and has a surface area of 5.2 ha. The maximum
depth of the lake is about 20 m. Figure 4 shows the
down-core profiles of excess 210Pb and 137Cs for Copper
Lake. The activities of both 137Cs and excess 210Pb are
high and indicative of large atmospheric fluxes being
delivered into a small basin.
Sinkhole lake – Lake Tulane (40 ha) is located in central
Florida at an elevation of 35 m above mean sea level and
contains one of the longest climate records for a lake in
the USA (Grimm et al., 1993). As a sinkhole lake, Lake
Tulane was formed when a limestone deposit catastrophi-
cally collapsed due to dissolution by groundwater. Today,
the water budget of the lake is balanced by surface and
groundwater contributions. Sustained groundwater
seepage can also transport radionuclides such as 226Ra.
Figure 5 shows the down-core profiles of excess 210Pb
and total (supported) 210Pb for Lake Tulane. There is a
pronounced mixed layer – ubiquitous in sediment that
contains an active biological community – which extends
down to ~10 cm. Excess 210Pb systematically declines
xs210Pb (dpm g–1)
0.1 1 10 100
0
surface mixed layer
10
b = –0.0.03248
20
Depth (cm)
Line 2 (squares)
30 b = –0.27013
Sedimentation rate:
40 Line 1
0.42 cm yr –1
Line 2
50 210
tot Pb (circles) 0.05 cm yr –1
Lake Tulane
210
Pb Dating, Figure 5 Excess 210Pb (xs210Pb) and total
(supported) 210Pb (tot210Pb, solid black circles) down-core
profiles and derived sedimentation rates in a sinkhole lake: Lake
Tulane, Florida. Open triangles (Line 1) and open squares (Line 2)
distinguish unique linear sediment rates. Summary parameters
of the linear regression are represented as the slope (b).
down core to yield two solid sedimentation rates, but at a
depth of ~40 cm, there is a notable inflection in the total
(supported) 210Pb activities. It is likely that this down-core
anomaly is produced from 226Ra-rich groundwater seep-
age (Brenner et al., 2004). Norton et al. (1985) concluded
that similar anomalies observed in several high-altitude
alpine lakes in Rocky Mountain National Park, Colorado,
were the result of groundwater movement that likely
impacted the distribution of 210Pb.
Summary
During the last decade, there has been an unprecedented
push to learn more about how the environment has
changed over time. This interest is grounded in trying to
better understand differences between natural- and
anthropogenic-driven changes and how our environment
may respond to such changes in the future. One of the
most fundamental steps in establishing a record of envi-
ronmental change is to first derive an accurate geochronol-
ogy. The geochronologies generated by 210Pb analysis can
resolve the very recent past (last 100–200 years), with
total (1s) uncertainties varying from just under 1 year to
around a decade (Baskaran, 2011). However, 210Pb
 210
Pb DATING
geochronologies are not without inherent problems and
uncertainties (Robbins and Edgington, 1975; von Gunten
and Moser, 1993). The development of a 210Pb geochro-
nology should never become a routine exercise (Appleby
and Oldfield, 1992; Appleby, 2008). Even some recent
studies that employ 210Pb as a geochronometer still fail
to meet the minimum requirements identified by Oldfield
and Appleby (1984). For example, some of these efforts
may fail to adequately evaluate inconsistencies in compet-
ing 210Pb models, or they may forego assessing 210Pb
model results against independent geochronologies (e.g.,
other radioisotope systems or varve counting). Lastly,
some 210Pb applications may ignore postdepositional
mixing (Benninger et al., 1979; Berner, 1980; Miguel
et al., 2003) that can yield a surprisingly plausible down-
core 210Pb profile but which produces a geochronological
model that is simply invalid (Kirchner, 2011). That said, a
carefully developed 210Pb geochronology, and one sub-
stantiated in the peer-reviewed literature per recommenda-
tions outlined in Smith (2001) and Hancock et al. (2002),
is invaluable in any recent environmental reconstruction.
Sediment dating using 210Pb methods will continue to be
refined with new improvements in analytical and model
capabilities.
Bibliography
Alexander, C. R., and Lee, H. J., 2009. Sediment accumulation on
the Southern California Bight continental margin during the
twentieth century. Geological Society of America, Special Paper,
54, 69–87.
Andrews, A. H., Stone, R. P., Lundstrom, C. C., and DeVogelaere,
A. P., 2009. Growth rate and age determination of bamboo corals
from the northeastern Pacific Ocean using refined 210Pb dating.
Marine Ecology Progress Series, 397, 173–185.
Appleby, P. G., 2008. Three decades of dating recent sediments by
fallout radionuclides: a review. The Holocene, 18, 83–93.
Appleby, P. G., and Oldfield, F., 1978. The calculation of 210Pb
dates assuming a constant rate of supply of unsupported 210Pb
to the sediment. Catena, 5, 1–8.
Appleby, P. G., and Oldfield, F., 1983. The assessment of 210Pb data
from sites with varying sediment accumulation rates.
Hydrobiologia, 103, 29–35.
Appleby, P. G., and Oldfield, F., 1992. Application of 210Pb to sed-
imentation studies. In Ivanovich, M., and Harmon, R. S. (eds.),
Uranium-series Disequilibrium: Application to Environmental
Problems. Oxford: Oxford Sciences Publications, pp. 731–778.
Appleby, P. G., Nolan, P. J., Gifford, D. W., Godfrey, M. J., Oldfield,
F., Anderson, N. J., and Battarbee, R. W., 1986. 210Pb dating by
low background gamma counting. Hydrobiologia, 141, 21–27.
Baskaran, M., 2011. Dating of Biogenic and Inorganic Carbonates
Using 210Pb-226Ra Disequilibrium Method: A Review. In
Baskaran, M. (ed.), Handbook of Environmental Isotope Geo-
chemistry. Berlin/Heidelberg: Springer, pp. 789–809.
Baskaran, M., and Swarzenski, P. W., 2007. Seasonal variations on
the residence times and partitioning of short-lived radionuclides
(234Th, 7Be and 210Pb) and depositional fluxes of 7Be and 210Pb
in Tampa Bay, Florida. Marine Chemistry, 104, 27–42.
Benninger, L. K., Aller, R. C., Cochran, J. K., and Turekian, K. K.,
1979. Effects of biological sediment mixing on the 210Pb chro-
nology and trace metal distribution in a Long Island Sound sed-
iment core. Earth Planetary Sciences Letter, 43, 241–259.
631
Benoit, G., and Rozan, T. F., 2001. 210Pb and 137Cs dating methods
in lakes: a retrospective study. Journal of Paleolimnology, 25,
455–465.
Berner, R. A., 1980. Early Diagenesis: A Theoretical Approach.
Princeton: Princeton University Press, p. 241.
Binford, M. W., 1990. Calculation and uncertainty analysis of 210Pb
dates for PIRLA project lake sediment cores. Journal of
Paleolimnology, 3, 253–267.
Binford, M. W., Kahl, J. S., and Norton, S. A., 1993. Interpretation
of 210Pb profiles and verification of the CRS dating model in
PIRLA project lake sediment cores. Journal of Paleolimnology,
9, 275–296.
Brenner, M., Schelske, C. L., and Kenney, W. F., 2004. Inputs of
dissolved and particulate 226Ra to lakes and implications for
210
Pb dating recent sediments. Journal of Paleolimnology, 32,
53–66.
Chanton, P. J., Martens, S. C., and Kipphut, W. G., 1983.
Lead-210 sediment geochronology in a changing coastal
environment. Geochimica et Cosmochimica Acta, 47,
1791–1804.
Crozaz, G., Picciotto, E., and De Breuck, W., 1964. Antarctic snow
chronology with 210Pb. Journal of Geophysical Research, 69,
2597–2604.
Crusius, J., and Anderson, R. F., 1995. Sediment focusing in six
small lakes inferred from radionuclide profiles. Journal of
Paleolimnology, 13, 143–155.
Gale, S. J., 2009. Dating the recent past. Quaternary Geochronol-
ogy, 4, 374–377.
Garcia-Orellana, J., Sanchez-Cabeza, J. A., Masque, P., Avila, A.,
Costa, E., Loye-Pilot, M. D., and Bruach-Menchen, J. M.,
2006. Atmospheric fluxes of 210Pb to the western Mediterranean
Sea and the Saharan dust influence. Journal of Geophysical
Research, 111, D15305.
Goldberg, E. D., 1963. Geochronology with 210Pb. In Radioactive
Dating. Proceedings of the Symposium on Radioactive Dating
Held by the International Atomic Energy Agency in
Co-operation with the Joint Commission on Applied Radioactiv-
ity, Athens, November 19–23, 1962, pp. 121–131.
Goldberg, E. D., and Bruland, K., 1974. Radioactive geochronol-
ogies. In Goldberg, E. D. (ed.), The Sea. New York, Wiley
Interscience, Vol. 5, pp. 451–489.
Grimm, E. C., Jacobson, G. L., Jr., Watts, W. A., Hansen, B. C., and
Maasch, K. A., 1993. A 50,000-year record of climate oscilla-
tions from Florida and its temporal correlation with the Heinrich
events. Science, 261, 198–200.
Hancock, G. Edgington, D. N., Robbins, J. A., Smith, J. N.,
Brunskill, G., and Pfitzer, J., 2002. Workshop on radiological
techniques in sedimentation studies: methods and applications.
In Fernandez, J. M., and Fichez, R. (eds.), Environmental
Changes and Radioactive Tracers. Proceedings of the
Sixth biennial Conference of the South Pacific Environmental
Radioactivity Association, Noumea, New Caledonia, June
19–23, 2000. Paris: IRD Editions, pp. 233–251. ISBN 2-7099-
1493-X.
Huh, C. A., Small, L. F., Niemnil, S., Finney, B. P., Hickey, B. M.,
Kachel, N. B., Gorsline, D. S., and Williams, P. M., 1990. Sedi-
mentation dynamics in the Santa Monica–San Pedro Basin off
Los Angeles: radiochemical, sediment trap and transmission
studies. Continental Shelf Research, 10, 137–164.
Kirchner, G., 2011. Pb-210 as a tool for establishing sediment chro-
nologies: examples of potentials and limitations of conventional
dating methods. Journal of Environmental Radioactivity, 102,
490–494.
Kirchner, G., and Ehlers, H., 1998. Sediment geochronology in
changing coastal environments: potentials and limitations of
the 137Cs and 210Pb methods. Journal of Coastal Research, 14,
483–492.
632 PEAT (14C)
Koide, M., Soutar, A., and Goldberg, E. D., 1972. Marine geochro-
nology with 210Pb. Earth and Planetary Science Letters, 14,
442–446.
Koide, M., Bruland, K. W., and Goldberg, E. D., 1973. 228Th/232Th
and 210Pb geochronologies in marine and lake sediments.
Geochimica et Cosmochimica Acta, 37, 1171–1187.
Krishnaswamy, S., Lal, D., Martin, J. M., and Meybeck, M., 1971.
Geochronology of lake sediments. Earth and Planetary Science
Letters, 11, 407–414.
Miguel, S., Bolívar, J. P., and Garcia-Tenorio, R., 2003. Mixing,
sediment accumulation and focusing using 210Pb and 137Cs.
Journal of Paleolimnology, 29, 1–11.
Nittrouer, C. A., DeMaster, D. J., McKee, B. A., Cutshall, N. H.,
and Larsen, I. L., 1983. The effect of sediment mixing on
210
Pb accumulation rates for the Washington continental shelf.
Marine Geology, 54, 201–221.
Norton, S. A., Hess, C. T., Blake, G. M., Morrison, M. L., and
Baron, J., 1985. Excess unsupported 210Pb in lake sediment from
Rocky Mountain lakes: a groundwater effect. Canadian Journal
of Fisheries and Aquatic Science, 42, 1249–1254.
Oldfield, F., and Appleby, P. G., 1984. Empirical test of 210Pb
models for lake sediments. In Haworth, Y., and Lund,
J. W. C. (eds.), Lake Sediments and Environmental History.
Leicester: Leicester University Press, pp. 93–124. ISBN
0-7185-1220-0.
Persson, B. R. R., and Helms, E., 2011. Polonium-210 and lead-210
in the terrestrial environment: a historical review. Journal of
Environmental Radioactivity, 102, 420–429.
Pourchet, M., Pinglot, J. P., and Melieres, M. A., 1989. Cesium-137
and lead-210 in alpine lake sediments: measurements and
modeling of mixing processes. Journal of Geophysical
Research, 94, 12,761–12,770.
Robbins, J. A., 1978. Geochemical and geophysical applications of
radioactive lead. In Nriagu, J. O. (ed.), Biogeochemistry of Lead
in the Environment. Amsterdam: Elsevier/North-Holland.
Topics in Environmental Health, Vol. 1A, pp. 285–393.
Robbins, J. A., and Edgington, D. N., 1975. Determination of recent
sedimentation rates in Lake Michigan using Pb-210 and Cs-137.
Geochimica et Cosmochimica Acta, 39, 285–304.
Sanchez-Cabeza, J. A., and Ruiz-Fernandez, A. C., 2012. 210Pb sed-
iment radiochronology: an integrated formulation and classifica-
tion of dating models. Geochimica et Cosmochimica Acta, 82,
183–200.
Sanchez-Cabeza, J. A., Ani-Ragolta, I., and Masqué, P., 2000. Some
considerations of the 210Pb constant rate of supply (CRS) dating
model. Limnology and Oceanography, 45, 990–995.
Sheibley, R. W., Foreman, J. R., Moran, P. W., and Swarzenski,
P. W., 2012. Atmospheric deposition, water-quality, and sedi-
ment data for selected lakes in Mount Rainier, North Cascades,
and Olympic National Parks, Washington, 2008–2010.
U.S. Geological Survey Data Series, 721, 34 p.
Shukla, B. S., and Joshi, S. R., 1989. An evaluation of the CIC
Model of 210Pb dating in sediments. Environmental Geology
and Water Sciences, 14, 73–76.
Smith, J. N., 2001. Why should we believe 210Pb sediment geochro-
nologies? Journal of Environmental Radioactivity, 55, 121–123.
Smith, J. N., and Walton, A., 1980. Sediment accumulation rates
and geochronologies measured in the Saguenay Fjord using the
210
Pb dating method. Geochimica et Cosmochimica Acta, 44,
225–240.
Swarzenski, P. W., Baskaran, M., Rosenbauer, R. J., and Orem,
W. H., 2006. Historical trace element distribution in sediments
from the Mississippi River delta. Estuaries and Coasts, 29,
1094–1107.
Turner, L. J., and Delorme, L. D., 1996. Assessment of 210Pb data
from Canadian Lakes using the CIC and CRS models. Environ-
mental Geology, 28, 78–87.
von Gunten, H. R., and Moser, R. N., 1993. How reliable is the
210
Pb dating method? Old and new results from Switzerland.
Journal of Paleolimnology, 9, 161–178.
Zaborska, A., Carroll, J., Papucci, C., and Pempkowiak, J., 2007.
Intercomparison of alpha and gamma spectrometry techniques
used in 210Pb geochronology. Journal of Environmental Radio-
activity, 93, 38–50.
Cross-references
Alpha Spectroscopy
Lacustrine Environments (14C)
Marine Varves
Sediment Mixing Rate, 210Pb and 234
Th
U-Series Dating
PEAT (14C)
Philippa Ascough
AMS Laboratory, Scottish Universities Environmental
Research Centre, East Kilbride, Scotland, UK
Definition
Peat. A highly organic soil that results when waterlogged
anaerobic conditions suppress the degradation of
vegetation.
Radiocarbon dating of peat. Measurement of the radiocar-
bon content in samples of either bulk peat or carbon-
bearing fractions isolated from bulk peat by physical
and/or chemical treatments. The objective is usually either
to provide an absolute chronology for environmental or
archaeological records contained within a peat deposit or
to understand the biogeochemical cycling of carbon
within the peat system.
Introduction
Peat is a highly organic deposit (e.g., 50–95 % organic
matter on a dry mass basis), comprising a thin, more rap-
idly cycling “acrotelm” layer, overlying a thick, saturated
“catotelm” layer, in which vegetation decay is very slow
(Ingram, 1978; Clymo, 1984). Peat develops as a result
of waterlogged anaerobic and acidic conditions that
inhibit vegetation decomposition (c.f. Yavitt and Lang,
1990; Johnson and Damman, 1991; Valentine
et al., 1994; Maltby and Proctor, 1996; Bridgham
et al., 1998). When the rate of net primary production at
the peat surface exceeds losses from decomposition, the
peat deposit grows in depth and/or extent. Peatlands
(also termed “bogs,” “mires,” “fens,” or “swamps”) are
hence commonly located on low-gradient slopes where
glacial deposits or bedrock impede drainage (Rabassa
et al., 1996; Coronato et al., 2006). Peat growth is an
important form of carbon storage; although peatlands
cover around 3 % of the Earth’s land area, they contain
20–30 % of the global soil carbon (C) pool, storing an esti-
mated 270–450 Pg (1 Pg ¼ 1015 g) of C, equivalent to
c. 50 % of C in atmospheric CO2 (Gorham, 1991; Zoltai
 PEAT (14C)
and Martikainen, 1996; Moore et al., 1998; Turunen
et al., 2002; Limpens et al., 2008). Carbon inputs in peat
include C fixed via photosynthesis from atmospheric CO2
or dissolved carbon and C delivered to the peat from atmo-
spheric deposition, surface runoff, or groundwater flow.
Peat is also a source of carbon, as anaerobic decomposi-
tion releases methane (CH4) to the atmosphere (Wang
et al., 2004), and water transport from peatlands carries
dissolved organic carbon to the oceans (Hope et al., 1994).
Bulk peat can be described as a “heterogeneous mix of
organic matter of different origins and different ages in
varying stages of biological degradation and humifica-
tion” (Brock et al., 2011, p. 550), meaning that radiocar-
bon dating of this material presents a range of practical
and theoretical problems and should be approached with
care in order to obtain accurate and precise results. Radio-
carbon dating of peat and its subfractions is desirable how-
ever, as this material forms a valuable long-term archive of
environmental and archaeological data extending over
timescales of up to 105 and even 106 years. Material held
within the peat deposit, as well as the peat itself, can serve
as a proxy for environmental conditions (e.g., tempera-
ture, precipitation, volcanic activity, fire histories) at the
time of deposition. Examples of proxies studied from peat
include macrofossils (e.g., Barber et al., 2003), geochem-
ical parameters (e.g., Shotyk, 1996), atmospheric cations/
particulates (e.g., Clymo et al., 1990), pollen (e.g.,
Edwards, 1979) and charcoal (e.g., Ohlson et al., 2006).
In addition, peat deposits frequently contain archaeologi-
cal material, including structural (e.g., Coles and Coles,
1986) and human remains (e.g., Stead et al., 1986). Ongo-
ing decomposition and compaction within peat deposits,
coupled with varying accumulation rates through time,
mean that depth is generally not linearly related to age
within a peat deposit (Clymo, 1984; Clymo et al., 1990)
and absolute dating methods are required to construct
a reliable timescale upon which to place proxy records
obtained from peat and compare between records (c.f.
Blaauw and Christen, 2005). The variations in net peat
accumulation rates through time that are identified
through radiocarbon dating also provide a useful climatic
proxy in themselves, as accumulation rates vary as
a response to past climate change (e.g., Mauquoy
et al., 2008).
In addition to being an environmental/archaeological
archive, peatlands form a key component of the global car-
bon cycle, and 14C is also used as a tracer to investigate the
biogeochemical cycling of carbon within the peat system.
Understanding peatland carbon dynamics is particularly
important in the light of potential future climatic change,
which is likely to be pronounced at higher latitudes, where
peat deposits are concentrated. This is achieved through
analysis of specific chemical fractions in peat to quantify
their turnover/transport rates and residence times within
peatlands. Examples include 14C measurement of gases
and dissolved organic carbon released from peatland sam-
ple sites (Billett et al., 2007; Tipping et al., 2010; Garnett
et al., 2011).
633
Sampling and pretreatment of peat for 14C dating
Prior to radiocarbon measurement, peat samples must
undergo laboratory pretreatment. The aim of the
pretreatment process is to isolate material of interest for
dating from contaminating carbon that has a different
14
C age. The first step in radiocarbon dating of peat
involves the selection of a suitable section of peat,
obtained from a deposit in the field. The aim of sampling
peat for 14C dating is to select the sample that is most rep-
resentative of the age of the peat layer itself. Larger sam-
ples may be subdivided in the laboratory, depending
upon the chronological resolution required. To date the
peat itself, initial processing (c.f. Harkness et al., 1989)
involves air-drying, followed by sieving (typically
<3 mm) to remove large root fragments (which may rep-
resent intrusive material of a different age to the peat
itself), and finally the sample is homogenized by grinding.
At this stage, it is possible to obtain a 14C measurement on
the sample. Several studies have, however, raised doubts
over the reliability of dates obtained on bulk peat (Kilian
et al., 1995; Shore et al., 1995; Nilsson et al., 2001), and
in practice it is more common to extract specific chemical
fractions from peat for dating. Radiocarbon dating labora-
tories consequently usually separate peat into three opera-
tionally defined fractions for dating on the basis of
pH-dependent solubility. These are humin (acid and alkali
insoluble), humic acids (alkali soluble, acid insoluble),
and fulvic acids (acid and alkali soluble) (Cook
et al., 1998). Humic and fulvic acids are a product of the
humification processes that occur as decomposing vegeta-
tion is transformed into peat, while humin is the fraction
most representative of the original plant material that
was converted to peat. Isolation of fulvic acids is
performed in order to remove this fraction from the humic
acids and humin and discard it. This is because the acid
solubility of fulvic acids renders them mobile in the peat
profile, meaning they are unreliable for dating (Shore
et al., 1995). Conversely, humic acids and humin are
thought to provide 14C ages that more accurately reflect
the time at which the peat sample formed. Pretreatment
to isolate humic acids involves an acid wash, followed
by an alkali wash, to extract both humic and fulvic acids.
The extract is filtered to isolate solids (i.e., the humin frac-
tion) and then acidified to precipitate the humic acids (e.g.,
Brock et al., 2010). Fulvic acids remain in solution and
can be discarded. Precise procedures used vary between
laboratories, with differences between factors such as
acid/alkali concentrations, extraction duration, and num-
ber of extractions (Cook et al., 1998). Along with peat
itself, intact organic remains contained within the peat
may be obtained for dating, most commonly plant macro-
fossils or charcoal fragments. The usual process for dating
such material is firstly isolation from the bulk peat matrix
by handpicking or wet-sieving. Isolated plant material or
charcoal is then subjected to an acid–base–acid
pretreatment (e.g., Hatté et al., 2001) to remove contami-
nating carbon, where removal of fulvic and humic acids
634 PEAT (14C)
by alkali extraction is followed by a final acid wash to
remove any contamination added through the absorption
of atmospheric CO2 during alkali treatment.
Considerations to maximize accuracy and
precision of 14C dates on peat
Peat presents difficulties when applying the 14C method,
particularly when the aim of the work is to date the time
since a specific layer in the peat was formed. For the date
to be accurate, the carbon that is dated must solely repre-
sent this formation event. Bulk peat is hence thought
unreliable for 14C dating because it contains a variable
mix of material of different ages (including aboveground
biomass, roots, fungi, insect remains, diatoms, spores, pol-
len, charred vegetation, etc.) (Blaauw et al., 2004a; Brock
et al., 2011), and any radiocarbon age will therefore sim-
ply be an integration of the ages of these substances,
weighted by mass in the sample, rather than reflecting an
actual date of deposition/final formation for a peat hori-
zon. However, multiple studies show that the humic acid
and humin fractions in peat can differ substantially in
age and radiocarbon content (Stout et al., 1981; Hammond
et al., 1991; Shore et al., 1995), raising questions over
which fraction is most suitable for dating. Whichever frac-
tion is selected, the accuracy of radiocarbon dates on peat
is improved by minimizing the possibility that the sample
contains contaminating carbon. Contamination can be
introduced either while the peat deposit is in situ or follow-
ing sampling. Contamination introduced in situ includes
modern roots and rhizomes penetrating into the peat from
above (Saarinen, 1996) and the movement of dissolved
organic carbon (including humic acids) through the peat
profile (Nilsson et al., 2001; Turetsky et al., 2004). Care
should be taken to identify the possibility that physical
movement through frost-heaving has not disturbed the
peat profile and buried surface peat with older deposits.
Another factor to consider is the possibility of whether
peat samples are affected by radiocarbon reservoir effects.
A 14C reservoir effect is a 14C age offset between
a particular carbon reservoir versus the atmosphere at
any particular point in time. This means that samples
grown or formed in the affected reservoir will have a 14C
age that is offset from their “true” atmospheric 14C age.
Radiocarbon reservoir effects typically result from the
introduction of “pre-aged” carbon into a reservoir, such
as carbon derived from dissolution of geological carbon-
ates, also known as the “hard-water” effect (Shotton,
1972). Aquatic mosses in peat have been found to have
a reservoir effect due to uptake of 14C-depleted carbon
during growth (MacDonald et al., 1987), meaning that
macrofossils or vegetation selected from peat for radiocar-
bon dating should be restricted to species that solely
obtain carbon from atmospheric CO2 (Turetsky
et al., 2004). Other potential sources of old carbon include
in-washing of ancient soil carbon to peat deposits via
groundwater or overland flow (Edwards and Rowntree,
1980) and organic matter released via thawing permafrost
(Damon et al., 1996). Following sampling, as with any
sedimentary or soil material, peat can be contaminated
by the growth of vegetation or bacterial action during stor-
age (e.g., Geyh et al., 1974). Appropriate preventative
measures include shortening storage time before analysis
and storing samples cold, frozen, or dried prior to analysis.
Constructing absolute peat chronologies:
calibration and age–depth modeling
considerations
The radiocarbon method requires that the 14C content of
the sample at the time of measurement (As) is compared
with the 14C content of the sample when it was living
(A0). A primary standard material (currently Oxalic Acid
II) is used to give a universal value for (A0). This, how-
ever, assumes that the 14C content of the atmosphere has
been constant through time. In fact, fluctuations in atmo-
spheric 14C content due to reasons including variations
in the Earth’s magnetic field, fluctuating intensity of the
solar wind, and human influence mean that radiocarbon
ages must be calibrated to convert to the calendar time-
scale. For samples formed in equilibrium with the 14C
content of atmospheric CO2, the current curve is IntCal09
(Reimer et al., 2013). Calibration of 14C ages in a peat
deposit can be performed in conjunction with age–depth
modeling to estimate the calendar age ranges of depths
within the peat profile that lie between 14C-dated layers
(e.g., Blaauw et al., 2007; Bronk Ramsey, 2008). Particu-
lar considerations are required when constructing
14
C-based chronologies in peat for periods after around
AD 1650, due to large fluctuations in atmospheric 14C con-
tent, due to both solar activity minima (the Maunder and
Dalton events) and anthropogenically induced changes.
The latter involve the reduction of atmospheric 14C content
due to release of 14C-dead CO2 from fossil fuel burning
(also known as the Suess effect) and an elevation in atmo-
spheric 14C content due to detonation of nuclear weapons
in the years up to AD 1963. With respect to the period after
c. AD 1950, the bomb 14C fluctuations can be used advan-
tageously to achieve highly precise dates of peat deposits
with uncertainties on the order of 1–2 years (e.g.,
Goodsite et al., 2001). For other periods where calibration
is problematic due to atmospheric 14C fluctuations, the
technique of “wiggle-matching” can be used to obtain
accurate chronologies (e.g., Blaauw et al., 2004b).
Wiggle-matching relies on the premise that a time-ordered
sequence of 14C dates from a peat deposit should match
the sequence represented in the 14C calibration curve and
can produce highly precise and accurate calibrated chro-
nologies, particularly when combined with Bayesian sta-
tistics (e.g., Blaauw et al., 2007). Radiocarbon dating
can also be usefully combined with other dating methods
in order to reduce uncertainties and improve the robust-
ness of age models. Such methods include 210Pb dating
(e.g., Clymo et al., 1990; Oldfield et al., 1997; Malmer
and Wallén, 1999; Loisel and Garneau, 2010; Le Roux
and Marshall, 2011), chrono-stratigraphic markers from
 PEAT (14C)
nuclear weapons testing, or the Chernobyl accident and
tephrochronology (e.g., Dugmore et al., 2000).
Summary and conclusions
Radiocarbon dating of peat presents particular problems,
not least due to the chemically and physically heteroge-
neous nature of this material, potentially comprising
a mix of carbon-containing compounds with ages span-
ning several thousand years. Despite this, 14C dates on
peat provide crucial chronological information for a key
source of climatic, archaeological, and paleoenvir-
onmental data. To obtain the best results, in which both
precision and accuracy of ages are maximized, careful
sample selection and chemical pretreatment are required,
which should be combined with consideration of potential
calibration issues. The application of statistical
approaches, such as wiggle-matching combined with
Bayesian analysis, is likely to improve chronological res-
olution in instances where sufficient 14C dates are
performed on a profile, as is the combination of 14C mea-
surements with other dating methods.
Bibliography
Barber, K. E., Chambers, F. M., and Maddy, D., 2003. Holocene
palaeoclimates from peat stratigraphy: macrofossil proxy cli-
mate records from three oceanic raised bogs in England and
Ireland. Quaternary Science Reviews, 22, 521–539.
Billett, M. H., Garnett, M. H., and Harvey, F., 2007. UK peatland
streams release old carbon to the atmosphere and young
dissolved organic carbon to rivers. Geophysical Research
Letters, 34(L23401), 2007, doi:10.1029/2007GL031797.
Blaauw, M., and Christen, J. A., 2005. Radiocarbon peat chronolo-
gies and environmental change. Applied Statistics, 54, 805–816.
Blaauw, M., van der Plicht, J., and van Geel, B., 2004a. Radiocar-
bon dating of bulk peat samples from raised bogs:
non-existence of a previously reported ‘reservoir effect’? Qua-
ternary Science Reviews, 23, 1537–1542.
Blaauw, M., van Geel, B., Mauquoy, D., and van der Plicht, J.,
2004b. Carbon–14 wiggle-match dating of peat deposits: advan-
tages and limitations. Journal of Quaternary Science, 19,
177–181.
Blaauw, M., Bakker, R., Christen, J. A., Hall, V. A., and van der
Plicht, J., 2007. A Bayesian framework for age modeling of
radiocarbon-dated peat deposits: case studies from the Nether-
lands. Radiocarbon, 49, 357–367.
Bridgham, S. D., Updegraff, K., and Pastor, J., 1998. Carbon, nitro-
gen, and phosphorus mineralization in northern wetlands. Ecol-
ogy, 79, 1545–1561.
Brock, F., Higham, T., Ditchfield, P., and Bronk Ramsey, C., 2010.
Current pretreatment methods for AMS radiocarbon dating at the
Oxford Radiocarbon Accelerator Unit (ORAU). Radiocarbon,
52, 103–112.
Brock, F., Lee, S., Housley, R. A., and Bronk Ramsey, C., 2011.
Variation in the radiocarbon age of different fractions of peat:
a case study from Ahrenshöft, northern Germany. Quaternary
Geochronology, 6, 550–555.
Bronk Ramsey, C., 2008. Deposition models for chronological
records. Quaternary Science Reviews, 27, 42–60.
Clymo, R. S., 1984. The limits to peat bog growth. Philosophical
Transactions of the Royal Society of London B, 303, 605–654.
Clymo, R. S., Oldfield, F., Appleby, P. G., Pearson, G. W., Ratnesar,
P., and Richardson, N., 1990. The record of atmospheric
635
deposition on a rainwater-dependent peatland. Philosophical
Transactions of the Royal Society of London B, 327, 331–338.
Coles, B., and Coles, J., 1986. Sweet Track to Glastonbury, the
Somerset Levels in Prehistory. London: Thames & Hudson.
Cook, G. T., Dugmore, A. J., and Shore, J. S., 1998. The influence
of pretreatment on humic acid yield and 14C age of Carex peat.
Radiocarbon, 40, 21–27.
Coronato, A., Roig, C., Collado, L., and Roig, F., 2006. Geomor-
phological emplacement and vegetational characteristics of
Fuegian peatlands, southernmost Argentina, South America. In
Martini, P., Martínez Cortízas, A., and Chesworth, W. (eds.),
Peatlands: Evolution and Records of Environmental and Cli-
mate Changes. Amsterdam: Elsevier. Developments in Earth
Surface Processes Series, pp. 111–129, Chapter 9.
Damon, P. E., Burr, G., Peristykh, A. N., Jacoby, G. C., and
D’Arrigo, R. D., 1996. Regional radiocarbon effect due to
thawing of frozen earth. Radiocarbon, 38, 597–602.
Dugmore, A. J., Newton, A. J., Larsen, G., and Cook, G. T., 2000.
Tephrochronology, environmental change and the Norse settle-
ment of Iceland. Environmental Archaeology, 5, 21–34.
Edwards, K. J., 1979. Palynological and temporal inference in the
context of prehistory, with special reference to the evidence from
lake and peat deposits. Journal of Archaeological Science, 6,
255–270.
Edwards, K. J., and Rowntree, K. M., 1980. Radiocarbon and
palaeoenvironmental evidence for changing rates of erosion at
a Flandrian stage site in Scotland. In Cullingford, R. A.,
Davidson, D. A., and Lewin, J. (eds.), Timescales in Geomor-
phology. Chichester: Wiley, pp. 207–223.
Garnett, M. H., Hardie, S. L., and Murray, C., 2011. Radiocarbon
and stable carbon analysis of dissolved methane and carbon
dioxide from the profile of a raised peat bog. Radiocarbon, 53,
71–80.
Geyh, M. A., Krumbein, W. E., and Kudrass, H. R., 1974.
Unreliable 14C dating of long-stored deep-sea sediments due to
bacterial activity. Marine Geology, 17, 45–50.
Goodsite, M. E., Rom, W., Heinemeier, J., Lange, T., Ooi, S.,
Appleby, P. G., Shotyk, W., van der Knaap, W. O., Lohse, C.,
and Hansen, T. S., 2001. High resolution AMS 14C dating of
post-bomb peat archives of atmospheric pollutants. Radiocar-
bon, 43, 495–515.
Gorham, E., 1991. Northern peatlands: role in the carbon cycle and
probable responses to global warming. Ecological Applications,
1, 182–195.
Hammond, A. P., Goh, K. M., Tonkin, P. J., and Manning, M. R.,
1991. Chemical pretreatments for improving the radiocarbon-
dates of peats and organic silts in a gley podzol environment –
Grahams Terrace, North Westland. New Zealand Journal of
Geology and Geophysics, 34, 191–194.
Harkness, G. T., Miller, B. F., Scott, E. M., and Baxter, M. S., 1989.
Design and preparation of sample for the International Collabo-
rative Study. Radiocarbon, 31, 407–413.
Hatté, C., Morvan, J., Noury, C., and Paterne, M., 2001. Is classical
acid–alkali–acid treatment responsible for contamination/an
alternative proposition. Radiocarbon, 43, 177–182.
Hope, D., Billett, M. F., and Cresser, M. S., 1994. A review of the
export of carbon in river water: fluxes and processes. Environ-
mental Pollution, 84, 301–324.
Ingram, H. A. P., 1978. Soil layers in mires: function and terminol-
ogy. Journal of Soil Science, 29, 224–227.
Johnson, L. C., and Damman, A. W. H., 1991. Species-controlled
Sphagnum decay on a South Swedish raised bog. Oikos, 61,
234–242.
Kilian, M. R., van der Plicht, J., and van Geel, B., 1995. Dating
raised bogs: new aspects of 14C AMS wiggle matching,
a reservoir effect and climatic change. Quaternary Science
Reviews, 14, 959–966.
636 PLANETARY SURFACES (CRATERING RATE)
Le Roux, G., and Marshall, W. A., 2011. Constructing recent peat
accumulation chronologies using atmospheric fall-out radionu-
clides. Mires and Peat, 7, 1–14.
Limpens, J., Berendse, F., Blodau, C., Canadell, J. G., Freeman, C.,
Holden, J., Roulet, N., Rydin, H., and Schaepman-Strub, G.,
2008. Peatlands and the carbon cycle: from local processes to
global implications – a synthesis. Biogeosciences, 5, 1475–1491.
Loisel, J., and Garneau, M., 2010. Late Holocene palaeoeco-
hydrology and carbon accumulation estimates from two boreal
peat bogs in eastern Canada: potential and limits of multi-proxy
archives. Palaeogeography, Palaeoclimatology, Palaeoecology,
291, 493–533.
MacDonald, G. M., Beukens, R. P., Kieser, W. E., and Vitt, D. H.,
1987. Comparative radiocarbon dating terrestrial plant macro-
fossils and aquatic moss from the “ice-free corridor” of western
Canada. Geology, 15, 837–840.
Malmer, N., and Wallén, B., 2004. Input rates, decay losses and
accumulation rates of carbon in bogs during the last millenium:
internal processes and environmental changes. The Holocene,
14, 111–117.
Maltby, E., and Proctor, M. C. F., 1996. Peatlands: their nature and
role in the biosphere. In Lappalainen, E. (ed.), Global Peat
Resources. Jyväskylä: Kukkalantie. International Peat Society,
pp. 11–19.
Mauquoy, D., Yeloff, D., Van Geel, B., Charman, D., and Blundell,
A., 2008. Two decadally resolved records from north-west Euro-
pean peat bogs show rapid climate changes associated with solar
variability during the mid-late Holocene. Journal of Quaternary
Science, 23, 745–763.
Moore, T. R., Roulet, N. T., and Waddington, J. M., 1998. Uncer-
tainty in predicting the effect of climatic change on the carbon
cycling of Canadian peatlands. Climatic Change, 40, 229–245.
Nilsson, M., Klarqvist, M., Bohlin, E., and Possnert, G., 2001. Var-
iation in 14C age of macrofossils and different fractions of minute
peat samples dated by AMS. The Holocene, 11, 579–586.
Ohlson, M., Korbøl, A., and Økland, R. H., 2006. The macroscopic
charcoal record in forested boreal peatlands in southeast Nor-
way. The Holocene, 16, 731–741.
Oldfield, F., Thompson, R., Crooks, P. R. J., Gedye, S. J., Hall,
Valerie A., Harkness, D. D., Housley, R. A., McCormac, F. G.,
Newton, A. J., Pilcher, J. R., Renberg, I., and Richardson, N.,
1997. Radiocarbon dating of a recent high-latitude peat profile:
Stor Åmyrân, northern Sweden. The Holocene, 7, 283-290.
Rabassa, J., Coronato, A., and Roig, C., 1996. The peat-bogs of
Tierra del Fuego, Argentina. In Lappalainen, E. (ed.), Global
Peat Resources. Jyskä: International Peat Society, pp. 261–266.
Reimer, P. J., Baillie, M. G. L., Bard, E., Bayliss, A., Beck, J. W.,
Bertrand, C. J. H., Blackwell, P. G., Buck, C. E., Burr, G. S.,
Cutler, K. B., Damon, P. E., Edwards, R. L., Fairbanks, R. G.,
Friedrich, M., Guilderson, T. P., Hogg, A. E., Hughen, K. A.,
Kromer, B., McCormac, G., Reimer, R. W., Manning, S., Bronk
Ramsey, C., Remmele, S., Southon, J. R., Stuiver, M., Talamo,
S., Taylor, F. W., van der Plicht, J., and Weyhenmeyer, C. E.,
2004. INTCAL04 terrestrial radiocarbon age calibration,
0–26 cal kyr BP. Radiocarbon, 46, 1029–1058.
Reimer, P. J., Bard, E., Bayliss, A., Beck, J. W., Blackwell, P. G.,
Bronk Ramsey, C., Buck, C. E., Cheng, H., Edwards, R. L., Frie-
drich, M., Grootes, P. M., Guilderson, T. P., Haflidason, H.,
Hajdas, I., Hatté, C., Heaton, T. J., Hoffmann, D. L., Hogg, A.
G., Hughen, K. A., Kaiser, K. F., Kromer, B., Manning, S. W.,
Niu, M., Reimer, R. W., Richards, D. A., Scott, E. M., Southon,
J. R., Staff, R. A., Turney, C. S. M., and van der Plicht, J., 2013.
IntCal13 and Marine13 radiocarbon age calibration curves
0–50,000 years cal BP. Radiocarbon, 55(4), 1869–1887.
Saarinen, T., 1996. Biomass and production of two vascular plants
in a boreal mesotrophic fen. Canadian Journal of Botany, 74,
934–938.
Shore, J. S., Bartley, D. D., and Harkness, D. D., 1995. Problems
encountered with the 14C dating of peat. Quaternary Science
Reviews, 14, 373–383.
Shotton, F. W., 1972. An example of hard-water error in radiocarbon
dating of vegetable matter. Nature, 240, 460–461.
Shotyk, W., 1996. Peat bog archives of atmospheric metal deposi-
tional: geochemical evaluation of peat profiles, natural variations
in metal concentrations, and metal enrichment factors. Environ-
mental Reviews, 4, 149–183.
Stead, I. M., Bourke, J., and Brothwell, D. R., 1986. Lindow Man.
The Body in the Bog. London: British Museum.
Stout, J. D., Goh, K. M., and Rafter, T. A., 1981. Chemistry and
turnover of naturally occurring resistant organic compounds in
soil. In Paul, E. A., and Ladd, J. N. (eds.), Soil Biochemistry.
New York: Marcel-Dekker, Vol. 5, pp. 1–73.
Tipping, E., et al., 2010. Sources and ages of dissolved organic mat-
ter in peatland streams: evidence from chemistry mixture model-
ling and radiocarbon data. Biogeochemistry, 100, 121–137.
Turetsky, M. R., Manning, S. W., and Wieder, R. K., 2004. Dating
recent peat deposits. Wetlands, 24, 324–356.
Turunen, J., Tomppo, E., Tolonen, K., and Reinikainen, A., 2002.
Estimating carbon accumulation rates of undrained mires in
Finland: application to boreal and subarctic regions. The
Holocene, 12, 69–80.
Valentine, D. W., Holland, E. A., and Schimel, D. S., 1994. Ecosys-
tem and physiological controls over methane production in north-
ern wetlands. Journal of Geophysical Research, 99, 1563–1571.
Wang, Z. P., Duan, Y., Yang, J. R., Li, L. H., and Han, X. G., 2004.
Plateau marsh methane oxidation as affected by inorganic
N. Pedosphere, 14, 195–204.
Yavitt, J. B., and Lang, G. E., 1990. Methane production in contrasting
wetland sites: response to organo-chemical components of peat
and to sulfate reduction. Geomicrobiology Journal, 8, 27–46.
Zoltai, S. C., and Martikainen, P. J., 1996. The role of forested
peatlands in the global carbon cycle. NATO ASI Series I, 40,
47–58.
Cross-references
14
C in Plant Macrofossils
210
Pb Dating
Accelerator Mass Spectrometry
Bomb Carbon
Groundwater Dating with Atmospheric Halogenated Compounds
Mass Spectrometry
Radiocarbon Dating
PLANETARY SURFACES (CRATERING RATE)
Stephanie C. Werner1 and Wolf Uwe Reimold2
1
Centre for Earth Evolution and Dynamics, University of
Oslo, Oslo, Norway
2
Museum für Naturkunde Berlin and Humboldt-
Universität zu Berlin, Berlin, Germany
Synonyms
Bombardment history; Cratering statistics; Projectile flux
Definitions
Planetary surface: Interface between the solid (or liquid)
material of a planet and either its atmosphere or the outer
 PLANETARY SURFACES (CRATERING RATE)
space. Actual planetary surfaces are found only on solid
planetary bodies, such as the terrestrial planets, moons,
or planetesimals.
Cratering rate: The rate at which craters of a given size
form per time interval and surface unit.
Introduction
Planetary surfaces carry the imprint of geological evolu-
tion of the specific planetary body and, when dated, allow
estimates of the timing and rates of geological processes.
Establishment of the geological evolutionary history
(chronostratigraphy) of solid-surface planetary bodies is
based on the most common surface-modifying geological
process, impact cratering. The crater density indicates
a relative sequence of events, and when calibrated by
radiometric dating of returned samples (e.g., Apollo sam-
ples from the Moon), crater density-based relative ages
can be converted to absolute temporal information. This
can allow dating of, for example, volcanic or structural
events and frame the thermal evolution of a planetary
body.
Cratering on planetary surfaces
Impact cratering occurs at all scales. Projectiles can range
from micrometer-sized dust particles to planetesimals of
10s or even 100s of kilometers in size. Planets without
atmosphere can record millimeter-sized craters to impact
basins of 2,500 km width. One of the largest impact events
during solar system evolution is probably the collision of
a large planetesimal with proto-Earth that resulted in the
Moon’s formation (e.g., Canup, 2004). Remnants of plan-
etesimals of the planet-forming period are found in the
asteroid main belt between Mars and Jupiter and beyond
the orbit of Neptune. From these belts, objects can be
Planetary Surfaces (Cratering Rate), Figure 1 Spatial impact crater distribution of all known craters on Earth (according to the Earth
Impact Database, June 2013, http://www.passc.net/EarthImpactDatabase/). The craters are plotted with symbols representing
gradually increasing crater sizes.
637
moved into planet-crossing orbits due to gravitational res-
onant interaction and collisions or due to solar radiation
pressure. These objects are the projectiles that have hit
the terrestrial planets and the Moon.
On Earth, craters are almost exclusively found on con-
tinents, with slightly denser crater populations on the old
cratonic units. Only few craters are detected on continental
shelf units. No crater is so far confirmed for oceanic units.
Craters are found on the seafloor with lower probability,
because of the relative young age of oceanic crust, which
is, on average, about 67 Ma old (calculated here according
to Müller et al., 1997), and because of the strong shielding
effect of the, on average, 5-km-deep water column.
Figure 1 shows the total known crater record on Earth with
crater diameters assigned by gradually changing symbol
sizes for visibility. A very different picture of cratering his-
tory is preserved on the Moon (Figure 2). The lack of
a significant atmosphere, very old surface ages, as well
as relatively limited endogenic geological activity suggest
that the lunar cratering record was accumulated during
nearly the entire evolution of the solar system.
The oldest solid material found on the Moon has been
dated at 4.47 Ga. Therefore, the lunar surface is commonly
considered a good basis for calibration of cratering rates
throughout solar system history. Additionally, the Moon
is the only other planetary body (besides Earth) for which
datable samples are available and that yielded ages deter-
mined in the laboratories on Earth using radiogenic
isotopes.
Dating of cratered planetary surfaces requires combin-
ing the information from geological mapping and crater
count statistics. Mapping delineates materials by geomor-
phological, multispectral, or reflectance features based on
natural breaks in the rock sequence, while crater count
638 PLANETARY SURFACES (CRATERING RATE)

Planetary Surfaces (Cratering Rate), Figure 2 Spatial impact crater distribution for the Moon (After Head et al., 2010). Craters larger
than 20 km (black) and lunar basins larger than about 250 km (gray) in diameter are plotted to scale.

statistics provide a quantitative tool for defining relative

ages and absolute model ages (calibrated by isotope ages).
The determination of absolute model ages on the Moon
(and other solid-surface objects) is based on the applica-
tion of a crater–production function (Figure 3) and
a cratering chronology model, which describes the inte-
grated cratering rate through time (Figure 4). The crater–
production function describes the expected crater size–
frequency distribution observed for a geological unit at
a specific time and is, when scaled, similar for all plane-
tary bodies (e.g., Ivanov et al., 2002).
For statistical evaluation and age determination, craters
are recorded by their size and plotted into histograms, such
as the crater size–frequency distribution curves. Graphs
used to display crater statistics plot cumulative crater fre-
quencies (number of craters greater than a given size per
square kilometer, Ncum) against the crater diameter D on
log–log plots. Distribution characteristics are defined by
polynomial functions or multiple-segment power laws,
where the cumulative frequency Ncum is proportional to
Db and b varies between 4 and 1.5 (e.g., Hartmann,
1999; Neukum et al., 2001, Figure 3). There are other rep-
resentations, such as incremental or relative (R) plots. The
latter describe the curve relative to a D2 distribution and
are used to compare distribution shape variations. The the-
oretical concept and mathematical background for dating
techniques based on crater counts were developed in the
1960s and 1970s, as summarized by the Crater Analysis
Techniques Working Group (Arvidson et al., 1979). Planetary Surfaces (Cratering Rate), Figure 3 Examples of four
The sketches in Figure 5 describe the crater size–fre- different but commonly used cumulative crater size–frequency
quency distribution modifications as a result of partial distributions in log–log space, determined for the Moon.
 PLANETARY SURFACES (CRATERING RATE)
Planetary Surfaces (Cratering Rate), Figure 4 Cratering chronology model according to Neukum et al. (1975). Calibration points
beyond 4.1 Ga (the currently best estimate for the age of the Nectaris Basin, Fernandes et al., 2013) are speculative, and,
therefore, earlier cratering rates are debated.
and complete resurfacing due to erosion or volcanic
emplacement. Both these geological processes can modify
the cratering record, and only thorough geological mapping
permits a meaningful interpretation of the cratering record.
The comparison of measured crater size–frequency distri-
butions and the crater–production function (see below, Fig-
ure 3) can aid the interpretation only if photogeological
mapping and crater statistics were performed well beyond
the image resolution limit. The relative differences of the
calculated areal crater densities are interpreted as relative
ages, and geological events at a global scale can be com-
pared even if no superposition relations are detected.
Cratering statistics and cratering rates
The approach of dating planetary surfaces by crater densi-
ties (i.e., the number of craters per square kilometer) was
first proposed by Öpik (1960), whereas linking crater den-
sities with isotopic ages was pioneered by Shoemaker
et al. (1962). The Apollo and Luna sampling missions pro-
vided lunar rock samples that were isotopically dated; these
isotopic ages were later correlated to crater densities on the
lunar surface (e.g., Hartmann, 1970). This permitted dating
of surface features so that the planetary evolution could be
studied through absolute geological time, and thus, the cra-
ter counting method is the only tool to date planetary sur-
face evolution when in situ samples are not available.
639
To determine absolute surface ages of planetary sur-
faces other than those for the Earth or Moon, some basic
assumptions are made: (1) the rate of impact crater forma-
tion is known, and the rate is size dependent as assigned by
the crater–production function and is stable with time
(Figure 3); (2) the cratering process is spatially and tempo-
rally random, i.e., no preferential bombardment seasons
exist; (3) target–property variations across the target body
are negligible; and (4) surface units are mapped according
to geological criteria. The conversion from crater frequen-
cies (relative ages) to absolute ages requires a cratering
chronology model. Such models link crater frequency
and absolute ages derived from isotope geochemistry
and were developed in the early 1970s (the most com-
monly adopted one is by Neukum et al., 1975, Figure 4).
The crater densities measured on different planetary sur-
faces that have been exposed for the same duration are
assumed to be similar for all planetary bodies.
For planetary bodies beyond the Earth and Moon, the
lunar chronology model is transferred to the impact condi-
tions on the other bodies. The scaling of crater sizes
between planetary bodies requires knowledge of (1) the
flux ratio of impactors at different sizes considering celes-
tial mechanical aspects and (2) how the conditions of cra-
ter formation differ from one planet to another. These
conditions are mainly defined by mass of the projectiles,
impact velocities, impact angles, gravity acceleration,
640 PLANETARY SURFACES (CRATERING RATE)

Planetary Surfaces (Cratering Rate), Figure 5 Cratering. A planetary clock: sequence A, B, and C and sequence A, F, and G depict the
effect of resurfacing activity on the crater distribution when (B) the entire unit is partially resurfaced, and only smaller craters were
erased, and (F) half of the unit is completely resurfaced. The resulting crater size–frequency distributions are shown in (E) in
comparison to artificial isochrons; event (B) as circles, event (F) as squares. Martian surface examples of resurfacing are shown in (D) as
caused by erosion and in (H) due to emplacement of volcanic flows.

and target properties of the impacted bodies, and are not
always fully understood (Werner and Ivanov, 2014).
Planetary system evolution and cratering
chronology models
The bombardment history of the Moon is closely related to
our understanding of planetary formation and evolution of
the solar system. The discovery of extrasolar planets has
revolutionized research on the dynamical evolution of
our own planetary system. The opportunity to observe
planetary systems in evolutionary stages, both younger
and older than ours, has inspired ideas of orbital migration
of the gaseous planets during their primordial accretion
phases. New dynamical orbital evolution models compris-
ing the orbital inward and outward migration of the gas-
eous giant planets have emerged, with important
implications for orbital changes of the surviving planetes-
imals (e.g., Tsiganis et al., 2005; Gomes et al., 2005).
These models suggest that the average impact velocities
have changed through time, making it difficult to relate
cratering statistics on old surfaces with crater counts on
young surfaces, as increased impact velocities would
result in larger crater sizes (Werner and Ivanov, 2014).
Traditional cratering statistics for planetary surface-age
determination assumes monotonic decay of the cratering
rate at constant average impact velocities.
Summary
Understanding the evolution of planets in the solar system
critically depends on accurate estimates of time and rates
at which processes, e.g., volcanism, occur. Absolute time
scales for planetary surface evolution in the inner solar
system (except the Earth and the Moon) can only be
derived by linking the lunar crater frequencies with isoto-
pically dated lunar samples. The crater counting method,
the only tool to date in absolute terms planetary surface
evolution when samples are not available, is challenged
by many observational uncertainties and the great diver-
sity of approaches. Diverse crater–production functions
are used to calibrate crater density measurements, which
has resulted in a variety of different lunar chronology
models. This diversity propagates to other planets such
 POLYMERASE CHAIN REACTION DNA AMPLIFICATION
as Mars and adds ambiguity to the related temporal inter-
pretations (e.g., Werner and Tanaka, 2011). Commonly
used cratering statistics for planetary surface-age determi-
nation assume monotonic cratering rate decay, but other
interpretations have been put forward due to the discovery
of extrasolar planetary systems. Cratering rates prior to
formation of the Nectaris Basin on the Moon are not
constrained and therefore debated in the light of solar sys-
tem evolution concepts (Werner and Ivanov, 2014). Nev-
ertheless, currently cratering statistics are the only
method to determine globally and across the solar system
relative and absolute ages and rates of planetary surface
evolution, until such time as datable samples will become
available from other planetary bodies.
Bibliography
Canup, R. M., 2004. Simulations of a late lunar-forming impact.
Icarus, 168(2), 433–456.
Crater Analysis Techniques Working Group, Arvidson, R.,
Boyce, J., Chapman, C., Cintala, M., Fulchignoni, M., Moore,
H., Neukum, G., Schultz, P., Soderblom, L., Strom, R.,
Woronow, A., and Young, R., 1979. Standard techniques for
presentation and analysis of crater size-frequency data. Icarus,
37, 467–474.
Fernandes, V. A., Fritz, J., Weiss, B. P., Garrick-Bethell, I., and
Shuster, D. L., 2013. The bombardment history of the Moon as
recorded by 40Ar-39Ar chronology. Meteoritics & Planetary
Science, 48(2), 241–269.
Gomes, R., Levison, H. F., Tsiganis, K., and Morbidelli, A., 2005.
Origin of the cataclysmic late heavy bombardment period of
the terrestrial planets. Science, 435, 466–469.
Hartmann, W. K., 1970. Note: lunar cratering chronology. Icarus,
13, 299.
Hartmann, W. K., 1999. Martian cratering VI. Crater
count isochrons and evidence for recent volcanism from Mars
Global Surveyor. Meteoritics and Planetary Science, 34,
167–177.
Head, J. W., III, Fassett, C. I., Kadish, S. J., Smith, D. E.,
Zuber, M. T., Neumann, G. A., and Mazarico, E., 2010.
Global distribution of large lunar craters: implications for
resurfacing and impactor populations. Science, 329(5998),
1504–1507.
Ivanov, B. A., Neukum, G., Bottke, W. F. Jr., Hartmann, W. K.,
2002. The comparison of size-frequency distributions of impact
craters and asteroids and the planetary cratering rate. In Bottke,
W. F. Jr., Cellino, A., Paolicchi, P., and Binzel, R. P. (eds.) Aster-
oids III. The University of Arizona Press: Tucson, Arizona,
USA, 89–101.
Müller, R. D., Roest, W. R., Royer, J.-Y., Gahagan, L. M., and
Sclater, J. G., 1997. Digital isochrons of the world’s ocean floor.
Journal of Geophysical Research, 102, 3211–3214.
Neukum, G., Koenig, B., and Arkani-Hamed, J., 1975. A study of
lunar impact crater size-distributions. Moon, 12, 201–229.
Neukum, G., Hartmann, W. K., and Ivanov, B. A., 2001. Cratering
records in the inner solar system in relation to the Lunar refer-
ence system. In Kallenbach, R., Geiss, J., and Hartmann, W. K.
(eds.), Chronology and Evolution of Mars. Dordrecht: Kluwer,
pp. 55–86.
641
Öpik, E. J., 1960. Interplanetary Encounters: Close-Range Gravi-
tational Interactions. Amsterdam: Elsevier, p. 155.
Shoemaker, E. M., Hackman, R. J., and Eggleton, R. E., 1962.
Interplanetary correlation of geologic time. Advances in the
Astronautical Sciences, 8, 70–89.
Tsiganis, K., Gomes, R., Morbidelli, A., and Levison, H. F., 2005.
Origin of the orbital architecture of the giant planets of the solar
system. Science, 435, 459–461.
Werner, S. C., and Tanaka, K. L., 2011. Redefinition of the crater-
density and absolute-age boundaries for the chronostratigraphic
system of Mars. Icarus, 215(2), 603–607.
Werner, S. C., and Ivanov, B. A., 2014. Exogenic
dynamics, cratering and surface ages. In Spohn, T., (ed.),
Treaties on Geophysics. Elsevier: Amsterdam, The Netherlands
(in press).
POLYMERASE CHAIN REACTION DNA
AMPLIFICATION
Martyna Molak
School of Biological Sciences, University of Sydney,
Sydney, Australia
Definition
Polymerase chain reaction (PCR) is a method of amplify-
ing DNA from biological samples. This makes it possible
to determine the DNA sequence, using techniques such as
Sanger sequencing.
Most genetic analyses involve the study of DNA
sequences. To obtain these data, the target DNA needs to
be amplified so that it can be distinguished from the rest
of the DNA present in a biological sample. PCR is
a technique that is routinely used to amplify DNA (Mullis
et al., 1986; Mullis and Faloona, 1987).
PCR requires the design of short DNA fragments
(primers) with sequences that match the regions flanking
the genetic marker being targeted (Figure 1). These are
added to a mixture of DNA (extracted from the bio-
logical sample), a DNA-replicating enzyme (polymer-
ase), basic units for building new DNA molecules
(dNTPs), and several other reagents. This mixture is
subjected to a cycle of temperature changes. First,
high temperature causes DNA to unwind and its
two strands to separate (denaturation). Then hydro-
gen bonding causes the primers to attach specifically
to matching sections of DNA (annealing). In the
final part of the cycle, polymerase uses the original
DNA strand as a template for copying the nucleotide
sequence and attaches dNTPs to the new strand that
has been started by the primers (elongation). This
process is repeated a large number of times, with
642 POTASSIUM–ARGON (ARGON–ARGON), STRUCTURAL FABRICS

Polymerase Chain Reaction DNA Amplification, Figure 1 Schematic presentation of the polymerase chain reaction. Double-
stranded DNA is denatured with high temperature. Primers bind to matching sections of the DNA. Polymerase assembles
a complementary DNA strand using dNTP building blocks. In each cycle the number of target DNA molecules is doubled.

each cycle causing the number of target DNA mole-

cules to double. Using PCR, the amount of target POTASSIUM–ARGON (ARGON–ARGON),
DNA can be massively increased, allowing the STRUCTURAL FABRICS
DNA sequence to be determined using other
techniques. Michael A. Cosca
United States Geological Survey (USGS), Denver Federal
Center, Denver, CO, USA
Bibliography
Mullis, K. B., and Faloona, F. A., 1987. Specific synthesis of DNA Definition
in vitro via a polymerase-catalyzed chain reaction. Methods in 40
Enzymology, 155, 335–350. Ar/39Ar geochronology of structural fabrics: The appli-
Mullis, K. B., Faloona, F. A., Scharf, S. J., Saiki, R. K., cation of 40Ar/39Ar methods to date development of struc-
Horn, G. T., and Erlich, H., 1986. Specific enzymatic tural fabrics in geologic samples.
amplification of DNA in vitro: the polymerase chain reaction.
Cold Spring Harbor Symposia on Quantitative Biology, 51,
263–273. Introduction
Structural fabrics develop during rock deformation at var-
iable pressures (P), temperatures (T), fluid compositions
Cross-references (X), and time (t). Structural fabrics are represented in
Gene Sequencing rocks by features such as foliations and shear zones
 POTASSIUM–ARGON (ARGON–ARGON), STRUCTURAL FABRICS 643

developed at the mm to km scale. In ideal cases, the P-T-X
history of a given structural fabric can be constrained
using stable isotope, cation exchange, and/or mineral
equilibria thermobarometry (Essene, 1989). The timing
of structural fabric development can be assessed qualita-
tively using geologic field observations or quantitatively
using isotope-based geochronology. High-precision geo-
chronology of the thermal and fluid flow histories associ-
ated with structural fabric development can answer
fundamental geologic questions including (1) when
hydrothermal fluids transported and deposited ore min-
erals, (2) fault activity and displacements, (3) the timing
and duration of orogenic stages, and (4) the rates and fre-
quency of these and other geologic processes.
Radioisotope systems (e.g., U–Pb, Rb–Sr, Sm–Nd)
that quantitatively retain radiogenic isotopes in minerals
that formed at relatively high temperatures (>500  C)
are necessary for dating of structural fabrics that devel-
oped deep within the crust. For example, isotope geochro-
nology of large-scale shear zones that define major crustal
boundaries in the Grenville Province of Ontario, Canada,
indicates that these well-developed structural fabrics
define narrow time intervals of Neo- and Mesoproterozoic
orogenic compression and extension (e.g., van Breemen
and Hanmer, 1986; van Breemen et al., 1986; Cosca
et al., 1991; Mezger et al., 1991; Cosca et al., 1992; van
der Pluijm et al., 1994).
Isotopic dating of structural fabrics developed at
shallower crustal levels requires isotope geochronometers
that are sensitive to rock deformation at lower tempera-
tures. The K–Ar geochronometer, including the 40Ar/39Ar
method (McDougall and Harrison, 1999), is well suited
for isotopic dating of structural fabrics developed close
to the brittle-ductile transition zone (~250– ~ 400  C)
because such rocks often contain potassium-bearing min-
erals including mixed-layer illite/smectite, biotite, and
muscovite that quantitatively retain radiogenic argon
(40Ar*) at or below ~400  C.
40
Ar/39Ar geochronology of structural
(deformation) fabrics
40
Ar/39Ar methods
Dating structural fabric using 40Ar/39Ar methods can be
approached either by incrementally heating separated
mineral grains with a furnace or infrared (IR) laser or by
in situ laser ablation within small polished rock samples
(thereby preserving textural context), with an IR or ultra-
violet (UV) laser. The U.S. Geological Survey (USGS)
system in Denver (Figure 1) consists of both an IR laser
(CO2, wavelength ¼ 10.6 mm) and an UV laser (ArF
exciplex, wavelength ¼ 193 nm) attached to an ultrahigh
vacuum (UHV) gas extraction system and mass spectrom-
eter equipped with five Faraday detectors and a single
compact discrete dynode ion multiplier. Samples for in
situ analysis are prepared from polished rock chips, sev-
eral mm on a side. One or more chips are irradiated by fast
neutrons in a nuclear reactor and are placed in a laser sam-
ple chamber (Figure 2) with a window that is transparent
to 193 nm UV light. For in situ analysis, UV lasers are pre-
ferred over IR lasers because the shorter wavelengths
results in minimal collateral heating of adjacent grains
within the sample. Because small gas concentrations are

Potassium–Argon (Argon–Argon), Structural Fabrics, Figure 1 Photograph of the USGS 40Ar/39Ar laboratory in Denver, CO. The
40
Ar/39Ar system shown includes two lasers (IR and UV) for liberating Ar from rock and mineral samples. The UV laser has been
removed for clarity (the position of the UV laser chamber (UVLC) is indicated). Setup and control of the 40Ar/39Ar experiments,
including automation of the mass spectrometer, lasers, and UHV gas extraction line, is done using computers.
644 POTASSIUM–ARGON (ARGON–ARGON), STRUCTURAL FABRICS

Potassium–Argon (Argon–Argon), Structural Fabrics, Figure 2 Photograph of (a) UV laser attached to the extraction line and (b)
expanded view from the top of the UVLC containing samples for analysis.

liberated during UV laser ablation (~1016 to ~1020
mol), isotopic measurements are performed by static mass
spectrometry using a single ion multiplier.
Multigrain 40Ar/39Ar geochronology
Determining when foliations and shear bands develop is
typically done by separating and analyzing pure,
multigrain mica samples using either K–Ar or 40Ar/39Ar
methods (e.g., Itaya and Takasugi, 1988; Kelley, 1988;
Hames and Cheney, 1997; Jaboyedoff and Cosca, 1999;
Mulch et al., 2004). Several multigrain 40Ar/39Ar studies
of naturally deformed rocks have demonstrated that
smaller grains yield younger ages (e.g., Kligfield et al.,
1986; Kelley, 1988; Goodwin and Renne, 1991; Hess
et al., 1993; Kirschner et al., 1996; Mulch et al., 2002;
Markley et al., 2002). These multigrain studies show that
the mean radius of physically separated grains reasonably
approximates the effective length scale of 40Ar* diffusion,
such that different-sized grains from the same sample
reflect progressive cooling of the rock and closure to
40
Ar* diffusion. Submicroscopic defects form in
minerals during rock deformation (Gapais and White,
1982; Passchier and Trouw, 2005) and exert a nontrivial
effect on intragrain 40Ar* retention and 40Ar/39Ar ages
(Lee, 1995; Kramar et al., 2003). In situ UV laser
ablation 40Ar/39Ar geochronology of individually
 POTASSIUM–ARGON (ARGON–ARGON), STRUCTURAL FABRICS 645

deformed mineral grains offers a direct means of

assessing the timing of deformation, the effective length
scales for 40Ar* diffusion, and thermal histories (Cosca
et al., 2011).

In situ 40Ar/39Ar geochronology

In situ 40Ar/39Ar geochronology has been applied to
investigate structural fabrics such as pseudotachylytes that
formed rapidly under brittle conditions (e.g., Sherlock and
Hetzel, 2001) as well as ductilely deformed, slowly cooled
rock (e.g., Hodges et al., 1994). The following examples,
one with mineral overgrowths and two involving internal
mineral deformation, illustrate how intragrain 40Ar/39Ar
age variations from different structural fabrics can be used
to interpret geologic histories.

Overgrowths and fluid flow

Isotopic dating of texturally and compositionally distinct
mineral overgrowths or strain fringes provides a direct
measure of rock deformation, fluid flow, and mineral
growth (e.g., Muller et al., 2000; Sherlock et al., 2005;
Wells et al., 2008). For example, the Proterozoic Telemark
detachment in southern Norway contains white mica
porphyroclasts (mica fish) that formed during an initial
period of upper amphibolite-grade deformation, and dur-
ing subsequent greenschist facies deformation these
grains developed compositionally distinct overgrowths
(Mulch et al., 2005). In situ UV laser ablation 40Ar/39Ar
geochronology of the white mica overgrowths revealed
that they developed over a period of approximately 10
my, and that occurred roughly 10 my after closure to argon
diffusion in the mineral cores. Together with oxygen iso-
tope values in deformed quartz, these observations indi-
cate that a compositionally distinct fluid was present
during deformation that facilitated growth or recrystalliza-
tion of the mica overgrowths.

White mica porphyroclasts in shear bands

In situ UV laser ablation 40Ar/39Ar geochronology of mica
porphyroclasts in shear bands provides a means of dating
past movement (e.g., Mulch and Cosca, 2004) and poten-
tially provides a means of reconstructing paleoelevations
(Mulch et al., 2004). Greenschist facies micaceous quartz
mylonites often contain large (mm-sized) porphyroclasts
of variably deformed white mica (Figure 3a, b). These
rocks often develop C’type shear bands (Berthé et al.,
1979) that cut across earlier formed foliations and surviv- Potassium–Argon (Argon–Argon), Structural Fabrics,
ing porphyroclasts. In some shear bands, the so-called Figure 3 Photomicrographs of white mica contained in typical
structural fabrics. (a) Large, white mica porphyroclast that
mica fish porphyroclasts develop and may undergo pro- rotated into a planar shear zone defined by the smaller micas.
gressive mechanical grain size reduction during shearing, (b) Strongly deformed white mica porphyroclasts with
resulting in smaller segments of mica (Figure 3c). This nonuniform optical properties suggest the presence of
process of grain size reduction has been dated in the submicroscopic defects. (c) White mica porphyroclasts (fish)
Pogallo shear zone of northern Italy (Mulch et al., 2002) displaying evidence of internal size reduction as they were
in which mica porphyroclasts yielded in situ UV laser rotated into a shear zone fabric. ms muscovite, qz quartz. All
scale bars are 1 mm.
ablation 40Ar/39Ar dates that became younger as both the
646 POTASSIUM–ARGON (ARGON–ARGON), STRUCTURAL FABRICS
physical grain size and the length scale for 40Ar* diffusion
decreased.
The extent of structural fabric development and defects
on 40Ar* retention in micas within shear bands at the sub-
grain scale can be assessed by in situ UV laser ablation of
individual mica porphyroclasts. For example, over 150
individual UV laser ablation 40Ar/39Ar analyses within
one large mica porphyroclast from a Carboniferous
quartz-mica pegmatite layer reveal 40Ar/39Ar dates that
correlate, and become younger, with position relative to
incipient shear bands within the mica fish (Kramar et al.,
2001). When examined by transmission electron micros-
copy, this mica fish contains abundant stacking faults or
microstructural distortions caused by partial dislocations
and localized charge imbalances (Kramar et al., 2003).
The interconnectivity of these line defects decreases reten-
tivity of 40Ar* by several orders of magnitude and can
explain the range of intragrain 40Ar/39Ar dates intermedi-
ate between known periods of Carboniferous and Oligo-
cene-Miocene deformation (Kramar et al., 2001, 2003).
Summary
40
Ar/39Ar studies of structural fabrics have shown that
retention of 40Ar* in naturally deformed rock is sensitive
to the intensity, duration, temperature, and presence of
fluids during rock deformation. Understanding the link
between 40Ar/39Ar ages and microstructures in deformed
rock remains a fertile ground for continued research.
Bibliography
Berthé, D., Choukroune, P., and Gapais, D., 1979. Orientations
préférentielles du quartz et orthogneissification progressive in
régime cisaillant: l’exemple du cisaillement sud-armoricain.
Bulletin du Minéralogie, 102, 265–272.
Cosca, M. A., Sutter, J. F., and Essene, E. J., 1991. Cooling and
inferred uplift/erosion history of the Grenville Orogen, Ontario:
constraints from 40Ar/39Ar thermochronology. Tectonics, 10,
959–977.
Cosca, M. A., Essene, E. J., Kunk, M. J., and Sutter, J. F., 1992. Dif-
ferential unroofing within the Central Metasedimentary Belt of
the Grenville Orogen: constraints from 40Ar/39Ar
thermochronology. Contributions to Mineralogy and Petrology,
110, 211–225.
Cosca, M. A., Stunitz, H., and Bourgeix, A.-L., 2011. 40Ar* loss in
experimentally deformed muscovite and biotite with implica-
tions for 40Ar/39Ar geochronology of naturally deformed rocks.
Geochimica et Cosmochimica Acta, 75, 7759–7778.
Essene, E. J., 1989. The current status of thermobarometry in meta-
morphic rocks. In Daly, J.S., Cliff, R.A., and Yardley, B.W.D.
(eds), Evolution of Metamorphic Belts, Geological Society,
London, Special Publications, 43, pp. 1–44.
Gapais, D., and White, S. H., 1982. Ductile shear bands in
a naturally deformed quartzite. Textures and Microstructures,
5, 1–17.
Goodwin, L. B., and Renne, P. R., 1991. Effects of progressive
mylonitization on Ar retention in biotites from the Santa Rosa
mylonite zone, California, and thermochronologic implications.
Contributions to Mineralogy and Petrology, 108, 283–297.
Hames, W. E., and Cheney, J. T., 1997. On the loss of 40Ar from
muscovite during polymetamorphism. Geochimica et
Cosmochimica Acta, 61, 3863–3872.
Hess, J. C., Lippolt, H. J., Gurbanov, A. G., and Michalski, I., 1993.
The cooling history of the late Pliocene Eldzhurtinskiy granite
(Caucasus, Russia) and the thermochronological potential of
grain-size/age relationships. Earth and Planetary Science
Letters, 117, 393–406.
Hodges, K. V., Hames, W. E., and Bowring, S. A., 1994. 40Ar/39Ar
age gradients in micas from a high-temperature-low-pressure ter-
rain: evidence for very slow cooling and implications for the
interpretation of age spectra. Geology, 22, 55–58.
Itaya, T., and Takasugi, H., 1988. Muscovite K-Ar ages of the
Sanbagawa schists, Japan and argon depletion during cooling
and deformation. Contributions to Mineralogy and Petrology,
100, 281–290.
Jaboyedoff, M., and Cosca, M. A., 1999. Dating incipient metamor-
phism using 40Ar/39Ar geochronology and XRD modeling:
a case study from the Swiss Alps. Contributions to Mineralogy
and Petrology, 135, 93–113.
Kelley, S., 1988. The relationship between K-Ar mineral ages, mica
grain sizes and movement on the Moine thrust zone, NW High-
lands, Scotland. Journal of the Geological Society of London,
145, 1–10.
Kirschner, D. L., Cosca, M. A., Hunziker, J. C., and Masson, H.,
1996. Staircase 40Ar/39Ar spectra of fine-grained white mica:
timing and duration of deformation and empirical constraints
on argon diffusion. Geology, 24, 747–750.
Kligfield, R., Hunziker, J. C., Dallmeyer, R. D., and Schamel, S.,
1986. Dating of deformation phases using K/Ar and 40Ar/39Ar
techniques: results from the northern Apennines. Journal of
Structural Geology, 8, 781–798.
Kramar, N., Cosca, M. A., and Hunziker, J. C., 2001. Heterogenous
40
Ar* distributions in naturally deformed muscovite: in situ UV-
laser ablation evidence for microstructurally controlled
intragrain diffusion. Earth and Planetary Science Letters, 192,
377–388.
Kramar, N., Cosca, M. A., Buffat, P. A., and Baumgartner, L. P.,
2003. Stacking fault-enhanced argon diffusion in naturally
deformed muscovite. In Vance, D., Muller, W., and Villa, I.
(eds.), Geochronology: Linking the Isotopic Record with
Petrology and Textures. London: Geological Society.
Geological Society, London, Special Publication, Vol. 220,
pp. 249–260.
Lee, J. K. W., 1995. Multipath diffusion in geochronology. Contri-
butions to Mineralogy and Petrology, 120, 60–82.
Markley, M., Teyssier, C., and Cosca, M. A., 2002. The relation
between grain size and 40Ar/39Ar date for Alpine white mica
from the Siviez-Mischabel Nappe, Switzerland. Journal of
Structural Geology, 24, 1937–1955.
McDougall, I., and Harrison, T. M., 1999. Geochronology and
Thermochronology by the40Ar/39Ar Method, 2nd edn. Oxford:
Oxford University Press.
Mezger, K., van der Pluijm, B. A., Essene, E. J., and Halliday, A. N.,
1991. Synorogenic collapse: a perspective from the middle crust,
the Proterozoic Grenville Orogen. Science, 254, 695–698.
Mulch, A., and Cosca, M. A., 2004. Crystallization or cooling ages?
In situ UV-laser 40Ar/39Ar geochronology of muscovite in
mylonitic rocks. Journal of the Geological Society of London,
161, 573–582.
Mulch, A., Cosca, M. A., and Handy, M. R., 2002. In situ UV-laser
40
Ar/39Ar geochronology of a micaceous mylonite: an example
of defect-enhanced argon loss. Contributions to Mineralogy
and Petrology, 142, 738–752.
Mulch, A., Teyssier, C., Cosca, M. A., Vanderhaeghe, O., and
Vennemann, T. W., 2004. Reconstructing paleoelevation in
eroded orogens. Geology, 32, 525–528.
Mulch, A., Cosca, M. A., Fiebig, J., and Andresen, A., 2005. Time
scales of deformation and exhumation in extensional detachment
systems determined by high-spatial resolution in situ UV-laser
 POTASSIUM–ARGON (ARGON–ARGON), STRUCTURAL FABRICS
40
Ar/39Ar dating. Earth and Planetary Science Letters, 233,
375–390.
Müller, W., Aerden, D., and Halliday, A. N., 2000. Isotopic dating
of strain fringe increments: duration and rates of deformation
in shear zones. Science, 288, 2195–2198.
Passchier, C. W., and Trouw, R. A. J., 2005. Micro-tectonics,
2nd edn. Berlin: Springer.
Sherlock, S. C., and Hetzel, R., 2001. A laser probe 40Ar/39Ar study
of pseudotachylyte from the Tambach fault zone, Kenya: direct
isotopic dating of brittle faults. Journal of Structural Geology,
23, 33–44.
Sherlock, S. C., Lucks, T., Kelley, S. P., and Barnicoat, A., 2005.
High spatial resolution 40Ar/39Ar laserprobe analysis of complex
polygenetic K-feldspar overgrowths: a record of sandstone
burial and tectonically-driven fluid migration. Earth and Plane-
tary Science Letters, 238, 329–341.
van Breemen, O., and Hanmer, S. K., 1986. Zircon morphology and
U/Pb geochronology in active shear zones: studies on
syntectonic intrusions along the northwest boundary of the Cen-
tral Metasedimentary Belt, Grenville Province, Ontario. Geolog-
ical Survey of Canada Paper, 86-1B, pp. 775–784.
647
van Breemen, O., Davidson, A., Loveridge, W. D., and Sullivan,
R. W., 1986. U-Pb zircon geochronology of Grenville tectonites,
granulites, and igneous precursors, Parry Sound, Ontario. Geo-
logical Association of Canada Special Paper, 31, pp. 191–207.
van der Pluijm, B. A., Mezger, K., Cosca, M. A., and Essene, E. J.,
1994. Determining the significance of high-grade shear zones by
using temperature-time paths, with examples from the Grenville
Orogen. Geology, 22, 743–746.
Wells, M. L., Spell, T. L., Hoisch, T. D., Arriola, T., and Zanetti,
K. A., 2008. Laser-probe 40Ar/39Ar dating of strain fringes:
mid-cretaceous synconvergent orogen-parallel extension in the
interior of the Sevier orogen. Tectonics, 27(TC3012), 1–20.
Cross-references
Alpine Terranes (K–Ar/Ar–Ar)
Ar–Ar and K–Ar Dating
Clays and Glauconites (K–Ar/Ar–Ar)
Metamorphic Terranes (K–Ar/40Ar/39Ar)
Minerals (40Ar–39Ar)
 Q
QUARTZ
Peter J. Heaney
Department of Geosciences, Pennsylvania State
University, University Park, PA, USA
Definition
On the Earth’s surface, quartz (or, more properly, alpha-
quartz) is by far the most common of the crystalline forms
of silicon dioxide (SiO2), constituting 12 % of the crust by
volume. Among the igneous rocks, quartz is abundant
within granites, granodiorites, pegmatites, and rhyolites.
The weathering and erosion of these rocks tend to concen-
trate quartz as sedimentary particles because of its low sol-
ubility (~10 ppm) and its mechanical durability (with
a Mohs hardness of 7 out of 10). As a result, quartz is
the major constituent of highly mature arenitic sandstones.
Quartz also can form in sedimentary environments
through the chemical precipitation of dissolved silica.
The dissolution of volcanic clays or of organosiliceous
tests (as are found in diatoms, radiolaria, and sponges)
produces a hydrous silica gel that will transform to quartz
over thousands of years through diagenetic dehydration at
temperatures up to 300  C and pressures up to 2 kb. Meta-
morphism of siliceous sedimentary and igneous rocks will
also induce quartz crystallization, and quartz is especially
abundant within quartzites, schists, and gneisses. Addi-
tionally, hydrothermal alteration and fracturing of rocks
may yield massive veins of quartz.
Quartz is differentiated from other silica minerals by its
crystal structure and its field of thermodynamic stability.
The low-temperature variety of quartz, called alpha-quartz
(or low quartz), is stable when heated up to 573  C and
when stressed to pressures of 20–30 kb. When heated at
atmospheric pressure above 573  C, quartz transforms
J.W. Rink, J.W. Thompson (eds.), Encyclopedia of Scientific Dating Methods, DOI 10.1007/978-94-007-6304-3,
© Springer Science+Business Media Dordrecht 2015
first to a structural variant called beta-quartz (or high
quartz). On further heating, beta-quartz transitions to
tridymite and thence to cristobalite before melting to silica
liquid at 1,727  C. When quartz at room temperature is
pressurized to ~20 kb, it transforms to coesite; above
75 kb stishovite is the stable silica phase.
Euhedral crystals of quartz appear as hexagonal prisms
capped at either or both ends by hexagonal pyramids, con-
sistent with the underlying trigonal symmetry of alpha-
quartz. The atomic framework of alpha-quartz consists
of Si4+ cations that are tetrahedrally coordinated by oxy-
gen anions (O2 ). Every oxygen anion is bonded to two
silicon cations, so that the tetrahedral units are corner-
linked to form continuous chains. In alpha-quartz, two dis-
tinct tetrahedral chains spiral about the crystallographic c-
axis, creating ditrigonal tunnels along this direction
(Figure 1).
Although the tight crystal structure of quartz is not ame-
nable to substitution by such radionuclides as U or Th, the
past half century has seen major progress in the use of
quartz for dating materials through luminescence-based
approaches. Despite its apparent chemical simplicity, nat-
ural quartz crystals host a wide variety of defects, includ-
ing vacant oxygen and silicon sites, as well as
substitutional and interstitial cations (particularly H, Li,
Al, and Fe). When quartz crystals are exposed to ionizing
radiation (alpha, beta, gamma, or cosmic), these defects
can serve as traps for free electrons generated by the radi-
ation. Irradiated quartz crystals will luminesce when an
external stimulation, such as from heat or light or stress,
is applied, and the electrons are evicted from their traps
to recombine with an appropriate defect center. In the
absence of sunlight, some electron traps in quartz are
thought to persist for hundreds of thousands to perhaps
millions or billions of years. The concentration of trapped
electrons (and hence the intensity of the stimulated
650 QUARTZ DEFECTS, OPTICALLY STIMULATED LUMINESCENCE AND THERMOLUMINESCENCE
Quartz, Figure 1 Projection of the crystal structure of alpha-quartz along the crystallographic c-axis. Silicon atoms are violet balls and
oxygen atoms are red balls.
luminescence) increases with higher radiation dose until
saturation is reached.
Through laboratory-based determinations of the rela-
tionship between radiation dose and light response, quartz
crystals have been used to date geologic materials that
were removed from sunlight and subsequently exposed
to constant doses of ionizing radiation. Quartz-based lumi-
nescence has proven particularly successful in the chro-
nometry of Quaternary sediments deposited by wind,
water, and ice, but scientists have employed the technique
for soils, volcanic rocks, and archaeological materials as
well. Although many different luminescent stimuli have
been experimentally investigated, heat and light are most
commonly applied. Thermoluminescent (TL) dating
involves heating quartz crystals at 5–20  C/s and monitor-
ing the temperature, wavelength, and intensity of the light
emission. Most electron traps are emptied when quartz is
heated above 500  C, and emission events at 110  C,
230  C, 270  C, 325  C, and 375  C are frequently
observed. Scientists have attempted to relate these
characteristic emissions to particular electron recombina-
tion centers. Optically stimulated luminescence
(OSL) requires stimulation by light, typically using
a laser emission in the blue-green portion of the electro-
magnetic spectrum, and the OSL emission occurs in the
ultraviolet.
Recent studies have increased the sophistication of
luminescence dating by combining these approaches
(e.g., thermally transferred optically stimulated lumines-
cence, or TT-OSL) and by analyses of multiple aliquots
of a given sample. In addition to improving precision
and accuracy, scientists hope that these approaches will
extend the reach of quartz luminescence dating to a few
million years.
Bibliography
Duller, G. A. T., and Wintle, A. G., 2012. A review of the
thermally transferred optically stimulated luminescence signal
from quartz for dating sediments. Quaternary Geochronology,
7, 6–20.
Frondel, C., 1962. The System of Mineralogy, 7th edn. New York/
London: Wiley, Vol. 3, p. 3.
Heaney, P. J., Prewitt, C. T., and Gibbs, G. V., 1994. Silica: Physical
Behavior, Geochemistry and Materials Applications. Washing-
ton, DC: Mineralogical Society of America.
Lian, O. B., and Roberts, R. G., 2006. Dating the quaternary: pro-
gress in luminescence dating of sediments. Quaternary Science
Reviews, 25, 2449–2468.
Preusser, F., et al., 2009. Quartz as a natural luminescence dosime-
ter. Earth-Science Reviews, 97, 184–214.
Cross-references
Luminescence Dating
Sediment, ESR
QUARTZ DEFECTS, OPTICALLY STIMULATED
LUMINESCENCE AND THERMOLUMINESCENCE
Marco Martini
Dipartimento di Scienza dei Materiali and Sezione INFN,
Universita’ degli Studi di Milano Bicocca, Milan, Italy
Definition
Point defects in solids: Any break in the regular pattern of
a crystal, due to absence of an atom (vacancy) or to an
atom of the crystal in a position out of the regular lattice
 QUARTZ DEFECTS, OPTICALLY STIMULATED LUMINESCENCE AND THERMOLUMINESCENCE
(interstitial) or an extra atom of different species
(impurity).
Luminescence: Energy emission in the form of light as
a consequence of absorption of energy of various types:
As an example radioluminescence occurs when a crystal
emits light after absorption of ionizing radiation. As
a general rule luminescence involves electrons that
increase their energy level in the absorption (excitation)
and return to the original level causing light emission
(de-excitation).
Delayed luminescence is a particular type of luminescence
occurring when on average there is light emission long
after the energy absorption. The most frequently reported
delayed luminescence is phosphorescence, caused by
transition rules that make scarcely probable the radiative
recombination.
Thermoluminescence (TL) is a kind of delayed lumines-
cence in a solid. The excited electrons are “trapped” in
defect sites. TL is stimulated by increasing the tempera-
ture of the emitting crystal, which causes a higher proba-
bility of “detrapping” the electrons, allowing them to
recombine at luminescence centers, with light emission.
Optically stimulated luminescence (OSL) is another kind
of delayed luminescence in a solid, where detrapping of
electrons is caused by light stimulation in place of the
heating of TL.
Radiation dosimetry: The measurement of the amount of
energy absorbed due to the exposition of a material to ion-
izing radiation. Delayed luminescences, TL and OSL, are
appropriate dosimetric techniques.
Electron spin resonance (ESR): Spectrometric technique
which detects unpaired electrons in a material. It is also
exploited as a dosimetric and dating technique.
Introduction
Quartz is one of the most abundant minerals in the conti-
nental crust of the Earth. It is also industrially synthesized
to exploit its properties as an electronic oscillator and is
present in many electronic devices for high-frequency
clocks.
Like all natural and synthetic materials, quartz contains
variable concentrations of defects which can have positive
and negative effects. In the case of quartz electronic oscil-
lators, the presence of low concentrations of Al ions,
which substitute Si ions, produces an unwanted radia-
tion-induced frequency shift, due to the charge-
compensating alkali ions at the Al sites. The same Al
defects together with other defects are responsible for
electronic trapping and luminescence recombination that
are very useful for dosimetric purposes: based on this
properties, radiation dosimetry and ceramics dating are
possible.
To correctly deal with these cases, the probability of
detrapping must be considered and the phenomenon gen-
erally follows an exponential decay and is temperature
dependent: at a fixed temperature the intensity of the emit-
ted light exponentially decreases (McKeever, 1985;
651
Martini and Meinardi, 1997). In most cases luminescence
is due to the presence of lattice defects that are generally
involved in the recombination of charge following
detrapping from defects of higher energy, regardless of
the original energy absorption: the emitted wavelengths
are very often the same in the various types of lumines-
cence of a certain material. Looking at delayed lumines-
cences, OSL and TL, further defects are to be
considered, specifically the ones responsible for charge
trapping.
Quartz defects
The luminescence properties of quartz defects have been
the subject of a large number of studies aimed at under-
standing the complex mechanisms governing the various
luminescence emissions and their modifications as
a consequence of irradiation and thermal treatments (Lieb
and Keinonen, 2006; Preusser et al., 2009).
Many defects, both intrinsic and impurity related, are
known to be present or produced in quartz, having been
identified through various spectroscopic techniques; how-
ever so far the role of these defects in producing lumines-
cence emissions is still under discussion. Although a great
number of studies have been carried out on the structure
properties of defects in quartz, mainly through ESR mea-
surements (Weil, 2000), only tentative correlations
between defects and luminescence emissions have been
put forth. Many defects are known to be present both in
the crystalline (quartz) and in the amorphous (silica) forms
of SiO2, but the luminescence properties of quartz and sil-
ica are largely different (Stevens Kalceff and Phillips,
1995). Nonetheless there is agreement on the role of some
impurity defects in quartz, particularly the so-called Al
centers in being related to the emission dynamics
(Halliburton et al., 1981; Martini et al., 2012a).
Some intrinsic or impurity-related defects have been
proposed as responsible for the luminescence emissions
in quartz, while only few proposals of trapping centers
are present in the scientific literature so far. It is to be
remembered that it happens often that traps and recombi-
nation centers are strongly spatially related and some
OSL and TL have been proposed to be of this kind (Itoh
et al., 2002; Martini et al., 2009).
Two main kinds of quartz defects are reported as related
to its luminescence properties: oxygen vacancies, the most
frequently occurring intrinsic defects, and Al centers, an
ubiquitous type of impurity center. Besides these defects,
other impurities, like Ti, Fe, and Ge ions, give origin to
many defects, and Si vacancies are known to be present
as a counterpart to oxygen vacancies; they give origin
to H4O4 centers, where the presence of H+ ions balances
the Si4+ missing charge.
Oxygen vacancies
The simplest type of O vacancy is the diamagnetic center
resulting from the removal of an O atom from the
652 QUARTZ DEFECTS, OPTICALLY STIMULATED LUMINESCENCE AND THERMOLUMINESCENCE
Quartz Defects, Optically Stimulated Luminescence and Thermoluminescence, Figure 1 Intrinsic defect centers in quartz (Taken
from Preusser et al., 2009).
otherwise perfect a-quartz, leaving a direct bond between
two Si atoms (Feigl et al., 1974).
This “neutral O vacancy” is often considered as the pre-
cursor of the paramagnetic E’1 center, which can be
described as a hole trapped at a neutral O vacancy. The
resulting defect features an unpaired electron on one of
the two Si atoms, while the other Si ion, being positively
charged, moves away from the vacancy into a nearby
plane configuration (Figure 1) (Rudra and Fowler,
1987). The neutral O vacancy capturing a hole is often
considered as the most likely precursor of the E’1; how-
ever, this relationship is controversial and the precursor
of E’1 is still to be identified.
Besides the E’1 center, E’2 and E’4 are known to be pre-
sent in quartz as modifications of the E’1, being related to
the presence of H ions (Fowler et al., 1988). A further
group of O deficient centers is the E”. They are very sim-
ilar to the E’ centers, wherein the number of primes
denotes how many unpaired electrons a center possesses.
Specifically E”1, E”2, and E”3 have been reported, charac-
terized by their slightly different ESR signal (Bossoli
et al., 1982). They are all quenched by heating the quartz
in the temperature range 50–100  C.
Al centers
Al3+ ion is the main ubiquitous impurity to which many
defects are associated. It is present as a substitutional ion
for Si4+ and is charge compensated by an alkali ion (M+)
or by an H+ ion, giving rise to [AlO4/M+]0, [AlO4/H+]0,
respectively. When neither M+ nor H+ is present near
a substitutional Al3+ ion, this latter can be charge compen-
sated by a positive hole, giving rise to the [AlO4]0 center
(Halliburton et al., 1981). The three kinds of Al centers
and their interrelations are shown in Figure 2.
Quartz luminescence, TL and OSL
Many useful results on quartz luminescence come not
only from TL and OSL, but also following other types of
stimulation, mainly cathode- and radioluminescence. In
these cases trapping is not considered and only lumines-
cence centers are investigated.
Three main emissions are known to be present in quartz
luminescence: a UV luminescence reported at 330–380
nm; a blue luminescence, generally reported at around
470 nm; and a red luminescence at around 620 nm. It must
be noted that the listed emissions (UV, blue and red) are
probably composite. This would explain why each of
them is often reported at different wavelengths. The blue
emission has been shown to be due to the sum of a band
at 485 nm (2.53 eV) and a band at 435 nm (2.85 eV). In
radioluminescence spectra the first is enhanced by irradia-
tion and the second is quenched by it (Martini et al.,
2012b).
A typical TL glow curve of quartz is given by three
main glow peaks at around the following temperatures
(using a 20  C/s heating rate): 110  C, 220  C, and
325–375  C. The emission wavelength at a given temper-
ature can vary among different types of quartz, but also in
some cases different glow peaks in the same specimen
emit at the same wavelength.
The 110  C TL peak in an untreated quartz emits in
the blue, but after repeated room temperature (RT) irradi-
ation, followed by heating to temperatures in the range
350–450  C, the sensitivity of the quartz crystal greatly
 QUARTZ DEFECTS, OPTICALLY STIMULATED LUMINESCENCE AND THERMOLUMINESCENCE 653

Quartz Defects, Optically Stimulated Luminescence and Thermoluminescence, Figure 2 Al centers in quartz (Taken from Preusser
et al., 2009).

increases, even of orders of magnitude: this is the phenom-
enon known as “pre-dose effect.” Zimmerman (1971)
interpreted the increased intensity as the result of
an increase in activated luminescence center concentra-
tion. It is to be noted that the 110  C TL emission of
a “pre-dosed” quartz is in the UV, at 360 nm; see Figure 3
(Martini et al., 2012b). OSL emits in the UV as well
(Huntley et al., 1996; Martini and Galli, 2007), and this
was supported by evidence of a pre-dose effect for the
OSL signal (Stoneham and Stokes, 1991; Koul and
Chougaonkar, 2007).
OSL, as well as TL, is also seen to greatly enhance its
sensitivity as a consequence of high temperature
annealing (Poolton et al., 2000; Schilles et al., 2001; Chen
et al., 2001; Lai et al., 2008). This effect is proposed to be
related to the variation of concentration of E’1 centers that
may act as non-radiative centers competing in the recom-
bination process (Poolton et al., 2000).
The emission wavelength of the main dosimetric TL
peak at 375  C is strongly dependent on the origin of
quartz: some quartzes have a dominant blue emissions,
others emit in the red (Shimizu et al., 2006).
654 QUARTZ DEFECTS, OPTICALLY STIMULATED LUMINESCENCE AND THERMOLUMINESCENCE

Quartz Defects, Optically Stimulated Luminescence and Thermoluminescence, Figure 3 (a) Sequence of radioluminescence
spectra of a natural quartz during irradiation (each spectrum after a dose of 20 Gy), followed by a heating up to 500  C. (b) Sequence
of thermoluminescence glow curves of a natural quartz after irradiation (each spectrum relative to a dose of 20 Gy), followed by
a heating up to 500  C, like in (a). (c) Dose response curve of the 360 nm (3.43 eV) radioluminescence emission band and of the 110  C
thermoluminescence glow peak. The data are normalized to the final value (Taken from Martini et al., 2012b).

It is to be noted that the most commonly used filters to
detect TL are centered at around 400 nm, with high effi-
ciency for detecting the blue emissions at about 470 nm,
while the red TL at about 640 nm is effectively excluded.
In most quartz a further glow peak is seen at 325  C.
This glow peak has been related to the 110  C glow peak
and to OSL (Kaylor et al., 1995; Wintle and Murray,
1997). It has been seen to have the same emission as the
OSL and the 110  C peak, in the range 380–420 nm.
Recombination models
So far it is not possible to definitely assign each emission
to a specific center; however some proposals find a good
deal of support. There is agreement on the fundamental
role of Al centers in participating in the luminescence of
quartz. In untreated quartz, Al is known to be present mainly
as charge compensated by an alkali ion, generally Li+. Irradi-
ation detaches Li+ which is then free to move along the chan-
nels parallel to the crystallographic c-axis. The effect is the
transformation of [AlO4/Li+]0 to [AlO4/H+]0 or to [AlO4]0
centers, depending on how many H+ ions are available.
By making specific electro-diffusion treatments
(“sweeping”), it is possible to modify the concentration
of alkali ions in quartz; if this sweeping is made in air at
RT, alkali ions are substituted by H+ at the Al3+ sites. By
sweeping out in vacuum at high temperature, Martini
et al. (1995) could eliminate from quartz crystals both
[AlO4/Li+]0 and [AlO4/H+]0, replacing most of them with
[AlO4]0 centers. They suggested that the 380 nm emission
 QUARTZ DEFECTS, OPTICALLY STIMULATED LUMINESCENCE AND THERMOLUMINESCENCE
is the result of recombination at [AlO4]0 centers, contrary
to Yang and McKeever (1990) who related the enhance-
ment of the intensity of the 380 nm emission to changes
in ESR active centers, assigning the emission to elec-
tron–hole recombination at [H3O4]0 centers. [AlO4]0 has
been also proposed as the defect responsible for the blue
TL at 470 nm (Yang and McKeever, 1990; Woda et al.,
2002).
Sweeping turned out to be extremely useful in the study
of defects in quartz. Jani et al. (1983) reported
a correlation between the growth of the E’1 center and
the decay of the [AlO4]0 center and observed that sweep-
ing out the alkali ions strongly reduces the intensity of the
E’1 ESR signal. Poolton et al. (2000) suggested that E’1
may act as a non-radiative center competing in OSL with
the [AlO4]0 recombination center.
Quartz types and luminescence
Quartzes of different origin present different luminescence
emissions: volcanic quartz generally emits in the red
(Westway, 2009), while slow solidification in plutonic
rocks results in strong blue emission (Hashimoto, 2008).
Hydrothermal quartz also emits at ~480 nm, but the emis-
sion has been reported as weak (Rink et al., 1993; Rendell
et al., 1994). Shimizu et al. (2006) detected a single emis-
sion peak in the red (~630 nm) for a sample of volcanic
quartz.
It should be noted however that red emission has been
detected in silica, but never in synthetic quartz. Its pres-
ence in volcanic quartz could be related to the amorphous
inclusions probably present in this kind of natural quartz.
Rink et al. (1993) also noticed a large red intensity in vol-
canic quartz and suggested an assignment to defect
clusters.
Conclusions
Many studies on the TL and OSL emissions in quartz pro-
vided evidence related to the defects involved in the radi-
ative recombinations, even if definite assignments of each
emission to a specific center are still lacking. The main
emissions are known to be in the UV (330–380 nm), in
the blue at around 470 nm, and in the red at around
620 nm. Precise measurements are clarifying that these
emissions are indeed composite, making it more difficult
to correctly follow their behavior as a function of irradia-
tion and thermal treatments. As a consequence the assign-
ments proposed so far are rather tentative, and a systematic
definition of luminescence emissions correlated with
defect dynamics must be reached as a first step to the
assignment-specific roles to defects.
Nonetheless an important and complex role of Al cen-
ters in participating in charge recombination can definitely
be concluded, but also that migrations by alkali and hydro-
gen ions promoted by irradiation and/or heating surely
play an essential part.
At the same time, the role of oxygen vacancies in indi-
rectly participating in the radiative recombination is clear.
655
Bibliography
Bossoli, R. B., Jani, M. G., and Halliburton, L. E., 1982. Radiation-
induced E” centers in crystalline SiO2. Solid State Communica-
tions, 44, 213–217.
Chen, G., Li, S. H., and Murray, A. S., 2001. Effect of heating on the
quartz dose–response curve. Radiation Measurements, 33,
59–63.
Feigl, F. J., Fowler, W. B., and Yip, K. L., 1974. Oxygen vacancy
model for E’1 center in SiO2. Solid State Communications, 14,
225–229.
Fowler, W. B., Rudra, J. K., Edwards, A. H., and Feigl, F. J., 1988.
Theory of oxygen vacancy defects in silicon dioxide. In Devine,
R. A. B. (ed.), The Physics and Technology of Amorphous SiO2.
New York: Plenum Press, pp. 107–112.
Halliburton, L. E., Koumvakalis, N., Markes, M. E., and Martin,
J. J., 1981. Radiation effects in crystalline SiO2: the role of alu-
minum. Journal of Applied Physics, 52, 3565–3574.
Hashimoto, T., 2008. An overview of red-thermoluminescence
(RTL) studies on heated quartz and RTL application to dosimetry
and dating. Geochronometria, 30, 9–16.
Huntley, D. J., Short, M. A., and Dunphy, K., 1996. Deep traps in
quartz and their use for optical dating. Canadian Journal of
Physics, 74, 81–91.
Itoh, N., Stoneham, D., and Stoneham, A. M., 2002. Ionic and elec-
tronic processes in quartz: mechanisms of thermoluminescence
and optically stimulated luminescence. Journal of Applied Phys-
ics, 92, 5036–5044.
Jani, M. G., Bossoli, R. B., and Halliburton, L. E., 1983. Further
characterization of the E1’ center in crystalline SiO2. Physical
Review B, 27, 2285–2293.
Kaylor, R. M., Feathers, J., Hornyak, W. F., and Franklin, A. D.,
1995. Optically stimulated luminescence in Kalahari quartz:
bleaching of the 325  C peak as the source of the luminescence.
Journal of Luminescence, 65, 1–6.
Koul, D. K., and Chougaonkar, M. P., 2007. The pre-dose phenom-
enon in the OSL signal of quartz. Radiation Measurements, 42,
1265–1272.
Lai, Z. P., Brückner, H., Fülling, A., and Zöller, L., 2008. Effects of
thermal treatment on the growth curve shape for OSL of quartz
extracted from Chinese loess. Radiation Measurements, 43,
763–766.
Lieb, K. P., and Keinonen, J., 2006. Luminescence of ion-irradiated
alpha quartz. Contemporary Physics, 47, 305–331.
Martini, M., and Galli, A., 2007. Ionic mechanisms in the optically
stimulated luminescence of quartz. Physica Status Solidi: Cur-
rent Topics in Solid State Physics, 4, 1000–1003.
Martini, M., and Meinardi, F., 1997. Thermally stimulated lumines-
cence: new perspectives in the study of defects in solids. La
Rivista del Nuovo Cimento, Serie 4, 20(8).
Martini, M., Paleari, A., Spinolo, G., and Vedda, A., 1995. Role of
[AlO4]0 centers in the 380-nm thermoluminescence of quartz.
Physical Review B, 52, 138–142.
Martini, M., Fasoli, M., and Galli, A., 2009. Quartz OSL emission
spectra and role of [AlO4] recombination centres. Radiation
Measurements, 44, 458–461.
Martini, M., Fasoli, M., Galli, A., Villa, I., and Guibert, P., 2012a.
Radioluminescence of synthetic quartz related to alkali ions.
Journal of Luminescence, 132, 1030–1036.
Martini, M., Fasoli, M., Villa, I., and Guibert, P., 2012b. Radiolumi-
nescence of synthetic and natural quartz. Radiation Measure-
ments, 47, 846–850.
McKeever, S. W. S., 1985. Thermoluminescence of Solids. Cam-
bridge: Cambridge University Press.
Poolton, N. R. J., Smith, G. M., Riedi, P. C., Bulur, E., Bøtter-
Jensen, L., Murray, A. S., and Adrian, M., 2000. Luminescence
sensitivity changes in natural quartz induced by high
656 QUARTZ DEFECTS, OPTICALLY STIMULATED LUMINESCENCE AND THERMOLUMINESCENCE
temperature annealing: a high frequency EPR and OSL study.
Journal of Physics D, 33, 1007–1017.
Preusser, F., Chithambo, M. L., Götte, T., Martini, M., Ramseyer,
K., Sendezera, E. J., Susino, G. J., and Wintle, A. G., 2009.
Quartz as a natural luminescence dosimeter. Earth Science
Reviews, 97, 184–214.
Rendell, H. M., Townsend, P. D., Wood, R. A., and Luff, B. J., 1994.
Thermal treatments and emission-spectra of TL from quartz.
Radiation Measurements, 23, 441–449.
Rink, W. J., Rendell, H. M., Marseglia, E. A., Luff, B. J., and
Townsend, P. D., 1993. Thermoluminescence spectra of igneous
quartz and hydrothermal vein quartz. Physics and Chemistry of
Minerals, 20, 353–361.
Rudra, J. K., and Fowler, W. B., 1987. Oxygen vacancy and the E1’
center in crystalline SiO2. Physical Review B, 35, 8223–8230.
Schilles, T., Poolton, N. R. J., Bulur, E., Bøtter-Jensen, L., Murray,
A. S., Smith, G. M., Riedi, P. C., and Wagner, G. A., 2001.
A multi-spectroscopic study of luminescence sensitivity changes
in natural quartz induced by high-temperature annealing. Jour-
nal of Physics D, 34, 722–731.
Shimizu, N., Mitamura, N., Takeuchi, A., and Hashimoto, T., 2006.
Dependence of radioluminescence on TL-properties in natural
quartz. Radiation Measurements, 41, 831–835.
Stevens Kalceff, M. A., and Phillips, M. R., 1995. Cathodolumi-
nescence microcharacterization of the defect structure of quartz.
Physical Review B, 52, 3122–3134.
Stoneham, D., and Stokes, S., 1991. An investigation of the rela-
tionship between the 110  C TL peak and optically stimulated
luminescence in sedimentary quartz. Nuclear Tracks and Radia-
tion Measurements, 18, 119–123.
Weil, J. A., 2000. A demi-century of magnetic defects in a-quartz.
In Pacchioni, G., Skuja, L., and Griscom, D. L. (eds.), Defects
in SiO2 and Related Dielectrics: Science and Technology.
Amsterdam: Kluwer, pp. 197–212.
Westway, K. E., 2009. The red, white and blue of quartz lumines-
cence: a comparison of De values derived for sediments from
Australia and Indonesia using thermoluminescence and optically
stimulated luminescence emissions. Radiation Measurements,
44, 462–466.
Wintle, A. G., and Murray, A. S., 1997. The relationship between
quartz thermoluminescence, photo-transferred thermolumines-
cence and optically stimulated luminescence. Radiation Mea-
surements, 27, 611–624.
Woda, C., Schilles, T., Rieser, U., Mangini, A., and Wagner, G. A.,
2002. Point defects and the blue emission in fired quartz at high
dose: a comparative luminescence and EPR study. Radiation
Protection Dosimetry, 100, 261–264.
Yang, X. H., and McKeever, S. W. S., 1990. Point-defects and the
predose effect in quartz. Radiation Protection Dosimetry, 33,
27–30.
Zimmerman, J., 1971. The radiation-induced increase of the 100  C
thermoluminescence sensitivity of fired quartz. Journal of Phys-
ics C: Solid State Physics, 4, 3265–3276.
Cross-references
Luminescence Dating
Sediment, ESR
 R
RADIATION DEFECT
Jeroen W. Thompson
Department of Medical Physics and Applied
Radiation Sciences, McMaster University, Hamilton,
ON, Canada
Definition
Radiation defects are divided into two types: (1) those pro-
duced by atomic displacements of normal lattice site ele-
ments during irradiation (intrinsic) which are later
modified by ionizing radiation and (2) those created ini-
tially during growth of the crystal lattice by substitution
of an impurity atom or molecule (extrinsic) that are later
modified by ionizing radiation. These types of defects
are radiation sensitive, that is, they accumulate with
increasing levels of ionizing radiation.
Radiation defects are utilized in all forms of trapped
charge or radiation exposure dating, i.e., thermolumines-
cence (TL), optically stimulated luminescence (OSL),
electron spin resonance (ESR), and other less common
techniques. Ionizing radiation from natural background
sources creates ions and free electrons; these ions and/or
electrons may be stabilized (trapped) as radiation-induced
defects before recombination can occur at normal lattice
sites. If a signal related to concentration of defects can
be measured through a suitable technique (such as TL,
OSL, or ESR) and if the signal can be related to the total
absorbed radiation dose, then an age may be inferred from
a measurement of signal intensity in combination with
a determination of past radiation dose rate.
In many cases, the precise identity of the defect(s)
responsible for a given dating signal may not be known.
J.W. Rink, J.W. Thompson (eds.), Encyclopedia of Scientific Dating Methods, DOI 10.1007/978-94-007-6304-3,
© Springer Science+Business Media Dordrecht 2015
One example of a known class of extrinsic radiation
defects is the Al center. The parent center is an Al3+ ion
substituting for Si4+ in quartz (crystalline SiO2), accompa-
nied by an interstitial cation M+ for reasons of charge neu-
trality: [AlO4/Li+]0, [AlO4/Na+]0, and [AlO4/H+]0 are
neutral parent centers (Ikeya, 1993). An electron released
due to the action of ionizing radiation may result in a hole
trapped at the Al defect site, with the cation diffusing away
to form the stable radiation defect [AlO4/h]0. The concen-
tration of Al defects may be determined through electron
spin resonance spectroscopy at low temperatures (77 K).
An example of an intrinsic radiation defect is the E'1
center in quartz. This is an oxygen vacancy charged with
a single unpaired electron located on an adjacent silicon
atom (Rudra and Fowler, 1987).
Bibliography
Ikeya, M., 1993. New Applications of Electron Spin Resonance.
Singapore: World Scientific.
McKeever, S. W. S., 1985. Thermoluminescence of Solids. Cam-
bridge: Cambridge University Press.
Rudra, J. K., and Fowler, W. B., 1987. Oxygen vacancy and the E'1
center in crystalline SiO2. Physical Review B, 35, 8223–8230.
Weil, J. A., Bolton, J. R., and Wertz, J. E., 1994. Electron Paramag-
netic Resonance: Elementary Theory and Practical Applica-
tions. New York: Wiley.
Cross-references
Electron Spin Resonance (ESR) Dating, General Principles
Feldspar, Infrared-Stimulated Luminescence
Luminescence Dating
Quartz
Quartz Defects, Optically Stimulated Luminescence and Thermo-
luminescence
Radiation Dose Rate
658 RADIATION DOSE RATE
RADIATION DOSE RATE
Grzegorz Adamiec
Institute of Physics, Centre for Science and Education,
GADAM Centre, Silesian University of Technology,
Gliwice, Poland
Definition
Radiation dose rate is ionizing radiation energy absorbed
per unit of time per unit of mass of the absorbing matter.
Introduction
When radiation travels through matter, it interacts with the
matter. As a result of those interactions, the energy of radi-
ation may be absorbed by the matter. Numerous physical
processes are responsible for the absorption of radiation
energy, depending on the type of radiation. In connection
with scientific dating methods, the term dose rate is partic-
ularly relevant to trapped charge dating techniques, i.e.,
luminescence (OSL – optically stimulated luminescence,
TL – thermoluminescence) or electron spin resonance
(ESR) dating. The determination of absorbed dose and
dose rate is the subject of dosimetry which plays
a particular role in medicine and the nuclear industry.
A comprehensive introduction to dosimetry and associ-
ated problems is given by Attix (2004).
In SI units the absorbed dose rate is expressed as 1 Gy/s,
where 1 Gy (Gray) is 1 J of radiation energy absorbed by
1 kg of the absorbing matter. In older work, the CGS unit
of dose rate of 1 rad/s may be encountered, which is equal
to 0.01 Gy/s. The absorbed energy concerns this part of the
radiation energy which is used to increase the internal
energy of matter, i.e., its temperature.
Ionizing radiation can be divided into two categories:
– Directly ionizing radiation – this includes charged par-
ticles (electrons, alpha particles, protons, nuclei,
muons) – which deposit energy in the matter through
which it travels, through Coulomb interactions.
– Indirectly ionizing radiation – neutral particles (e.g., pho-
tons in the form of gamma radiation, neutrons) – which
deposits its energy in a two-step process. In the first step,
the particles lose some of their kinetic energy in colli-
sions with secondary charged particles (electrons) pre-
sent in the matter through which the radiation passes,
and in the second step, these secondary charged particles
lose their energy either through collisions or through
radiation emission (e.g., bremsstrahlung – X-rays emit-
ted as a result of deceleration of charged particles).
_ in a given finite vol-
The average absorbed dose rate (D)
ume (element) of the considered matter is defined by
De
D_ ¼ ð1Þ
DmDt
where De is the difference between radiation energy
imparted to the considered volume and radiation energy
leaving that volume, Dm is the mass of the considered vol-
ume, and Dt is the time over which this energy has been
absorbed.
A device which allows the measurement of absorbed
dose in a given volume is called a dosimeter. This includes
devices for direct measurements (e.g., an ionization cham-
ber with the necessary electronics) or materials whose
properties allow storage of radiation energy (e.g., in the
form of effects the radiation causes) and a later readout
of the accumulated signal and post-determination of the
total dose accumulated over the recording time. Examples
of the latter may include small chips made of Al2O3:C
(carbon-doped aluminum oxide) used in OSL dosimetry
or capsules with doped lithium fluoride (e.g., LiF:Mg)
used in thermoluminescence dosimetry.
Dose rate determination
Direct measurements of dose rate within matter are
a complex undertaking, especially if the knowledge of dis-
tribution on a small scale within the material and with tem-
poral resolution is required (small volume or mass and
very short time in Eq. (1)). This is due to the fact that radi-
ation is absorbed differently depending on the density and
the atomic number of the matter through which radiation
passes and on the type of radiation and its energy. There-
fore, if the dosimeter, or dose rate meter, is made of
a different material than the one in which we want to deter-
mine the dose rate, a need arises to apply conversion cal-
culations. In addition, if different types of radiation, of
a range of energies, are present (a so-called mixed radia-
tion field), it is difficult, if not impossible, to calibrate
the devices in an absolute fashion to provide readings in
Gy/s (or Gy in the case of accumulated dose). Therefore,
the application of dosimeters always requires careful con-
sideration in a given experimental setup.
On the other hand, in applications like trapped charge
dating techniques, of interest is the average dose rate over
the burial time of the dated geological sediment or
ceramics rather than momentary dose rate. There exists
a possibility to use Monte Carlo simulations to determine
the distribution of dose within matter on the basis of sim-
ulating the physical phenomena during the passage of
radiation through matter and the transport of secondary
charged particles in matter (e.g., see Nathan and Mauz,
2008; Guérin and Mercier, 2011, 2012, example of appli-
cation in Grine et al., 2007).
Example: dose rates originating from various
natural sources
As already mentioned in the introduction, in relation to sci-
entific dating methods, the concept of dose rate has
a particular relevance to trapped charge dating techniques.
In the Earth’s crust, the major radioactive elements present
are in the uranium and thorium decay series (natural ura-
nium series consists of 99.29 % 238U and 0.71 % 235U by
weight of natural uranium) and potassium (40K). All these
isotopes are emitters of gamma rays and beta particles while
 RADIATION DOSE RATE
the two decay series also contain alpha particle emitters. In
addition, the Earth is exposed to a constant flux of muons
created in upper layers of atmosphere by very high energy
protons traveling from outside the solar system. Hence,
the total dose rate experienced by objects on the Earth’s sur-
face consists of the contributions of alpha, beta, gamma,
and cosmic radiations from these sources. The concentra-
tion of U, Th, and K is different at different points of the
Earth’s crust; hence, the natural dose rate varies from place
to place.
Each of the types of radiation has specific issues when
its dose rate in trapped charge dating is considered. These
issues are related to their range and the way in which they
interact with matter. The palaeodose, i.e., the total dose
that would be determined by trapped charge dating tech-
niques, is given by


P ¼ age  kpa D_ a þ pb D_ b þ D_ g þ D_ c ð2Þ
where pa and pb are factors (values between 0 and 1)
related to the size of grains which are being dated (these
factors take into account the dose rate being different at
the surface of the grains and in their interiors); k is the
effectiveness of a particles in inducing a luminescence
(or ESR) signal and D_ a , D_ b , and D_ g dose rates of a, b, g
radiation as calculated from the infinite matrix assumption
(see next paragraph), respectively; and D_ c is the
cosmic radiation dose rate. The quantity in the brackets in
Eq. (2) is the total so-called annual dose, i.e., dose
rate expressed in Gy/a or Gy/ka (Gy per annum or
kiloannum). It needs to be stressed that the actual radiation
dose rate experienced by the grains does not contain the
factor k.
The dose rates of alpha, beta, and gamma radiation are
determined with the use of the infinite matrix assumption
(see Fano, 1954; Roesch and Attix, 1968; Aitken, 1985;
and Guérin et al., 2012 for a critical review), i.e., under
the assumption that radioisotopes are distributed uni-
formly in a homogenous matter surrounding the dated
material (e.g., grains). In such a case, every portion of
the matter absorbs as much radiation energy as it emits,
as all the radiation energy is absorbed in the matter. Under
this assumption, the specific activities (e.g., in Bq/kg, i.e.,
parent disintegrations per second per kilogram of matter)
of uranium (natural U) and thorium (232Th) chains and
potassium (40K) are determined in the material surround-
ing the dated grains using a chosen method and they are
then converted into dose rates using factors derived under
the infinite matrix assumption; the latest conversion factor
values are given by Guérin et al. (2011, 2012).
In Eq. (2) the term D_ c denotes the contribution of cos-
mic radiation attributed mainly to the flux of muons
(Prescott and Hutton, 1994). The dose rate from cosmic
radiation varies with latitude, altitude, and depth below
the Earth’s surface. This contribution can be calculated
from the geographical coordinates and depth of the dated
sample. Although often this contribution is small, it is
non-negligible, and in environments with low
659
concentrations of radioisotopes considerable, and has to
be taken into account.
The dose rate received by dated mineral grains
(or artifacts) also depends on the amount of moisture, or,
occasionally, carbonates present in the sediment. Water
and carbonates attenuate the radiation field, and the more
of these substances present, the more strongly attenuated
the radiation. The effects of water content may be taken
into account as suggested by Aitken (1985) and Guerin
and Mercier (2012). The influence of carbonates is more
difficult to assess as discussed by Nathan and Mauz
(2008) and applied by Grine et al. (2007).
The assumptions of an infinite matrix are not precisely
met in real contexts and they have to be considered in
determining the annual dose. Particularly, in complex
environments, these assumptions ought to be considered
carefully. These problems include the presence of highly
radioactive grains, different mineral layers of differing
radioactivity, and difficult or impossible to determine
levels of moisture present in the dated context.
Summary
The concept of dose rate is vital in trapped charge dating
techniques with uncertainty or error being directly
reflected in the calculated age. The determination of the
radiation dose that has been acquired, e.g., by mineral
grains, may be a complex task, requiring a range of
assumptions to be made.
Bibliography
Aitken, M. J., 1985. Thermoluminescence Dating. London:
Academic. 378 pp.
Attix, F. H., 2004. Introduction to Radiological Physics and Radia-
tion Dosimetry. Weinheim: Wiley-VCH, p. 607.
Fano, U., 1954. A note on the Bragg-Gray cavity principle for mea-
suring energy dissipation. Radiation Research, 1, 237–240.
Grine, F. E., Bailey, R. M., Harvati, K., Nathan, R. P., Morris, A. G.,
Henderson, G. M., Ribot, I., and Pike, A. W. G., 2007. Late
Pleistocene human skull from Hofmeyr, South Africa, and mod-
ern human origins. Science, 315(5809), 226–229.
Guérin, G., and Mercier, N., 2011. Determining gamma dose rates
by field gamma spectroscopy in sedimentary media: results of
Monte Carlo simulations. Radiation Measurements, 46,
190–195.
Guérin, G., and Mercier, N., 2012. Preliminary insight into dose
deposition processes in sedimentary media on a grain scale:
Monte Carlo modelling of the effect of water on gamma dose
rates. Radiation Measurements, 47, 541–547.
Guérin, G., Mercier, N., and Adamiec, G., 2011. Dose rate conver-
sion factors: update. Ancient TL, 29(1), 5–8. http://www.aber.ac.
uk/en/media/departmental/iges/ancienttl/pdf/vol29no1/guerin_
atl29(1)_5-8.pdf.
Guérin, G., Mercier, N., Nathan, R., Adamiec, G., and Lefrais, Y.,
2012. On the use of the infinite matrix assumption and associated
concepts: a critical review. Radiation Measurements, 47,
778–785.
Nathan, R. P., and Mauz, B., 2008. On the dose rate estimate of
carbonate-rich sediments for trapped charge dating. Radiation
Measurements, 43, 14–25.
Prescott, J. R., and Hutton, J. T., 1994. Cosmic ray contributions to
dose rates for luminescence and ESR dating: large depths and
660 RADIATION AND RADIOACTIVITY
long-term time variations. Radiation Measurements, 23(2–3),
497–500.
Roesch, W. C., and Attix, F. H., 1968. Basic concepts of dosimetry.
In Attix, F. H., and Roesch, W. C. (eds.), Radiation Dosimetry,
Volume I: Fundamentals. New York/London: Academic.
Cross-references
Alpha Spectroscopy
Electron Spin Resonance (ESR) Dating, General Principles
Luminescence Dating, Dose Rates
Radiation and Radioactivity
RADIATION AND RADIOACTIVITY
Regina DeWitt
Department of Physics, East Carolina University,
Greenville, NC, USA
Definition
Radiation: The term radiation includes the whole electro-
magnetic spectrum as well as all atomic and subatomic
particles. Ionizing radiation refers to the ability of certain
types of radiation to ionize the medium they traverse. Only
ionizing radiation will be considered here.
Radioactivity: The spontaneous emission of particles or
electromagnetic rays from unstable atomic nuclei.
Introduction
In 1895 Wilhelm Conrad Röntgen experimented with
a Crooke’s tube, an evacuated glass enclosure in which
current can be passed through a high vacuum from one
electrode to another. He noticed that a small screen placed
at some distance started fluorescing when the tube was on.
Röntgen named the unknown radiation X-rays. He
observed that they could penetrate paper and wood, but
would be absorbed by aluminum and tin foil. Encouraged
by Röntgen’s observations, Henri Becquerel
experimented with uranium minerals which he placed on
photographic plates. He found that the plates showed signs
of exposure even in the absence of light and eventually
named the uranium emissions a- and b-rays. In 1897
Marie and Pierre Curie made similar observations in
experiments with other elements and suggested the term
“radioactivity” to describe the phenomenon. Many further
discoveries have led to an increased understanding of the
atomic and nuclear structure and the processes leading to
the emission of ionizing radiation. Today radiation and
radioactivity have found application in fields as varied as
medicine, energy production, and dating.
Structure of the nucleus
Atoms consist of a positively charged nucleus around
which negative electrons revolve in stable orbits, bound
by electrostatic coulomb forces due to their opposite
charges. The nuclei consist of positively charged protons
and electrically neutral neutrons which are collectively
referred to as nucleons. The atomic number, Z, is defined
as the number of protons in a nucleus. The number of neu-
trons is designated by the neutron number, N. The mass
number, A, is the total number of nucleons in a nucleus.
An arbitrary element X with atomic number Z and mass
number A is denoted by AZ X, e.g., carbon-14 with its 6 pro-
tons and 8 neutrons is written as 146 C. All nuclei of a given
element have the same number of protons, but may have
different numbers of neutrons. Such nuclei are referred
to as isotopes (e.g., 12 13 14
6 C, 6 C, 6 C).
The positive protons in the nucleus repel one another.
However, the nucleus is held together as a result of the
strong nuclear force. Nuclei can only exist at certain dis-
crete energy states, similar to the energy states of the
atoms themselves. The distance between these nuclear
energy levels is on the order of keV or MeV while atomic
levels are on the order of eV. The number of levels
increases with the mass number, A. The lowest energy
state is called ground level. If the nucleus is excited above
this level, it can de-excite by emitting a photon and/or one
or several nucleons.
The stability of a nucleus depends on its number of pro-
tons and neutrons. Nuclei of relatively small atomic num-
bers are most stable when the numbers of protons and
neutrons are comparable. 12 13
6 C and 6 C are both stable.
Heavy stable nuclei contain considerably more neutrons
than protons, such as the stable bismuth isotope 209 83 Bi,
which contains 83 protons and 126 neutrons. Bismuth is
the heaviest element with a stable isotope. For nuclei with
Z > 83 the strong nuclear force can no longer compensate
the repulsive forces between the protons and the nuclei
disintegrate by a process of decay.
Radioactivity
Unstable nuclei emit particles until they reach a stable
composition, and until stable they are “radioactive.” This
process is often accompanied by a rearrangement of the
nucleons within the nucleus to a lower-energy state, which
leads to the emission of a high-energy photon – compara-
ble to the emission of photons during the de-excitation of
electrons. The most important decay modes are a-, b-, g-
decay, and fission.
a-decay is predominantly observed for heavier nuclei
with Z > 83. The nucleus decays by emitting an a-particle,
4
2He, in which process it loses two protons and two
neutrons:
A
ZX ! A4
Z2 Y þ 42 He ð1Þ
X is referred to as parent nuclide and Y is the daughter
nuclide. For example, the parent 238
92 U decays to its daugh-
ter 234
90 Th by emitting an a-particle. The decay often leaves
the daughter nuclide in one of several possible excited
states, where each state is reached with different probabil-
ity (see Figure 1). a-Particles can therefore be emitted at
several discrete energies characteristic to the parent
nuclide.
 RADIATION AND RADIOACTIVITY 661

(compare Figure 1). The discrete energies of these g-rays

are characteristic for the parent nuclide and are in the
keV-MeV range. Neither atomic number nor mass number
is changed during g-decay of a nucleus.
Spontaneous fission is only observed for high mass
numbers. In the process the nucleus splits into two smaller
nuclei. Several neutrons and large amounts of energy are
released during this process. For 238U the probability to
decay via a-decay is 106 times larger than the probability
to undergo spontaneous fission, but this process varies
among nuclides. In some nuclei such as 235U, fission can
be induced by absorption of a slow neutron, resulting in
an excited 236U nucleus, which ultimately splits into two
smaller nuclei.

Activity and half-life

Radiation and Radioactivity, Figure 1 a-Decay scheme of U 238 Activity, A, is the rate at which unstable nuclei decay, i.e.,
to 234Th. After emission of the a-particle, 234Th can be left in the the number of decays per second. Two units are com-
ground state or one of the two excited states. Decay to an monly used for activity:
excited state is followed by de-excitation and emission of g-rays.
The scheme shows possible a-energies and their emission 1 curie ¼ 1 Ci ¼ 3:7  1010 decays=s
probabilities in percent, as well as possible g-energies (Data from
Table of Nuclides, Korea Atomic Energy Research Institute). which is the activity of 1 g of radium and the SI unit
1 becquerel ¼ 1 Bq ¼ 1 decay=s:
In b-decay a nucleus emits an electron or positron. The conversion factor is 1 Ci ¼ 3.7  1010 Bq.
b-decay is the conversion of a neutron to a proton, under The number of unstable parent nuclei decreases expo-
the emission of an electron and an electron antineutrino: nentially over time. Given an initial number N0 of parent
nuclei at start time t ¼ 0, the number N of nuclei remaining
1
0n ! 11 pþ þ e þ ve ð2Þ
at time t is given by
It is observed for nuclides with an excess of neutrons
N ¼ N 0 elt ð6Þ
and the process can be described by
l is referred to as the decay constant with unit s1. The
A
ZX ! A
Zþ1 Y þ e þ ve ð3Þ
larger the decay constant, the more rapidly the number of
A prominent example is the decay of radiocarbon to parent nuclei decreases with time. The decay constant is
 characteristic for each radioactive nuclide. The activity
6 C ! 7 N þ e þ ve .
nitrogen: 14 14
+
b -decay refers to the conversion of a proton to a neutron, depends on the number N of nuclei and the decay constant:
under emission of a positron (i.e., the positively charged A ¼ lN ð7Þ
antiparticle of the electron) and an electron neutrino:
1 þ Related to the decay constant is the half-life t½ of
1p ! 10 n þ eþ þ ve ð4Þ a radioactive nucleus:
A
ZX ! A
Z1 Y þ eþ þ ve ð5Þ ln 2
t 1=2 ¼ ð8Þ
An example is the decay of 23 23 l
12Mg
to 11 Na.
Similar to
a-decay, the daughter nuclide is often left in one of several The half-life is the time in which the number of parent
possible excited states. Unlike a-particles, however, elec- nuclides decreases by a factor 2, i.e., from N0 to ½N0
trons and positrons do not have discrete energies. b-decay and from ½N0 to ¼N0.
results in two particles and the energy can be divided By determining the current number of parent and daugh-
between the particles in an infinite number of ways. ter nuclides in a system, it is possible to use radioactive
Therefore b-particles are emitted with continuous ener- decay in a variety of different methods to determine the
gies; the spectrum is a curve with a maximum at interme- age of materials. The mass m of radioactive atoms can be
diate energies and extends to an energy maximum. calculated from their activity A, their nuclide mass M, their
a- and b-decays often leave the parent nucleus in an half-life, and Avogadro’s number NAv ¼ 6.022  1023:
excited state, similar to an excited atom. The nucleus
reaches a lower-energy state by rearranging the protons AM
m¼ t1=2 ð9Þ
and neutrons and by emitting a high-energy photon, a g-ray N Av ln 2
662 RADIATION AND RADIOACTIVITY
Radiation and Radioactivity, Figure 2 Thorium decay series. The nuclides, their type of decay (a or b), and their half-lives are
indicated.
Radioactive decay series
When a parent nucleus decays, it may decay into another
radioactive nucleus. As a result, the original parent nuclide
can produce a series of related nuclides which are referred
to as a radioactive decay series (see “U-Series Dating”).
A famous example is the decay series of 238 206
92 U to 82 Pb
(also known as the uranium-radium series). The mass
number of 23892 U is 238 which can be written as 4n + 2
(n ¼ 59). By variation of n all possible mass numbers in
the uranium-radium series can be obtained, which is there-
fore often labeled A ¼ 4n + 2. Other examples are 235 92 U to
207
82 Pb (uranium-actinium series, with mass numbers
90 Th to 82 Pb (mass numbers A ¼ 4n).
A ¼ 4n + 3) and 232 208
Some nuclides can decay in two different ways, such as
212 212 
83 Bi that has two daughter nuclides: 84 Po via b -decay
208
and 81 Tl via a-decay. Thus within each decay series there
may be multiple pathways to reach the last member of the
series. Half-lives can vary considerably within one decay
series. Extreme examples are 232 90 Th with a half-life of
12
84 Po with a half-life of 0.3  10
1.4  1010 year and 212
s in the same decay series. Figure 2 shows the complete
thorium decay series with type of decay and half-life for
each nuclide.
Radioactive equilibria and disequilibria
In a decay series nuclides are transformed into each other:
nuclide 1 ! nuclide 2 ! nuclide 3 ! . . .
Nuclide 2 (N2) is produced through the decay of
nuclide 1 (N1), but decays simultaneously at a different
rate into nuclide 3. The production rate of nuclide 2 is
given by the decay rate of nuclide 1 diminished by the
decay rate of nuclide 2.
In a system where initially only the parent nuclide N1 is
present, daughter nuclides build up over time. Eventually
an equilibrium state will be reached, where the ratio of the
numbers of nuclei N1/N2 is constant.
l1
N2 ¼ N1 ð10Þ
l2  l1
The time necessary to attain this radioactive equilib-
rium depends on the decay constants of the two nuclides.
Two important cases include systems where the mother
nuclide has a much longer half-life than the daughter and
systems where the half-lives are comparable.
Secular radioactive equilibrium refers to cases where
the half-life of the parent nuclide is much larger than that
of the daughter: t1/2(1)  t1/2(2). In secular radioactive
equilibrium, the activities of the parent nuclide and the
daughter nuclide are the same. A prominent example is
the decay of 226Ra (1,600 years) to 222Rn (3.825 days).
The gaseous radon is easily separated from its parent
nuclide. Once radon is prevented from escaping, radioac-
tive equilibrium will be established after about eight
half-lives of the daughter nuclide, i.e., approximately
1 month (compare Figure 3). In equilibrium 226Ra,
222
Rn, and all other short-lived daughter nuclides have
the same activities.
Transient radioactive equilibrium can be attained when
the half-life of the parent nuclide is larger than that of the
daughter, but the decay of the parent nuclide cannot be
neglected: t1/2(1) > t1/2(2). Once equilibrium is
established, the activity of the daughter nuclide is larger
than that of the parent nuclide (compare Figure 4). The
ratio of the activities is constant.
As the difference between the half-lives becomes
smaller, the time required to reach equilibrium increases
continuously. If the half-life of the parent is shorter than
that of the daughter, no equilibrium can be attained.
Radioactive equilibria have several practical applica-
tions. By measuring the activity of the daughter nuclides,
the activity of the parent nuclide can be determined, e.g.,
the amount of 238U in a sample can be determined by mea-
suring the activity of 234Th. Alternatively, if activities of
parent and daughter nuclides can be measured, it is possi-
ble to determine if an equilibrium state has been attained.
 RADIATION AND RADIOACTIVITY
Radiation and Radioactivity, Figure 3 Activities A (1) of nuclide
1, A (2) of nuclide 2, and sum of activities when the half-life of the
parent nuclide is much larger than the half-life of the daughter
nuclide. Activities are normalized to the initial activity of the
parent nuclide. The timescale is normalized to the half-life of the
daughter nuclide. Secular radioactive equilibrium, i.e., parent
and daughter have same activities, is achieved after a time
equivalent to approximately eight half-lives of the daughter.
Radiation and Radioactivity, Figure 4 Activities A (1) of nuclide
1, A (2) of nuclide 2, and sum of activities when the half-life of the
parent nuclide is five times the half-life of the daughter nuclide.
Activities are normalized to the initial activity of the parent
nuclide. The timescale is normalized to the half-life of the
daughter nuclide. Transient radioactive equilibrium is achieved
when the ratio of parent and daughter activities is constant.
663
A deviation from the equilibrium state, i.e.,
a disequilibrium, points toward transport of nuclides into
or out of the system. Nuclides of the decay series can be
separated by natural processes if parent and daughter have
different chemical behavior (e.g., uranium is soluble, tho-
rium is particle adhesive) or if they have different proba-
bilities of escape (the gas radon has a very high
probability to escape out of a system). Recoil due to
a-decay can lead to lower binding of daughters such as
is the case for 234U and 238U.
Interaction of radiation with matter
Ionizing radiation can be divided into three groups –
charged particles (heavy charged particles and electrons),
photons, and neutrons – each of which has different mech-
anisms of interacting with matter.
Heavy charged particles, i.e., protons and heavier
nuclei such as a-radiation, interact predominantly through
coulomb forces between their positive charge and the neg-
ative charge of the electrons in the material being tra-
versed. This leads to ionization and excitation of the
target atoms. Energies required for the ionization process
are on the order of several eV, i.e., the energy of visible
and ultraviolet light. Radioactive decays produce radiation
with energies of 0.1–10 MeV, so that a-particles produce
a track with large numbers of ions and electrons. The sec-
ondary electrons released in these ionization processes are
often energetic enough to cause further ionization and are
often referred to as d-radiation. Some of the charges can
become trapped instead of recombining with the ions
(see “Luminescence Dating”). The ionization density of
heavy charged particles is on the order of several thousand
ion pairs per mm travelled. It depends on the energy lost
per unit distance travelled by the ionizing particle,
which is also called linear energy transfer or LET. The
LET of heavy charged particles increases with distance
travelled and decreases sharply after reaching a peak
value, meaning a-particles interact more strongly as their
velocity decreases. Due to their large mass, they are not
easily deflected and their trajectory is usually a straight
line with a well-defined range. A 3 MeV a-particle has
a range of 1.7 cm in air and can barely penetrate a sheet
of paper.
b-Radiation consists of electrons or positrons and can
interact with matter in different ways including ionization
and excitation of electrons in the absorbing material. Elec-
trons can also interact with the electric field of positive
nuclei under emission of bremsstrahlung. If they move
through a substance with a velocity higher than the speed
of light in that substance, they emit Čerenkov radiation,
which is observed as deep blue light. This light is most
commonly seen in open, pool-type reactors. Positrons
usually annihilate with electrons in which process two
gamma photons of 511 keVare emitted. Opposed to a-par-
ticles, electrons are significantly deflected by collisions
with other electrons and therefore follow a zigzag path.
Their range is less clearly defined. 3 MeV electrons have
664 RADIATION AND RADIOACTIVITY
a range of approximately 20 m in air with an ionization
density of ~ 4 ion pairs per mm and can penetrate a few
millimeters of aluminum. Electrons with low energies
can also experience backscattering in high-Z absorbers
so that absorbers of low atomic number such as
plastic or aluminum are most effective in absorbing
b-radiation.
g- and X-rays both consist of photons and have similar
properties, but are distinguished by their origins: X-rays
are emitted by the electron shell of atoms when electrons
change from a higher- to a lower-energy orbit (characteris-
tic X-rays) or when electrons (or other particles) are
slowed down in the electric field of nuclei (bremsstrah-
lung). X-ray energies range from 100 eV to 100 keV.
g-Radiation is emitted during de-excitation of a nucleus,
and energies range from 10 keV to 10 GeV. While parti-
cles usually lose their energy in a succession of collisions,
photons mostly lose their energy in a single process. Pho-
tons interact in three different ways: In the photo effect the
g-photon transfers its complete energy to an electron
which is then emitted as a photoelectron. In the Compton
effect the photon gives off only part of its energy to an
electron and is then deflected into a different direction. If
its energy is larger than 1.02 MeV, the photon can be
transformed into an electron-positron pair (pair produc-
tion). The probability for each process depends on the
energy of the photon and the absorbing material. Due to
the probabilistic nature of the interactions, photons have
no clearly defined range. They are either removed from
a beam or not. Instead of giving a range in a specific mate-
rial their interaction is usually described by an absorption
coefficient or their reduction in intensity: 1 MeV
g-radiation is reduced to an intensity of 1 % by 5 cm of
lead or 25 cm concrete.
Neutrons are electrically neutral and interact only very
weakly with electrons. They lose energy through colli-
sions with nuclei. The neutron transfers part of its energy
to the nucleus and changes its direction. This process is
called scattering and leads to a slowing down of the neu-
trons. Scattering is most efficient in materials that contain
a high amount of hydrogen, such as paraffin. Neutrons can
also be absorbed by a nucleus (neutron capture). This pro-
cess can lead to the formation of a stable or unstable
nucleus and is used in neutron activation where radioac-
tive nuclides are created by bombardment with neutrons
(see “Neutron Activation Analysis”). Neutrons are classi-
fied by their energy. Thermal and slow neutrons (energies
>0.1 eVand 0.1–100 eV, respectively) are effectively cap-
tured by nuclei. Thermal neutron scattering is used for
material characterization. Fast neutrons (0.1–10 MeV)
typically lose their energy through a series of scattering
events which can lead to recoil or excitation of the target
nucleus. There is also a small probability for fast neutron
capture. Neutrons with energies >10 MeV can cause
spallation when colliding with a nucleus, in which process
the target nucleus emits several nucleons with the result of
leaving a lighter nucleus.
Measurement of radiation and radiation
dosimetry
Radiation is detected through its interaction with matter.
Different methods include the ionization of gases or semi-
conductors and excitation of scintillators.
1. In gas-filled detectors radiation creates a trail of ion
pairs. Ions and electrons are separated by applying an
electric field and give rise to a current. This process is
used in ionization chambers which are most effective
for a- and fission fragments, proportional counters
which allow quantitative measurement of a- and
b-radiation, or Geiger-Muller counters for high-energy
b- or X-rays (see “Beta Counter”).
2. Scintillation detectors consist of transparent solids or
liquids which emit photons when radiation is absorbed.
The emitted photons are subsequently measured.
Examples include NaI(Tl), CsI(Tl), and ZnS(Ag).
Scintillation counters are more efficient detectors for
g-radiation than Geiger-Muller counters and are also
suitable for a- and b-radiation.
3. Semiconductor detectors, such as high-purity Germa-
nium crystals, are widely used for g- and X-ray spec-
troscopy. Radiation creates charge carriers which are
separated by an electric field. They are able to produce
further electron hole pairs, creating an electric pulse.
The pulse height is proportional to the energy of the
g-photon.
Detectors of types 2 and 3 are frequently used for spec-
troscopy of a- and g-radiation to allow identification of
radionuclides (see “Alpha Spectroscopy”). Germanium
detectors and NaI(Tl) scintillators are commonly used
for gamma spectroscopy while alpha spectroscopy is car-
ried out with silicon surface barrier detectors. Due to the
continuous nature of b-particles beta spectroscopy is less
frequently applied. X-ray spectroscopy is carried out with
Si(Li) detectors.
Other detectors such as photographic film and lumines-
cence detectors are based on trapped electrons created by
the passage of radiation. These detectors are commonly
used for radiation dosimetry. Radiation dosimetry quanti-
tatively relates radiation exposure and resulting physical,
chemical, or biological changes. The primary physical
quantity used is the absorbed dose. It describes the energy
deposited by radiation per kg mass. Units used for dose are
the SI-unit gray (Gy) and the rad:
1 gray ¼ 1Gy ¼ 1 J=kg ¼ 100 rad
“Luminescence Dating” is based on measurement of
absorbed dose.
Types of nuclides
Radiation and radioactivity form the basis of many dating
methods for which applications different types of nuclides
and their origins have to be distinguished.
 RADIATION AND RADIOACTIVITY
Primordial nuclides have been present from early times
of the universe. During early stages of the universe, light
elements such as H, He, Li, and Be (number of nucleons
A  7) formed in primordial nucleosynthesis. After the
formation of stars, fusion processes in stellar nucleosyn-
thesis resulted in nuclides with A  60. Nuclides with
A > 60 formed through neutron capture. Some of these
primordial nuclides are radioactive. Radiogenic nuclides
can be stable or radioactive and result from the decay of
a parent nuclide.
The number of radioactive nuclides varies over time and
their variation can be described by the laws of radioactive
decay. While the concentration of the parent nuclide(s)
decreases, the concentration of stable decay products
increases. The decay constants are known (see “Radioac-
tive Decay Constants: A Review”). This forms the basis
of a variety of nuclear dating methods. Suitable timescales
depend on the half-lives. Examples for terrestrial parent-
daughter pairs used in dating are 40K/40Ar (see “Ar–Ar
and K–Ar Dating”), 87Rb/87Sr (see “Rb–Sr Dating”),
147
Sm/143Nd (see “Sm–Nd Dating”), 176Lu/176Hf (see
“Lu-Hf Dating: The Lu-Hf Isotope System”), and
187
Re/187Os (see “Rhenium–Osmium Dating
(Meteorites)”).
When using these pairs it is important to take into
account that daughter nuclides might already be present
at t ¼ 0. Usually a stable isotope of the daughter nuclide
is used to calibrate initial concentrations of the decay
product (see “Ar–Ar and K–Ar Dating”).
A similar approach is used for nuclides in natural decay
series. If parent and daughter nuclides show different
chemical behaviors, radioactive disequilibria can be used
to determine the date of separation. It is important to deter-
mine if a system is closed or open, i.e., to ensure that no
losses have occurred. See for comparison “210Pb Dating,”
“U-Series Dating,” “U–Th/He Dating,” and “Uranium–
Lead Dating.”
Nuclides in the uranium and thorium chains are also
used for “Fission Track Dating and Thermochronology”
and alpha-recoil track dating. Spontaneous or induced fis-
sion of heavy nuclei, mainly 238U, can leave tracks in
solids that can be made visible under a microscope by
treatment with chemicals. Similarly, recoiling of atoms
due to a-decay can leave a track. The track density
depends on the concentration of 238U and other nuclides
and the time elapsed.
Another large group of nuclides that are used for dating
are of cosmogenic origin (Lal, 2000; Gosse and Phillips,
2001; see “Terrestrial Cosmogenic Nuclide Dating”). Cos-
mogenic radionuclides are continuously formed by the
interaction of cosmic rays with the atmosphere and the
Earth’s surface. Cosmic radiation includes galactic cosmic
rays and solar energetic particles. The galactic cosmic ray
spectrum consists of approximately 85 % protons, 12 %
alpha particles, 1 % heavier nuclei, and about 2 % elec-
trons and positrons (Benton and Benton, 2001). The ener-
gies range from several 10 up to 1012 MeV with a broad
665
peak between 0.1 and 1 GeV. Solar energetic particles
include electrons, protons, and heavier ions, and ~ 50
events can be expected in the 11-year solar cycle (for
a detailed description Benton and Benton, 2001).
Primary cosmic rays interact with the atmosphere and
have been completely attenuated at 20 km height. They
cause spallation reactions and result in secondary protons,
neutrons, and muons. These secondary particles lead to
further nuclear reactions in the atmosphere and can even
reach the surface. In the process new nuclides are formed
which can be radioactive (3H, 14C, 10Be, 26Al, 32Si, 36Cl,
39
Ar, 41Ca, 81Kr) or stable (3He and 21Ne). Prerequisite
for dating based on cosmogenic nuclides is the knowledge
of production rates over the time span to be dated.
Summary
The decay of primordial and radiogenic nuclides and the
production of cosmogenic nuclides form the basis of vari-
ous dating methods. Radioactive decay can be described
by an exponential decay law. The radiation emitted in
the decay process can be used to characterize the nuclides
present in a system and to determine their concentrations.
Radiation is detected through the interaction processes
with the material it traverses.
Bibliography
Benton, E. R., and Benton, E. V., 2001. Space radiation dosimetry in
low-earth orbit and beyond. Nuclear Instruments and Methods in
Physics Research B, 184, 255–294.
Gosse, J. C., and Phillips, F. M., 2001. Terrestrial in situ cosmogenic
nuclides: theory and application. Quaternary Science Reviews,
20, 1475–1560.
Lal, D., 2000. Cosmogenic nuclide production rate systematics in
terrestrial materials: present knowledge, needs and future actions
for improvement. Nuclear Instruments and Methods in Physics
Research B, 172, 772–778.
General references
Ivanovich, M., and Harmon, R. S., 1992. Uranium-Series Disequi-
librium: Applications to Earth, Marine, and Environmental Sci-
ence. Oxford: Clarendon.
Knoll, G. F., 2010. Radiation Detection and Measurement.
New York: Wiley.
Krane, K. S., 1988. Introductory Nuclear Physics. New York:
Wiley.
Lieser, K. H., 2001. Nuclear and Radiochemistry. Weinheim:
Wiley-VCH.
Tsoulfanidis, N., and Landsberger, S., 2011. Measurement and
Detection of Radiation. Boca Raton: CRC Press.
Turner, J. E., 2007. Atoms, Radiation, and Radiation Protection.
Weinheim: Wiley-VCH.
Cross-references
Alpha Spectroscopy
Ar–Ar and K–Ar Dating
Beta Counter
Fission Track Dating and Thermochronology
Lu-Hf Dating: The Lu-Hf Isotope System
Luminescence Dating, Dose Rates
666 RADIOACTIVE DECAY CONSTANTS: A REVIEW

Neutron Activation Analysis

210
Pb Dating
Radiation Defect
Radiation Dose Rate
Radioactive Decay Constants: A Review
Radiocarbon Dating
Rb–Sr Dating
Rhenium–Osmium Dating (Meteorites)
Sm–Nd Dating
Terrestrial Cosmogenic Nuclide Dating
Uranium–Lead Dating
U-Series Dating
U–Th/He Dating
RADIOACTIVE DECAY CONSTANTS: A REVIEW
W. Jack Rink1 and Larry M. Heaman2
1
School of Geography and Earth Sciences, McMaster
University, Hamilton, ON, Canada
2
Department of Earth and Atmospheric Sciences,
University of Alberta, Edmonton, AB, Canada
Definition
The radioactive decay constant (l) is a characteristic of
unstable radionuclides (see chart of the nuclides) that
spontaneously decay at different rates to a more stable
atomic configuration; the larger the decay constant, the
more rapidly the parent radionuclide is depleted with time.
The basic formulation of the radioactive decay equation is
based on the principle that the number of decay events of
any radioactive parent atom is proportional to the number
of these atoms present at any given time (Rutherford and
Soddy, 1902). This basic principle has been transformed
into the law of radioactive decay N ¼ N0 elt, which states
that given an initial number N0 of radioactive parent nuclei
at start time t ¼ 0, the number of radioactive nuclei
remaining at time t is given by N. Related to the decay
constant is the half-life t½ of a radioactive nucleus: t½ ¼ ln
2/l. The half-life is the time in which half the number of
parent nuclides have decayed.
Introduction
The determination of decay constants and/or corresponding
half-lives of radionuclides is a fundamental aspect of geo-
chronologic methods. A wide range of scientific questions
can be addressed with dating techniques. A variety of ques-
tions can be addressed using dating methods. These range
from the age of archaeological materials and human fossils
to the age of the oldest materials in our solar system and
many geological processes on earth.
The process of radioactive decay proceeds through two
main mechanisms: (1) a-decay, which is alpha particle
emission (two protons plus two neutrons) from the
nucleus, and (2) b-decay, when a nucleus emits an electron
or positron. The accuracy and precision of the decay con-
stant measurements are a determinant in the accuracy
and precision of the age result. Progress in this area for
long-lived radioactive nuclides in part depends on the
difficulty in directly measuring the decay events and
the range in mass and energy of the different particles
emitted during decay. As better methods become
available to determine the decay constants for the various
methods, more accurate and precise values become
available.
Approaches to determining decay constants
There are essentially three different approaches used to
determine the decay constant of long-lived radionuclides:
(1) direct counting techniques, (2) age comparison of geo-
logical materials using multiple dating techniques, and
(3) the in-growth method. The direct counting technique
involves directly measuring the alpha, beta, or gamma
activity, commonly using liquid scintillation counting for
beta emitters, and dividing by the total number of

87
Radioactive Decay Constants: A Review, Table 1 Summary of Rb decay constant determinations

Reference Year T1/2(1010 a) l (1011 a1)

Age comparisons
Aldrich et al. 1956 5.0  0.2 1.39  0.06
Steiger and Jäger 1977 4.88  0.03 1.420  0.010
Minster et al. 1982 4.94  0.03 1.402  0.008
Shih et al. 1985 4.94  0.04 1.402  0.011
Amelin and Zaitsev 2002 1.396  0.006
Nebel et al. 2011 1.393  0.004
Absolute counting experiments
Flynn and Glendenin 1959 4.70  0.10 1.475  0.031
Neumann and Huster 1976 4.88  0.06 1.420  0.030
Kossert 2003 4.967  0.032 1.396  0.009
Ingrowth experiments
McMullen et al. 1966 4.72  0.04 1.468  0.012
Davis et al. 1977 4.89  0.04 1.419  0.012
Rotenberg et al. 2012 4.9624  0.0065 1.3968  0.0017

Modified after Begemann et al. (2001)

 RADIOACTIVE DECAY CONSTANTS: A REVIEW 667

Radioactive Decay Constants: A Review, Table 2 Radioactive decay constants and half-lives for selected radionuclides in current
use in geochronology

Nuclide Daughter Decay constant (a1) Half-life (a) Reference Note

Cosmogenic
10
Be 10
B (4.997  0.043)  1007 (1.387  0.012)  106 Korschinek et al., 2010
14
C 14
N (1.2448  0.0067)  1004 5568  30 Libby, 1952 a
(1.2097  0.0084)  1004 5730  40 Godwin, 1962 b
26
Al* 26
Mg (9.832  0.335)  1007 (7.05  0.24)  105 Norris et al., 1983
Nishiizumi, 2004
32
Si 32
P (4.813  0.368)  1003 144 +/ 11 Fifield and Morgenstern, 2009
36
Cl 36
Cl (2.3028  0.0153)  1006 (3.01  0.02)  105 NNDC, 2012 c
Naturally occurring
40
K 40
Ar (5.757  0.016)  1011 (l«) (1.2040  0.0034)  1010 Renne et al., 2011
40
K 40
Ca (4.955  0.013)  1010 (lb) (1.3989  0.0037)  1009 Renne et al., 2011
87
Rb 87
Sr (1.3968 + 0.0027/0.0018)  1011 (4.9624  0.0065)  1010 Rotenberg et al., 2012
147
Sm 143
Nd (6.478  0.054)  1012 (1.070  0.009)  1011 Kossert et al., 2009
176
Lu 176
Hf (1.904  0.018)  1011 (3.640  0.035)  1010 Kossert et al., 2013 d
187
Re 187
Os (1.6668  0.0034)  1011 (4.1586  0.0085)  1010 Selby et al., 2007
190
Pt 186
Os (1.414  0.012)  1012 (4.90  0.04)  1011 Cook et al., 2004 e
(1.78  0.14)  1012 (3.9  0.3)  1011 Tavares et al., 2006
232
Th 208
Pb (4.948  0.028)  1011 (1.401  0.008)  1010 LeRoux and Glendinin, 1963 f
235
U 207
Pb (0.98485  0.00067)  109 (7.0381  0.0048)  108 Jaffey et al., 1971 g
(0.98571  0.00012)  109 (7.03196  0.00086)  108 Mattinson, 2010
238
U 206
Pb (1.55125  0.00083)  1010 (4.4683  0.0024)  109 Jaffey et al., 1971 f
238
U (8.453  0.103)  1017 (8.2  0.1)  1015 Holden and Hoffman, 2000 h
U-series
210
Pb 210
Bi (3.122  0.031)  1002 22.20  0.22 Audi et al., 2003
230
Th 226
Ra (9.170  0.013)  1006 (75.584  0.110)  103 Cheng et al., 2013
231
Pa 227
Ac (2.116  0.007)  1005 (32.760  0.110)  103 Robert et al., 1969
234
U 230
Th (2.822  0.003)  1006 (245.620  0.260)  103 Cheng et al., 2013
Extinct (i)
244
Pu Fission (1.050  0.032)  1011 (6.6  0.2)  1010 Holden and Hoffman, 2000 h
146
Sm 142
Nd (1.019  0.105)  108 (68  7)  106 Kinoshita et al., 2012
(6.75  0.32)  109 (102.6  4.8)  106 Friedman et al., 1966
182
Hf 182
W (7.788  0.087)  108 (8.9  0.1)  106 Vockenhuber et al., 2004
41
Ca 41
K (0.102  0.007)  106 Audi et al., 2003
(6.3  1.7)  106 (0.11  0.03)  106 Brown et al., 1953
60
Fe 60
Ni (2.64  0.04)  107 (2.62  0.04)  106 Rugel et al., 2009
53
Mn 53
Cr (1.87  0.20)  107 (3.7  0.4)  106 Honda and Imamura, 1971

General: Bold values are the values reported in the indicated publications. The non-bold value was calculated from it, and its error was cal-
culated using the percentage error of the published value
*Extinct nuclide
a
Known as the “Libby half-life,” though less accurate, it is still used in conventional radiocarbon dating to calculate conventional radiocar-
bon years before present
b
The refined value resulting from three later independent measurements and agreed in conference as the most probable value
c
National Nuclear Data Center Chart of the Nuclides
d
See reference for discussion on the “l 176Lu dilemma”
e
Due to wide variance in recent determinations, two recent values are given
f
Value accepted by the IUGS-SOG, Steiger and Jäger, 1977
g
Value reported here is relative to the widely accepted 238U/235U ratio of 137.88. A recent study by Hiess et al. (2012) indicates a more accu-
rate estimate of the 238U/235U ratio in terrestrial samples is 137.818  0.045 (2s)
h
Spontaneous fission decay mode
i
A more complete list of extinct radionuclides used in meteorite research is reported by Dauphas and Chaussidon (2011)

radiogenic atoms present (e.g., Neumann and Huster,
1976; Kossert, 2003). The age comparison approach
involves selecting relatively simple geological materials
and determining their age using multiple dating tech-
niques where the decay constant for the parent isotope in
one dating system (e.g., 238U) is considered to be well
established and accurate (e.g., Amelin and Zaitsev,
2002). The in-growth method involves storing a quantity
of purified radioactive parent in a closed container, and
after a period of time (typically years to decades), the
quantity of daughter isotope is determined using isotope
dilution (e.g., Rotenberg et al., 2012).
668 RADIOACTIVE DECAY CONSTANTS: A REVIEW
Tabulations
The strengths and weaknesses of each approach to deter-
minations of decay constants or half-lives are nicely sum-
marized by Begemann et al. (2001), and the range of decay
constant values obtained for 87Rb using all three methods
are summarized in Table 1. An important aspect of these
decay constant determinations is that with improved
detection systems and more robust in-growth experiments,
there is convergence and general agreement between all
three methods for the value of the 87Rb decay constant.
Here, we report on the most recent determinations of
the values relevant to geochronology (Table 2). The values
are given as either the decay constant with uncertainty or
the half-life with uncertainty. The published reported
value is shown in bold; the other is calculated from
it. Emphasis here was given to reporting not only the
values but the research works that actually made the most
recent determinations alongside those which have
remained in current use for long periods of time. All of
the values reported in the tables are revisions of earlier
determinations with the exception of Libby, 1952. We note
that more than half of the values in the table have been
reported since 2000. Table 2 is for reference and not meant
as a tabulation of “recommended values” for use. Rather,
we aim to provide the most up-to-date compilation of the
values and let geochronologists themselves decide if the
revised values are to become routinely used in practice.
Bibliography
Amelin, Y., and Zaitsev, A. N., 2002. Precise geochronology of
phoscorites and carbonatites: the critical role of U-series disequi-
librium in age interpretations. Geochimica et Cosmochimica
Acta, 66, 2399–2419.
Audi, G., Bersillon, O., Blachot, J., and Wapstra, A. H., 2003. The
NUBASE evaluation of nuclear and decay properties. Nuclear
Physics A, 729, 3–128.
Begemann, F., Ludwig, K. R., Lugmair, G. W., Min, K., Nyquist,
L. E., Patchett, P. J., Renne, P. R., Shih, C.-Y., Villa, I. M., and
Walker, R. J., 2001. Call for an improved set of decay-constants
for geochronological use. Geochimica et Cosmochimica Acta,
65, 111–121.
Brown, F., Hanna, G. C., and Yaffe, L., 1953. The radioactive decay
of 41Ca. Proceedings of the Royal Society of London Series A,
220, 203–219.
Cheng, H., Edwards, R. L., Shen, C.-C., Polyak, V. J., Asmerom, Y.,
Woodhead, J., Hellstrom, J., Wang, Y., Kong, X., Spötl, C.,
Wang, X., and Alexander, E. C., Jr., 2013. Improvements in
230Th dating, 230Th and 234U half-life values, and U–Th isoto-
pic measurements by multi-collector inductively coupled plasma
mass spectrometry. Earth and Planetary Science Letters,
371–372, 82–91.
Cook, D. L., Walker, R. J., Horan, M. F., Wasson, J. T., and Morgan,
J. W., 2004. Pt-Re-Os systematics of group IIAB and IIIAB iron
meteorites. Geochimica et Cosmochimica Acta, 68, 1413–1431.
Dauphas, N., and Chaussidon, M., 2011. A perspective from extinct
radionuclides on a young stellar object: the sun and its accretion
disk. Annual Reviews of Earth and Planetary Science, 39,
351–386.
Fifield, L. K., and Morgenstern, U., 2009. Silicon-32 as a tool for
dating the recent past. Quaternary Geochronology, 4, 400–405.
Friedman, A. M., Milsted, J., Metta, D., Henderson, D., Lerner, J.,
Harkness, A. L., and Rok Op, D. J., 1966. Alpha decay
half lives of 148Gd 150Gd and 146Sm. Radiochimica Acta, 5,
192–194.
Godwin, H., 1962. Half-life of radiocarbon. Nature, 195, 984.
Hiess, J., Condon, D. J., McLean, N., and Noble, S. R., 2012.
238 235
U/ U systematics in terrestrial uranium-bearing minerals.
Science, 355, 1610–1614.
Holden, N. E., and Hoffman, D. C., 2000. Spontaneous fission half-
lives for ground state nuclides (Technical report). Pure and
Applied Chemistry, 72, 1525–1562.
Honda, M., and Imamura, M., 1971. Half-life of Mn53. Physical
Review C, 4, 1182–1188.
Jaffey, A. H., Flynn, K. F., Glendenin, L., Bentley, W. C., and Essling,
A. M., 1971. Precision measurement of half-lives and specific
activities of 235U and 238U. Physical Review C, 4, 1889–1906.
Kinoshita, N., Paul, M., Kashiv, Y., Collon, P., Deibel, C. M.,
DiGiovine, B., Greene, J. P., Henderson, D. J., Jiang, C. L.,
Marley, S. T., Nakanishi, T., Pardo, R. C., Rehm, K. E., Robert-
son, D., Scott, R., Schmitt, C., Tang, X. D., Vondrasek, R., and
Yokoyama, A., 2012. A shorter 146Sm half-life measured and
implications for 146Sm-142Nd chronology in the solar system.
Science, 335, 1614–1617.
Korschinek, G., Bergmaier, A., Faestermann, T., Gerstmann, U. C.,
Knie, K., Rugel, G., Wallner, A., Dillmann, I., Dollinger, G.,
Lierse von Gostomski, C., Kossert, K., Maiti, M., Poutivtsev,
M., and Remmert, A., 2010. A new value for the half-life of
10
Be by heavy-ion elastic recoil detection and liquid scintillation
counting. Nuclear Instruments and Methods in Physics Research
B, 268, 187–191.
Kossert, K., 2003. Half-life measurements of 87Rb by liquid scintil-
lation counting. Applied Radiation and Isotopes, 59, 377–382.
Kossert, K., Jörg, G., Nähle, O., and Lierse v. Gostomski, C., 2009.
High-precision measurement of the half-life of 147Sm. Applied
Radiation and Isotopes, 67, 1702–1706.
Kossert, K., Jörg, G., and Lierse v. Gostomski, C., 2013. Experi-
mental half-life determination of 176Lu. Applied Radiation and
Isotopes, 81, 140–145.
LeRoux, L. J., and Glendenin, L. F., 1963. Half-life of 232Th. In Pro-
ceedings of the National Meeting on Nuclear Energy, Pretoria,
South Africa, April 1963, pp. 83–94.
Libby, W. F., 1952. Radiocarbon Dating. Chicago: University of
Chicago Press.
Mattinson, J. M., 2010. Analysis of the relative decay constants of
235
U and 238U by multi-step CA_TIMS measurements of
closed-system natural zircon samples. Chemical Geology, 275,
186–198.
Minster, J.-F., Birck, J.-L., and Allègre, C. J., 1982. Absolute age of
formation of chondrites studied by the 87Rb-87Sr method.
Nature, 300, 414–419.
National Nuclear Data Center Chart of the Nuclides (revision June
1, 2012), Nuclear Wallet Data, Brookhaven National Laboratory.
http://www.nndc.bnl.gov/chart/
Neumann, W., and Huster, E., 1976. Discussion of the 87Rb half-life
determined by absolute counting. Earth and Planetary Science
Letters, 33, 277–288.
Nishiizumi, K., 2004. Preparation of 26Al AMS standards. Nuclear
Instruments and Methods in Physics Research B, 223–224,
388–392.
Norris, T. L., Gancarz, A. J., Rokop, D. J., and Thomas, K. W., 1983.
Half-life of 26Al. Journal of Geophysical Research, 88, B331–B333.
Renne, P. R., Balco, G., Ludwig, K. R., Mundil, R., and Min, K.,
2011. Response to the comment by W.H. Schwarz et al. on “Joint
determination of 40K decay constants and 40Ar*/40K for the fish
Canyon sanidine standard, and improved accuracy for 40Ar/39Ar
geochronology by P.R. Renne et al. (2010). Geochimica et
Cosmochimica Acta, 75, 5097–5100.
 RADIOCARBON DATING
Robert, J., Miranda, C. F., and Muxart, R., 1969. Mesure de la
période du protactinium-231 par microcalorimétrie.
Radiochimica Acta, 11, 104–108.
Rotenberg, E., Davis, D., Amelin, Y., Ghosh, S., and Bergquist,
B. A., 2012. Determination of the decay-constant of 87Rb by lab-
oratory accumulation of 87Sr. Geochimica et Cosmochimica
Acta, 85, 41–57.
Rugel, G., Faestermann, T., Knie, K., Korschinek, G., Poutivtsev,
M., Schumann, D., Kivel, N., Günther-Leopold, I., Weinreich,
R., and Wohlmuther, M., 2009. New Measurement of the 60Fe
Half-Life. Physical Review Letters, 103, 072502.
Rutherford, E., and Soddy, F., 1902. LXXXIV. —The radioactivity
of thorium compounds. II. The cause and nature of radioactivity.
Journal of the Chemical Society, Transactions, 81, 837–860.
Selby, D., Creaser, R. A., Stein, H. J., Markey, R. J., Hannah, J. L.,
2007. Assessment of the 187 Re decay constant by cross calibra-
tion of Re–Os molybdenite and U–Pb zircon chronometers in
magmatic ore systems. Geochimica et Cosmochimica Acta,
71(8), 1999–2013.
Steiger, R. H., and Jäger, E., 1977. Subcommission on geochronol-
ogy: convention on the use of decay constants in geo- and
cosmochronology. Earth and Planetary Science Letters, 36,
359–362.
Tavares, O. A. P., Terranova, M. L., and Medeiros, E. L., 2006. New
evaluation of alpha decay half life of 190Pt isotope for the Pt-Os
dating system. Nuclear Instruments and Methods in Physics
Research B, 243, 256–260.
Vockenhuber, C., Oberli, F., Bichler, M., Ahmad, I., Quitte, G.,
Meier, M., Halliday, A. N., Lee, D. C., Kutschera, W., Steier,
P., Gehrke, R. J., and Helmer, R. G., 2004. New half-life mea-
surement of 182Hf: improved chronometer for the early solar sys-
tem. Physical Review Letters, 93, 172501.
Cross-references
Radiation and Radioactivity
RADIOCARBON DATING
A. J. Timothy Jull1 and George S. Burr2,3
1
NSF-Arizona AMS Laboratory, Physics Building,
University of Arizona, Tucson, AZ, USA
2
NSF Arizona Accelerator Mass Spectrometry
Laboratory, Department of Geosciences and Physics,
University of Arizona, Tucson, AZ, USA
3
Department of Geosciences, National Taiwan University,
Taipei, Taiwan
Synonyms
14
C dating; Carbon dating; Carbon-14 dating; Radiocar-
bon dating
Definition
Radiocarbon dating is a method of measuring the age of
organic or carbonate phases in radiocarbon years by the
level of the nuclide 14C remaining in the sample. Calibra-
tion of the radiocarbon age is done by comparison to
known-age records such as tree rings. Stuiver and Polach
(1977) define a radiocarbon date to include five things:
(a) the use of the 5,568-year half-life (mean life 8,033
669
years); (b) the assumption of constancy of 14C atmo-
spheric level during the past; (c) direct or indirect compar-
ison to a known standard, oxalic acid; (d) isotopic
fractionation normalization of all sample activities to the
base of d13C ¼ 25 %; and (e) the year 1950 which is
automatically the base year, with ages given in years BP
(i.e., where present is defined as AD 1950).
Introduction
Radiocarbon (carbon-14 or 14C) is produced in the
upper atmosphere by the action of cosmic ray-induced
radiation on nitrogen in the atmosphere. The nuclear reac-
tion 14N(n, p)14C has a high cross section of 2.5 barns.
The production rate of 14C is estimated to average
~2.2 atoms/cm2/s (Usoskin and Kromer, 2005). Once pro-
duced, 14C is rapidly incorporated into the atmosphere as
CO and then oxidized to CO2. In this form it makes its
way into the terrestrial biosphere through photosynthesis,
through the carbon cycle. The time scale of this process is
rapid, so that 14C in the atmosphere and the biosphere are
in equilibrium on a time scale of <1 year (see Figure 1).
Radiocarbon dating is one of the most important isotopic
geochronological methods and has an immense range of
applications to scientific studies including archaeology,
geology, and oceanography. The development of radiocar-
bon dating changed the way that geologists and archaeol-
ogists look at chronology (Renfrew, 1979).
The radioactive decay equation gives the relationship
between the decay rate and the number of 14C atoms that
will decay with time according to the radioactive decay
equation (Rutherford and Soddy, 1902):
dn=dt ¼ lN ð1Þ
In this equation the decay rate (dn/dt) is directly propor-
tional to the number of radioactive atoms present (N). The
“decay constant” l defines the rate of decay. The decay con-
stant is related to the “half-life” t½ of a radionuclide as the
time required for ½ of the remaining atoms to decay. The
half-life is related to the decay constant by the relation:
ln2
t 1=2 ¼ ð2Þ
l
By convention, a radiocarbon date implies a half-life of
5,568 years, as originally proposed by Willard Libby
(1955). The useful limit of the technique is around
50,000 years in a typical radiocarbon laboratory. This
explains the great facility of the method which can be
used, for example, to date the most recent global glacial-
interglacial advances and retreats, the development of
modern humans, and the rise of civilizations. Hence, it
covers a time scale of fundamental importance to geology
and archaeology. It should be noted that Libby’s half-life
determination has since been revised to 5,730 years, but
radiocarbon dates are still calculated with the 5,568-year
value by convention. This allows for unambiguous com-
parisons between radiocarbon dates, regardless of when
the measurements were made. This difference introduces
670 RADIOCARBON DATING

PRODUCTION OF 14 C
COSMIC
P RAY PROTON
R
O
D 14N NUCLEUS
U
C (σ ≈ 2 barns)
T THERMAL
I SPALLATION NEUTRON
O PRODUCTS 14C

N
OXIDATION

14CO
2

D
PHOTOSYNTHESIS
I
S
T
R
I
B
U
T
I
O DISSOLVED CO2
N CABONATES
BICARBONATES
Equilibrium 14C
Concentration: 12C
≈ 10−12
D
E
C Then: 14C → 14N + e− + υ
A τ1/2 = 5700 years
Y
One Gram → ~ 10 counts/minute

Radiocarbon Dating, Figure 1 A diagram of radiocarbon dating.

a 2 % offset which is compensated for when the dates are
calibrated to calendar years (see below).
We can integrate Eq. 1 with appropriate limits to obtain
the equation:
N and now, most measurements are made by AMS (Burr
¼ elt ð3Þ
N0
Where N0 is the number of atoms at time t ¼ 0 and N is the
number of atoms at any time t. This is the common expres-
sion of how the remaining amount of carbon-14 relates to
time. It is equally valid to use the activity (or the decay
rate), since from Eq. 1, these would obey the same
equation.
Originally, radiocarbon measurements were performed
by counting the radioactive decay of the 14C isotope
(Cook and van der Plicht, 2007). We can count either the
decays of the 14C, because it emits a b-particle, or we
can count the atoms themselves. In recent decades, the
method of accelerator mass spectrometry (AMS) has
gradually taken over the measurement of radiocarbon,
and Jull, 2009). The older counter methodology is still
used and can be applied to counting of both liquid samples
and gas, especially when sufficient material is available
(Longin, 1971; Cook and van der Plicht, 2007).
If we look at Eq. 1, it should be clear that it is more
effective to measure atoms than decays: for modern 14C,
the decay rate (A0) is 13.59 decays per minute per gram
of C. In AMS, we typically measure a sample of
0.5 mg C, and we can count over 120,000 atoms from a
modern sample. If we take a counting rate of 120,000
 RADIOCARBON DATING
atoms in 1,080 s, or 111 counts/s – from 0.5 mg of car-
bon – then this translates to 111  60/
0.0005 ¼ 1.33  107 counts per minute per gram.
A basic assumption of radiocarbon dating is that we
know the 14C age of the starting material at the time when
the sample was formed. This makes it possible to solve the
radioactive decay equation (3), which can be expressed as
a ratio of the number of atoms, the isotopic ratio relative to
a stable isotope, or the activity (decay rate) ratio. We can
define the “fraction of modern carbon,” F:
14
.
C 13
Cs½25
F ð25Þ ¼ 14 
ð4Þ
C 13 C
1950½25
where the subscript S [25] refers to the sample
renormalized to a d13C value of 25% and 1950 refers
to the standards discussed above, renormalized to 1950
AD, and d13C ¼ 19% for OXI and 25% for OXII, or
other secondary standards referenced to the oxalic acid
standard (Stuiver and Polach, 1977). An equivalent
expression could be written using atomic ratios (N/N0) or
activity ratios (decay rate, A/A0) for the sample and stan-
dard instead of the isotope ratio 14C/13C, in which case
N0 is the number of atoms at time zero, N is the number
of atoms remaining, t is the time in years, and l is the
decay constant. We could also substitute the isotopic ratio
14 12
C/ C, also equivalent to activity (A), referenced to a
zero-aged value, 14C/12C0, or A0. All these values would
be normalized to d13C at 25%. As noted above, the
decay constant l used here assumes a half-life of 5,568
years.
To calculate an age, the decay of 14C is written in terms
of F as follows:
F ð25Þ ¼ elt ð5Þ
and rearranging this equation gives an expression for cal-
culating the uncalibrated radiocarbon age of a sample:
age ¼ 8033lnF ð25Þ ð6Þ
This age is an uncalibrated result, since it assumes that
radiocarbon decays from a constant value. This is usually
called the “radiocarbon age.” We know that, due to fluctu-
ations in atmospheric production of 14C, we need to cor-
rect for the difference from the “zero” age value. In order
to do this, we need a suite of materials of known age and
their corresponding apparent “radiocarbon ages.” This
process is called “calibration” and is an important step
for precision dating using carbon-14.
Calibration
One of the potential problems of radiocarbon dating is that
it is assumed that the activity (A0) or number of atoms (N0)
at the time of production of the sample is always the same.
This is not the case. We know that 14C in known-age tree
rings fluctuates with time due to several factors, including
the level of CO2 in the atmosphere and cosmic-ray
671
fluctuations due to changes in the Earth’s magnetic field
and other effects (Burr, 2007). To correct for these prob-
lems, an international calibration team has made a cross-
calibration of 14C against other records – predominantly
dendrochronology, U–Th dating of corals, U–Th-dated
speleothems, and varved-sediment records. These calibra-
tions are published and revised every 4–5 years. The most
recent one is INTCAL13 (Reimer et al., 2013), which
refines the curves presented previously derived by
INTCAL09 (see Figure 2).
Fossil-fuel and bomb carbon-14
Another complication is that fossil fuel burning added
14
C-free carbon dioxide to the atmosphere resulting in
decreased levels of 14C from about 1780 to 1950
AD. Since 1950 AD, we have experienced an increase
due to the addition of high levels of 14C from atmospheric
nuclear weapons testing. Although most testing was
banned after 1963, this is an important signal added to
the atmosphere in the last 60 years. The bomb-carbon
record is well established and can be used to pinpoint spe-
cific dates in the time period since 1950 AD (Burr, 2013).
Indeed, this signal increases to a maximum of almost
twice the pre-bomb level and has since decreased. Various
plots of the bomb-curve profile are shown in the bomb-
carbon section of this encyclopedia. It is very useful in
applications where we need to resolve if a material was
produced in the last 60 years – for example, in determin-
ing forgeries in the artwork or in forensic applications.
Using radiocarbon dating
We can use radiocarbon dating to estimate the age of a
wide variety of carbon-containing materials. Both organic
and inorganic materials at the Earth’s surface and in the
oceans form in equilibrium or in steady-state conditions
with the atmospheric 14C. This allows us to determine
the initial level of 14C and their ages. When an animal or
plant dies, the level of 14C decays according to Eq. 3.
There are a vast number of applications of 14C, and more
details of 14C dating are given in other sections of this
encyclopedia.
Sample preparation
Before measurement of 14C, it is important to clean the
sample, so that any 14C that we measure comes from the
sample itself and not from contaminants introduced from
surrounding environment where the material was found.
After we clean the material, we need to convert it to a form
suitable to the measurement technique employed: this
could be CO2 gas or benzene for liquid-scintillation
counting or solid graphite or CO2 gas for AMS.
Pretreatment protocols
The most basic of all pretreatments is the acid-base-acid
(ABA) or acid-alkali-acid (AAA) cleaning method for
treating a wide range of organic materials, which can
include charcoal, wood, cellulose, plant material, and
672 RADIOCARBON DATING
Radiocarbon Dating, Figure 2 The calibration curve for the period 10,000–26,000 years (Reimer et al., 2013). The horizontal axis
shows the true age based on various records, and the vertical axis displays the radiocarbon age.
animal remains. After a physical inspection, samples are
cleaned with the addition of a sequence of steps, including
(1) HCl acid with a sufficient concentration to dissolve
carbonates, usually 1 N HCl, (2) 0.1 % NaOH, and
(3) HCl acid. Following each step, the samples are neutral-
ized with distilled water. Following the last step, they are
dried. The dried sample is then combusted at
900–1,100  C with CuO or oxygen gas. There are many
refinements of this method. For example, Hatté
et al. (2001) identified potential contamination in the
ABA pretreatment which can affect wood and paleosol
dates and suggested improvements to avoid them.
Although charcoals are generally excellent material for
dating, Bird et al. (1999) recommended the use of a more
severe oxidation step using chromic acid for purification
of some old charcoals (Bird et al., 1999). Rebollo
et al. (2008) looked at the chemical and structural changes
in charcoals pretreated with the ABA method, and they
suggested several modifications to the ABA method based
on their findings, including the use of weaker acids, elim-
inating the initial acid step, treatment with distilled water,
and agitation in the alkali step. In another case, Ascough
et al. (2011) reported that for some charcoals the alkali-
soluble fraction normally removed from the charcoals
may be degradation products of the charcoal itself, so
removing them does not necessarily improve the cleanli-
ness of the sample.
Other types of samples need other specialized treatment.
Carbonate samples are etched using 100 % H3PO4 to remove
50–85 % of the carbonate, then dried and acidified with
H3PO4, and the carbon is collected as CO2. Sediments are
treated in a number of ways. One approach is to use selective
combustion. After cleaning and drying the sample, the sam-
ple is combusted stepwise, at successively higher tempera-
tures to liberate different organic components according to
their combustion temperatures (McGeehin et al., 2001).
Radiocarbon dating of works of art and other artifacts is
an important application. Usually, textiles, parchment,
canvas, and artworks (e.g., canvas from a painting) are
cleaned in a Soxhlet apparatus, which utilizes hot solvents
to remove contaminants. Typically, a series of solvents
such as hexane, ethanol, methanol, and acetone are used
(Jull et al., 2004; Hajdas, 2008, 2009; Brock, 2013). The
samples can then be cleaned with distilled water and dried,
before combustion at up to 1,100  C. Iron artifacts can be
dated either by melting the sample in an oxygen atmo-
sphere and extracting CO2 or by using wet-chemical tech-
niques (Park et al., 2010).
There are some excellent examples that illustrate the
use of radiocarbon dating for artworks. The most famous
is the Shroud of Turin (Damon et al., 1989), but the list
also includes the Dead Sea Scrolls and a more recent
example is the Voynich manuscript (Zyats et al., 2012)
and the Egyptian Book of the Dead (Figure 3).
 RADIOCARBON DATING
Radiocarbon Dating, Figure 3 An example of radiocarbon dating – Egyptian Book of the Dead. This document dates to 1620–1430
BC based on the calibrated age, or 3,249  49 radiocarbon years before present (Image courtesy Brooklyn Museum of Art).
There are many specialized approaches to dating
bones. Much attention is given to bone pretreatment
because of their critical role in archaeological studies
and the complexities of bone chemistry. The standard
pretreatment method for bone is the collagen extraction
technique of Longin (1971). Collagen is a fibrous pro-
tein and the principal constituent of bone. This material
degrades with time as a result of postdepositional alter-
ation and eventually breaks down into individual amino
acids. Bone collagen may interact with exogenous
organics through reactions in the burial environment.
As a result, a number of screening procedures have
been developed to characterize the degree of preserva-
tion for individual samples (van Klinken, 1999). The
most common screening procedure for AMS is the
determination of percent collagen in a sample, usually
by measuring the elemental C/N ratio. Modern bone con-
sists of ~22 wt% collagen, and this value decreases with
burial age. Samples with as low as 0.5 % collagen may be
datable, but often require specialized chemical
pretreatment, such as the isolation and analysis of individ-
ual amino acids (Van Klinken, 1999). Samples with
<0.5 % collagen are generally considered unsuitable for
obtaining a useful date. Another refinement of the collagen
extraction technique has become standard in the past
decade, using ultrafiltration to isolate the >30 kD molecu-
lar weight fraction. The method is highly selective and has
673
been shown to remove contaminants better than the stan-
dard collagen extraction technique (Brown et al., 1988).
Brock (2013) has also applied the collagen approach to
parchment dating.
Groundwater and surface waters can be dated by
extracting dissolved inorganic carbon by acidification, in
the same manner as solid carbonates are processed. Addi-
tionally dissolved organic carbon can be extracted using
either KMnO4 or K2Cr2O7 as an oxidizing agent
(Leonard et al., 2013). We can use these values of 14C to
better understand the circulation of groundwater and car-
bon cycling in lakes (e.g., Yu et al., 2007) and determine
reservoir ages.
These are just a few examples of pretreatment methods
for typical samples. Specialized techniques that are
beyond the scope of this discussion have been developed
to pretreat a wide range of samples (see, e.g., Hajdas,
2009). The goal of the pretreatment in these cases is to iso-
late a carbon fraction that was in equilibrium with the
atmosphere when it formed and has not been contami-
nated. For AMS, once the carbon has been extracted, it
can be converted to graphite (a common target material
in AMS machines) by means of a catalytic decomposition
reaction, using Fe as the catalyst. For liquid-scintillation
spectrometry, the CO2 must be converted to benzene,
which involves several additional chemistry steps (Cook
and van der Plicht, 2007).
674 RADIOCARBON DATING
Radiocarbon Dating, Figure 4 (a, b) The result of a calibration of a radiocarbon date using the Oxcal calibration program (c14.ox.ac.
uk/oxcal) for two values of slightly different precision.
Measurement and calculations
AMS measurements are made by comparing the count rate
of 14C from a sample to the count rate in a known-age stan-
dard. For activity (counting) measurements, the normal-
ized activity (A) is compared to that of a known
standard. In both cases, we can then compare the mea-
sured ratio of the sample to a known standard.
Standards
For 14C, we use a primary (SRM-4990B, OXI) and sec-
ondary (SRM-4990C, OXII) standard from the US
National Institute of Standards and Technology (NIST)
to determine radiocarbon dates. The secondary standard
is normalized to the consensus value of OXI. The ratio
of OXII/OXI is well defined (Mann, 1983). Many other
well-established secondary standards are also available,
which can be referred back to the primary OXI value
(Scott et al., 2007). Some laboratories use their own inter-
nal standards and reference them to the primary oxalic
acid standards.
Uncertainties
Burr et al. (2007) have summarized how the uncertainties
in radiocarbon dates are determined at the Arizona labora-
tory. There are several potential problems when consider-
ing radiocarbon dating uncertainties. For example,
radiocarbon dates are quoted with one standard deviation
uncertainties by convention, whereas calibrated radiocar-
bon dates are normally quoted at two standard deviations.
Some laboratories estimate additional errors above
counting errors from their laboratory procedures (these
are always present), but others do not. Often radiocarbon
age uncertainties are published, especially by nonspecial-
ists, with little discussion of how they were determined,
and uncertainties in reservoir corrections are often
neglected. It is therefore important to quote the original
“radiocarbon ages” because they are unambiguous assum-
ing they adhere to the definition given above.
Using calibration curves
There are a number of programs available for calibration of
the “radiocarbon age” to a calendar age. Some good exam-
ples easily accessible on the Internet are Calib (www.calib.
org) and OxCal (c14.arch.ox.ac.uk/oxcal). The result of
applying the calibration to a radiocarbon result can be
shown in Figure 4a, b, where we see the difference between
a radiocarbon date of 2,320  20 and 2,320  30 years. The
difference in the precision of the calibration age is due to the
structure of the calibration curve.
D14C notation
For certain samples, the initial value for modern (1950 AD)
will not be the same as the atmosphere. We know that there
are offsets in marine samples (e.g., Dumond and Griffin,
2002), and there can also be offsets in other samples such
as lakes (Yu et al., 2007), where the radiocarbon content of
the carbon in the lake may not be “modern” (i.e., at the level
of A0) due to storage or input of old carbon from other
sources. Age offsets of this type are described in terms of
the difference between the radiocarbon content of the reser-
voir and contemporary atmosphere. This is the age-corrected
D14C value. For sample prior to 1950 AD, D14C is defined as


D14 C ¼ 1, 000 FelðaÞ  1 % ð7Þ
where a is the absolute age of the material in years before
1950 AD (often stated as calendar years BP in radiocarbon
literature), which should be independently obtained. An
example of this use is Miyake et al. (2012).
A different situation is the case of a sample containing
14
C from after 1950 AD, such as a sample containing
bomb 14C. In this case, the true D14C value should be
corrected for the decay of the standard, as the standard
has some age relative to the material being studied. This
increases F to the correct value.


D14 C ¼ 1, 000 Felða1950ADÞ  1 % ð8Þ
In Eqs. 5 and 6, the l used is 1.21  104, representing the
true half-life (5,730 years) of 14C.
 RADIOCARBON DATING
Marine reservoir effects
Stuiver et al. (1986) defined the “marine reservoir effect”
as the difference between the measured radiocarbon age
and the atmospheric age determined independently at a
known time (t):
Rðt Þ ¼ 14 C age of marine sample
 14 C age of atmospheric sample ð9Þ
This correction factor can be determined directly from
coexisting marine and terrestrially derived plant material
at the same location. The value of R can be related to
D14C by the equation (Burr and Jull, 2009):
! Patagonia (Argentina). Radiocarbon, 45, 9–15.
1, 000 þ 1
14
D C atm
R ¼ 8033ln D C ð10Þ
1, 000 þ 1
14
mar
Also, Stuiver et al. (1986) introduced the concept of DR. It
is important to understand the definition of DR, to quote
from Stuiver et al. (1986), who defined a graphical model
for this calculation, “The difference DR in reservoir age of
the regional part of the ocean from which the users sample
is derived, and the reservoir age of our model ocean, is
determined through the use of figure 10A.”
Stuiver et al. (1986) paper defines the oceanic reservoir
age value DR as
DR ¼ Rmeasured  Rexp ð11Þ
where Rmeasured is the R actually measured in the ocean
water and Rexp is the expected value derived from the
model of Stuiver and Braziunas (1993). It is important to
be aware that this is a model-dependent expression.
Because this is not always easily established, many
authors report Rexp as the average oceanic value of
R which is 400 years (e.g., Cordero et al., 2003). So, the
definition of DR (as opposed to R) is dependent on the
oceanographic model used. This marine model is now
incorporated into the calibration software (e.g.,
MARINE09, Reimer et al., 2009).
Bibliography
Ascough, P. L., Bird, M. I., Francis, S. M., and Lebl, T., 2011. Alkali
extraction of archaeological and geological charcoal: evidence
for diagenetic degradation and formation of humic acids. Jour-
nal of Archaeological Science, 38, 69–78.
Bird, M. I., Ayliffe, L. K., Fifield, L. K., Turney, C. S. M.,
Cresswell, R. G., Barrows, T. T., and David, B., 1999. Radiocar-
bon dating of “old” charcoal using a wet oxidation, stepped-
combustion procedure. Radiocarbon, 41(2), 127–140.
Brock, F., 2013. Radiocarbon dating of historical parchments.
Radiocarbon, 55, 353–363.
Brown, T. A., Nelson, D. E., Vogel, S. J., and Southon, J. R., 1988.
Improved collagen extraction by modified Longin method.
Radiocarbon, 30, 171–178.
Burr, G. S., 2007. Causes of temporal 14C variations. In Elias, S. A.
(ed.), Encyclopedia of Quaternary Science. Oxford: Elsevier,
pp. 2931–2940.
Burr, G. S., 2013. Bomb carbon. This volume.
675
Burr, G. S., and Jull, A. J. T., 2009. Accelerator mass spectrometry
for radiocarbon research. In Gross, M. L., and Caprioli, R. (eds.),
Encyclopedia of Mass Spectrometry. Amsterdam: Elsevier,
Vol. 5, pp. 656–669.
Burr, G. S., and Jull, A. J. T., 2013. Accelerator mass spectrometry.
This volume.
Burr, G. S., Donahue, D. J., Tang, Y., Beck, J. W., McHargue, L.,
Biddulph, D., Cruz, R., and Jull, A. J. T., 2007. Error analysis
at the NSF-Arizona AMS facility. Nuclear Instruments and
Methods in Physics Research B, 259, 149–153.
Cook, G. T., and van der Plicht, J., 2007. Radiocarbon dating: con-
ventional method. In Elias, S. (ed.), Encyclopedia of Quaternary
Science. Amsterdam: Elsevier, pp. 2899–2911.
Cordero, R. R., Panarello, H., Lanzelotti, S., and Favier Dubois,
C. M., 2003. Radiocarbon age offsets between living organisms
from the marine and continental reservoir in coastal localities of
Damon, P. E., Donahue, D. J., Gore, B. H., Hatheway, A., Jull,
A. J. T., Linick, T. W., Sercel, P. J., Toolin, L. J., Bronk, C. R.,
Hall, E. T., Hedges, R. E. M., Housley, R., Law, I. A., Perry,
C., Bonani, G., Trumbore, S., Wölfli, W., Bowman, S. G. E.,
Leese, M. N., and Tite, M. S., 1989. Radiocarbon dating of the
Shroud of Turin. Nature, 337, 611–615.
Dumond, D. E., and Griffin, D. G., 2002. Measurements of the
marine reservoir effect on radiocarbon ages in the eastern Bering
Sea. Arctic, 55, 77–86.
Hajdas, I., 2008. Radiocarbon dating and its application in quater-
nary studies. Quaternary Science Journal, 57(1–2), 2024.
Hajdas, I., 2009. Applications of radiocarbon dating method. Radio-
carbon, 51, 79–90.
Hatté, C., Morvan, J., Noury, C., and Paterne, M., 2001. Is classical
Acid-Alkali-Acid treatment responsible for contamination? An
alternative proposition. Radiocarbon, 43(2A), 177–182.
Hatté, C., and Jull A. J. T., 2013. Radiocarbon dating: Plant macrofos-
sils. In: Elias, S., and Mock, C. (eds.), Encyclopedia of Quaternary
Science, Second Edition. Elsevier: Amsterdam. pp. 361–367.
Jull, A. J. T., Donahue, D. J., Broshi, M., and Tov, E., 1995. Radio-
carbon dating of scrolls and linen fragments from the Judean
Desert. Radiocarbon, 37(1), 11–19.
Jull, A. J. T., Burr, G. S., McHargue, L. R., Lange, T. E., Lifton,
N. A., Beck, J. W., and Donahue, D. J., 2004. New frontiers in
dating of geological, paleoclimatic and anthropological applica-
tions using accelerator mass spectrometric measurements of 14C
and 10Be in diverse samples. Global and Planetary Change, 41,
309–323.
Jull, A. J. T., Burr, G. S., Zhou, W., Cheng, P., Song, Y., Leonard,
A. G., Cheng, L., and An, Z. S., 2014. Radiocarbon measure-
ments of dissolved inorganic and organic carbon in Lake Qing-
hai and inflowing rivers (NE Tibet-Qinghai Plateau), China.
Radiocarbon 56(3): 1113–1127.
Leonard, A., Castle, S., Burr, G. S., Lange, T., and Thomas, J., 2013.
A wet oxidation method for AMS radiocarbon analysis of
dissolved organic carbon in water. Radiocarbon, 55, 545–552.
Libby, W. F., 1955. Radiocarbon Dating, 2dth edn. Chicago:
University of Chicago Press.
Longin, R., 1971. New method of collagen extraction for radiocar-
bon dating. Nature, 230, 241–242.
Mann, W. B., 1983. An international reference material for radiocar-
bon dating. Radiocarbon, 25, 519–527.
McGeehin, J., Burr, G. S., Jull, A. J. T., Reines, D., Gosse, J., Davis,
P. T., Muhs, D., and Southon, J., 2001. Stepped-combustion 14C
dating of sediment. Radiocarbon, 43(2A), 255–262.
Miyake, F., Nagaya, K., Masuda, K., and Nakamura, T., 2012. A
signature of cosmic-ray increase in AD 774-775 from tree rings
in Japan. Nature, 486, 240–242.
Park, J. S., Burr, G. S., and Jull, A. J. T., 2010. A thermal and acid
treatment for carbon extraction from cast iron and its application
676 RADIOCARBON DATING OF MARINE CARBONATES
to AMS dating of cast iron objects from ancient Korea. Radio-
carbon, 52(3), 1312–1321.
Rebollo, N. T., Cohen-Ofri, I., Popovitz-Biro, R., Bar-Yosef, O.,
Meignen, L., Goldberg, P., Weiner, S., and Boaretto, E., 2008.
Structural characterization of charcoal exposed to high and low
pH: implications for 14C sample preparation and charcoal preser-
vation. Radiocarbon, 50(2), 289–307.
Reimer, P. J., Baillie, M. G. L., Bard, E., Bayliss, A., Beck, J. W.,
Blackwell, P. G., Bronk Ramsey, C., Buck, C. E., Burr, G. S.,
Edwards, R. L., Friedrich, M., Grootes, P. M., Guilderson,
T. P., Hajdas, I., Heaton, T. J., Hogg, A. G., Hughen, K. A., Kai-
ser, K. F., Kromer, B., McCormac, F. G., Manning, S. W.,
Reimer, R. W., Richards, D. A., Southon, J. R., Talamo, S.,
Turney, C. S. M., van der Plicht, J., and Weyhenmeyer, C. E.,
2009. INTCAL09 and MARINE09 radiocarbon age
calibration curves, 0–50,000 cal BP. Radiocarbon, 51(4),
1111–1150.
Reimer, P. J., Bard, E., Bayliss, A., Beck, J. W., Blackwell, P. G.,
Bronk Ramsey, C., Buck, C. E., Cheng, H., Edwards, R. L., Frie-
drich, M., Grootes, P.M., Guilderson, T. P., Haflidason, H.,
Hajdas, I., Hatté, C. Heaton, T. J., Hoffmann, D. L., Hogg,
A. G., Hughen, K. A., Kaiser, K. F., Kromer, B., Manning,
S. W., Niu, M., Reimer, R. W., Richards, D. A., Scott, E. M.,
Southon, J. R., Staff, R. A., Turney, C. S. M., and van der Plicht,
J., 2013. INTCAL13 and MARINE13 radiocarbon age calibra-
tion curves 0–50,000 years cal BP. Radiocarbon 55(4): 1869–
1887. doi:10.2458/azu_js_rc.55.16947
Renfrew, C., 1979. Before Civilization: The Radiocarbon Revolu-
tion and Prehistoric Europe. Cambridge: Cambridge University
Press.
Rutherford, E., and Soddy, F., 1902. The causes and nature of radio-
activity, part I. Philosophical Magazine, 6(4), 370–396.
Scott, E. M., Cook, G. T., Naysmith, P., Bryant, C., and O’Donnell,
D., 2007. A report on phase 1 of the 5th international radiocar-
bon intercomparison (VIRI). Radiocarbon, 49, 409–426.
Scott, E. M., Cook, G. T., and Naysmith, P., 2010. A report on phase
2 of the fifth international radiocarbon intercomparison. Radio-
carbon, 52, 846–858.
Stuiver, M., and Braziunas, T. F., 1993. Modeling atmospheric 14C
influences and 14C ages of marine samples to 10,000
BC. Radiocarbon, 35(1), 137–189.
Stuiver, M., and Polach, H. A., 1977. Reporting of 14C data. Radio-
carbon, 19(3), 355–363.
Stuiver, M., Pearson, G. W., and Braziunas, T., 1986. Radiocarbon
age calibration of marine samples back to 9000 cal yr
BP. Radiocarbon, 28(2B), 980–1021.
Usoskin, I. G., and Kromer, B., 2005. Reconstruction of the 14C pro-
duction rate from measured relative abundance. Radiocarbon,
47, 31–37.
van Klinken, G. J., 1999. Bone collagen quality indicators for
palaeodietary and radiocarbon measurements. Journal of
Archaeological Science, 26, 687–695.
Yu, S. Y., Shen, J., and Colman, S. M., 2007. Modeling the radiocar-
bon reservoir effect in lacustrine systems. Radiocarbon, 49,
1241–1254.
Zyats, P., Barabe, J., and Hodgins, G. W. L., 2012. The Mysterious
Voynich Manuscript: Collaboration yields new insights. Oral
Presentation American Institutes for Conservation 40th Annual
Meeting, Albuquerque, NM, May, 2012.
Cross-references
Accelerator Mass Spectrometry
Bomb Carbon
14
C in Plant Macrofossils
Dendrochronology, Progress
U–Th/He Dating
RADIOCARBON DATING OF MARINE CARBONATES
Quan Hua
Australian Nuclear Science and Technology Organisation,
Kirrawee DC, NSW, Australia
Definition
Contaminant: Carbon-containing materials that do not
originally belong to the sample.
Pretreatment: A process applying physical and chemical
treatments to remove contaminants.
Contamination and sample pretreatment
Marine carbonates such as shells and corals mainly consist
of aragonite. Before samples are processed for radiocar-
bon dating, all contaminants must be removed; otherwise
the determination of correct radiocarbon ages may not be
achieved. Contaminants are derived from the surrounding
environment if samples were buried in soils or sediments
(e.g., secondary carbonates derived from groundwater
and recrystallization of sample carbonate due to chemical
exchange between the sample and the surrounding envi-
ronment). These carbonate contaminants are mostly in
the form of calcite. Contaminants can also be conservation
materials in the case of museum specimens. Contamina-
tion cannot always be seen by naked eye. In such cases,
samples should be screened for secondary carbonates
and recrystallization and diagenetic alteration using
X-ray diffraction, thin-section examination (McGregor
and Gagan, 2003), or scanning electron microscopy
(Webb et al., 2007; Nothdurft and Webb, 2009).
If the outer layers of carbonate samples are contami-
nated by secondary carbonates, the outer surface of the
samples is usually cleaned physically by abrasion or
sandblasting and/or chemically by acid etching. Routine
pretreatment of shell samples involves removal of the
outer 20 % (by weight) of the samples by acid etching
(Walker, 2005). If contamination appears not only in the
outer layers but also in the inner layers of carbonate sam-
ples (e.g., recrystallization and diagenetic alteration), acid
etching may not be useful. This pretreatment method may
result in an increase of the proportion of diagenetic calcite
in the samples because aragonite is more soluble than cal-
cite in acid media (Douka et al., 2010). This means poorly
preserved carbonate samples may not be adequate for 14C
dating. Recently, Douka et al. (2010) and Russo et al.
(2010) have tried to separate original aragonite from dia-
genetic calcite of poorly preserved fossil shells and corals
by density separation using lithium heteropolytungstate
and bromoform, respectively. Their initial test results were
very encouraging.
Pretreated samples free of contamination are converted
to CO2 by reaction with acid solutions (dilute HCl or 85 %
H3PO4). The sample CO2 is usually converted to benzene
or graphite for radiocarbon analysis using radiometric and
accelerator mass spectrometry (AMS) method, respec-
tively (Walker, 2005; Hua, 2009; Sloss et al., 2013).
 RADIOCARBON DATING OF MARINE CARBONATES
Marine reservoir effect and age calibration
After its production in the atmosphere, radiocarbon is oxi-
dized to 14CO2. In this form, 14C is transferred from the
atmosphere to the surface ocean via air-sea gas exchange
and then incorporated into the deep ocean. The residence
time of carbon in the deep ocean is quite long
(~800–1,000 years), and during this time 14C in deep
ocean waters decays without replenishment from the sur-
face. As a result, the deep ocean has a much lower 14C
content than that of the atmosphere, or materials drawing
carbon from the deep ocean reservoir are much older than
contemporaneous materials of terrestrial origin. For the
surface ocean, the aging effect is smaller because the sur-
face ocean, exchanging with both the atmosphere and the
14
C-depleted deep ocean, has a 14C level intermediate
between these two reservoirs. In other words, marine
organisms living in the surface ocean (e.g., shells, corals,
and planktonic foraminifera) appear older than contempo-
raneous terrestrial samples. The age offset between sur-
face ocean and terrestrial samples, ~400 years on
average, is known as the marine reservoir age (R). Marine
reservoir effect is corrected during the process of age
calibration.
As radiocarbon and calendar ages are not the same, the
former ages have to be converted to the latter ones using
an age calibration process. To calibrate a radiocarbon date
for a surface ocean sample, one can use an internationally
ratified terrestrial calibration curve (e.g., IntCal13 for the
Northern Hemisphere (Reimer et al., 2013) or SHCal13
for the Southern Hemisphere (Hogg et al., 2013)) with
a known value of R. Alternatively, one can use the current
internationally ratified marine calibration curve Marine13
(Reimer et al., 2013) with a known value of regional
offset from the global marine model age for that sample,
defined as DR. The latter method is generally preferred.
Databases of R and DR values for different regions are
available online (e.g., http://calib.qub.ac.uk/marine/ and
http://radiocarbon.ldeo.columbia.edu/research/resage2.htm).
Calibration of 14C ages is undertaken using a computer
program. Several programs are available online (e.g.,
CALIB at http://radiocarbon.pa.qub.ac.uk/calib/, OxCal
at http://c14.arch.ox.ac.uk/embed.php?File¼oxcal.html).
Dating of recent materials
Atmospheric nuclear weapon tests, mostly in the late
1950s and early 1960s, released a considerable amount
of 14C into the atmosphere, leading to an enormous
increase in atmospheric 14C concentration. Via air-sea
exchange of CO2, 14C levels in the surface ocean also
increased significantly. This means bomb radiocarbon
can also be used to date recent marine carbonate samples.
However, unlike the simple zonal pattern of bomb 14C in
the atmosphere (Hua et al., 2013), the distribution of bomb
14
C in the surface ocean is sensitive to local and regional
patterns of ocean circulation including horizontal advec-
tion and vertical transport such as ocean upwelling. In
other words, 14C levels in the surface ocean have local
677
and regional characteristics such as those depicted in
Figure 1 for the Pacific Ocean. Therefore, recent marine
carbonate samples from a particular location can accu-
rately be dated only if a surface ocean bomb 14C curve
for such location/region, which serves as a radiocarbon
calibration curve, is well determined. Based on Figure 1,
the highest dating accuracy can be achieved for marine
organisms formed in the period from ~1960 to the mid-
1970s, when there was a significant increase in surface
ocean 14C.
Advances in dating of marine carbonates
Small quantity of materials required by the AMS method
(tens of micrograms to a couple of milligrams of carbon;
Hua, 2009; Sloss et al., 2013) has advanced radiocarbon
dating in general and dating of marine carbonates in par-
ticular. For example, depositional chronologies of reef
islands are traditionally based on radiocarbon dating of
bulk carbonate sediments using the radiometric or conven-
tional method. However, Woodroffe et al. (2007) dated by
AMS different skeletal components of carbonate sedi-
ments (corals, foraminifera, and gastropod shells)
collected in a transect across Warraber Island in Torres
Strait (Australia) and compared these AMS dates to
those of corresponding bulk carbonate sediments. The
authors found that gastropods appeared to be the
youngest among the samples and concluded that 14C dates
on the most dominant skeletal organism (gastropods in
this case) rather than those on bulk carbonate sediments
should deliver reliable information on reef island
evolution.
The ability to analyze microgram-sized samples by
AMS also helped to answer one of the important questions
on the study of reef islands: How quick after their death
are skeletal organisms transported from reef flats, where
they are produced, to deposit to build reef islands? This
question becomes more important as climate change pro-
gresses. In this scenario, the oceans become more acidic
due to their uptake of anthropogenic CO2; therefore it is
likely that the skeletal organisms are vulnerable and
consequently the development of reef islands is
unsustainable. Dawson et al. (2013) studied Raine Reef
in northwestern Australia and dated individual foraminif-
era, the most dominant skeletal organism at Raine, at dif-
ferent locations including the reef flat and beaches.
A surface ocean 14C bomb curve for the study region
derived from known-age corals was reconstructed to facil-
itate the dating of these recent foraminifera. The AMS
results on pristine samples indicated that it took a short
period of time (maximum of 13–27 years) to transport
the samples from the reef flat to the island, over distance
up to 2 km.
Another example of dating of small-sized carbonate
samples is the use of bomb 14C in the surface ocean to val-
idate fish ages (Ewing et al., 2007; Andrews et al., 2011).
This method was pioneered by Kalish (1993) when he
showed a good comparison between 14C values recorded
678 RADIOCARBON DATING OF MARINE CARBONATES
Radiocarbon Dating of Marine Carbonates, Figure 1 The 14C content in the surface waters of the Pacific Ocean for the period
1950–1984 recorded in corals. Error bars are 1s (Adapted from Hua (2009)).
in known-age otolith carbonate of a marine fish species
and regional surface ocean 14C records from corals of
the Great Barrier Reef. In this application, the birth dates
of fish specimens were first estimated by counting annual
increments visible in otolith thin sections. These dates
were then checked or validated by comparing the 14C con-
tent of the first annual otolith increment with a regional
surface ocean 14C bomb curve.
Difficulties in dating of marine carbonates
The 14C content in some shellfish species may not reflect
that of dissolved inorganic carbon (DIC) of surrounding
sea waters at the time of accretion of their calcium carbon-
ate skeletons, leading to problems in radiocarbon dating of
these species. Petchey (2009) attributed these issues to the
dietary and habitat preferences of shellfish. In particular,
deposit feeders consume detritus in areas dominated by
limestone geologies containing much older carbon or
digest younger carbon from a terrestrial source in areas
influenced by local freshwater runoff. As a result, the
14
C ages of these shellfish samples are older or younger
than the true ages, respectively. Dating of carnivorous
shellfish is also problematic because of the uncertainty in
their 14C content, which is dependent on the carbon reser-
voirs of their prey. By contrast, suspension feeders, which
consume suspended phytoplankton and seawater DIC,
reflect the 14C content of the surface ocean and are there-
fore considered as the most reliable shellfish for dating
(Petchey, 2009).
The 14C content of organisms living in atoll lagoons
and estuaries may also be different from that of open ocean
waters. For shallow lagoons with limited exchange with
open ocean waters, 14C levels of lagoonal waters are
strongly influenced by air-sea gas exchange and are there-
fore much higher than those of nearby open sea waters
(Petchey, 2009). Similarly, 14C levels of estuarine waters
reflect the interaction and mixing of terrestrial and marine
carbons (Ulm, 2002). The estuarine effect is significant
where water circulation is restricted. Thus accurate dating
of these organisms can be achieved only if R and/or DR
values relevant to their environment are known.
Radiocarbon dating of surface ocean samples involves
estimates of marine reservoir effect (R and DR). These
values for a given location are generally assumed to be
constant with time when calibrating marine 14C ages.
However, recent studies have reported large variability in
the marine reservoir effect of several hundred to a couple
of thousand years for various regions during the Late
Glacial and Holocene (e.g., Siani et al., 2001; Bondevik
et al., 2006; Yu et al., 2010; Ortlieb et al., 2011). These
variations resulted from changes in ocean circulation and
the carbon cycle associated with climate change. For these
regions, if temporal variability in R and DR does not take
into account when calibrating marine 14C ages, the resul-
tant calibrated 14C ages may be erroneous. Ortlieb et al.
(2011) reported high DR values for the northern Chile-
southern Peru coast of up to 1,000 years during the period
7,000–10,000 cal years BP (early to mid-Holocene) in
comparison with their modern value of 200–300 years.
 RADIOCARBON DATING OF MARINE CARBONATES
The authors showed that the calibrated 14C ages during the
early to mid-Holocene based on the constant modern DR
value were older than the true calendar ages of up to
800–1,000 years.
Summary
Radiocarbon dating has widely been used to reconstruct
chronologies for marine carbonates for the past 50 ka for
studies in archaeology, earth sciences, paleoceanography,
marine ecology, and Quaternary climate change. For
example, radiocarbon-dated corals have been used to
investigate changes in sea surface temperature and salin-
ity, and variations in El Niño-Southern Oscillation
through time. Radiocarbon dates on boring bivalves
occurring in dislocated boulders and on mollusk shells
associated with gravel deposits along coast lines have also
been employed to constrain the timing of tsunami events.
In addition, radiocarbon dating of cultural shells has been
used for the study of human occupation and colonization
of Pacific atolls.
With the advent of AMS radiocarbon dating on micro-
gram-sized samples, some radiocarbon applications are
available (e.g., validation of fish ages by 14C analysis of
fish otoliths) or become more reliable (e.g., 14C-based
chronologies for the study of reef island evolution). How-
ever, the dietary and habitat preferences of shellfish, the
lagoonal and estuarine effects, and possible temporal var-
iations in the marine reservoir effect should be carefully
considered when dating marine samples.
Bibliography
Andrews, A. H., Kalish, J. M., Newman, S. J., and Johnston, J. M.,
2011. Bomb radiocarbon dating of three important reef-fish spe-
cies using Indo-Pacific D14C chronologies. Marine and Fresh-
water Research, 62, 1259–1269.
Bondevik, S., Mangerud, J., Birks, H. H., Gulliksen, S., and Reimer,
P., 2006. Changes in North Atlantic radiocarbon reservoir ages
during the Allerød and Younger Dryas. Science, 312,
1514–1517.
Dawson, J. L., Smithers, S. G., and Hua, Q., 2013. The importance
of large benthic foraminifera to reef island sediment budget and
dynamics at Raine Island, northern Great Barrier Reef. Geomor-
phology, in press.
Douka, K., Hedges, R. E. M., and Higham, T. F. G., 2010. Improved
AMS 14C dating of shell carbonates using high-precision X-ray
diffraction and a novel density separation protocol (CarDS).
Radiocarbon, 52(2–3), 735–751.
Ewing, G. P., Lyle, J. M., Murphy, R. J., Kalish, J. M., and Ziegler,
P. E., 2007. Validation of age and growth in a long-lived temper-
ate reef fish using otolith structure, oxytetracycline and bomb
radiocarbon methods. Marine and Freshwater Research, 58,
944–955.
Hogg, A. G., Hua, Q., Blackwell, P. G., Niu, M., Buck, C. E.,
Guilderson, T. P., Heaton, T. J., Palmer, J. G., Reimer, P. J.,
Reimer, R. W., Turney, C. S. M., and Zimmerman, S. R. H.,
2013. SHCAL13 Southern Hemisphere calibration,
0–50,000 cal yr BP. Radiocarbon, 55(4), 1889–1903.
Hua, Q., 2009. Radiocarbon: a chronological tool for the recent past.
Quaternary Geochronology, 4, 378–390.
679
Hua, Q., Barbetti, M., and Rakowski, A. Z., 2013. Atmospheric
radiocarbon for the period 1950–2010. Radiocarbon, 55(4),
2059–2072.
Kalish, J. M., 1993. Pre- and post-bomb radiocarbon in fish otoliths.
Earth and Planetary Science Letters, 114, 549–554.
McGregor, H. V., and Gagan, M. K., 2003. Diagenesis and geo-
chemistry of Porites corals from Papua New Guinea: implica-
tions for paleoclimate reconstruction. Geochimica et
Cosmochimica Acta, 67, 2147–2156.
Nothdurft, L. D., and Webb, G. E., 2009. Earliest diagenesis in
scleractinian coral skeletons: implications for palaeoclimate-
sensitive geochemical archives. Facies, 55, 161–201.
Ortlieb, L., Vargas, G., and Saliège, J.-F., 2011. Marine radiocarbon
reservoir effect along the northern Chile-southern Peru coast
(14–24 S) throughout the Holocene. Quaternary Research, 75,
91–103.
Petchey, F., 2009. Dating marine shell in Oceania: issues and pros-
pects. In Fairbairn, A., O’Connor, S., and Marwick, B. (eds.),
Terra Australis 28: New Directions in Archaeological Science.
Canberra: ANU E Press, pp. 157–172.
Reimer, P. J., Bard, E., Bayliss, A., Beck, J. W., Blackwell, P. G.,
Bronk Ramsey, C., Buck, C. E., Cheng, H., Edwards, R. L., Frie-
drich, M., Grootes, P. M., Guilderson, T. P., Haflidason, H.,
Hajdas, I., Hatté, C., Heaton, T. J., Hoffman, D. L., Hogg,
A. G., Hughen, K. A., Kaiser, K. F., Kromer, B., Manning,
S. W., Niu, M., Reimer, R. W., Richards, D. A., Scott, M.,
Southon, J. R., Staff, R. A., Turney, C. S. M., and van der
Plicht, J., 2013. IntCal13 and Marine13 radiocarbon age calibra-
tion curves 0–50,000 years cal BP. Radiocarbon, 55(4),
1869–1887.
Russo, C. M., Tripp, J. A., Douka, K., and Higham, T. F. G., 2010.
A new radiocarbon pretreatment for molluscan shell using den-
sity fractionation of carbonates in bromoform. Radiocarbon,
52(2–3), 1301–1311.
Siani, G., Paterne, M., Michel, E., Sulpizio, R., Sbrana, A., Arnold,
M., and Haddad, G., 2001. Mediterranean sea surface radiocar-
bon reservoir age changes since the Last Glacial Maximum. Sci-
ence, 294, 1917–1920.
Sloss, C. R., Westaway, K. E., Hua, Q., and Murray-Wallace, C. V.,
2013. An introduction to dating techniques: a guide for geomor-
phologists. In Shroder, J., Switzer, A. D., and Kennedy, D. M.
(eds.), Treatise on Geomorphology. San Diego:
Academic. Methods in Geomorphology, Vol. 14, pp. 346–369.
Ulm, S., 2002. Marine and estuarine reservoir effects in central
Queensland, Australia: determination of DR values.
Geoarchaeology, 17(4), 319–348.
Walker, M., 2005. Quaternary Dating Methods. Chichester: Wiley.
Webb, G. E., Price, G. J., Nothdurft, L. D., Deer, L., and Rintoul, L.,
2007. Cryptic meteoric diagenesis in fresh water bivalves: impli-
cations for radiocarbon dating. Geology, 35, 803–806.
Woodroffe, C. D., Samosorn, B., Hua, Q., and Hart, D. E., 2007.
Incremental accretion of a sandy reef island over the past 3000
years indicated by component-specific radiocarbon dating. Geo-
physical Research Letters, 34, L03602, doi: 10.1029/
2006GL028875.
Yu, K., Hua, Q., Zhao, J., Hodge, E., Fink, D., and Barbetti, M.,
2010. Holocene marine 14C reservoir age variability: evidence
from 230Th-dated corals from South China Sea. Paleocea-
nography, 25, PA3205, doi: 10.1029/2009PA001831.
Cross-references
Accelerator Mass Spectrometry
Bomb Carbon
14
C in Plant Macrofossils
Radiocarbon Dating
Sedimentary Rocks (Rb-Sr Geochronology)
680 RADIOCARBON DATING OF TERRESTRIAL CARBONATES
RADIOCARBON DATING OF TERRESTRIAL
CARBONATES
Jeffrey S. Pigati
Denver Federal Center, U.S. Geological Survey, Denver,
CO, USA
Introduction
Terrestrial carbonates encompass a wide range of mate-
rials that potentially could be used for radiocarbon (14C)
dating. Biogenic carbonates, including shells and tests of
terrestrial and aquatic gastropods, bivalves, ostracodes,
and foraminifera, are preserved in a variety of late Quater-
nary deposits and may be suitable for 14C dating. Primary
calcareous deposits (marls, tufa, speleothems) and second-
ary carbonates (rhizoliths, fracture fill, soil carbonate)
may also be targeted for dating when conditions are favor-
able. This chapter discusses issues that are commonly
encountered in 14C dating of terrestrial carbonates, includ-
ing isotopic disequilibrium and open-system behavior, as
well as methods used to determine the reliability of ages
derived from these materials. Recent methodological
advancements that may improve the accuracy and preci-
sion of 14C ages of terrestrial carbonates are also
highlighted.
Dating terrestrial carbonates
All materials that yield reliable 14C ages have two com-
mon characteristics. First, the initial 14C activity of
a material must be in isotopic equilibrium with atmo-
spheric 14C at the time it was alive (for living materials)
or precipitated (for nonliving materials). Second, it must
behave as a closed system with respect to carbon after
death or precipitation; that is, secondary (contaminant)
carbon is neither added to nor exchanged with carbon that
is original to the material itself. If both of these criteria are
met, then the measured 14C activity is a function of only
two parameters: the initial 14C activity of the atmosphere
and the amount of time elapsed since the death of the
organism.
The measured 14C activity and age of the material are
related by the familiar decay equation
A ¼ Ao elt
where A and Ao are the measured and initial 14C activities
of the material, respectively; l is the decay constant of
14
C; and t is the time elapsed since death or precipitation.
Conventional radiocarbon ages (incorrectly) assume that
atmospheric 14C activity is invariant through time
(Ao ¼ 1) and are calculated using the Libby half-life for
14
C of 5,568  30 years (Libby, 1955). Radiocarbon ages
can be converted to calendar year ages via calibration to
account for temporal variations in the 14C activity of the
atmosphere that are known to have occurred (Reimer
et al., 2013). Calibration also accounts for the differences
between the Libby half-life and the Cambridge half-life
for 14C of 5,730  40 years (Godwin, 1962). If the equilib-
rium and closed-system behavior criteria are met, and the
calibration curves are accurate, then the calibrated age of
a material should represent its true age in calendar years.
Isotopic equilibrium
Organisms obtain carbon to build their shells or tests
from up to four different sources: air, water, food, and
rocks. The 14C activity of these sources ranges from mod-
ern (atmospheric CO2, live plants, modern water) to
14
C-deficient (organic detritus, “old” water) to 14C-free
or “dead” (carbonate rocks, coal, some peat deposits).
The influence that each of these sources has on the overall
14
C content of a shell or test depends on the relative pro-
portion that they contributes to the total carbon pool. For
example, if a gastropod obtains all of its shell carbon from
air, live plants, and modern water, then the isotopic equi-
librium criterion is satisfied and its shell may be suitable
for 14C dating. However, if the same gastropod consumed
old organic detritus or peat, drank water that was deficient
in 14C, or ingested limestone and incorporated the dead
carbon in its shell, then the isotopic equilibrium criterion
would be violated and the resulting shell age would be
too old. Common causes of isotopic disequilibrium in bio-
genic carbonates are discussed below in order to demon-
strate the importance of identifying and quantifying the
various sources that contribute carbon to the dated
material.
Terrestrial gastropods and the “limestone problem”
Terrestrial gastropods acquire carbon to build their shells
from air, water, plants (live and decayed), and carbonate
rocks (limestone, dolomite, soil carbonate). Respired
CO2 from the atmosphere is introduced to the bicarbonate
pool in the gastropod’s hemolymph and passed along to
the extrapallial fluid, from which the shell carbonate is
ultimately precipitated (Wilbur, 1972). Carbon from food
sources (e.g., living plants, fungi, organic detritus) and
water (dissolved inorganic carbon; DIC) is also incorpo-
rated into the extrapallial fluid and incorporated into the
shell material (Balakrishnan and Yapp, 2004). Estimates
of the individual contributions of atmospheric CO2,
plants, and water toward gastropod shell carbon vary con-
siderably, but together, in the absence of limestone or car-
bonate rocks, they account for 100 % of the total input
(Goodfriend and Hood, 1983; Stott, 2002; Romaniello
et al., 2008).
In most environments, the 14C activities of live plants
and water sources are in equilibrium with atmospheric car-
bon, which is ideal for 14C dating. However, gastropods
also consume organic detritus (i.e., decaying plant litter),
which potentially could cause their 14C ages to be anoma-
lously old. In most cases, however, the amount of time
elapsed between the death of a plant, its incorporation into
decomposition products, and consumption by gastropods
is usually quite short, on the order of a few years, which
would not affect the resulting 14C ages significantly. Gas-
tropods living in or around coal seams and peatlands are
 RADIOCARBON DATING OF TERRESTRIAL CARBONATES
an exception, as they may have access to carbon that is
hundreds or even thousands of years old. Similarly, gastro-
pods living in areas of active volcanism, where 14C-
deficient carbon in plants may be present, can also yield
ages that are much too old. Shell ages derived from gastro-
pods living in these unusual settings should be evaluated
carefully.
These exceptions aside, the primary concern with dat-
ing terrestrial gastropod shells is the “limestone problem.”
A term coined by Goodfriend and Stipp (1983), the lime-
stone problem refers to the fact that many gastropods
scrape or abrade carbonate rocks and ingest the powder
or granules which then dissolve in their stomach acid to
produce CO2. As with carbon from other sources, the dead
carbon from rocks is introduced to the hemolymph, passed
on to the extrapallial fluid, and ultimately incorporated in
the shell carbonate. Dead carbon from limestone can
account for up to ~30 % of the total carbon in shells of
some large terrestrial gastropods, which would cause
apparent 14C ages of the shell carbonate to be ~3,000
14
C years too old (Tamers, 1970; Goodfriend and
Stipp, 1983; Zhou et al., 1999; Pigati et al., 2004). Unfor-
tunately, a simple correction factor that accounts for the
limestone effect cannot be applied universally because
the amount of old carbon incorporated by gastropods is
highly variable, both between taxa and between
populations of a single taxon living in different habitats
(Goodfriend et al., 1999; Pigati et al., 2010). Thus, to be
confident in 14C ages derived from terrestrial gastropod
shells, researchers should avoid dating gastropods that
incorporate dead carbon from rocks altogether.
Aquatic organisms and hard-water (or reservoir) effects
The 14C activity of dissolved inorganic carbon (DIC) in
springs and lakes can be lower than atmospheric values
for two reasons: (1) interaction between ground water
and carbonate rocks, and (2) isotopic decay of DIC that
occurs as water molecules travel between areas of
recharge and discharge. The magnitude of the local
“hard-water” or “reservoir” effect can be highly variable,
ranging from negligible to extreme. For water bodies that
freely exchange carbon with the atmosphere, the 14C
activity of DIC should be in isotopic equilibrium with
atmospheric carbon. On the other hand, the 14C activity
of ground water flowing through large carbonate aquifers
can be very depleted, reaching values as low as ~2–3 % of
modern values (e.g., Winograd and Pearson, 1976; Riggs,
1984). Fully aquatic organisms, such as gill-breathing
aquatic gastropods (subclass Prosobranchia), ostracodes,
and foraminifera, acquire all of their carbon from aqueous
sources, either directly from the water or indirectly from
aquatic plants. Thus, the full magnitude of the local
hard-water effect is passed on to shells or tests of these
organisms. Lung-breathing aquatic and semiaquatic gas-
tropods (subclass Pulmonata) obtain their shell carbon
from both water and air and, therefore, are subject to par-
tial hard-water effects. For these gastropods, the offset
between true and apparent ages depends on both the
681
magnitude of the local hard-water effect and the propor-
tion of shell carbon derived from aqueous sources and
the atmosphere.
Closed-system behavior
Even if organisms avoid the limestone problem and hard-
water effects, their shells or tests must remain closed sys-
tems with respect to carbon after death if they are to yield
reliable 14C ages. To do so, carbon atoms measured during
the 14C dating process must consist solely of those that
originally resided in the shell or test. Biogenic carbonates
that exhibit open-system behavior typically yield 14C ages
that are too young, and the magnitude of the error depends
upon the degree of such behavior. Unlike limestone or
hard-water effects, which result in discrepancies between
measured and actual 14C ages that remain constant over
time, the impact of open-system behavior increases with
sample age (Figure 1). For example, if a gastropod shell
that is 10,000 14C years old is contaminated by the equiv-
alent of 1 % modern carbon, then the resulting age would
be 9,730 14C years, a difference of only 270 14C years. In
contrast, if a shell that is 50,000 14C years old is contami-
nated by modern carbon to the same extent, the apparent
age would be only 35,540 14C years, a difference of
14,460 14C years!
Reliable or not? Evaluating carbonates for
potential use in 14C dating
If terrestrial carbonates are to be used for 14C dating, it is
critical that they are evaluated repeatedly in different
depositional settings, habitats, and climate regimes to
determine if they meet the isotopic equilibrium criterion
and behave as closed systems with respect to carbon dur-
ing burial. Evaluations can be done using at least three dif-
ferent methods: (1) evaluating the 14C content of modern
specimens, (2) comparing 14C ages of terrestrial carbon-
ates with independent ages, and (3) analytical
assessments.
Evaluating modern specimens
Evaluating modern specimens for the presence of old car-
bon from either limestone or hard water can be done with
relative ease and minimal cost. Moreover, such evalua-
tions provide worst-case scenarios because the influence
of old carbon is most apparent in young/modern samples.
In modern evaluations, the 14C activities of live-collected
specimens are measured by accelerator mass spectrometry
(AMS) and compared against atmospheric values
(Hua and Barbetti, 2004). If old carbon from limestone
or hard water is present, then the measured 14C activity
will be lower than the atmosphere. It is critical to perform
such evaluations on the specific taxon of interest. For
example, Goodfriend and coworkers found evidence of
significant (10–30 %) old carbon in large (1–2 cm diame-
ter) gastropods collected live in Texas and elsewhere in the
southwestern United States (Goodfriend, 1987;
Goodfriend et al., 1999). In contrast, Pigati et al. (2004,
682 RADIOCARBON DATING OF TERRESTRIAL CARBONATES

Radiocarbon Dating of Terrestrial Carbonates, Figure 1 Deviation from the true 14C age for four scenarios: (1) closed-system
behavior and no dead carbon (thick solid line), (2) closed-system behavior and 10 % dead carbon (dashed line), (3) open-system
behavior equivalent to 1 % modern carbon contamination and 10 % dead carbon (thin solid line), and (4) open-system behavior
equivalent to 1 % modern carbon contamination and no dead carbon (dotted line) (After Pigati et al., 2010).

2010) examined the 14C activities of small (<1 cm diam-
eter) terrestrial gastropods collected live from limestone
or carbonate terrains throughout North America. They
found that nearly 80 % of the taxa analyzed did not show
evidence of measurable limestone effects. Results from
these studies show the importance of evaluating the spe-
cific taxon of interest rather than extrapolating results
across taxonomic lineages.
Obviously, evaluating modern specimens is only possi-
ble if the taxa of interest are still living today, which is not
always the case. Moreover, although modern evaluations
usually give clear indication of whether old carbon is
a problem in the material to be dated, researchers must
assume that the results observed remain invariant over
time. In general, this appears to be the case for terrestrial
gastropods. Terrestrial taxa such as Succineidae, Discus,
and a few other common small gastropods that appear to
avoid limestone in modern settings also yield ages that
are similar to independent ages in the fossil record (Pigati
et al., 2013). Similarly, Mesodon shells are offset from
independent ages by the equivalent of ~3,000 14C years
or more in both modern systems and the fossil record
(Rakovan et al., 2013).
In aquatic environments, researchers can evaluate mod-
ern aquatic or semiaquatic organisms living near extant
springs and lakes to determine the influence of potential
hard-water effects, which are also assumed to be invariant
over time. However, hard-water or reservoir effects in
some hydrologic systems (mostly lakes) can vary signifi-
cantly over time because of changes in the balance
between hydrologic inputs, say, for example, between run-
off and ground water. Thus, if there is reason to believe
that the relative contributions of different inputs have var-
ied in the past, then the results of modern evaluations may
not be directly applicable to aquatic systems in the geo-
logic record.
Comparison against known ages
One of the best – and most difficult – evaluations to con-
duct is to compare terrestrial carbonate ages against inde-
pendent ages to determine if the shells or tests remained
closed systems after death. If the evaluations can be
conducted for the same time period and environmental
conditions represented in the geologic record of interest,
then we can be reasonably confident in the results. How-
ever, establishing strong independent chronologies is
often difficult and costly.
Thus far, there have been relatively few studies
conducted in which terrestrial gastropod shell ages have
been systematically compared against robust independent
ages. Results from loess deposits in Alaska and the Great
Plains demonstrate that shell ages (mostly from
Succineidae) closely track independent 14C ages derived
from plant macrofossils and luminescence ages over the
past 40,000 year or so (Figure 2). Shell ages of terrestrial
gastropods from soils and wetland deposits in the south-
western United States and Europe, as well as glacial
deposits in the Midwest United States, are also indistin-
guishable from independent ages, which suggests that
many small terrestrial gastropod shells do, in fact, yield
reliable 14C ages for the fossil record, regardless of the
local rock type, depositional setting, or climate regime
(e.g., Preece, 1991, 1994; Pigati et al., 2010, 2011).
 RADIOCARBON DATING OF TERRESTRIAL CARBONATES 683

Radiocarbon Dating of Terrestrial Carbonates, Figure 2 (a) Independent and gastropod shell ages from a Holocene loess section in
Wrangell-St. Elias National Park near Chitina, AK (Muhs et al., 2013b; Pigati et al., 2013). The independent ages consist of calibrated 14C
ages from well-preserved wood, and the shell ages are from Succineidae. (b) Independent and gastropod shell ages from loess
deposits at Loveland, IA (Muhs et al., 2013a; Pigati et al., 2013). The independent ages consist of optically stimulated luminescence
(OSL) ages, and the shell ages are from Succineidae.

Chronologies based on 14C dating of aquatic carbonates
can be compared to 14C ages of vascular plant material or
charcoal, as well as independently dated chronometric
markers such as tephra layers, unique geochemical signa-
tures (e.g., presence of industrialization by-products,
anthropogenic indicators), or geomagnetic excursions. In
ideal situations, researchers working in aquatic systems
should use multiple tie points between the carbonates
and independent ages to determine whether the local
hard-water effect remained constant over time.
Analytical assessments
Analytical tools, including scanning electron microscopy
(SEM) images and X-ray diffraction (XRD), can be used
to determine the nature and physical location of contami-
nants in terrestrial gastropod shells (Yates, 1986; Rech
et al., 2011). SEM images can show the position and phys-
ical structure of potential contaminants, especially if they
are present on the surface of the material dated. However,
if chemical exchange has occurred without concomitant
morphologic alteration, then SEM imagery may not reveal
the contaminants. XRD methods can detect the presence
of calcite contaminants in aragonitic shell material, but
in most cases the amount of calcite contamination must
be at least ~1 % of the total to be detected reliably. If
SEM or XRD techniques cannot be applied, then, at
a minimum, the shells or tests of terrestrial and aquatic
organisms should be broken and inspected under
magnification to ensure that they are free of detritus before
processing.
Methodologic innovations
One of the primary assumptions in radiocarbon dating is
that the carbon atoms isolated for dating consist solely of
atoms that are original to the material itself (i.e., no con-
tamination). Some terrestrial carbonates, such as ostra-
codes, have exceptionally high surface area-to-volume
(SA/V) ratios, which increases the likelihood of chemical
exchange with the atmosphere during pretreatment. Stan-
dard laboratory procedures for dating carbonate shells or
tests often include etching the surfaces with dilute acid,
which removes dust and other particulate contaminants.
However, recent studies have shown that modern atmo-
spheric carbon can be incorporated into ostracode tests
by either surface adsorption or exchange during laboratory
processing (Hajdas et al., 2004). Such contamination may
not necessarily be removed during the etching process.
Stepped-dissolution experiments using ostracode tests
show that even after etching, younger carbon is preferen-
tially released during initial acid dissolution steps
(i.e., contaminants plus sample material), whereas older
carbon is released in later steps, presumably all original
to the sample (Zimmerman et al., 2012). These results
imply that if the SA/V ratio of a material is high enough,
contaminants can be missed by standard laboratory
pretreatment protocols and the resulting 14C ages would
be too young. Although stepped-dissolution procedures
684 RADIOCARBON DATING OF TERRESTRIAL CARBONATES
can be time consuming, technically challenging, and
expensive if multiple aliquots are measured by AMS, they
may ultimately yield 14C ages that are more accurate and
precise than those obtained using standard methods.
Conclusions
Terrestrial carbonates encompass a wide variety of mate-
rials that can potentially be used for 14C dating. However,
issues related to isotopic disequilibrium and open-system
behavior must be addressed before 14C ages derived from
these materials are considered reliable. Removing con-
taminating materials from terrestrial carbonates can be dif-
ficult because the contaminant species is often calcite,
which may be the same as, or similar to, the sample mate-
rial. For biogenic carbonates that are composed of arago-
nite, small amounts of calcite contamination may be
detected using scanning electron microscopy or X-ray dif-
fraction. In the fossil record, 14C ages derived from terres-
trial carbonates should be compared against 14C ages of
independent materials (ideally charcoal or plant macrofos-
sils) in depositional settings and time periods that are sim-
ilar to those in the geologic record of interest. If the
terrestrial carbonates consistently meet the isotopic equi-
librium and closed-system behavior criteria, then the
materials should yield reliable 14C ages. New methodo-
logical and technological innovations may improve the
accuracy and precision of 14C ages of terrestrial carbon-
ates, which is particularly important as high-resolution
chronologies are increasingly given high priority in
paleoclimate and other studies.
Acknowledgments
This manuscript benefited from constructive reviews from
Gene Ellis, Dan Muhs, and Tim Jull. This project was
funded by the U.S. Geological Survey Climate and Land
Use Change Research and Development Program.
Bibliography
Balakrishnan, M., and Yapp, C. J., 2004. Flux balance models for
the oxygen and carbon isotope compositions of land snail shells.
Geochimica et Cosmochimica Acta, 68, 2007–2024.
Godwin, H., 1962. Half-life of radiocarbon. Nature, 195, 984.
Goodfriend, G. A., 1987. Radiocarbon age anomalies in shell car-
bonate of land snails from semi-arid areas. Radiocarbon, 29,
159–167.
Goodfriend, G. A., and Hood, D. G., 1983. Carbon isotope analysis
of land snail shells: implications for carbon sources and radiocar-
bon dating. Radiocarbon, 25, 810–830.
Goodfriend, G. A., and Stipp, J. J., 1983. Limestone and the prob-
lem of radiocarbon dating of land-snail shell carbonate. Geology,
11, 575–577.
Goodfriend, G. A., Ellis, G. L., and Toolin, L. J., 1999. Radiocarbon
age anomalies in land snail shells from Texas: ontogenetic, indi-
vidual, and geographic patterns of variation. Radiocarbon, 41,
149–156.
Hajdas, I., Bonani, G., Zimmerman, S. R. H., Mendelson, M., and
Hemming, S. R., 2004. 14C ages of ostracodes from Pleistocene
lake sediments of the western Great Basin, USA – results of pro-
gressive acid leaching. Radiocarbon, 46, 189–200.
Hua, Q., and Barbetti, M., 2004. Review of tropospheric bomb 14C
data for carbon cycle modeling and age calibration purposes.
Radiocarbon, 46, 1273–1298.
Libby, W. F., 1955. Radiocarbon Dating, 2nd edn. Chicago, IL:
University of Chicago Press, p. 175.
Muhs, D. R., Bettis, E. A., Roberts, H. M., Harlan, S. S.,
Paces, J. B., and Reynolds, R. L., 2013a. Chronology and
provenance of last-glacial (Peoria) loess in western Iowa and
paleoclimatic implications. Earth and Planetary Science Letters,
80, 468–481.
Muhs, D. R., Budahn, J. R., McGeehin, J. P., Bettis, E. A. I., Skipp,
G., Paces, J. B., and Wheeler, E. A., 2013b. Loess origin, trans-
port, and deposition in southern Alaska over the past 10,000
years, Wrangell-St. Elias National Park, Alaska. Journal of Aeo-
lian Research, 11, 85–99.
Pigati, J. S., Quade, J., Shanahan, T. M., and Haynes, C. V., 2004.
Radiocarbon dating of minute gastropods and new constraints
on the timing of spring-discharge deposits in southern Arizona,
USA. Palaeogeography, Palaeoclimatology, Palaeoecology,
204, 33–45.
Pigati, J. S., Rech, J. A., and Nekola, J. C., 2010. Radiocarbon dat-
ing of small terrestrial gastropods in North America. Quaternary
Geochronology, 5, 519–532.
Pigati, J. S., Miller, D. M., Bright, J., Mahan, S. A., Nekola, J. C.,
and Paces, J. B., 2011. Chronology, sedimentology, and micro-
fauna of ground-water discharge deposits in the central Mojave
Desert, Valley Wells, California. Geological Survey of America
Bulletin, 123, 2224–2239.
Pigati, J. S., McGeehin, J. P., Muhs, D. R., and Bettis, E. A. I., 2013.
Radiocarbon dating late Quaternary loess deposits using small
terrestrial gastropod shells. Quaternary Science Reviews, 76,
114–128.
Pigati, J. S., McGeehin, J. P., Muhs, D. R., Grimley, D. C., Wang,
H., and Nekola, J. C., (in press-b). Radiocarbon dating loess
deposits in the Mississippi Valley using terrestrial gastropod
shells (Polygyridae, Helicinidae, and Discidae). Aeolian
Research.
Preece, R. C., 1991. Radiocarbon-dated molluscan successions
from the Holocene of central Spain. Journal of Biogeography,
18, 409–426.
Preece, R. C., 1994. Radiocarbon dates from the ‘Allerød soil’
in Kent. Proceedings of the Geologists’ Association, 105,
111–123.
Rakovan, M. T., Rech, J. A., Pigati, J. S., Nekola, J. C., and
Wiles, G. C., 2013. An evaluation of Mesodon and other
large terrestrial gastropod shells for dating late Holocene and his-
toric alluvium in the Midwestern USA. Geomorphology, 193,
47–56.
Rech, J. A., Pigati, J. S., Lehmann, S. B., McGimpsey, C. N.,
Grimley, D. A., and Nekola, J. C., 2011. Assessing open-system
behavior of carbon-14 in terrestrial gastropod shells. Radiocar-
bon, 53, 325–335.
Reimer, P. J., Bard, E., Bayliss, A., Beck, J. W., Blackwell, P. G.,
Bronk Ramsey, C., Grootes, P. M., Guilderson, T. P., Haflidason,
H., Hajdas, I., Hatte, C., Heaton, T. J., Hoffmann, D. L., Hogg,
A. G., Hughen, K. A., Kaiser, K. F., Kromer, B., Manning, S.
W., Niu, M., Reimer, R. W., Richards, D. A., Scott, E. M.,
Southon, J. R., Staff, R. A., Turney, C. S. M., van der Plicht,
J., 2013. IntCal13 and Marine13 Radiocarbon Age
Calibration Curves 0–50,000 Years cal BP. Radiocarbon, 55,
1869–1887.
Riggs, A. C., 1984. Major carbon-14 deficiency in modern snail
shells from southern Nevada springs. Science, 224, 58–61.
Romaniello, L., Quarta, G., Mastronuzzi, G., D'Elia, M., and
Calcagnile, L., 2008. 14C age anomalies in modern land snails
shell carbonate from Southern Italy. Quaternary Geochronology,
3, 68–75.
 RADIOLUMINESCENCE (RL)
Stott, L. D., 2002. The influence of diet on the d13C of shell carbon
in the pulmonate snail Helix aspersa. Earth and Planetary Sci-
ence Letters, 195, 249–259.
Tamers, M. A., 1970. Validity of radiocarbon dates on terrestrial
snail shells. American Antiquity, 35, 94–100.
Wilbur, K. M., 1972. Shell formation in mollusks. In Florkin, M.,
and Scheer, B. T. (eds.), Chemical Zoology. New York:
Academic. Mollusca, Vol. 7, pp. 103–145.
Winograd, I. J., and Pearson, F. J., 1976. Major carbon 14 anomaly
in a regional carbonate aquifer: possible evidence for megascale
channeling, south central Great Basin. Water Resources
Research, 12, 1125–1143.
Yates, T., 1986. Studies of non-marine mollusks for the selection of
shell samples for radiocarbon dating. Radiocarbon, 28,
457–463.
Zhou, W., Head, W. J., Wang, F., Donahue, D. J., and Jull, A. J. T.,
1999. The reliability of AMS radiocarbon dating of shells from
China. Radiocarbon, 41, 17–24.
Zimmerman, S. R. H., Steponaitis, E., Hemming, S. R., and
Zermeño, P., 2012. Potential for accurate and precise radiocar-
bon ages in deglacial-age lacustrine carbonates. Quaternary
Geochronology, 13, 81–91.
RADIOLUMINESCENCE (RL)
Tobias Lauer
Leibniz Institute for Applied Geophysics, Hannover,
Germany
Synonyms
Infrared radiofluorescence; Infrared radioluminescence;
IR-RF
Definition
The radioluminescence (RL) phenomenon is based on the
emission of fluorescent light at a wavelength range of
865 nm occurring when potassium (K-) feldspar grains
are exposed to a radioactive source (e.g., Trautmann
et al., 1998; Erfurt et al., 2003). The ionization leads to
electron transfers from an excited state to the ground state
of the electron center (Trautmann et al., 2000), and RL is
emitted while electrons are captured in free optically
active traps. RL is therefore not linked to recombination
centers and reflects the charge density within electron
traps (Trautmann, 2000). RL is most likely related to elec-
tron transitions in Pb+ centers (Erfurt, 2003). Contrary to
the luminescence signal, the RL signal decreases with irra-
diation time until a saturation limit is reached, and the RL
signal can be increased by exposing the K-feldspars to
light <500 nm ( >2.5 eV).
The radioluminescence phenomenon can be used for
dating applications. Infrared radioluminescence or infra-
red radiofluorescence (IR-RF) dating is a method which
can be used to determine the depositional age of Quater-
nary sediments (e.g., Trautmann et al., 1999; Krbetschek
et al., 2000) and therefore provides an alternative to the
685
luminescence dating method. As IR-RF dates the last sun-
light exposure of feldspar grains before burial,
a sufficiently long sunlight exposure of the investigated
material during transport (e.g., aeolian, fluvial) is
mandatory.
The saturation dose of the IR-RF is at about
1,200–1,500 Gy (Erfurt and Krbetschek, 2003), and with
an IR-RF single-aliquot regenerative-dose dating proto-
col, it was possible to date sediments >600 ka
(Wagner et al., 2010; Lauer et al., 2011). However,
bleaching properties, and sensitivity changes of the
IR-RF signal in between measurement cycles, have to be
checked to obtain accurate age estimates (Buylaert et al.,
2012).
Bibliography
Buylaert, J. P., Jain, M., Murray, A. S., Thomsen, K. J., and Lapp, T.,
2012. IR-RF dating of sand-sized K-feldspar extracts: a test of
accuracy. Radiation Measurements, 47, 759–765.
Erfurt, G., 2003. Infrared luminescence of Pb+ centres in potassium-
rich feldspar. Physica Status Solidi A, 200(2), 429–438.
Erfurt, G., and Krbetschek, M. R., 2003. IRSAR – a single-aliquot
regenerative-dose dating protocol applied to the infrared
radiofluorescence (IR-RF) of coarse-grain K feldspar. Ancient
TL, 21, 21–28.
Erfurt, G., Krbetschek, M. R., Bortolot, V. J., and Preusser, F., 2003.
A fully automated multi-spectral radioluminescence reading sys-
tem for geochronometry and dosimetry. Nuclear Instruments
and Methods in Physics Research Section B, 207(4), 487–499,
doi:10.1016/S0168-583X(03)01121-2.
Krbetschek, M. R., Trautmann, T., Dietrich, A., and Stolz, W., 2000.
Radioluminescence dating of sediments: methodological
aspects. Radiation Measurements, 32(5–6), 493–498,
doi:10.1016/S1350-4487(00)00122-0.
Lauer, T., Krbetschek, M. R., Frechen, M., Tsukamoto, S.,
Hoselmann, C., and Weidenfeller, M., 2011. Infrared
Radiofluorescence (IR-RF) dating of Middle Pleistocene fluvial
archives of the Heidelberg Basin (Southwest Germany).
Geochronometria, 38, 23–33.
Trautmann, T., 2000. A study of radioluminescence kinetics of nat-
ural feldspar dosimeters: experiments and simulations. Journal
of Physics D-Applied Physics, 33, 2304–2310.
Trautmann, T., Krbetschek, M. R., Dietrich, A., and Stolz, W., 1998.
Investigations of Feldspar radioluminescence: potential for
a new dating technique. Radiation Measurements, 29(3–4),
421–425, doi:10.1016/S1350-4487(98)00012-2.
Trautmann, T., Krbetschek, M. R., Dietrich, A., and Stolz, W., 1999.
Feldspar radioluminescence: a new dating method and its phys-
ical background. Journal of Luminescence, 85(1–3), 45–58,
doi:10.1016/S0022-2313(99)00152-0.
Trautmann, T., Krbetschek, M. R., Dietrich, A., and Stolz, W., 2000.
The basic principle of radioluminescence dating and a localized
transition model. Radiation Measurements, 32(5–6), 487–492,
doi:10.1016/S1350-4487(00)00119-0.
Wagner, G. A., Krbetschek, M. R., Degering, D., Bahain, J. J., Shao,
Q., Falguères, C., Voinchet, P., Dolo, J. M., Garcia, T., and
Rightmire, G. P., 2010. Radiometric dating of the type-site for
Homo-heidelbergensis at Mauer, Germany. Proceedings of the
National Academy of Sciences of the United States of America
(PNAS), 107(46), 19726–19730.
Cross-references
Luminescence Dating
686 Rb–Sr DATING
Rb–Sr DATING
Oliver Nebel
Research School of Earth Sciences, The Australian
National University, Acton, ACT, Australia
Synonyms
Rb–Sr geochronology
Definition
Parent–daughter ratio: The ratio of rubidium (Rb) to
strontium (Sr). The ratio is commonly expressed as
87
Rb/86Sr, where the unstable Rb-87 isotope is referred
to as the parent nuclide, being approximately 27 % of all
rubidium. The daughter nuclide (Sr-87) is represented by
Sr-86, which is stable and not subject to radiogenic
ingrowth and constitutes approximately 9.9 % of all
strontium.
Nicolaysen diagram: A diagram of Rb-87 (abscissa) ver-
sus Sr-87 (ordinate), both normalized to Sr-86
(expressed as 87Rb/86Sr and 87Sr/86Sr) illustrating the
effect of age on the radiogenic decay of Rb-87 to Sr-87.
Initial isotope equilibrium: The assumption that at the time
of formation of a rock, all phases therein share the same Sr
isotope composition; a prerequisite for an isochron.
Isochron: A best-fit line of three or more phases in
a Nicolaysen diagram with its slope corresponding to an
age of phases that are in initial isotope equilibrium.
Two-point isochrons may be used and can have geologic
significance but do not allow constraints on their
reliability.
Errorchron: A median through three or more phases in
a Nicolaysen diagram that do not plot on a single regres-
sion line within the limits of their uncertainty. An age
corresponding to the slope of this median may have geo-
logic relevance but cannot be treated as an actual geologic
event.
Isochron initial: Common expression of the intercept of
the isochron with the ordinate in a Nicolaysen diagram
that marks the strontium isotope composition at the time
of formation which, by definition, is that of all phases con-
tributing to the isochron.
MSWD: Mean square of weighted deviates; a value
describing the degree of correlation of individual points
on an isochron with an ideal value of 1.
Historical background
The Rb–Sr dating technique is among the most widely
used and most powerful dating tools available in Earth sci-
ences. It is an effective means of dating igneous rocks or
metamorphic events and, under special circumstances,
can be applied to sedimentary sequences, ore deposits,
and past volcanic eruptions or to date fluid events in the
deep crust. The natural radioactivity of Rb-87 was discov-
ered by Campbell and Wood (1906) and first described as
a dating technique by radiochemist pioneer Otto Hahn
(Hahn and Walling, 1938; Hahn et al., 1943). Subsequent
improvements in precision and accuracy have contributed
significantly to a better understanding of Earth’s history.
With the advent and distribution of mass spectrometers,
the Rb–Sr technique has become the most important tool
in dating igneous rocks throughout Earth’s history and to
the infant stages of the solar system. It has been used to
date meteorites (e.g., Minster and Allègre, 1976; Minster
et al., 1982; Misawa et al., 1993), was among the first dat-
ing techniques applied to lunar samples returned from the
Apollo missions (e.g., Papanastassiou and Wasserburg,
1970; Papanastassiou et al., 1970; Murthy et al., 1971),
and has identified the oldest rocks of Earth (e.g., Hurst
et al., 1975; Moorbath et al., 1977a, Moorbath et al.,
1977b; De Laeter et al., 1985). Hence, Rb–Sr dating has
greatly contributed to developing an emerging idea of
our environment: an evolving and ever changing planet.
In recent years, the Rb–Sr dating technique has lost
some of its popularity to other dating tools. Other chro-
nometers, such as U–Pb in zircon, can, in principle, date
similar events to higher precision and, in the case of
high-temperature igneous emplacement events, higher
accuracy. Nonetheless, taking the properties of alkali and
alkaline earth elements into account, and refinements in
analytical techniques for Rb–Sr, it remains a powerful tool
to date rocks and metamorphic assemblages and is
re-establishing its place among the top dating techniques.
The concept of Rb–Sr dating
Rb–Sr dating is based on the radioactive decay of the iso-
tope Rb-87 (that today accounts for ~28 % of all rubidium)
to Sr-87 (~7 % of all strontium, Table 1) by beta decay, i.e.,
the emission of an electron. Rb-87 decays to Sr-87 with
a half-life of approximately 49 billion years. This makes
the Rb–Sr dating technique, in principle, suitable to date
samples from the infant stages of the solar system to very
recent igneous events, i.e., a few million years before
present.
For dating with the Rb–Sr technique, at least two
phases (either whole rocks or minerals) with different
Rb–Sr ratios are required. To directly compare Sr isotope
compositions of these phases, irrespective of their abso-
lute amount of Rb and Sr, all Sr-87 values are reported
Rb–Sr Dating, Table 1 Relative abundances of the stable
isotopes of Sr (average atomic mass of 87.62  1) and Rb
(average atomic mass of 85.4678  3) and the radioactive
isotope Rb-87 (De Laeter et al., 2003). Note that relative
abundances of Sr are variable in individual reservoirs due to the
radiogenic ingrowth of Sr-87. All uncertainties refer to the last
significant digit
Strontium (Sr) Rubidium (Rb)
Sr-84 0.0056  1
– Rb-85 0.7217  2
Sr-86 0.0986  1
Sr-87 0.0700  1 Rb-87 0.2783  2
Sr-88 0.8258  1
 Rb–Sr DATING 687

Rb–Sr Dating, Figure 1 (a) 87Rb/86Sr isochron diagram (Nicolaysen, 1961); showing the time-dependent evolution of three different
phases, (A–C) with variable 87Rb/86Sr that formed in initial isotopic equilibrium and thus have the same initial Sr isotope
composition. Points A0 –C0 reflect the present-day isotope composition of phases A–C, as measured by mass spectrometry. (b) The
same isotope evolution in an Sr isotope versus time plot with the relative evolution of the three different phases as a consequence of
their variable 87Rb/86Sr.

relative to a stable Sr isotope (Sr-86) that is not subject to
radiogenic ingrowth. Thus, all Sr isotope compositions are
reported as 87Sr/86Sr; 87Rb/86Sr ratios are reported
accordingly as 87Rb/86Sr following the equation:
AðRb-87Þ  M ðRbÞ  ½ppm Rb
ð1Þ
AðSr-86Þ  M ðSrÞ  ½ppm Sr
where A is the relative abundance, M the atomic mass, and
[ppm] the concentration of a species, usually given in parts
per million. The amount of radiogenic ingrowth of Sr-87
in at least two phases with different Rb/Sr corresponds to
the amount of decayed Rb-87. With knowledge of the
exact decay rate (the Rb decay constant, l87Rb), this
enables the calculation of radiogenic ingrowth and ulti-
mately, by integrating both phases, the time required for
this decay. This relationship is displayed in Figure 1,
where three phases with different 87Rb/86Sr (A, B, and
C) start with the same Sr isotope composition and with
time evolve to different 87Sr/86Sr (A0 , B0 , and C0 ), based
on their 87Rb/86Sr. Today, these phases plot on
a regression line (within analytical uncertainty), which is
usually highlighted by an MSWD ~ 1.
The isotope composition of a sample after a given time
t is expressed by the decay equation:
87
Sr 87
Sr 87
Rb  lt  higher external reproducibility, i.e., the variability in iso-
86 Sr
ðt Þ ¼ 86 Sr
ðt 0 Þ þ 86 Sr
e  1 ð2Þ
where 87Sr/86Sr(t0) is the initial isotope composition at
a time t0. If two or more Sr isotope compositions and their
87
Rb/86Sr are known, a time interval Dt (t1t0) and the Sr
isotope composition at t0 can be calculated using
a Nicolaysen diagram (Nicolaysen, 1961), where the age
difference Dt (or simply t) corresponds to the slope of
the regression line through those samples. Ideally more
than two samples are used for this concept to ensure that
the regression line indeed reflects an age; the more sam-
ples that are used in a Nicolaysen diagram plot, the more
likely it is that the regression is geologically meaningful.
Obviously, for two samples a regression can be calculated,
even though both samples may not be related to each
other, rendering any calculated age (or initial isotope com-
position) of such “two-point” isochrons potentially dubi-
ous without any independent age indicators.
Analytical techniques and limitations
The precision of an age derived from an isochron is con-
trolled by (i) the number of points used to obtain
a regression line, (ii) the spread in 87Rb/86Sr, (iii) the
degree to which the phases involved formed in isotope
equilibrium, and (iv) the errors on each individual analysis
of 87Rb/86Sr and Sr isotope composition. Where the first
three parameters are defined by the sample, the latter is
a matter of analytical capability. Strontium isotope com-
positions (87Sr/86Sr) vary from subtle variations among
whole rocks or mineral phases in the fifth decimal place
(i.e., 0.0000X) to large variations, especially in older sam-
ples. The 87Sr/86Sr can be routinely analyzed with high
precision and accuracy in samples with >50 ng Sr by ther-
mal ionization mass spectrometry (TIMS) to within
40 ppm. Modern TIMS instruments can achieve preci-
sions that are far better (up to 5 ppm); however, other
sources of error usually result in poorer accuracy and
tope compositions of a single sample from multiple ana-
lyses (including sample digestion). Error sources can
include procedural contaminants (commonly referred to
as “blank”) or isotope tracer calibrations. The procedural
blank can be minimized to insignificant levels by the use
of ultraclean reagents (e.g.., distilled acids, de-ionised
water, or purified chromatographic extraction resin) and
by chemical purifications in a clean room environment,
and modern laboratories generally fulfill this requirement.
The second parameter required for age determinations
is the 87Rb/86Sr, which is commonly determined by iso-
tope dilution using an isotope tracer. To ensure accurate
688 Rb–Sr DATING
results for isochron relationships, the Rb/Sr of a sample
should directly be determined in the same sample cut as
the one used for Sr isotope analysis so that the amount of
Rb directly corresponds to the radiogenic ingrowth of
Sr. The method of isotope dilution is currently the most
precise and accurate way to obtain this ratio in conjunction
with Sr isotope analyses; a precision of ~0.1–0.2 % on
87
Rb/86Sr values can be obtained. Alternative methods,
e.g., using single-collector mass spectrometry with induc-
tively coupled plasma source (ICP-MS), can be used, but
currently yield much poorer precision (~3–5 %) on the
elemental ratio. For isotope dilution, both Sr and Rb
(resident in the same material) are doped with a mixed iso-
tope tracer artificially enriched in one Sr and one Rb iso-
tope, preferably a low natural abundance isotope. This
“spike” is added to the sample prior to digestion to ensure
full spike-sample equilibrium. Knowledge of the concen-
trations and isotopic compositions of the spike, coupled
with accurate weights of sample and spike, can be used
to calculate Rb and Sr concentrations. Spiked Sr isotopes
are simultaneously analyzed with the radiogenic isotope
composition; spiked Rb isotopes are analyzed individu-
ally. For this, both Rb and Sr must be separated from the
rock matrix and from each other to avoid matrix effects
during the isotope analyses and, more importantly, iso-
baric interferences of Sr-87 and Rb-87, which will both
be detected as mass 87. The common method for this sep-
aration is chromatographic extraction of dissolved (and
spiked) sample material.
The calibration of the isotope tracer, a prerequisite to
ensure accurate determinations of 87Rb/86Sr, is not
straightforward for both Rb and Sr. Whereas for other
radiogenic systems, such as U–Pb or Sm–Nd, metals
can be weighed very accurately and used to prepare a well-
characterized standard solution, Rb metal is highly reac-
tive in the atmosphere and does not permit this approach.
Alternatives are Rb salts, which must be dried prior to ana-
lyses (due to their hygroscopic nature) and stoichiometry
must be assumed. A good example of a modern calibration
of a Rb–Sr spike using different salts for cross-calibration
is given in Nebel et al. (2011). A way to avoid
interlaboratory bias due to different spike calibrations is
through analyses of a Rb–Sr isotope standard. Unfortu-
nately, there are currently not many well-characterized
standards available. The most commonly used standard
is the NIST-607 K-feldspar standard. There are several
problems associated with K-feldspar (see below); how-
ever, Nebel and Mezger (2006) presented a possible solu-
tion using this standard to cross-calibrate isotope tracer
solutions. To obtain the most accurate result and to
eliminate error magnification by mixing disproportional
spike-sample ratios, general knowledge of sample concen-
trations for both, Rb and Sr, is desirable and often different
spikes with variable 87Rb/86Sr are used for high or low
Rb/Sr phases, respectively.
The conventional method for Rb isotope dilution anal-
ysis is by thermal ionization mass spectrometry (TIMS).
Mass spectrometric-induced isotope fractionation on
a TIMS instrument can only be controlled to within
0.5–1 % for elements with only two stable isotopes, such
as Rb, and for decades this has been a limiting factor in
Rb–Sr chronology. Recent developments in multi-
collector ICP-MS have enabled analyses of Rb isotopes
to <0.1 % (Waight et al., 2002a; Nebel et al., 2005), which
enables the determination of 87Rb/86Sr to a precision of
~0.2 % (Waight et al., 2002a; Nebel and Mezger, 2006),
which corresponds to precisions achieved for other dating
tools, e.g., Sm–Nd or Lu–Hf. In contrast to other dating
methods, a major disadvantage for Rb–Sr dating is that
no in-situ dating can be performed. To date, ion micro-
probe and laser ablation ICP-MS analyses, common
in-situ techniques applied to date high U/Pb minerals, do
not yield precise and accurate combined Sr isotope ratios
and parent–daughter ratios in high 87Rb/86Sr phases
required to obtain geologically significant ages (Waight
et al., 2002b; Vroon et al., 2008). The major reason for this
is the inability to adequately correct for the isobaric inter-
ference of Rb-87 on Sr-87.
The last key parameter to accurately date rocks and
mineral assemblages with Rb–Sr is the decay constant
(l87Rb). By successfully addressing all of the above
mentioned parameters, a precision in Rb–Sr ages can be
achieved that is comparable to other dating techniques.
However, a systematic bias can be introduced by insuffi-
cient knowledge of the Rb decay constant (Begemann
et al., 2001), i.e., the rate at which Rb decays. Recent
efforts to refine l87Rb have improved the precision and
accuracy of the constant to 0.1–0.2 % (Nebel et al., 2011;
Rotenberg et al., 2012); an error in the order of the analyt-
ical precision of the measurements (see Figure 2 for a com-
pilation of reported Rb-87 decay constants and their
precisions). Accordingly, all ages in this article have been
recalculated using an Rb-87 decay constant of
l87Rb ¼ 1.393  1011 (Nebel et al., 2011). The value of
l87Rb ¼ 1.42  1011, which was commonly used before
this revision, was proposed in 1977 by an international
commission on geochronology (Steiger and Jäger, 1977).
The revision of this value implies that all previously
reported Rb–Sr ages are ca. 2 % older than previously
thought.
Dating of igneous rocks (including volcanic ash)
The most widely used application of the Rb–Sr dating
technique is in determining the age of crystallization of
igneous rocks. For this, either multiple whole rocks
(with a genetic, often comagmatic relationship) or mineral
separates from a single rock that formed contemporane-
ously are used to date the emplacement event. This implies
that all species used for the dating share an identical Sr iso-
tope composition at the time of formation. Rock suites and
minerals need to be carefully evaluated to fulfill this crite-
rion, which is often not the case (Field and Raheim, 1979),
or to assess subsequent alteration by secondary processes.
Examples of such cases are detailed below. However, in
the ideal case of isotope coherence between samples and
 Rb–Sr DATING 689

Rb–Sr Dating, Figure 2 Compilation of reported decay constants with details on the methods of determination (Modified after
Nebel et al. (2011)). Gray dotted lines are recalculated values as reported in Begemann et al. (2001) (Data sources: Neumann and
Huster, 1976; Davis et al., 1977; Steiger and Jäger, 1977; Minster et al., 1982; Williams et al., 1982; Shih et al., 1985; Amelin and Zaitsev,
2002; Kossert, 2003; Rotenberg et al., 2012).

further textural or petrogenetic evidence for an igneous

origin of a rock, an age for a suite of samples can be calcu-
lated. For this, in the isochron diagram of Figure 1a, the
dashed lines correspond to a family of lines leading from
A to A’, B to B’, and so on, with the equation
y ¼ mx þ n ð3Þ
where all x–y coordinates in a Nicolaysen diagram
plot as points on a straight line: the isochron. In this dia-
gram the following parameters define the age:
x ¼ 87Rb/86Sr and y ¼ 87Sr/86Sr(t) for any given sample,
and n ¼ 87Sr/86Sr(t0) or initial 87Sr/86Sr is common to all
samples. The age t (in years) of the suite of samples then
corresponds to the slope m of this isochron with

m ¼ elt  1
The initial ratio n is the intercept of the isochron with
the ordinate; l is the Rb-87 decay constant. Obviously,
two or more co-genetic species are required for this dating
method. An example for one of the first dating attempts of
a suite of Archean igneous whole rocks from the Qorqut
igneous complex in Greenland (Moorbath et al., 1981) is
presented in Figure 3. These samples, assumed to be in ini-
tial isotope equilibrium, correspond to a calculated age
ð4Þ
Rb–Sr Dating, Figure 3 Isochron diagram of whole rocks from
a late Archean granite complex in Greenland (Moorbath
et al., 1981). Rb–Sr and Pb–Pb yield identical results, and all rocks
are presumably non-altered, indicating coherence of the isotope
systems and an emplacement event at 2.58  0.03 Ga.
690 Rb–Sr DATING
that is similar to the Pb–Pb age of the same samples, not-
ing that the MSWD of this example indicates some isotope
disturbance and thereby technically is an errorchron
(MSWD ~ 13).
Other samples that are seemingly in isotope equilibrium
(assuming no post-emplacement disturbance) are volcanic
ash horizons. These represent important time markers in
the Phanerozoic era because of (i) mineral species that
allow precise dating and (ii) their widespread spatial distri-
bution that allow correlations of stratigraphic horizons.
Although U–Pb in zircon is the most precise and com-
monly used method to date these events with extremely
precise resolution on timescales of thousands of years in
Quaternary eruptions (Crowley et al., 2007), Rb–Sr is, in
principle, applicable to dating of volcanic ash (Lanphere
and Baadsgaard, 2001). Ash often contains apatite,
K-feldspar, and biotite, which, in combination, share
a large spread in 87Rb/86Sr and are thus ideally suited for
Rb–Sr dating. However, in agreement with high-precision
U–Pb analyses of zircon and titanite from tuff (Schmitz
and Bowring, 2001), Rb–Sr on single grains from the Fish
Canyon Tuff, an enormous volcanic eruption approxi-
mately 28 Ma ago and often used as a dating standard in
40
Ar/39Ar chronology, has revealed that Rb–Sr systemat-
ics can yield different age information. This difference
may reflect crystal residence times in the underlying
magma chamber prior to eruption or that some crystals
are xenocrysts, i.e., assimilated from other sources, and
are thereby not in isotope equilibrium with the host melt
(Charlier et al., 2007), as is so often observed for U–Pb
in zircon. In young volcanic systems where timescales of
magma storage contribute more substantially to the abso-
lute age, this can be problematic and needs to be carefully
evaluated to obtain precise and accurate times of eruption.
This is of particular importance for cross-calibration of
geochronologic tools. However, in older systems,
a pristine sample suite provided, precise and accurate
Rb–Sr ages on bentonites are a useful alternative to
U–Pb or 40Ar/39Ar dating in obtaining absolute ages of
key stratigraphic horizons in Earth’s history (Williams
et al., 1982; Baadsgaard et al., 1988). This can be expected
to be more important in future studies, especially in light
of recent analytical improvements of Rb–Sr dating of
mica as an alternative to 40Ar/39Ar dating (Willigers
et al., 2004).
Isotope mixing versus isochron relationship
A potential complication for accurate age determinations
of igneous rocks with the Rb–Sr technique may result
from the mixing of two isotopically distinct magmatic res-
ervoirs at or close to the time of emplacement, where iso-
tope equilibrium is not achieved prior to crystallization.
Isotope mixing between two reservoirs cannot easily be
identified in a magmatic suite without additional age con-
straints or other petrologic genetic information. Mixing of
the two end-members produces a straight line in an isotope
evolution diagram, because both abscissa and ordinate are
normalized to the same isotope (Sr-86). Mixing thus
mimics isochron relationships and indicates an age that
is older than the actual igneous-rock-forming event. This
is particularly important for bulk rock analyses where geo-
chemical heterogeneity can be preserved more easily than
for mineral species derived from a single sample. This
effect is illustrated in Figure 4a for a granitic suite from
southern Germany (Siebel, 1994). The slope of the iso-
chron using whole rocks calculated to the time of emplace-
ment of ~325 Ma indicates a pseudo-chron age of
~100 Ma (with a typical, almost ideal MSWD ¼ 1.4),
which is superimposed on the actual age. Because mixing
between two different reservoirs on a whole rock scale is
not a perfect process on a larger scale (kilometers), the cor-
relation may exhibit an MSWD > 1 resulting in an
errorchron relationship. Mixing may be identified in a
3-dimensional plot of 87Rb/86Sr versus 87Sr/86Sr versus
1/Sr [ppm] (Wendt, 1993), as shown in Figure 4b.
Isotopic closure and “cooling ages”
Isotope dating requires all phases, whole rocks or min-
erals, to be in isotopic equilibrium at the time of formation.
However, for large igneous bodies, cooling of magma and
the mineral phases that crystallized from it do not occur
instantly and contemporaneously. Because conductive
heat transfer in igneous rocks is an inefficient process,
the thermal gradient between the intrusive magma and
the ambient rocks may persist for a long time, often for
millions of years. Minerals in the igneous body remain
open systems with exchange of parent–daughter isotopes
by volume diffusion. Only when a certain temperature is
reached are minerals considered “closed” for an isotope
system. This temperature is called the closure temperature,
TC (Dodson, 1973). Alternative scenarios are possible, in
which a system is closed rather quickly after emplacement
but is subject to interdiffusion of elements between min-
erals (Jenkin et al., 2001, 1995). For phases with high
87
Rb/86Sr, such as phengite, biotite, or K-feldspar, relative
TC has been empirically determined (e.g., Von
Blanckenburg et al., 1989). They vary between 500  C
and 300  C and may be dependent on grain size, chemis-
try, pressure, or fluid content (Villa, 1998). The TC for
Rb–Sr is lower than for most other isotope systems includ-
ing K–Ar or 40Ar/39Ar. This effect has long been known
(Armstrong et al., 1966; Verschure et al., 1980; Delmoro
et al., 1982) and used to calculate cooling paths for igne-
ous and metamorphic systems (e.g., Von Blanckenburg
et al., 1989).
Rb–Sr analyses of different mineral phases can still
yield coherent ages with a well-defined isochron relation-
ship among different phases, even if all phases closed long
after the actual emplacement event. This dates the time of
cooling through isotope closure. Although initially con-
sidered a matter of grain size and diffusional processes
(Dodson, 1973), element exchange between coexisting
phases may be equally important for Rb–Sr (Jenkin
et al., 2001). Thus, Rb–Sr analyses of different grain
 Rb–Sr DATING 691

Rb–Sr Dating, Figure 4 (a) Pseudo-isochron produced by a two-component mixing between isotopically distinct reservoirs
preserved in granites from Southern Germany (Data from Siebel, 1994). The data are age corrected to the time of formation at 320 Ma,
as deduced independently by 40Ar/39Ar isotope dating. (b) The reciprocals of the Sr concentration show a straight line correlation
with isotopic composition that indicates mixing of two components with variable Sr concentrations (Wendt, 1993). The combined
pseudo-chron and isochron slopes yield an apparent, yet meaningless “age” of 423  41 Ma (MSWD ¼ 1.7; n ¼ 7; (Siebel, 1994).

fractions should not be used to determine cooling paths by
Rb–Sr dating alone and are better accompanied by ana-
lyses of other phases with other isotope systems, which,
in turn, have other closure temperatures. A cooling path
may then be determined in a temperature–time diagram.
Additional complications may, however, arise, due to dif-
fusion below the TC. Detailed micro-drill investigations
(on a sub-crystal scale) in K-feldspar grains from
a slowly cooled plutonic body indicate Sr isotope hetero-
geneity, likely induced by diffusion (Siebel et al., 2005),
which can compromise bulk mineral analyses.
Timescales of isotope closure vary considerably and, as
noted, depend on a number of parameters. A good exam-
ple is cooling of the Great Dyke, a layered mafic intrusion
in the Zimbabwe Craton. This massive mantle-derived
igneous body is an extreme case of a magmatic body that
intruded into presumably hot country rock, such that the
thermal gradient between the intrusion and the ambient
country rocks was small. Accordingly, the intrusion
cooled very slowly, and reported Rb–Sr ages are
~23 Myr younger than ages determined for the same rocks
using systems with higher closure temperatures (e.g.,
U–Pb in zircon or sphene) (Nebel and Mezger, 2008;
Figure 5). The age deduced from ortho- and
clinopyroxenes is identical to the age calculated from
combined pyroxene–biotite, which indicates a similar TC
for both minerals. Summing information from the avail-
able literature then allows a relative TC for Rb–Sr that
is phengite > K-fsp > biotite  clinopyroxene.
Dating metamorphic rocks
For the dating of rocks, it is required that, after the time of
its formation (termed t0), a system remains undisturbed
with respect to Rb–Sr systematics, i.e., that both Rb and
Sr elemental and isotope abundances remain a closed sys-
tem. That means that the amount of the parent nuclide
Rb-87 must not be altered (by removal or addition) by
exchange with the surrounding environment and that the
daughter product Sr-87 must correspond to the amount
of Rb responsible for the ingrowth. If a rock is subject to
metamorphism, i.e., physical and/or chemical processes
that lead to a change in its texture and mineralogical and/or
chemical composition, this requirement may be
compromised.
Metamorphic conditions can reach temperatures far
exceeding TC for Rb–Sr so that a system can “reset.”
Metamorphism and the associated formation of minerals
at the expense of existing ones often also include the
release of fluids or involve fluids from outside the rock
(metasomatism), which promote the diffusion of ions
and/or the mobility of elements. This is especially the case
for the classical “mobile” large ion lithophile elements
(LILE) that include most alkali and alkaline earth elements
(including Rb and Sr), and often the Rb and Sr budget in
such rock associations can be significantly altered. Isotope
systems are very sensitive to these processes, even without
an apparent ( i.e., visible) metamorphic overprint, and they
may even affect a rock below the igneous TC for an isotope
system, e.g., because of dislocation effects in the crystal
lattice that occur during the radioactive decay and isotope
ingrowth, up to the point when a crystal becomes
metamict. This susceptibility is often reflected in age rela-
tionships of old rocks that yield younger Rb–Sr ages than
their presumed emplacement age (deduced from other
chronometers) or no age information at all (isotope scatter
that does not permit construction of an isochron). With the
already relatively low TC for Rb–Sr in most minerals
(300–500  C) compared to temperatures in metamorphic
reactions that can easily exceed these values, this feature
692 Rb–Sr DATING
Rb–Sr Dating, Figure 5 (a) Rb–Sr age for the Great Dyke, Zimbabwe. (b) Cooling path of the intrusion by comparison with other
chronometers (Modified after (Nebel and Mezger, 2008)).
can have considerable impact on many rock assemblages
in old crustal terranes and has been a long-standing prob-
lem in Rb–Sr dating (Fairbairn et al., 1961; Riley and
Compston, 1962).
However, this seemingly strong limitation of the Rb–Sr
chronometer can, in turn, be used advantageously to date
“secondary events,” i.e., the time of metamorphic reset-
ting, hereafter termed tm. Internal mineral isochrons for
individual whole rocks can yield age information on the
thermal event, whereas on a whole rock scale the original
emplacement age of an igneous rock can be preserved. For
this to occur, the system is closed on a bulk rock scale and
preserves its amount of accumulated Sr-87 and Rb con-
centration, and only a redistribution between mineral
phases took place. This can be the case if little or no fluids
are involved in the metamorphic reactions, as fluids often
act as a transport medium for Rb and Sr beyond the scales
of bulk rock samples. The combination of igneous whole
rock and metamorphic mineral isochrons is illustrated in
Figure 6a, where whole rocks 1, 2, and 3 yield the original
emplacement age (t0), whereas the constituent minerals of
whole rock 2 in combination with the bulk rock yield the
time of metamorphic reset (tm). At tm, minerals achieve
isotope equilibrium with 87Sr/86Sr similar to that of their
respective whole rock, and radiogenic ingrowth then starts
anew as a function of the Rb–Sr in each mineral. This
ideal case scenario of metamorphic reset on a mineral
scale and preservation of primary age information on
a whole rock scale has been observed in the Carn
Chuinneag Intrusion in Scotland (Long, 1964; Pidgeon
and Johnson, 1974). Unfortunately, in most cases the
cooling of metamorphic terranes can be very slow such
that individual mineral phases do not close contemporane-
ously, so that their isotopic clock does not start ticking at
the same time. This is then reflected in an errorchron
relationship that, at best, only allows the broad determina-
tion of cooling paths or an approximation of tm that often
fails modern requirements on precision.
The concept of isotope resetting is illustrated in
Figure 6b (redrawn after Faure (1986) and following the
concept of Riley and Compston (1962)), where the isoto-
pic evolution over time of different mineral phases with
variable 87Rb/86Sr is shown. It is important to note that
some phases reset earlier than others (or may solely be
affected by the metamorphic events), all of which are
a function of their individual metamorphic TC (which as
stated earlier can also include isotope diffusion)
(Ganguly and Ruiz, 1987). In most cases, Rb–Sr isotope
systematics yield errorchron ages that may be geologically
meaningless. However, in the worst case (often if only two
or three data points are employed or they have a relatively
small spread in 87Rb/86Sr), Rb–Sr data do not allow dis-
tinctions between a true isochron or meaningless mixing
age populations (Riley and Compston, 1962). This can
be the case if whole rocks are reset at different times or ele-
mental and isotope exchange beyond whole rock scales
occurred, but data coincidentally plot on a straight line
within their analytical uncertainty. In the absence of
other petrologic or isotopic parameters, the determination
of model ages of, for example, K-feldspar and mica
(tmica, tK-fsp) can assist in clarification of the geologic his-
tory of a sample, as, for a given initial Sr isotope composi-
tion, these ages should be identical. These model ages then
should correspond to the correct igneous age. However,
the initial 87Sr/86Sr for this calculation is often unknown,
and iterative approaches by multiple phase analyses are
often too imprecise to yield useful age information.
In many cases, one (or more) metamorphic event(s)
results in a complete redistribution and homogenization
of Rb–Sr systematics in all phases and on whole rock
 Rb–Sr DATING 693

Rb–Sr Dating, Figure 6 (a, b) Modified after Faure (1986). (a) Concept of isotope resetting. Whole rocks 1, 2, and 3 preserve an
isochron of the original igneous event with the age t0, whereas minerals of whole rock 2 give the metamorphic age tm. Half-open
symbols are metamorphic assemblages; open symbols are the reset assemblages of WR2 at the time tm. At this time WR2 is internally
homogenized with an initial 87Sr/86Sr of SrI-m B: isotope evolution of various minerals of a whole rock during a metamorphic event. All
phases evolve as a function of their 87Rb/86Sr. At the time tm, all phases are reunited and homogenized in their Sr isotope composition
with a new initial Sr isotope composition Sr(I)m. The whole rock remains a closed system and is not affected by this event. Corrections
of metamorphic phases to an incorrect age can yield false initials (Sr(I)false); reciprocally, an incorrect initial Sr isotope assumption will
provide a biased model age (tmica, tK-fsp).

scales for very large metamorphic terranes. This has been
shown by many early Rb–Sr studies on Archean metamor-
phic terranes in old cratons (e.g., Hurst et al., 1975;
Moorbath et al., 1977a; Moorbath et al., 1977b; Kalsbeek,
1981). A complete reset is also possible during the meta-
morphic crystallization of sediments, so that metamorphic
Rb–Sr ages tm can be determined for para- and
orthogneisses alike.
Obviously, without additional constraints it is not possi-
ble to predict to which extent isotopic exchange and
homogenization occurred, and the possibility of isotope
“inheritance” has been reported even for isotope systems
with very high TC, such as the Sm–Nd system (Roth
et al., 2013). On the other hand, isotope systems with
a high closure temperature may not yield metamorphic
ages as they may only be partially reset, whereas Rb–Sr
may have been fully re-equilibrated. Such a case is
reported for the Yilgarn block in Western Australia
(Figure 7), where Rb–Sr systematics give a precise timing
of metamorphic resetting (McCulloch et al., 1983).
Sm–Nd data, however, preserved its isotope character
during this metamorphic event and can be used to
694
Rb–Sr Dating, Figure 7 Rb–Sr whole rock isochron for
metamorphic gneisses from the Yilgarn Craton in Western
Australia yields a precise metamorphic age, whereas Sm–Nd
isotope systematics do not permit tight age constraints
(2800  150 Ma). When age corrected to the Rb–Sr age, the Nd
isotope data is interpreted to reflect variable juvenile protoliths
of the gneiss (McCulloch et al., 1983).
reconstruct the history of the precursor of the metamor-
phic terrane. Therefore, if placed in a geologic context,
the Rb–Sr systematics of metamorphic rocks have the
potential to yield precise and accurate age information
and may otherwise simply be used to test the extent of iso-
tope equilibrium (for a given age) by correction for radio-
genic ingrowth to the respective initial ratios.
Dating fluid events and ore deposition
The fluid mobility of LILE implies that Rb and subse-
quently Sr can be mobilized in crustal fluids during low-
to high-grade metamorphic events. Whereas this is often
considered a problem for igneous rocks (they do not stay
as a closed isotope system), there is a potential for dating
such events with the Rb–Sr technique. In addition, mobi-
lization of elements is often related to the formation of
ores, and knowledge of the timing of ore formation is of
great importance in economic geology or in regional geo-
logic investigations. The Rb–Sr technique has long been
exploited in dating base metal sulfide deposits, such as
Mississippi Valley-type deposits (Nakai et al., 1990;
Brannon et al., 1992; Nakai et al., 1993; Christensen
et al., 1995), which are associated with large-scale brine
migrations under low temperatures (i.e., <200  C; Nakai
et al., 1993), or lode gold deposits that are either porphyric
or epithermal in nature (Yang and Zhou, 2001). These ores
often contain mineral assemblages with parent–daughter
ratios that are unsuitable for dating, which makes age
determinations of the mineralization notoriously difficult.
Rb–Sr DATING
Rb–Sr Dating, Figure 8 Rb–Sr isochron for two individual veins
from a gold deposit in China (Yang and Zhou, 2001). Both
isochrons indicate an identical age of deposition, yet display
variations in their initial Sr isotope composition, indicative of
distinct fluid sources. If combined, both veins would display an
errorchron relationship.
Sphalerite and pyrite are common sulfides formed dur-
ing mineralization events and appear to contain
a reasonable spread in Rb/Sr and enough material to
obtain isotope results on mineral separates of reasonable
size and quantity. This is of particular importance as sam-
ple heterogeneity (in 87Sr/86Sr) can be significant (i.e.,
outside analytical reproducibility) in crystals that grow
in the presence of fluids as fluid sources and compositions
can vary in these systems. Hence, variations in 87Rb/86Sr
and 87Sr/86Sr may occur between neighboring samples
or within growth domains of single samples. In the latter
case, full isotope equilibrium of all domains is required,
which, under hydrothermal conditions, is not always the
case. However, some fluid systems can be deposited on
timescales of minutes (e.g., Weatherly and Hanley 2013)
so that isotope equilibrium can likely be assumed in these
systems.
An example of such an application is shown in Figure 8
that shows pyrite-only isochrons for a lode gold deposit
from China (Yang and Zhou, 2001), where the spread in
87
Rb/86Sr within individual crystals and/or batches of
crystals is used as an advantage for dating. Pyrite separates
from two individual veins yield two well-defined iso-
chrons with a mean age of 125.2  0.9 Ma. Both individ-
ual determinations give confidence that gold deposition
from fluids in the veins occurred at this time. However,
as illustrated in the figure, the source of the fluids is distin-
guished by their initial isotope compositions, so that
a joint-age determination would inevitably result in an
errorchron.
 Rb–Sr DATING 695

Rb–Sr Dating, Figure 9 Whole rock–mineral isochron (including a melt vein but excluding a leachate fraction) from lunar meteorite
LAP02205 (Data taken from Rankenburg et al. (2007)). The Rb–Sr isochron yields an almost ideal MSWD of 1.07 indicating isotope
equilibrium. (b) Comparison of the Rb–Sr age with other age data for the same meteorite; the Rb–Sr age is the most precise and likely
indicates the time of formation of the rock. 40Ar/39Ar ages (Fernandes et al. (2009)), Sm–Nd age (Rankenburg et al. (2007)), U–Pb age
(Anand et al. (2006)).

Dating extraterrestrial material
The Rb–Sr technique was widely applied to meteorite and
lunar research, predominantly in the 1970s and 1980s,
with a wide number of applications (e.g., Papanastassiou
et al., 1970; Papanastassiou and Wasserburg, 1970;
Allègre et al., 1975; Birck et al., 1975; Minster and
Allègre, 1976; Birck and Allègre, 1978; Minster et al.,
1982; Nyquist et al., 1986; Shih et al., 1992; Misawa
et al., 1993). Even though extraterrestrial whole rock
samples and mineral species have limited spread in
87
Rb/86Sr, their extreme age makes them suitable for dat-
ing with this technique. However, the use of Rb–Sr in
696 Rb–Sr DATING
cosmo-chronology has become less popular in recent
years. The reason for this is the improvement in other dat-
ing methods and intense isotope tracing of reservoirs that
has led to a much more refined chronology of early solar
system events and the identification of multiple processes
and events. This renders most whole rock dating
approaches incorrect as they would lead to comparison
of mixed populations of rocks producing geologically
meaningless ages (Field and Raheim, 1979). First analyses
of a series of bulk meteorites yielded a good correlation
among different groups of meteorites. Whereas absolute
age determinations were hindered by the issue of decay
constant uncertainties, these authors used the data to cal-
culate an Rb decay constant of 1.402  1011 year1
(Minster et al., 1982), a value that is close to the recently
refined value (Nebel et al., 2011; Rotenberg et al., 2012).
However, it is now clear that, while this approach yielded
important information at the time, replicate analyses using
modern Rb–Sr methods would most certainly reveal iso-
tope disequilibrium among these samples, as decades of
meteorite research has revealed substantial isotope hetero-
geneity among early solar system reservoirs (cf. Hans
et al., 2013).
A second serious problem for Rb–Sr dating of extrater-
restrial samples is that of terrestrial alteration in most
meteorites (if not reported as “falls”). Rb–Sr dating of
such material should be accompanied by other chronome-
ters, intense evaluation of trace element abundances that
allow identification of primary igneous processes or sec-
ondary alteration, and/or careful treatment of samples
(such as leaching). If these conditions are met, modern
Rb–Sr dating can provide very precise and accurate infor-
mation on the history of solar system material
(Rankenburg et al., 2007). In addition, if an isochron rela-
tionship can be established, the initial Sr isotope ratio of
this isochron can provide invaluable information on the
time-integrated evolution of the greater source domain,
such as the Moon (Papanastassiou and Wasserburg,
1970; Papanastassiou et al., 1970). In addition, despite
the difference in age and origin for many meteorites, isoto-
pic closure also applies to metamorphic events in space,
and similar to terrestrial analogues, the Rb–Sr technique
is capable of obtaining ages for high-temperature collision
events, for example, on Mars (Nyquist et al., 1979), or
meteorites that have been subject to metamorphic
overprinting (e.g., Birck and Allègre, 1978).
A good example of modern Rb–Sr dating of extrater-
restrial material is the lunar meteorite La Paz Icefield
02205 (LAP02205). Rankenburg et al. (2007) analyzed
bulk rocks, pyroxene, and feldspar separates (and addi-
tional phases) and obtained a Rb–Sr age of
3010  14 Ma (illustrated in Figure 9a). This age agrees
with, but is more precise than, the corresponding U–Pb
age of 2.93  0.15 Ga (Anand et al., 2006) or the Sm–
Nd age of 2992  85 Ma (Rankenburg et al., 2007) on
the same sample (Figure 9b). The latter system has also
been subject to disturbance in phosphates so that Rb–Sr
appears to be more robust with respect to the nonsilicates
that often occur in meteorites. Notably, 40Ar/39Ar results
of 2985  16 Ma and 2874  56 Ma (Fernandes
et al., 2009) on this meteorite appear to be somewhat
younger, which at least in the first instance may be related
to inaccuracies in the Ar decay constants (Begemann
et al., 2001). Irrespective of the cause of difference in
absolute ages using various chronometers, it seems appar-
ent that Rb–Sr can provide accurate and very precise age
information that cannot otherwise be determined.
Conclusion
The Rb–Sr dating technique is a widely used technique in
obtaining ages of igneous or metamorphic events and may
be applied to dating volcanic sequences, ore deposition, or
fluid-related events. In comparison with other dating tech-
niques, there are several disadvantages, for example, its
relatively low closure temperature and consequent suscep-
tibility to isotope disturbance. However, if applied in
a petrogenetic context or on samples that allow other
means of identifying (hydro-)thermal overprints, it is
a useful and reliable dating tool. In addition, it can also
be used to date low-temperature events or the prolonged
cooling history of igneous bodies or greater crustal ter-
ranes. The advent of modern analytical instrumentation
now enables determination of high-precision Rb–Sr ages
that are comparable in precision and accuracy with other
dating techniques. Limitations such as poorly constrained
thermal histories of samples and the necessity of species
(mineral or rock) with a considerable spread in Rb/Sr
ratios remain.
Bibliography
Allègre, C. J., Birck, J. L., Fourcade, S., and Semet, M. P., 1975.
Rubidium-87/Strontium-87 age of juvinas basaltic achondrite
and early igneous activity in solar-system. Science, 187(4175),
436–438.
Amelin, Y., and Zaitsev, A. N., 2002. Precise geochronology of
phoscorites and carbonatites: the critical role of U-series disequi-
librium in age interpretations. Geochimica et Cosmochimica
Acta, 66(13), 2399–2419.
Anand, M., et al., 2006. Petrology and geochemistry of LaPaz
Icefield 02205: a new unique low-Ti mare-basalt meteorite.
Geochimica et Cosmochimica Acta, 70(1), 246–264.
Armstrong, R. L., Jäger, E., and Eberhardt, P., 1966. A comparison
of K-Ar and Rb-Sr ages in Alpine biotites. Earth and Planetary
Science Letters, 1, 19.
Baadsgaard, H., Lerbekmo, J. F., and McDougall, I., 1988.
A radiometric age for the cretaceous-tertiary boundary based
on K–Ar, Rb–Sr, and U–Pb ages of bentonites from Alberta,
Saskatchewan and Montana. The Canadian Journal of Earth
Sciences, 25, 1088–1097.
Begemann, F., et al., 2001. Call for an improved set of decay con-
stants for geochronological use. Geochimica et Cosmochimica
Acta, 65(1), 111–121.
Birck, J. L., and Allègre, C. J., 1978. Chronology and chemical his-
tory of parent body of basaltic achondrites studied by Rb-87-Sr-
87 method. Earth and Planetary Science Letters, 39(1), 37–51.
Birck, J. L., Fourcade, S., and Allègre, C. J., 1975. Rb-87 Sr-86 age
of rocks from Apollo-15 landing site and significance of internal
isochrons. Earth and Planetary Science Letters, 26(1), 29–35.
 Rb–Sr DATING
Brannon, J. C., Podosek, F. A., and McLimans, R. K., 1992. Alle-
ghenian age of the Upper Mississippi Valley zinc-lead deposit
determined by Rb–Sr dating of sphalerite. Nature, 356(6369),
509–511.
Campbell, N. R., and Wood, A., 1906. Radioactivity of the alkali
metals. Proceedings of the Cambridge Philosophical Society,
14, 15–21.
Charlier, B. L. A., Bachmann, O., Davidson, J. P., Dungan, M. A.,
and Morgan, D. J., 2007. The upper crustal evolution of a large
silicic magma body: evidence from crystal-scale rb–sr isotopic
Heterogeneities in the fish canyon magmatic system, Colorado.
Journal of Petrology, 48(10), 1875–1894.
Christensen, J. N., Halliday, A. N., Leigh, K. E., Randell, R. N.,
and Kesler, S. E., 1995. Direct dating of sulfides by Rb–Sr:
a critical test using the polaris Mississippi valley-type Zn–Pb
deposit. Geochimica et Cosmochimica Acta, 59(24),
5191–5197.
Crowley, J. L., Schoene, B., and Bowring, S. A., 2007. U–Pb dating
of zircon in the Bishop Tuff at the millennial scale. Geology,
35(12), 1123–1126.
Davis, D. W., Gray, J., Cumming, G. L., and Baadsgaard, H., 1977.
Determination of Rb-87 decay constant. Geochimica et
Cosmochimica Acta, 41(12), 1745–1749.
De Laeter, J. R., et al., 1985. Rb–Sr, Sm–Nd and Pb–Pb geochro-
nology of ancient gneisses from Mt. Narryer, Western Australia.
Australian Journal of Earth Sciences, 32(4), 349–358.
De Laeter, J. R., et al., 2003. Atomic weights of the elements:
review 2000. Pure and Applied Chemistry, 75(6), 683–800.
Delmoro, A., Puxeddu, M., Dibrozolo, F. R., and Villa, I. M., 1982.
Rb–Ar and K–Ar ages in minerals at temperatures of
300 C–400 C from deep wells in the Larderello geothermal
field (Italy). Contributions to Mineralogy and Petrology, 81(4),
340–349.
Dodson, M. H., 1973. Closure temperature in cooling geochrono-
logical and petrological systems. Contributions to Mineralogy
and Petrology, 40(3), 259–274.
Fairbairn, H. W., Hurley, P. M., and Pinson, W. H., 1961. The rela-
tion of discordant Rb–Sr mineral and whole rock ages in an igne-
ous rock to its time of crystallization and subsequent Sr-87-Sr86
metamorphism. Geochimica et Cosmochimica Acta, 23(1–2),
135–144.
Faure, G., 1986. Principles of Isotope Geology. New York: Wiley,
p. 589.
Fernandes, V. A., Burgess, R., and Morris, A., 2009. Ar-40-Ar-39
age determinations of lunar basalt meteorites Asuka 881757,
Yamato 793169, Miller Range 05035, La Paz Icefield 02205,
Northwest Africa 479, and basaltic breccia Elephant Moraine
96008. Meteoritics & Planetary Science, 44(6), 805–821.
Field, D., and Raheim, A., 1979. Geologically meaningless Rb–Sr
total rock isochron. Nature, 282(5738), 497–499.
Ganguly, J., and Ruiz, J., 1987. Time-temperature relation of min-
eral isochrons – a thermodynamic model, and illustrative exam-
ples for the Rb–Sr system. Earth and Planetary Science Letters,
81(4), 338–348.
Hahn, O., and Walling, E., 1938. Über die Möglichkeit geologischer
Altersbestimmungen rubidiumhaltiger Mineralen und Gesteine.
Zeitschrift für Anorganische und allgemeine Chemie, 236,
78–82.
Hahn, O., Strassman, F., Mattauch, J., and Ewald, H., 1943.
Geologische Altersbestimmungen mit der Strontiummethode.
Chemische Zeitung, 67, 55–56.
Hans, U., Kleine, T., and Bourdon, B., 2013. Rb–Sr chronology of
volatile depletion in differentiated protoplanets: BABI, ADOR
and ALL revisited. Earth and Planetary Science Letters, 374,
204–214.
Hurst, R. W., Bridgwater, D., Collerson, K. D., and Wetherill, G. W.,
1975. 3600 m.y. Rb–Sr ages from very early Archean gneisses
697
from Saglek Bay, Labrador. Earth and Planetary Science Let-
ters, 27(3), 393–403.
Jenkin, G. R. T., Rogers, G., Fallick, A. E., and Farrow, C. M., 1995.
Rb–Sr closure temperatures in bi-mineral rocks – a mode effect
and test for different diffusion models. Chemical Geology, 122
(1–4), 227–240.
Jenkin, G. R. T., Ellam, R. M., Rogers, G., and Stuart, F. M., 2001.
An investigation of closure temperature of the biotite Rb–Sr sys-
tem: the importance of cation exchange. Geochimica et
Cosmochimica Acta, 65(7), 1141–1160.
Kalsbeek, F., 1981. The northern extend of the Archean basement of
Greenland – a review of Rb–Sr whole rock ages. Precambrian
Research, 14(3–4), 203–219.
Kossert, K., 2003. Half-life measurements of Rb-87 by liquid scintilla-
tion counting. Applied Radiation and Isotopes, 59(5–6), 377–382.
Lanphere, M. A., and Baadsgaard, H., 2001. Precise K–Ar, Ar-40/
Ar-39, Rb–Sr and U/Pb mineral ages from the 27.5 Ma Fish
Canyon Tuff reference standard. Chemical Geology, 175(3–4),
653–671.
Long, L. E., 1964. Rb–Sr chronology of carn Chuinneag intrusion,
Ross Shire, Scotland. Journal of Geophysical Research, 69(8),
1589–1597.
McCulloch, M. T., Compston, W., and Froude, D., 1983. Sm-Nd
and Rb–Sr dating of Archean gneisses, eastern Yilgarn Block,
Western Australia. Journal of the Geological Society of Austra-
lia, 30(1–2), 149–153.
Minster, J. F., and Allègre, C. J., 1976. Rb-87-Sr-87 history of
Norton-County Enstatite Achondrite. Earth and Planetary
Science Letters, 32(2), 191–198.
Minster, J. F., Birck, J. L., and Allègre, C. J., 1982. Absolute age of
formation of chondrites studied by the Rb-87-Sr-87 method.
Nature, 300(5891), 414–419.
Misawa, K., Tatsumoto, M., Dalrymple, G. B., and Yanai, K., 1993.
An extremely low U/Pb source in the Moon – U–Th–Pb, Sm–Nd,
Rb–Sr, and Ar-40 Ar-39 isotopic systematics and age of lunar
meteorite Asuka 881757. Geochimica et Cosmochimica Acta,
57(19), 4687–4702.
Moorbath, S., Allaart, J. H., Bridgwater, D., and McGregor, V. R.,
1977a. Rb–Sr ages of early Archean Supracrustal rocks and
Amitsoq gneisses at Isua. Nature, 270(5632), 43–45.
Moorbath, S., Wilson, J. F., Goodwin, R., and Humm, M., 1977b.
Further Rb–Sr age and isotope data on early and late Archean
rocks from Rhodesian Craton. Precambrian Research, 5(3),
229–239.
Moorbath, S., Taylor, P. N., and Goodwin, R., 1981. Origin of gra-
nitic magma by crustal remobilization – Rb–Sr and Pb–Pb geo-
chronology and isotope geochemistry of the late Archean Qorqut
granite complex of southern West Greenland. Geochimica et
Cosmochimica Acta, 45(7), 1051–1060.
Murthy, V. R., Evensen, N. M., Jahn, B., and Coscio, M. R., 1971.
Rb–Sr ages and elemental abundances of K, Rb, Sr, and Ba in
samples from ocean of storms. Geochimica et Cosmochimica
Acta, 35(11), 1139–1153.
Nakai, S., Halliday, A. N., Kesler, S. E., and Jones, H. D., 1990.
Rb–Sr dating of sphalerites from Tennessee and the genesis of
Mississippi Valley-type ore deposits. Nature, 346(6282),
354–357.
Nakai, S., et al., 1993. Rb–Sr dating of sphalerite from mississippi
valley-type (MVT) ore deposits. Geochimica et Cosmochimica
Acta, 57(2), 417–427.
Nebel, O., and Mezger, K., 2006. Reassessment of the NBS
SRM-607 K-feldspar as a high precision Rb/Sr and Sr isotope
reference. Chemical Geology, 233(3–4), 337–345.
Nebel, O., and Mezger, K., 2008. Timing of thermal stabilization of
the Zimbabwe Craton deduced from high-precision Rb–Sr chro-
nology, Great Dyke. Precambrian Research, 164(3–4),
227–232.
698 Rb–Sr GEOCHRONOLOGY (IGNEOUS ROCKS)
Nebel, O., Mezger, K., Scherer, E., and Münker, C., 2005. High pre-
cision determinations of 87Rb/85Rb in geologic materials by
MC-ICP-MS. International Journal of Mass Spectrometry,
246, 10–18.
Nebel, O., Scherer, E. E., and Mezger, K., 2011. Evaluation of the
Rb-87 decay constant by age comparison against the U–Pb sys-
tem. Earth and Planetary Science Letters, 301(1–2), 1–8.
Neumann, W., and Huster, E., 1976. Discussion of the 87Rb half-life
by absolute counting. Earth and Planetary Science Letters, 33,
277–288.
Nicolaysen, L. O., 1961. Graphic interpretations of discordant age
measurements on metamorphic rocks. Annals of the New York
Academy of Sciences (Geochronology of Rock Systems), 91,
198–206.
Nyquist, L. E., Wooden, J., Bansai, B., Wiesmann, H.,
McKay, G., and Bogard, D. D., 1979. Rb–Sr age of the
Shergotty achondrite and implications for metamorphic resetting
of isochron ages. Geochimica et Cosmochimica Acta, 43,
1057–1074.
Nyquist, L. E., et al., 1986. Rb–Sr and Sm–Nd Internal isochron
ages of a subophitic basalt clast and a matrix sample from the
Y75011 eucrite. Journal of Geophysical Research-Solid Earth
and Planets, 91(B8), 8137–8150.
Papanastassiou, D. A., and Wasserburg, G. J., 1970. Rb–Sr ages from
ocean-of-storms. Earth and Planetary Science Letters, 8(4), 269–278.
Papanastassiou, D. A., Wasserburg, G. J., and Burnett, D. S., 1970.
Rb–Sr ages of lunar rocks from the sea of tranquillity. Earth and
Planetary Science Letters, 8(1), 1–19.
Pidgeon, R. T., and Johnson, M. R. W., 1974. Comparison of zircon
U–Pb and whole rock Rb–Sr systems in three phases of Carn
Chuinneag granite, Northern Scotland. Earth and Planetary Sci-
ence Letters, 24(1), 105–112.
Rankenburg, K., Brandon, A. D., and Norman, M. D., 2007.
A Rb–Sr and Sm–Nd isotope geochronology and trace element
study of lunar meteorite LaPaz Icefield 02205. Geochimica et
Cosmochimica Acta, 71(8), 2120–2135.
Riley, G. H., and Compston, W., 1962. Theoretical and technical
aspects of Rb–Sr geochronology. Geochimica et Cosmochimica
Acta, 26(DEC), 1255–1281.
Rotenberg, E., Davis, D. W., Amelin, Y., Ghosh, S., and Bergquist,
B. A., 2012. Determination of the decay-constant of Rb-87 by
laboratory accumulation of Sr-87. Geochimica et Cosmochimica
Acta, 85, 41–57.
Roth, A. S. G., et al., 2013. Inherited Nd-142 anomalies in Eoarchean
protoliths. Earth and Planetary Science Letters, 361, 50–57.
Schmitz, M. D., and Bowring, S. A., 2001. U–Pb zircon and titanite
systematics of the Fish Canyon Tuff: an assessment of high-
precision U–Pb geochronology and its application to young volca-
nic rocks. Geochimica et Cosmochimica Acta, 65(15), 2571–2587.
Shih, C. Y., et al., 1985. Chronology and petrogenesis of a 1.8-g
lunar granitic clast – 14321,1062. Geochimica et Cosmochimica
Acta, 49(2), 411–426.
Shih, C. Y., Nyquist, L. E., Bansal, B. M., and Wiesmann, H., 1992.
Rb–Sr and Sm–Nd chronology of an Apollo-17 Kreep basalt.
Earth and Planetary Science Letters, 108(4), 203–215.
Siebel, W., 1994. Inferences about magma mixing and thermal
events from isotopic variations in redwitzites near the KTB site,
Projektleitung Kontinentales Tiefbohrprogramm der
Bundesrepublik Deutschland im Niedersächsischen Landesamt
für Bodenforschung.
Siebel, W., Reitter, E., Wenzel, T., and Blaba, U., 2005. Sr isotope sys-
tematics of K-feldspars in plutonic rocks revealed by the Rb–Sr
microdrilling technique. Chemical Geology, 222(3–4), 183–199.
Steiger, R. H., and Jäger, E., 1977. Subcommission on geochronol-
ogy – convention on use of decay constants in geochronology
and cosmochronology. Earth and Planetary Science Letters,
36(3), 359–362.
Verschure, R. H., et al., 1980. On the thermal stability of Rb–Sr and
K–Ar biotite systems – evidence from coexisting
Sveconorwegian (ca. 870 ma) and Caledonian (ca. 400 Ma) bio-
tites in SW Norway. Contributions to Mineralogy and Petrology,
74(3), 245–252.
Villa, I. M., 1998. Isotopic closure. Terra Nova, 10(1), 42–47.
Von Blanckenburg, F., Villa, I. M., Baur, H., Morteani, G., and
Steiger, R. H., 1989. Time calibrations of a PT-path from the
western Tauern window, eastern Alps – the problem of closure
temperatures. Contributions to Mineralogy and Petrology,
101(1), 1–11.
Vroon, P. Z., van der Wagt, B., Koornneef, J. M., and Davies, G. R.,
2008. Problems in obtaining precise and accurate Sr isotope
analysis from geological materials using laser ablation
MC-ICPMS. Analytical and Bioanalytical Chemistry, 390(2),
465–476.
Waight, T., Baker, J., and Willigers, B., 2002a. Rb isotope dilution
analyses by MC-ICPMS using Zr to correct for mass fraction-
ation: towards improved Rb–Sr geochronology? Chemical
Geology, 186(1–2), 99–116.
Waight, T., Baker, J., and Peate, D., 2002b. Sr isotope ratio measure-
ments by double-focusing MC-ICPMS: techniques, observa-
tions and pitfalls. International Journal of Mass Spectrometry,
221(3), 229–244.
Weatherly, D. K., and Hanley, R. W., 2013. Flash vaporization dur-
ing earthquakes evidenced by gold deposits. Nature Geoscience,
6, 294–298.
Wendt, I., 1993. Isochron or mixing line. Chemical Geology, 104
(1–4), 301–305.
Williams, I. S., Tetley, N. W., Compston, W., and McDougall, I.,
1982. A Comparison of K–Ar and Rb–Sr Ages of rapidly cooled
igneous rocks – two points in the Paleozoic time scale
re-evaluated. Journal of the Geological Society, 139(SEP),
557–568.
Willigers, B. J. A., Mezger, K., and Baker, J., 2004. Development of
high precision Rb–Sr phlogopite and biotite geochronology; an
alternative to 40Ar/39Ar tri-octahedral mica dating. Chemical
Geology, 213(4), 339–358.
Yang, J. H., and Zhou, X. H., 2001. Rb–Sr, Sm–Nd, and Pb isotope
systematics of pyrite: Implications for the age and genesis of
lode gold deposits. Geology, 29(8), 711–714.
Rb–Sr GEOCHRONOLOGY (IGNEOUS ROCKS)
Tod Waight
Department of Geosciences and Natural Resource
Management, Geology Section, University of
Copenhagen, Copenhagen, Denmark
Synonyms
Mica chronology; Mica dating; Rb–Sr dating; Sr dating;
Sr geochronology
Definition
Application of the radioactive decay of 87Rb to 87Sr to
determining the age of crystallization of igneous rocks.
Introduction and background
Both rubidium (Rb) and strontium (Sr) are large ion lithophile
elements and over the course of Earth history have become
 Rb–Sr GEOCHRONOLOGY (IGNEOUS ROCKS)
relatively concentrated in crustal lithologies, e.g., gneisses
and granites, and sediments derived from these. Because of
this, application of the Rb–Sr decay scheme was one of the
first radiometric dating systems used to determine the age
of rocks and remains a widely used system for dating geolog-
ical units formed at relatively high temperatures, in particular
volcanic, plutonic, and metamorphic rocks formed in the
crust. Relatively low closure temperatures for commonly
used minerals and susceptibility for disturbance of Rb and
Sr complicate the use of the system as a geochronometer,
and application of methods such as U–Pb in zircon is often
preferred to yield more reliable age information.
Age determination with the Rb–Sr system revolves
around the construction of an isochron using minerals or
rocks with variable Rb/Sr ratios. Igneous rocks form by
the cooling, crystallization, and solidification of magmas.
During crystallization, different minerals form depending
on pressure and temperature and the composition of the
magma. Elements present at trace concentrations (<1,000
ppm) in magmas will not normally form their own minerals
but will instead be incorporated into minerals by substituting
for major elements with similar ionic radii and charge. Rb is
monovalent with an ionic radius similar to K+ so will tend to
be concentrated in K-rich minerals, whereas Sr is bivalent
and substitutes for Ca2+ in mineral lattices. Minerals with
high K contents (e.g., biotite, alkali feldspar) will therefore
have high Rb/Sr, whereas Ca-rich minerals (e.g., plagioclase
feldspar, clinopyroxene, calcite, apatite) will have low Rb/
Sr. During construction of an isochron, various mineral
phases with variable Rb/Sr ratios are separated from
a rock. These minerals are then dissolved and Rb and Sr
purified by ion exchange chromatography and analyzed
for their Rb and Sr concentrations and Sr isotope composi-
tion, typically by thermal ionization mass spectrometry
(TIMS). Mineral selections are made so as to aim for as wide
a range in Rb/Sr as possible. A whole rock composition
(representing an average of all the minerals present in the
rock) is also often included in the age determination.
At the time of crystallization, the minerals inherit the
original Sr isotope composition of the parental magma.
With time the 87Rb in the minerals will decay to 87Sr, such
that K-rich (Rb-rich) minerals will become progressively
more radiogenic (higher 87Sr/86Sr) with time, whereas
the amount of ingrowth in the Rb-poor minerals will be
lower or, in the case of Rb-free minerals, absent. For
samples that behave as a closed system, plotting the mea-
sured 87Sr/86Sr of the different mineral phases versus their
87
Rb/86Sr composition (isochron diagram) will define
a line (an isochron) with a slope that is proportional to
the age of the rock (see Figure 1). The age of the rock
can then be calculated using the equation:
1 crystallizes and becomes more evolved, at least until gen-
t ¼ lnð1 þ slopeÞ
l erally late-stage K-bearing minerals such as alkali feldspar
The intersection of the isochron with the Y-axis
(or alternatively analysis of a Rb-free phase) provides
the initial Sr isotopic composition of the system
699
87
Sr/86Sr(i), a parameter that is useful to geochemists for
understanding the geological history of the source rocks
that melted to generate the magma. Chemical differentia-
tion of the Earth over its history due to partial melting
has resulted in the mantle having relatively low Rb/Sr,
and thus, over time, it has evolved to relatively
unradiogenic 87Sr/86Sr (ca. 0.7035). In contrast, the conti-
nental crust has relatively high Rb/Sr and has thus evolved
to have highly variable and typically radiogenic 87Sr/86Sr
(e.g., >0.710). As Sr isotopic compositions are not modi-
fied by processes such as melting and crystallization, these
signatures can be used to identify the source(s) of magma
and to identify processes such as crustal contamination of
mantle-derived magmas.
An example of a mineral isochron constructed for the
Habu Granodiorite in southwest Japan is shown in
Figure 1. Various mineral phases were separated from
the rock and, together with a whole rock powder and
a felsic mineral fraction (dominated by feldspar), were
analyzed for their Sr isotopic composition and Rb and Sr
concentration. Biotite, with a high Rb/Sr, has evolved with
time to have a high 87Sr/86Sr, whereas the felsic mineral
fraction with a lower Rb/Sr has a much less radiogenic
87
Sr/86Sr. The apatite analyses fall off the isochron and
were not included in the age calculation as the aim of this
study was to assess their suitability for determining the ini-
tial isotopic composition of a magma.
In order to obtain meaningful age information from
a mineral isochron, three fundamental assumptions must
be fulfilled:
(a) The minerals all formed in equilibrium with each
other.
(b) The minerals all formed from a reservoir of homoge-
neous isotopic composition (87Sr/86Sr(i)).
(c) The minerals have been a closed system (i.e., not been
subject to thermal or chemical disturbance) subse-
quent to crystallization.
It is possible to assess the validity of these assumptions
statistically through the degree of fit of the analyses to
a straight line regression (MSWD) and in the extrapolated
error on the initial isotopic composition.
A Rb–Sr isochron for igneous rocks can also be
constructed using whole rock compositions. Magmas
become more evolved (i.e., enriched in SiO2) during crys-
tallization through the process of fractional crystallization,
where minerals crystallize and become isolated from the
surrounding residual magma. During fractional crystalli-
zation, Rb generally behaves as an incompatible element
(preferring to remain in the melt phase rather than going
into the crystalline phase). Therefore, Rb concentrations
will increase in residual melts as the magma progressively
and biotite begin to crystallize. In contrast, plagioclase
feldspar, the major reservoir for Sr in an igneous system,
crystallizes over a broad range of temperatures, and
therefore, Sr concentrations in residual magmas will
700 Rb–Sr GEOCHRONOLOGY (IGNEOUS ROCKS)

Rb–Sr Geochronology (Igneous Rocks), Figure 1 A combined mineral whole rock isochron for the Habu Granodiorite in Japan; the
inset shows the compositions of the whole rock powder, a felsic mineral fraction (mostly feldspar and quartz), and three apatite
separates (Data from Tsuboi and Suzuki, 2003). 87Sr/86Sr(i) is the intercept of the isochron with the Y-axis and represents the initial
isotopic composition of the magma at the time of emplacement. MSWD (mean square of weighted deviates) is a statistical parameter
describing the degree of fit of the data points to a straight line.

tend to decrease with increasing differentiation. As
a consequence, Rb/Sr will generally increase in residual
melts with increasing fractional crystallization and evolu-
tion. A series of whole rock samples with variable Rb/Sr
can thus be analyzed from a single magmatic series and
plotted on an isochron diagram to yield an age. An impor-
tant assumption in constructing whole rock Rb–Sr iso-
chrons is that the magma crystallized as a closed system,
that is, there was no input from external sources such as
contamination by the surrounding country rock, and thus,
all rocks crystallized from a magma with a homogeneous
initial isotopic composition. An example of a whole rock
isochron based on a sequence of variably evolved rocks
from a granitic pluton in China is shown in Figure 2.
The studied pluton outcrops over an area of 160 km2 and
ranges in composition from granodiorite, through
monzogranite, to syenogranite and granophyres and is
considered to have evolved by closed system fractional
crystallization.
In some cases, it is possible to generate Rb–Sr ages are
within error when realistic model ages by analysis of
a single, highly radiogenic mineral phase; in fact, this is
how the Rb–Sr dating system was first developed. Min-
erals such as biotite and alkali feldspar with high Rb/Sr
will evolve to extremely radiogenic 87Sr/86Sr (values
>3) in relatively old rocks. At such radiogenic composi-
tions, extrapolation to an assumed and geologically rea-
sonable initial 87Sr/86Sr (e.g., between values of 0.7 and
0.75) results in minimal change in the isochron slope
and thus enables calculation of a model age based on
a single data point. For example, alkali feldspar from the
Illímaussaq Intrusion in southern Greenland has
a measured 87Rb/86Sr of 407.1 and 87Sr/86Sr ¼ 7.4727. If
an initial 87Sr/86Sr of 0.70 is assumed, this results in
a calculated age of 1,161.9 Ma, whereas assuming an ini-
tial composition of 0.71 produces an age of 1,160.2 Ma;
both ages are within error when realistic analytical errors
are applied to the measurements (Waight et al., 2002).
Rb–Sr geochronology of igneous rocks is most often
applied to evolved lithologies (e.g., rhyolites and granites)
as their evolved nature promotes crystallization of the
high-K (and Rb) phases ideally suited for dating. Applica-
tion to mafic rocks is more limited as the rocks them-
selves, and the minerals that crystallize from them, have
limited and generally low Rb/Sr. Nevertheless, it is possi-
ble as is illustrated, for example, in the study by Paces
et al. (1991) on Apollo 17 lunar basalts, where a Rb–Sr
mineral isochron yielded a Rb–Sr age of 3.67  0.10 Ga,
in good agreement with a Sm–Nd mineral age on the same
material, despite a variation in 87Rb/86Sr of only
0.001–0.02 (Figure 3).
The Rb–Sr system has been invaluable in determining
the ages of ultramafic rocks, such as kimberlite, which
are U-Pb (see entries “Uranium–Lead Dating” and
“Uranium–Lead, Chemical Isochron U–Pb Method
(CHIME)”, Rb-Sr (see entry “Rb–Sr Dating”), kimberlite
(see entry “Kimberlites (K-Ar/Ar-Ar)”), which are the
ultimate source of most of the world’s diamonds, and the
related ultramafic lamprophyres. These rocks are amena-
ble to Rb–Sr dating because, despite their very primitive
compositions, they can crystallize significant quantities
of phlogopite, the Mg-rich end-member of the biotite
group of minerals with high Rb/Sr (see also entry “Mete-
orites (U–Pb), Uranium–Lead, Rubidium-Strontium,
Kimberlite” Heaman, 2013; this volume). In contrast, bulk
rock kimberlite compositions have relatively low Rb/Sr.
This difference between bulk rock and phlogopite Rb/Sr
was utilized in the study of Tappe et al. (2009) who dated
 Rb–Sr GEOCHRONOLOGY (IGNEOUS ROCKS) 701

Rb–Sr Geochronology (Igneous Rocks), Figure 2 Rb–Sr whole rock isochron for the Xinhuatun Granite, northeastern China, redrawn
from Wu et al. (2003). Two ages are presented, one excluding sample X-5 which is a microgranular enclave and may not be directly
related to the magmatic system; both ages are identical within error.

Rb–Sr Geochronology (Igneous Rocks), Figure 3 Rb–Sr mineral isochron for a lunar basalt, redrawn from Paces et al. (1991). The
different minerals analyzed are plagioclase feldspar (Plag1 and Plag2), ilmenite (Ilm1 and Ilm2), and pyroxene (Px). Also included
are a nonmagnetic mineral fraction (NMag1) dominated by plagioclase and pyroxene, a magnetic mineral fraction (Mag1)
dominated by ilmenite and magnetic pyroxene, a whole rock powder (WR), and whole rocks of two additional samples (70138 and
71095). Ilm1 is not included in the regression line calculation.

an ultramafic lamprophyre (aillikite) from Greenland by
analyzing five phlogopite phenocrysts and a bulk rock to
yield a well-defined Rb–Sr age of 157.9  1.1 Ma
(Figure 4), in good agreement with U–Pb perovskite dates
obtained on related aillikite dikes from the same region.
Chemical disturbance
An important source of error in Rb–Sr dating of igneous
rocks is post-crystallization disturbance, particularly in
the presence of fluids. Both Rb and Sr are large ion
lithophile elements and thus relatively mobile in fluids
during, for example, hydrothermal alteration and
weathering. The 87Sr produced by radioactive decay of
Rb is generally not strongly bound into the crystal struc-
ture of the host mineral and thus can be relatively easily
mobilized. If this occurs, it will modify the 87Rb/86Sr
and 87Sr/86Sr composition of the investigated phase and
potentially move it away from the ideal isochrons, increas-
ing the error on the calculated age or erasing any original
age information.
Closure temperature
All geochronological systems are controlled by closure or
blocking temperatures, the temperature at which a mineral
702 Rb–Sr GEOCHRONOLOGY (IGNEOUS ROCKS)

Rb–Sr Geochronology (Igneous Rocks), Figure 4 Rb–Sr isochron for an aillikite dike from West Greenland, comprising five phlogopite
phenocrysts (phl-A to phl-D) and a whole rock analysis (Reproduced with permission from Tappe et al. (2009)).

Rb–Sr Geochronology (Igneous Rocks), Figure 5 Geochronological data using various minerals and isotopic systems for the Rønne
Granite (Bornholm, Denmark). The dashed line is a cooling curve assuming all heat loss is via one-dimensional heat conduction
from the intrusion depth (10 km) toward the surface. The gray symbols represent various permutations of closure temperatures and
decay constants for the 40Ar-39Ar system (Reproduced with permission from Waight et al. (2012)).

system becomes closed to diffusional loss of radiogenic
daughter atoms and therefore can begin to accumulate
the radiogenic isotopes produced by decay and record
age information. In addition to temperature, closure tem-
peratures are also constrained by the diffusivity of the ele-
mental species involved, cooling rates, crystal size, as well
as the presence or absence of fluids and tectonic strain.
The closure temperatures for common materials inves-
tigated using the Rb–Sr system are relatively well known
and are typically relatively low. The high Rb–Sr minerals
are most important in constraining geochronological data,
and muscovite is typically considered to have a closure
temperature for the Rb–Sr system at ca. 500  C and biotite
at ca. 300  C (Dodson, 1973; Villa, 1998). Alkali feldspar
is less well constrained with a closure temperature of
around 475–575  C (Giletti, 1991). In rapidly cooled
igneous systems, Rb–Sr ages can be considered to be crys-
tallization ages and generally agree with other geochrono-
logical systems with higher closure temperatures (e.g.,
U–Pb in zircon). However, in slowly cooled systems
this is not the case, for example, the U–Pb zircon system
(closure temperature ca. 900  C) yields an age of
1,456  5 Ma for the Rønne Granite (Bornholm, Den-
mark), yet biotite from the same sample yields Rb–Sr
 RHENIUM–OSMIUM DATING (METEORITES)
model and two-point isochron ages of about 1,370 Ma and
indicates either slow cooling or subsequent disturbance of
the Rb–Sr system (Figure 5). Similarly, the Rb–Sr mineral
age presented in Figure 2 is ca. 10 Myr. younger than
U–Pb zircon on the same rock, a discrepancy also attrib-
uted to slow cooling by Wu et al. (2003). Age determina-
tions from multiple geochronometers based on different
minerals and isotopic systems can thus be used to determine
a cooling history for a region. One important consequence
of the different closure temperatures for Sr isotopes in var-
ious minerals is that slow cooling may result in different
minerals passing through their closure temperatures at dif-
ferent times, causing scatter on isochron diagrams. While
minerals with higher closure temperatures typically behave
as closed systems, other minerals will still be open to diffu-
sional exchange and ongoing radioactive decay prior to clo-
sure; thus, the minerals will not have crystallized from
a homogeneous reservoir (e.g., Giletti, 1991). Similarly, rel-
atively low-temperature thermal disturbances may cause
open system behavior and diffusional exchange of elements
in some mineral systems, resetting the isotopic systematics
and thus erasing any information on the original crystalliza-
tion age of the magma. However, the low closure tempera-
ture for biotite especially has been particularly useful for
dating metamorphic events.
Summary
The preferential incorporation of Rb (following K) or Sr
(following Ca) into different mineral structures, as well
as their differing behaviors during closed system evolu-
tion of magmas by fractional crystallization, results in
a wide range in Rb/Sr in mineral and whole rock composi-
tions. The radioactive decay of 87Rb to 87Sr can then be
exploited to generate an isochron and yield age informa-
tion about a magmatic system, providing all the analyzed
materials crystallized from a homogeneous parental melt,
and the system has not been subsequently thermally or
chemically disturbed. Some caution is required when
applying the Rb–Sr chronometer to igneous systems;
evaluation of the mobility of elements (especially Rb)
and whether the strontium isotopic composition has been
compromised by fluids (e.g., hydrothermal alteration and
weathering) is required and the consequences of the rela-
tively low closure temperature for Sr diffusion in many
minerals, especially biotite, must be considered.
Bibliography
Dickin, A. P., 2005. Radiogenic Isotope Geology, 2nd edn. Cam-
bridge: Cambridge University Press.
Dodson, M. H., 1973. Closure temperatures in cooling geochrono-
logical and petrological systems. Contributions to Mineralogy
and Petrology, 40, 259–274.
Giletti, B. J., 1991. Rb and Sr diffusion in alkali feldspars, with
implications for cooling histories of rocks. Geochimica et
Cosmochimica Acta, 55, 1331–1343.
Heaman, L. M., 2013. Uranium-Lead and Rubidium-Strontium
dating (Kimberlites). In Rink, W. J., and Thompson, J. (eds.),
Encyclopedia of Scientific Dating Methods. Springer (Online
publication Sept 2013).
703
Paces, J. B., Nakai, S., Neal, C. R., Taylor, L. A., Halliday, A. N.,
and Lee, D.-C., 1991. A strontium and neodymium isotopic
study of Apollo 17 High-Ti Mare Basalts: resolution of ages,
evolution of magmas, and origins of source heterogeneities.
Geochimica et Cosmochimica Acta, 55, 2025–2043.
Tappe, S., Steenfelt, A., Heaman, L. M., and Simonetti, A., 2009.
The newly discovered Jurassic Tikiusaaq Carbonatite-Aillikite
occurrence, West Greenland, and some remarks on
Carbonatite-Kimberlite relationships. Lithos, 112S, 385–399.
Tsuboi, M., and Suzuki, K., 2003. Heterogeneity of initial 87Sr/86Sr
ratios within a single pluton: evidence from apatite strontium iso-
topic study. Chemical Geology, 199, 189–197.
Villa, I., 1998. Isotopic closure. Terra Nova, 10, 42–47.
Waight, T. E., Baker, J. A., and Willigers, B. J. A., 2002. Rb isotope
dilution analyses by MC-ICPMS using Zr to correct for mass
fractionation: towards improved Rb-Sr geochronology? Chemi-
cal Geology, 186, 99–116.
Waight, T. E., Frei, D., and Storey, M., 2012. Geochronological con-
straints on granitic magmatism, metamorphism, cooling and
uplift on Bornholm, Denmark. Bulletin of the Geological Society
of Denmark, 60, 23–46.
Wu, F.-y., Jahn, B.-m., Wilde, S. A., Lo, C.-H., Yui, T.-F., Lin, Q.,
Ge, W.-c., and Sun, D.-y., 2003. Highly fractionated I-type gran-
ites in NE China (I): geochronology and petrogenesis. Lithos,
66, 241–273.
Cross-references
Apatite
Chromatography
Feldspars
Rubidium–Strontium Dating, Hydrothermal Events
Thermal Ionization Mass Spectrometer (TIMS)
Uranium–Lead, Zircon
RHENIUM–OSMIUM DATING (METEORITES)
Richard J. Walker
Isotope Geochemistry Laboratory, Department of
Geology, University of Maryland, College Park, MA, USA
Synonyms
187
Re–187Os isotopic dating; Osmium isotopic dating
Definition
The Re–Os isotopic system is used for determining forma-
tion ages and constraining ages of secondary alteration of
meteorites and meteorite components, especially in rela-
tion to metals and sulfides.
Introduction
The rhenium-osmium isotopic dating system as applied to
geo- and cosmochemical materials is based on the
negatron (-b) transition of 187Re to 187Os. The decay con-
stant (l) for 187Re is 1.67  1011 year1, which is equiv-
alent to a half-life (t1/2) of approximately 42 billion years.
This value has been determined through the comparisons
of precisely defined isochrons for cosmochemical and ter-
restrial materials to absolute ages for the same materials
704 RHENIUM–OSMIUM DATING (METEORITES)
determined using the U–Pb system (Smoliar et al., 1996;
Selby et al., 2007). The Re–Os isotopic system has been
of interest to the cosmochemical community for more than
50 years because of the unique geochemical characteris-
tics of both elements, as compared to elements comprising
other long-lived chronometers. Both elements partition
nearly quantitatively into metal relative to silicates; hence,
they are termed highly siderophile elements (HSE). In
addition, both elements also tend to strongly partition into
sulfides and sulfide melts relative to silicates; hence, they
are also chalcophile elements. Consequently, the system
has been long pursued as a means of determining the ages
of early solar system metals, such as iron meteorites, and
the metal and sulfide components of some primitive mete-
orites. Because of the relatively high abundances of Re
and Os in primitive and iron meteorites, and their rela-
tively low abundances in the silicate mantles and crusts
of differentiated planetary bodies, the system can also
potentially serve as a tracer of the chemical natures and
quantities of late accreted materials to planetary bodies
(Chou, 1978; Morgan, 1986; Morgan et al., 2001).
Historical development of system to dating
meteorites
Early studies of the Re–Os system, as applied to both geo-
and cosmochemical problems, were hindered by extreme
analytical limitations. These limitations mainly stemmed
from the very high first ionization potentials of Re and
Os, resulting in extremely low thermal ionization yields,
when using the standard thermal ionization mass spectro-
metric methods available in the 1950s–1980s. Despite the
analytical limitations, the first detailed application of the
Re–Os isotopic system to dating meteorites was published
by Hirt et al. (1963). In that study the authors examined
variations in 187Os/186Os and 187Re/186Os for a variety
of iron meteorites. The low-precision data were found to
form a linear trend that was interpreted to be an isochron.
The generation of isochrons in this and subsequent studies
of iron meteorites were advantaged by the large range of
Re/Os ratios typically present within iron meteorite
groups. Most such variations are consistent with solid
metal-liquid metal fractionation on the parent bodies of
the meteorites (e.g., McCoy et al., 2011). Hirt et al.
(1963) applied a 187Re half-life of 43 billion years to their
iron meteorite isochron and suggested a general crystalli-
zation age for irons of approximately 4.0 billion years.
Subsequent studies of the Re–Os isotopic systematics of
iron meteorites and metal components from chondritic
meteorites were published by Luck et al. (1980) and Luck
and Allègre (1983). These studies, which utilized second-
ary ion mass spectrometry of the chemically purified ele-
ments, reported much higher precision isotopic and
elemental compositions, compared to the prior studies.
They also reported good linear trends for meteorites on
plots of 187Re/186Os versus 187Os/186Os, interpreted the
correlations to be isochrons, and redefined the decay con-
stant for 187Re so that the isochrons would give ages close
to the 4.57 Ga canonical age of the solar system. Regres-
sion uncertainties for isochrons slopes in these studies,
however, were no better than 2 %, equivalent to uncer-
tainties of 90 million years for early solar system
objects.
Analyses of bulk chondrites, as well as metal separates
from chondrites, were found to be characterized by limited
variations in Re/Os and 187Os/186Os, compared to iron
meteorites (Luck and Allègre, 1983; Walker and Morgan,
1989), so the generation of precise isochrons for these
types of materials was not possible. Further, analyses of
some bulk chondrites, determined by resonance ionization
mass spectrometry, did not plot within analytical uncer-
tainties of the iron meteorite reference isochron, indicating
either incorporation of isotopically diverse Os into primi-
tive meteorites at the time of formation or relatively late-
stage open-system behavior (Walker and Morgan, 1989).
Iron meteorites
The development of negative thermal ionization mass
spectrometry (NTIMS) techniques in the early 1990s, as
applied to the Re–Os system, permitted the measurement
of these elements to much higher precision than was previ-
ously possible (Creaser et al., 1991; Völkening et al.,
1991). Also, at about this time, the common reference iso-
tope for Os was changed from 186Os to 188Os, because of
the recognition that 186Os is the decay product of 190Pt
and that its atomic abundance, relative to the non-
radiogenic isotopes of Os, varies in nature. Coupled with
improved chemical processing methods (Shirey and
Walker, 1995), the NTIMS technique permitted the gener-
ation of high-precision isochrons for iron meteorite groups
representing five different parent bodies (e.g., Horan et al.,
1992; Shen et al., 1996; Smoliar et al., 1996; Cook et al.,
2004; Walker et al., 2008; McCoy et al., 2011). For exam-
ple, Smoliar et al. (1996) reported an isochron age for
group IIAB irons of 4,537  8 Ma, with a precision of bet-
ter than 0.2 %. Isochron ages with nearly comparable
precision have been reported for other major iron meteor-
ite groups (IAB, IIIAB, IVA, and IVB), constraining the
ages of crystallization to the first 50 Ma of solar system
history (Figure 1). Unfortunately, crystallization ages with
uncertainties of 10–20 million years are of limited utility
with regard to understanding the chronological sequence
of planetesimal formation, melting, differentiation, and
crystallization, given that many such bodies evidently
underwent some or all of these processes within the first
10 million years of solar system history (e.g., Lugmair
and Galer, 1992; Misawa et al., 2005; Day et al., 2012b).
Application of the short-lived 182Hf–182W isotopic system
(t1/2 of 182Hf is 8.9 million years) has revealed that, in most
cases, metallic cores segregated from silicates within just
the first ~3 million years of solar system history (e.g.,
Horan et al., 1998; Kruijer et al., 2012). It should be noted,
however, that 182W model ages of iron meteorites define
the timing of metal-silicate segregation, not crystalliza-
tion. Hence, the Re–Os system retains utility for
 RHENIUM–OSMIUM DATING (METEORITES)
Rhenium–Osmium Dating (Meteorites), Figure 1 Isochron
diagram for group IVA iron meteorites. Meteorites from this
group display a relatively large range in 187Re/188Os and
corresponding 187Os/188Os. The crystallization age derived from
this isochron is 4,540  17 Ma (Figure is from McCoy et al.
(2011)).
understanding the cooling rates of cores of larger, more
slowly cooled bodies. Further, the system has been used
to examine whether or not the HSE within a given iron
meteorite have remained closed to elemental migration,
due to impact resetting or other processes, since the time
of formation (e.g., McCoy et al., 2011).
Primitive meteorites
The application of the Re–Os system to dating bulk chon-
dritic meteorites and their components has met with mixed
success. Even with improved analytical methods, most
bulk samples of chondrites from the major groups do not
define sufficient spread in Re/Os to permit the generation
of precisely defined group isochrons, even assuming that
the requirements for isochron dating of different primitive
meteorites could be met (same formation ages; derived
from isotopically uniform sources). Hence, the system
has virtually no utility for dating the formation of primi-
tive meteorites. Further, a major proportion of bulk chon-
drites analyzed using NTIMS do not plot within analytical
uncertainties of a primordial reference isochron (Chen
et al., 1998; Walker et al., 2002; Fischer-Gödde et al.,
2010; van Acken et al., 2011) (Figure 2a). Such deviations
have largely been interpreted to reflect open-system
behavior tens of millions of years to billions of years after
the parent bodies formed. Open-system behavior on prim-
itive parent bodies may reflect late-stage impact heating
and/or aqueous alteration.
As the nature and intensity of open-system behavior
appears to be mostly specific to individual meteorites,
not chondritic groups or subgroups (Walker et al., 2002),
the study of components from a meteorite is a potentially
important means for investigating the processes involved
705
in creating the open-system behavior specific to the sam-
pled portion of the parent body. Common components
present in primitive meteorites include chondrules, cal-
cium-aluminum-rich inclusions (CAIs), and matrix. As
with bulk chondrites, several studies have documented
open-system behavior among chondritic components
from the same meteorites. For example, Becker et al.
(2001) analyzed CAIs from the carbonaceous chondrite
Allende and found moderate-scale open-system behavior.
Analyses of multiple pieces of a single CAI were charac-
terized by sufficient spread in 187Re/188Os and
187
Os/188Os to define an errorchron that defined a reset
age of ~1,600 Ma. Presumably a metamorphic event led
to the isotopic homogenization of Os within the CAI at
that time. Horan et al. (2009) reported large variations in
187
Re/188Os, and consequently 187Os/188Os, between
magnetic and nonmagnetic portions of chondrules and
matrix separated from the relatively low metamorphic
grade ordinary chondrites Dhajala and Ochansk. In con-
trast to the Allende CAIs, they reported relatively minor
open-system behavior for these components. Collectively,
these studies show that the Re–Os system will continue to
play an important role in constraining the timing of late-
stage events affecting primitive meteorites.
Perhaps the greatest utility of the Re–Os isotopic sys-
tem, as applied to primitive meteorites, is for genetic trac-
ing. On average, carbonaceous chondrites have less
radiogenic 187Os/188Os than ordinary or enstatite chon-
drites (Figure 2b). This reflects the lower long-term
187
Re/188Os of carbonaceous chondrites, which, in turn,
may reflect a nebular fractionation compared with the
formational environments of enstatite and ordinary chon-
drites. The lower 187Os/188Os of carbonaceous chondrites
has been useful, for example, in assessing their role (or
lack thereof) in the generation of terrestrial and lunar
impact melt rocks (e.g., Puchtel et al., 2008; Fischer-
Gödde and Becker, 2012).
Achondrite meteorites
Rhenium-Os isotopic data have been published for
a number of types of achondritic meteorites including
the shergottite-nahklite-chassignite (SNC) suite, pre-
sumed to be derived from Mars (Birck and Allègre,
1994; Brandon et al., 2012); the howardite-eucrite-
diogenite (HED) suite, presumed to be derived from the
asteroid Vesta (Dale et al., 2012; Day et al., 2012b); as
well as angrites and ureilites (Rankenburg et al., 2007;
Riches et al., 2012). The system has also been examined
in various primitive achondrites, such as brachinites
(Day et al., 2012a), as well as lunar meteorites (Puchtel
et al., 2008). Given the inherent problems relating differ-
ent meteorites to one another through a common process
and formation time, the system has proved useful from
a geochronologic standpoint only for dating individual
meteorites from which bulk pieces or separated minerals
define substantial variation in Re/Os. Consequently, the
Re–Os isotopic system has been only sparingly applied
706 RHENIUM–OSMIUM DATING (METEORITES)

Rhenium–Osmium Dating (Meteorites), Figure 2 (a) Isochron diagram for primitive (chondritic) meteorites. Bulk samples of
chondrites from the carbonaceous, enstatite, and ordinary chondrite groups are characterized by limited variations in 187Re/188Os
and 187Os/188Os, compared to most within-group variations for iron meteorites (e.g., Figure 1 above). Analytical uncertainties are
typically smaller than the symbols. The considerable variance of data around the 4.57 Ga reference isochrons reflects relatively late-
stage open-system behavior of Re and/or Os. (b) Histogram showing the distributions of 187Os/188Os in the three major groups of
chondrites (Data are from Walker et al. (2002), Fischer-Gödde et al. (2010), and van Acken et al. (2011)).

to dating achondrites. For example, Brandon et al. (2012)
reported a Re–Os isochron age of 164  12 Ma for the
Antarctic shergottite EETA 79001. This age is in good
agreement with Sm–Nd and Rb–Sr isochrons ages for
the same meteorite and provides important supporting evi-
dence that relatively young ages for some shergottites are
true igneous crystallization ages, rather than reset ages
associated with impact events.
Most Re–Os data as well as accompanying concentra-
tion data for other HSE in achondrites have been used to
consider the origins of HSE to the silicate portions of the
parent bodies to these meteorites. For example, Day
et al. (2012b) concluded that diogenites were variably
enriched in the HSE via a process termed late accretion,
whereby additional meteoritic material accreted to the par-
ent body after cessation of core segregation, leading to
a re-enrichment of HSE in the silicate portion of the body
(Walker, 2009). Evidence for variable extents of late
accretion is present in the terrestrial mantle (Meisel
et al., 2001; Becker et al., 2006), the lunar mantle (Day
et al., 2007), and the Martian mantle (Brandon et al.,
2012).
Summary and conclusions
Since the 1960s the Re–Os isotopic system has been used
for dating various cosmochemical materials. The system
has primarily been applied to constraining crystallization
ages of large metal masses, such as asteroidal cores, via
the analysis of chemically related iron meteorites. How-
ever, due to the limited precision of isochron ages for
irons, the system has proved more important for demon-
strating long-term closed-system behavior for the HSE in
those metals studied. The system has little utility for
precisely dating the absolute formation ages of bulk chon-
drites and their components, but is quite valuable for
constraining periods of open-system behavior in chon-
dritic components, such as CAIs and chondrules.
Bibliography
Becker, H., Walker, R. J., MacPherson, G. J., Morgan, J. W., and
Grossman, J. N., 2001. Rhenium-osmium systematics of cal-
cium-aluminum-rich inclusions in carbonaceous chondrites.
Geochimica et Cosmochimica Acta, 65, 3379–3390.
Becker, H., Horan, M. F., Walker, R. J., Gao, S., Lorand, J.-P., and
Rudnick, R. L., 2006. Highly siderophile element composition
of the Earth’s primitive upper mantle: constraints from new data
on peridotite massifs and xenoliths. Geochimica et
Cosmochimica Acta, 70, 4528–4550.
Birck, J. L., and Allègre, C. J., 1994. Contrasting Re/Os magmatic
fractionation in planetary basalts. Earth and Planetary Science
Letters, 124, 139–148.
Brandon, A. D., Puchtel, I. S., Walker, R. J., Day, J. M. D., Irving,
A. J., and Taylor, L. A., 2012. Evolution of the martian mantle
inferred from the 187Re–187Os isotope and highly siderophile
element systematics of shergottites meteorites. Geochimica et
Cosmochimica Acta, 76, 206–235.
Chen, J. H., Papanastassiou, D. A., and Wasserburg, G. J., 1998.
Re-Os systematics in chondrites and the fractionation of the plat-
inum group elements in the early solar system. Geochimica et
Cosmochimica Acta, 62, 3379–3392.
Chou, C.-L., 1978. Fractionation of siderophile elements in the
Earth’s upper mantle. In Proceedings of the 9th Lunar and Plan-
etary Science Conference, pp. 219–230.
Cook, D. L., Walker, R. J., Horan, M. F., Wasson, J. T., and Morgan,
J. W., 2004. Pt-Re-Os systematics of group IIAB and IIIAB iron
meteorites. Geochimica et Cosmochimica Acta, 68, 1413–1431.
Creaser, R. A., Papanastassiou, D. A., and Wasserburg, G. J., 1991.
Negative thermal ion mass spectrometry of osmium, rhenium,
and iridium. Geochimica et Cosmochimica Acta, 55, 397–401.
 RHENIUM–OSMIUM GEOCHRONOLOGY: SULFIDES, SHALES, OILS, AND MANTLE
Dale, C. W., Burton, K. W., Greenwood, R. C., Gannoun, A., Wade,
J., Wood, B. J., and Pearson, D. G., 2012. Late accretion on the
earliest planetesimals revealed by the highly siderophile ele-
ments. Science, 336, 72–75.
Day, J. M. D., Pearson, D. G., and Taylor, L. A., 2007. Highly
siderophile element constraints on accretion and differentiation
of the Earth-Moon system. Science, 315, 217–219.
Day, J. M. D., Walker, R. J., Qin, L., and Rumble, D., 2012b. Early
timing of late accretion in the solar system. Nature Geoscience,
5, 614–617.
Day, J. M. D., Walker, R. J., Ash, R. D., Liu, Y., Rumble, D., Irving,
A. J., Goodrich, C. A., Tait, K., McDonough, W. F., and Taylor,
L. A., 2012a. Origin of Graves Nunataks 06128 and 06129,
brachinites and brachinite-like achondrites by partial melting of
volatile-rich primitive parent bodies. Geochimica et
Cosmochimica Acta, 81, 94–128.
Fischer-Gödde, M., and Becker, H., 2012. Osmium isotope and
highly siderophile element constraints on ages and nature of
meteoritic components in ancient lunar impact rocks.
Geochimica et Cosmochimica Acta, 77, 135–156.
Fischer-Gödde, M., Becker, H., and Wombacher, F., 2010. Rho-
dium, gold and other highly siderophile element abundances in
chondritic meteorites. Geochimica et Cosmochimica Acta, 74,
356–379.
Hirt, B., Herr, W., and Hoffmeister, W., 1963. Age determinations
by the rhenium-osmium method. In Radioactive Dating. Vienna:
International Atomic Energy Agency, pp. 35–43.
Horan, M. F., Morgan, J. W., Walker, R. J., and Grossman, J. N.,
1992. Re-Os isotope constraints on the age of iron meteorites.
Science, 255, 1118–1121.
Horan, M. F., Smoliar, M. I., and Walker, R. J., 1998. 182W and
187
Re-187Os systematics of iron meteorites: chronology for melt-
ing, differentiation and crystallization in asteroids. Geochimica
et Cosmochimica Acta, 62, 545–554.
Horan, M. F., Alexander, C. M. O.’. D., and Walker, R. J., 2009.
Highly siderophile element evidence for early solar system pro-
cesses in components from ordinary chondrites. Geochimica et
Cosmochimica Acta, 73, 6984–6997.
Kruijer, T. S., Fischer-Gödde, M., Kleine, T., Sprung, P., Leya, I.,
and Wieler, R., 2013. Neutron capture on Pt isotopes in iron
metereorites and the Hf-W chronology of core formation in plan-
etesimals. Earth and Planetary Science Letters, 361, 162–172.
Luck, J.-M., and Allègre, C. J., 1983. 187Re-187Os systematics in mete-
orites and cosmochemical consequences. Nature, 302, 130–133.
Luck, J.-M., Birck, J.-L., and Allègre, C.-J., 1980. 187Re-187Os sys-
tematics in meteorites: early chronology of the solar system and
age of the Galaxy. Nature, 283, 256–259.
Lugmair, G. W., and Galer, S. J. G., 1992. Age and isotopic relation-
ships among the angrites Lewis Cliff 86010 and Angra dos Reis.
Geochimica et Cosmochimica Acta, 50, 2601–2614.
McCoy, T. J., Walker, R. J., Goldstein, J. I., Yang, J., McDonough,
W. F., Rumble, D., Chabot, N. L., Ash, R. D., Corrigan, C. M.,
Michael, J. R., and Kotula, P. G., 2011. Group IVA irons: new
constraints on the crystallization and cooling history of an aster-
oidal core with a complex history. Geochimica et Cosmochimica
Acta, 75, 6821–6843.
Meisel, T., Walker, R. J., Irving, A. J., and Lorand, J.-P., 2001.
Osmium isotopic compositions of mantle xenoliths: a global per-
spective. Geochimica et Cosmochimica Acta, 65, 1311–1323.
Misawa, K., Yamaguchi, A., and Kaiden, H., 2005. U-Pb and
207
Pb-206Pb ages of zircons from basaltic eucrites: implications
for early basaltic volcanism on the eucrite parent body.
Geochimica et Cosmochimica Acta, 69, 5847–5861.
Morgan, J. W., 1986. Osmium isotope constraints on Earth’s late
accretionary history. Nature, 317, 703–705.
Morgan, J. W., Walker, R. J., Brandon, A. D., and Horan, M. F.,
2001. Siderophile elements in Earth’s upper mantle and lunar
707
breccias: data synthesis suggests manifestations of the same late
influx. Meteoritics and Planetary Science, 36, 1257–1275.
Puchtel, I. S., Walker, R. J., James, O. B., and Kring, D. A., 2008.
Osmium isotope and highly siderophile element systematics of
lunar impact melt rocks: implications for the late accretion his-
tory of the Moon and Earth. Geochimica et Cosmochimica Acta,
72, 3022–3042.
Rankenburg, K., Brandon, A. D., and Humayun, M., 2007. Osmium
isotope systematics of ureilites. Geochimica et Cosmochimica
Acta, 71, 2402–2413.
Riches, A. J. V., Day, J. M. D., Walker, R. J., Simonetti, A., Liu, Y.,
Neal, C. R., and Taylor, L. A., 2012. Rhenium-osmium isotope
and highly-siderophile element abundance systematics of angrite
meteorites. Earth and Planetary Science Letters, 353–354,
208–218.
Selby, D., Creaser, R. A., Stein, H. J., Markey, R. J., and Hannah,
J. L., 2007. Assessment of the 187Re decay constant by cross cal-
ibration of Re-Os molybdenite and U-Pb zircon chronometers in
magmatic ore systems. Geochimica et Cosmochimica Acta, 71,
1999–2013.
Shen, J. J., Papanastassiou, D. A., and Wasserburg, G. J., 1996.
Precise Re-Os determinations and systematics of iron meteorites.
Geochimica et Cosmochimica Acta, 60, 2887–2900.
Shirey, S. B., and Walker, R. J., 1995. Carius tube digestion for low-
blank rhenium-osmium analysis. Analytical Chemistry, 34,
2136–2141.
Smoliar, M. I., Walker, R. J., and Morgan, J. W., 1996. Re-Os ages
of group IIA, IIIA, IVA and IVB iron meteorites. Science, 271,
1099–1102.
Van Acken, D., Brandon, A. D., and Humayun, M., 2011. High-
precision osmium isotopes in enstatite and Rumuruti chondrites.
Geochimica et Cosmochimica Acta, 75, 4020–4036.
Völkening, J., Walczyk, T., and Heumann, K. G., 1991. Osmium
isotope ratio determinations by negative thermal ionization mass
spectrometry. International Journal of Mass Spectrometry and
Ion Processes, 105, 147–159.
Walker, R. J., 2009. Highly siderophile elements in the Earth, Moon
and Mars: update and implications for planetary accretion and
differentiation. Chemie der Erde, 69, 101–125.
Walker, R. J., and Morgan, J. W., 1989. Rhenium-osmium isotope
systematics of carbonaceous chondrites. Science, 243, 519–522.
Walker, R. J., Horan, M. F., Morgan, J. W., Becker, H., Grossman,
J. N., and Rubin, A., 2002. Comparative 187Re-187Os systemat-
ics of chondrites: implications regarding early solar system pro-
cesses. Geochimica et Cosmochimica Acta, 66, 4187–4201.
Walker, R. J., McDonough, W. F., Honesto, J., Chabot, N. L.,
McCoy, T. J., Ash, R. D., and Bellucci, J. J., 2008. Origin and
chemical evolution of group IVB iron meteorites. Geochimica
et Cosmochimica Acta, 72, 2198–2216.
RHENIUM–OSMIUM GEOCHRONOLOGY: SULFIDES,
SHALES, OILS, AND MANTLE
Holly Stein1,2 and Judith Hannah1,2
1
AIRIE Program, Colorado State University, Fort Collins,
CO, USA
2
Centre for Earth Evolution and Dynamics (CEED),
University of Oslo, Oslo, Norway
Synonyms
Re–Os geochronology; 187Re–187Os geochronology
708 RHENIUM–OSMIUM GEOCHRONOLOGY: SULFIDES, SHALES, OILS, AND MANTLE
Definition
Rhenium–osmium geochronology is based on radioactive
decay of 187Re to 187Os with a half-life of 41.6 b.y. – about
10 times the age of the earth. Both Re and Os are
siderophile-chalcophile elements, that is, they are both
strongly partitioned into metals or sulfides rather than sili-
cates. This distinguishes them from other widely used
geochronometers whose parent-daughter elements reside
in silicates. Re–Os geochronology underpins dating of
materials from meteorites, the mantle, and metallic ore
deposits. Furthermore, both Re and Os are redox-sensitive
metals, soluble when oxidized, and fixed by reduction.
Thus, both are mobile in Earth’s presently oxidized surface
environments but are concentrated in sulfides and organic
matter in anoxic-euxinic sediments. This is the basis for
Re–Os dating of the depositional age of organic-rich sedi-
mentary rocks and provides a temporally constrained
record of changing redox conditions through earth history.
Fundamentals of Re–Os geochemistry
There are two naturally occurring isotopes of Re: stable
185
Re and radioactive 187Re. Mass fractionation of the two
Re isotopes is readily observed during the extreme condi-
tions imposed by mass spectrometry (e.g., Suzuki
et al., 2004; Zimmerman et al., 2007). The few reported nat-
ural variations, however, are less than 0.3 % (Miller
et al., 2009) and thus introduce errors less than other
sources of uncertainty for Re–Os geochronology. As with
other high-mass elements used in geochronology, we there-
fore assume uniform present-day abundances of 185Re
(37.398 %) and 187Re (62.602 %; Gramlich et al., 1973)
as an underpinning assumption for radiometric dating.
There are seven naturally occurring isotopes of
Os: 184Os, 186Os, 187Os, 188Os, 189Os, 190Os, and 192Os.
Isotope 187Os is the product of beta decay of 187Re, with
a decay constant of 1.666  1011 year1 (Smoliar
et al., 1996). Isotope 186Os is the product of alpha decay
of 190Pt with a decay constant of 1.542  1012 years1
(Walker et al., 1997). The abundances of the remaining
isotopes may be considered constant for present-day Re–
Os isotope geochemistry and geochronology. Prior to rec-
ognition of small, but measurable contributions of radio-
genic 186Os, Os isotopic compositions were reported as
187
Os/186Os. Current convention is to normalize abun-
dances of 187Os to 188Os, reporting isotopic variations in
terms of the ratio 187Os/188Os.
Os is highly compatible in mantle sulfides and metal
alloys and is thus strongly sequestered in the earth’s mantle
relative to the crust. Re, in contrast, is only mildly compat-
ible and is therefore transferred to the crust by partial melts
far more efficiently than Os (e.g., Reisberg and Lorand,
1995). Thus, 187Re/188Os ratios in fertile (undepleted) man-
tle are low (about 0.4), whereas 187Re/188Os ratios in the
crust are much higher and extremely variable; crustal
187
Re/188Os averages about 50 and ranges from one to three
orders of magnitude (or more) higher than the mantle
(Shirey and Walker, 1998). The extreme variability in
187
Re/188Os ratios offers two key opportunities. First,
because 187Re/188Os ratios are markedly higher in the crust
than in the mantle, decay of 187Re over time produces an
enormous range of 187Os/188Os ratios in crustal materials.
This provides a sensitive tracer of crustal input to mantle
or crustal contamination in mantle-derived materials. Sec-
ond, extremely high 187Re/188Os ratios in some materials
make them suitable for dating very young samples, despite
the long half-life of 187Re.
Figure 1 illustrates Earth’s Re and Os reservoirs and
Figure 2 details the Re–Os budgets in various reservoirs,
along with the potential for cycling among those reservoirs.
Whereas subduction cycles crustal Re and Os into the man-
tle, partial melting and slab dehydration introduce the ele-
ments into the crust. The oceans receive Re and Os from
seafloor hydrothermal processes, seafloor weathering, riv-
erine inputs from continental weathering, rain of cosmic
dust, and the occasional meteorite. Present-day input from
continental weathering constitutes about 80 % of the Os
budget of seawater (Sharma et al., 1997).
Fundamentals of Re–Os dating: from
sample to age
Analyses for high-precision Re–Os ages are acquired using
the isotope dilution method, in which a known amount of
one isotope of each element is added to the sample prior
to analysis. Given the amount and composition of the isoto-
pic “spike” added and the measurement of isotopic ratios in
the mixture, it is straightforward to calculate the ratios in the
sample. For example, the weight in mg of Re in a sample can
be calculated as follows:
 
AWsam 185 Respk  RM 187 Respk
ReW ¼ SpkW ð1Þ
AWspk RM 187 Resam  185 Resam
where Rew ¼ weight of Re in the sample, Spkw ¼ weight
of Re in the spike (same units as sample),
Rhenium–Osmium Geochronology: Sulfides, Shales, Oils, and
Mantle, Figure 1 Reservoirs important to the terrestrial Re–Os
budget.
 RHENIUM–OSMIUM GEOCHRONOLOGY: SULFIDES, SHALES, OILS, AND MANTLE 709

Aeolian dust
Low [Re] and [Os]
187Os/188Os ≈ 1.05 today Cosmic dust
Low Re/Os, high [Os]
187Os/188Os= chondritic
Present day 187Os/188Os = 0.127

Eroding mafic/ultramafic rocks

Eroding organic-rich shales
Low Re/Os, high [Os] High Re/Os,
≈ chondritic
187Os/188Os
High [Re] and [Os]
187Os/188Os variable
Riverine input
187Os/188Os ≈ 1.26 today Anoxic muds
High Re/Os
High [Re] and [Os]
Modern seawater
187Os/188Os ≈ seawater

[Re] ≈ 7 ppt, [Os] ≈ 0.01 ppt

187Os/188Os = 1.06

Oxide crusts
Hydrothermal input Low Re/Os, high [Os]
Low Re/Os, high [Os] 187Os/188Os ≈ seawater
Mantle Sulfide deposits
187Os/188Os ≈ chondritic Re/Os variable
Low Re/Os, high [Os]
187Os/188Os ≈ chondritic
187Os/188Os≈ chondritic

Rhenium–Osmium Geochronology: Sulfides, Shales, Oils, and Mantle, Figure 2 Re and Os reservoirs with sources and sinks for
seawater (From Hannah and Stein, 2012). Riverine concentrations of Re and Os depend strongly on the composition of bedrock in the
drainage area; flux to the oceans is a mixture of dissolved and suspended load. Therefore, global averages are not available. Data are
summarized from a variety of sources, all cited in the main text.

AWsam ¼ atomic weight of Re in the sample,
AWspk ¼ atomic weight of Re in the spike, 187Respk and
185
Respk ¼ fraction of each isotope in the spike, 187Resam
and 185Resam ¼ fraction of each isotope in the sample,
and RM ¼ measured 185Re/187Re ratio. Dividing Rew by
the weight of the sample in grams yields the concentration
of Re in mg per gram, or ppm. Derivations, explanations,
and examples of applications for this and many other calcu-
lations based on isotopic data are given in Faure and
Mensing (2005).
Today, isotopic analyses of Re and Os are most com-
monly acquired in three steps:
(1) dissolution of the sample and equilibration with isoto-
pic spikes in a closed vessel;
(2) chemical purification of Re and Os, separating each
from the matrix and from each other;
(3) isotopic analysis of each element by negative thermal
ionization mass spectrometry (NTIMS).
For Step 1, the sample and spikes are loaded into thick-
walled Pyrex tubes (Carius tube) with inverse aqua regia,
sealed, and equilibrated for at least 24 h at 240  C
(Shirey and Walker, 1995). To isolate organic matter or
sulfides in shale samples without dissolving or signifi-
cantly leaching detrital minerals, a solution of CrO3–
H2SO4 is used in place of inverse aqua regia (Selby and
Creaser, 2003; Kendall et al., 2004).
For Step 2, Os is separated from the solution in the
Carius tube either by distillation (Morgan and Walker,
1989) or dissolution in an organic phase (CCl4 or CHCl3)
and back extraction into HBr (Shen et al., 1996; Cohen
and Waters, 1996; Brauns, 2001), followed by microdis-
tillation (Roy-Barman and Allègre, 1995; Birck
et al., 1997); both utilize the volatility of Os under
oxidizing conditions and its quantitative drawdown into
HBr. Re is subsequently purified on ion exchange col-
umns (Morgan and Walker, 1989).
For Step 3, although NTIMS is almost universally used
for high-precision measurement of Os isotopes, some labs
use solution multi-collector inductively coupled plasma
mass spectrometry (MC-ICP-MS) to determine the Re iso-
topic ratio. Spot analyses, such as laser ablation
MC-ICP-MS for Re–Os isotopic analyses, are not feasible
in most cases for one or both of two reasons. First, most Os
concentrations and some Re concentrations are too low,
below detection limits for LA-MC-ICP-MS. Second, in
some materials – notably in molybdenite – the radiogenic
daughter isotope 187Os may diffuse and collect in crystal
defects, so that it is spatially decoupled from its parent
187
Re (Stein et al., 2001, 2003; Košler et al., 2003; Selby
and Creaser, 2004). The resulting ages may be highly erro-
neous, either too old or too young (see discussion in later
section).
Prior to interpretation of the isotopic data, raw measure-
ments of isotopic ratios from mass spectrometry are
corrected for instrumental mass fractionation, and all ana-
lytical uncertainties must be identified and appropriately
calculated. Mass fractionation of Os isotopes is readily
corrected for each measurement by using
190
Os/188Os ¼ 3.08271 (Platzner, 1999). Because Re has
only two isotopes, however, and 185Re is introduced as
an isotopic spike, there is no stable, unchanging isotopic
ratio in the sample to measure. If MC-ICP-MS is used,
however, the sample can be doped with an element of sim-
ilar atomic weight that has at least two stable isotopes, typ-
ically Ir or W, and the mass fractionation of those isotopes
used to correct for that of 185Re and 187Re (e.g., Schoen-
berg et al., 2000; Li et al., 2010). For NTIMS, Re isotope
710 RHENIUM–OSMIUM GEOCHRONOLOGY: SULFIDES, SHALES, OILS, AND MANTLE

fractionation is determined from repeated measurements
of a standard and assumed to be the same for sample ana-
lyses. Measured ratios are corrected for mass fractionation
and isobaric interferences. Other sources of uncertainty in
Re–Os analyses arise from weighing, standard and spike
calibrations, and blanks. The 2s uncertainties for Re and
Os abundances and the isotopic ratios are calculated by
numerical propagation of all uncertainties.
Given the measured abundances of the parent (187Re)
and daughter (187Os) isotopes, each normalized to 188Os,
a stable isotope, the age and initial composition of the
Os can be calculated from the basic age equation:
187
187
187
consanguineous samples and fitting a linear regression
on an isochron diagram, a plot of the measured
187
Re/188Os versus 187Os/188Os (Figure 3).
Some samples have very little or effectively no com-
mon Os on formation. The mineral molybdenite (MoS2),
for example, takes in large amounts of parent Re, but
almost quantitatively excludes Os at the time of crystalli-
zation. Other sulfides may also have unusually high
187
Re/188Os ratios in the thousands or tens of
thousands – so-called low-level highly radiogenic
(LLHR) samples (Stein et al., 2000). In such cases, for
which the Os is nearly 100 % radiogenic, the traditional

isochron diagram is meaningless. Any presumed measure-

Os Os Re  lt  ment of 188Os clearly carries an extreme uncertainty (since
188
¼ 188 þ 188 e 1 there is too little to measure!). This is described in more
Os measured Os initial Os measured
detail below in the sections that describe dating molybde-
ð2Þ nite and LLHR samples.
where the measured ratios are determined from mass spec-
trometry, the initial ratio is the isotopic composition of Os Historical background
at the time the sample equilibrated with and derived from The promise of highly accurate and precise Re–Os chro-
its source, l is the 187Re decay constant, and t is the age of nology was realized in the mid-1990s after four major,
the sample in years. In any sample with measureable analytical issues were overcome: (1) lack of sensitive
“common” Os (non-radiogenic Os incorporated into the mass spectrometry techniques, (2) poor stoichiometry of
sample when it formed), the initial Os isotopic composi- Os standard materials, (3) difficulty equilibrating spike
tion of the common Os must be known. Thus there are and sample Re and Os, and (4) uncertainty in the 187Re
two unknowns: the initial Os isotopic composition and decay constant. Resolving these issues opened the
the age. This is overcome by measuring multiple door to a wide range of applications for geologic problems

Rhenium–Osmium Geochronology: Sulfides, Shales, Oils, and Mantle, Figure 3 Example of a Re–Os isochron plot. Four shale
samples (or four related sulfide minerals), for example, were deposited at time zero, and all acquired Os from seawater with the same
initial 187Os/188Os ratio. Over time, as 187Re decays to 187Os, the position of each point shifts up and to the right. The increase in
187
Os/188Os is proportional to the magnitude of 187Re/188Os for each point. Linear regression of the four points at t >0 yields a
straight line with the equation shown. The slope (m ¼ elt1) is a function of the age of the sample and the decay constant. The
y-intercept is the initial 187Os/188Os ratio, the ratio that would exist in a sample that had no parent 187Re. The uncertainties on the
slope and initial 187Os/188Os ratio depend on the analytical uncertainties for each data point and the goodness of fit (MSWDs, or mean
square of weighted deviations) for the regression.
 RHENIUM–OSMIUM GEOCHRONOLOGY: SULFIDES, SHALES, OILS, AND MANTLE
from the mantle to the crust, from magmatic systems to
surface processes.
Negative TIMS
Many isotopic methods rely on measurements of positive
ions by thermal ionization mass spectrometry (TIMS).
Recognition that both Re and Os are much more effi-
ciently ionized as negative ions enabled far more precise
measurements of Re and Os isotopic compositions on
picogram quantities (Creaser et al., 1991; Völkening,
et al., 1991). The switch to negative TIMS (NTIMS) was
critical to the development of Re–Os isotope geochemis-
try of crustal materials, as many labs turned to this new
and more precise analytical tool. Analysis of Re isotopes
by MC-ICP-MS has continued in some labs because it
affords the opportunity to correct for instrumental mass
fractionation of Re isotopes. In addition, a few labs have
developed high-precision methods for analysis of Os iso-
topes by direct sparging of Os from the reaction vessel into
an MC-ICP-MS (e.g., Hassler et al., 2000; Nozaki
et al., 2012; Sen and Peucker-Ehrenbrink, 2014).
Stoichiometry of Os standards
Accurate concentration analyses using isotope dilution
depend critically on the accuracy of the gravimetric stan-
dards used to calibrate spikes. The materials used to make
Os standards are notoriously nonstoichiometric, leading to
inherent errors in the calculated Os concentrations
(Morgan et al., 1995). Alternative materials have been
used to overcome this (Shen et al., 1996; Yin
et al., 2001), but the inconsistencies remained. Selby and
Creaser (2001) adapted an earlier method (Gilchrist,
1932) to reduce ammonium hexachloroosmate to a pure
osmium metal sponge and thereby directly measure the
Os content of the salt. A further adaptation of this method
yields a robust measurement for the Os standard used
today at AIRIE (Markey et al., 2007). Nevertheless, the
persistent problem with stoichiometry of Os gravimetric
standards remains challenging.
Spike challenges: equilibration difficulties and the
mixed Re-double Os spike
A long-standing challenge was achieving isotopic equilib-
rium between the sample and Os spike without loss of
Os. Loss of Os is especially problematic for open-vessel
methods or use of PTFE vessels, through which Os may
diffuse. External precision (reproducibility) was greatly
improved as laboratory methods moved from less
reliable equilibration of spike and sample Re and Os by
oxidizing alkaline fusion (Morgan and Walker, 1989:
Markey et al., 1998) to high-pressure and high-
temperature sample dissolution and spike equilibration
using inverse aqua regia in a sealed Carius tube (Shirey
and Walker, 1995).
A particular challenge with spiking is presented by
samples with low concentrations of common Os and very
high concentrations of radiogenic 187Os. There are no
711
well-measured ratios of Os isotopes that are neither radio-
genic nor impacted by the addition of the spike. Thus,
there is no mechanism for monitoring mass fractionation
or determining the amount of common Os. These prob-
lems were resolved simultaneously with development of
a mixed Re-double Os spike (185Re–188Os–190Os;
Markey et al., 2003). Sufficient 188Os spike is added to
match the abundance of 187Os, so that the 187Os/188Os
ratio is well measured. In the same spike solution, a small
amount of 190Os is used to achieve a well-measured ratio
with 192Os, the most abundant isotope in the common
Os spectrum. The well-determined 188Os/190Os ratio is
used for the fractionation correction, while the balanced
190
Os/192Os ratio is used to determine the amount of com-
mon Os in the sample. At the same time, use of a mixed
Re–Os spike eliminates weighing errors associated with
spike addition. This breakthrough has dramatically
improved the precision and accuracy of Re–Os ages for
young molybdenites and LLHR samples.
Half-life hurdles: the 187Re decay constant
Establishing the half-life of a parent isotope with certainty
is imperative for radiometric dating. Molybdenite, with its
extraordinarily high Re concentrations and negligible
common Os, contains essentially pure 187Os, thereby pro-
viding the first reliable estimates of the half-life of 187Re at
about 43.5 b.y. (Herr and Merz, 1955; Hirt et al., 1963).
Fifteen years later, the 187Re half-life was determined at
45.6  1.2 b.y. by assuming the slope of a meteorite iso-
chron represented an age of 4.550 Ga (Luck and Allègre,
1982). Re–Os ages for 11 molybdenites based on this
revised half-life showed that, in a few cases, ages were
too old (some older than the age of the earth). These highly
inaccurate results were attributed to Re loss due to meta-
morphic and/or hydrothermal processes (Luck and
Allègre, 1982) – a major setback for Re–Os dating of
molybdenite, as the chronometer was labeled as
unreliable.
The half-life of 187Re was again redetermined as 42.3 
1.3 Ga by measuring ingrowth of 187Os over several years
in a purified kilogram of Re (as perrhenic acid, HReO4;
Lindner et al., 1989). Analysis of the same molybdenites
that previously gave aberrant ages (Luck and Allègre,
1982) now produced ages in good agreement with wall
rock ages determined by other isotopic methods, leading
to the conclusion that earlier discordant results reflected
sample-spike equilibration problems rather than issues
with the molybdenite chronometer (Suzuki et al., 1992,
1993).
The 3 % uncertainty on the 187Re half-life determined
by Lindner et al. (1989) placed serious limits on Re–Os
geochronology. Two labs simultaneously undertook
improvements by back-calculating the 187Re decay con-
stant from Re–Os isotope systematics in iron meteorites
well dated by other methods. Shen et al. (1996) calculated
a 187Re decay constant of 1.66  1011 year1; while they
do not report an uncertainty for their decay constant, it is
712 RHENIUM–OSMIUM GEOCHRONOLOGY: SULFIDES, SHALES, OILS, AND MANTLE
tied to the 1.2 % uncertainty on the calibration of their Os
tracer. Smoliar et al. (1996) examined Re–Os systematics
in group IIIA iron meteorites; by assuming these formed
at the same time as angrite meteorites that yield a concor-
dant Pb–Pb model age (Lugmair and Galer, 1992), they
were able to calculate the 187Re decay constant at (1.666
 0.017)  1011 year1. The assumption that the analyzed
group IIIA and angrite meteorites are the same age is
supported by agreement within 5 m.y. of the 53Mn–53Cr
formation intervals. The 1.02 % uncertainty calculated
for the Smoliar et al. (1996) 187Re decay constant can be
reduced to 0.31 % if the lab’s Os standard can be tied back
to the University of Maryland standard used by Smoliar
et al. (1996).
Persistent misperceptions
It is not about closure temperatures! Rather, it is the “con-
tainer” that assures isotopic closure. The geochemical
behavior of Re and Os controls the reliability of the
Re–Os chronometers. Neither element is accommodated
by silicate minerals, and neither element is soluble in
reducing aqueous fluids. Thus, there are only two routes
for either Re or Os to escape from its hosting phase:
(1) sit adjacent to another phase with a comparable affinity
for either Re or Os under conditions that allow diffusion or
(2) interact with an oxidizing fluid. An isolated sulfide
mineral surrounded by silicates will not lose Re or
Os. Surface weathering or attack by any oxidizing fluid,
in contrast, promotes disturbance of the Re–Os isotopic
system (e.g., Georgiev et al., 2012).
A “Model 1” isochron does not mean “precise and
accurate”! Many labs plot isochrons from their data using
the freely available software Isoplot (Ludwig, 2012). The
first model for linear regression used by Isoplot is termed
“Model 1” and assumes that scatter of data from the
straight line is attributable only to the uncertainties of the
individual data points (following York, 1969). The pro-
gram also calculates a “probability of fit” and rejects the
Model 1 regression if the probability is <15 %. (The user
may change this value to any probability between 5 % and
30 %.) If the Model 1 regression is rejected, Isoplot calcu-
lates a “Model 3” age, which assumes an unknown, but
normally distributed variation in the initial 187Os/188Os
ratio. This change in regression model results in a larger
age uncertainty. In fact, the size of the uncertainties for
individual data points strongly impacts the choice of
model – the larger the uncertainties, the higher the proba-
bility of fit, and therefore, the greater likelihood that the
Model 1 fit will be accepted. Thus, “better” data – that
is, more precise analyses and/or underestimated
analytical uncertainties – are less likely to result in a
Model 1 regression. Conversely, less precise measure-
ments and/or overestimated analytical uncertainties for
each data point readily result in an unwarranted Model
1 regression. Examples of improbably high analytical
uncertainties and overly optimistic isochrons are all too
common.
A lower MSWD does not always mean a better iso-
chron! Ludwig (2012, footnote on p. 22) eloquently states
the meaning of “MSWD”:
MSWD ¼ Mean Square of Weighted Deviates. This does not
refer to porcine perverts; it is, roughly, a measure of the ratio
of the observed scatter of the points (from the best-fit line) to
the expected scatter (from the assigned errors and error corre-
lations). The MSWD parameter cannot be compared to the
classical R2 parameter, and is not a measure of how highly
correlated the X- and Y-values are. If the assigned errors
are the only cause of scatter, the MSWD will tend to be near
unity. MSWD values much greater than unity generally indi-
cate either underestimated analytical errors, or the presence
of non-analytical scatter. MSWD values much less than unity
generally indicate either overestimated analytical errors or
unrecognized error-correlations.
In short, a low MSWD may result from an
overestimation of analytical uncertainties. Equally impor-
tant, an MSWD 1 suggests that analytical uncertainties
have been underestimated – though it can also indicate an
extraordinarily colinear data set. Ideally, for a precise iso-
chron based on good measurements of well-behaved sam-
ples, the MSWD will be close to 1.
Applications
The many improvements made during the 1990s for ana-
lytical methods and standards opened a world of new
applications for Re–Os geochronology. Precision and
accuracy of measurements of meteoritic material and man-
tle xenoliths greatly increased. With the Smoliar
et al. (1996) 187Re decay constant, the first NTIMS ana-
lyses of molybdenites aligned with U–Pb ages for related
zircons (Stein et al., 1997, 2001). High-precision and
accurate analyses of low-level basalts, crustal sulfides,
and organic matter became a reality. The next sections pro-
vide an overview of these applications, each of which is
also discussed in separate focused and condensed chapters
in this volume. Here we pay particular tribute to the devel-
opment and application of the Re–Os molybdenite chro-
nometer, as its value is no less than that of zircon dating.
Meteorites and mantle
Much of the early work on Re–Os systematics in earth
materials focused on xenoliths from the earth’s mantle
and on meteorites (see also entry “Rhenium–Osmium
Dating (Meteorites)”). Prior to the 1990s, analytical limi-
tations required the use of materials with relatively high
concentrations of Re and Os. More importantly, the strong
fractionation of both Re between earth’s mantle and
mantle-derived partial melts, and of both Re and Os
between silicates and metals, affords ample opportunities
to explore the origins of planetary bodies and the evolu-
tion of planetary interiors. Re–Os systematics can reveal
the age and probable origin of meteorites from the solar
nebula, the formation of Fe–Ni cores, and timing of sec-
ondary alteration (cf. Walker, this volume). Studies
focused on shergottite meteorites, of Martian origin,
 RHENIUM–OSMIUM GEOCHRONOLOGY: SULFIDES, SHALES, OILS, AND MANTLE
reveal the evolution of the Martian mantle, early magma
ocean, and late accretion (e.g., Brandon et al., 2012).
Re–Os isotopic studies of mantle xenoliths, exposed
mantle peridotites, and mantle-derived basalts provide
high-quality information on melt-depletion events in the
mantle, the evolution of the mantle through time, and the
preservation of subcontinental mantle lithosphere.
Reisberg et al. (2004) provide an example of multiple pro-
cesses impacting Os isotope composition of mantle sam-
ples preserved as xenoliths, deriving useful model ages
for melt extraction, yet questioning the interpretation of
these data as mantle stabilization ages. Rudnick and
Walker (2009) give an excellent review of the fundamen-
tals, potential pitfalls, and lessons learned in two decades
of Re–Os applications in the mantle.
Crustal sulfides
Molybdenite geochronology
Molybdenum disulfide (MoS2) or molybdenite is a silver-
colored mineral with a distinct bluish tint. The blue tint is
particularly notable in samples with high Re. This hexag-
onal (rarely trigonal) mineral is soft, has a distinct platy
cleavage, and can easily be bent with the fingers
(Figure 4). Molybdenite occupies a unique niche – it is
the only naturally occurring mineral that takes in signifi-
cant Re, usually at the level of tens to hundreds of ppm,
other than its exceedingly rare Re-rich solid-solution
cousin, rheniite (ReS2; Korzhinsky et al., 2004; Tessalina
et al., 2008). Molybdenite occurs in a wide range of geo-
logic environments. Almost invariably, it is present in
porphyry-style ore deposits worldwide. It may also be
found as a minor accessory mineral in silicic-felsic meta-
morphic rocks and biotite gneisses. Molybdenite is most
Rhenium–Osmium Geochronology: Sulfides, Shales, Oils, and
Mantle, Figure 4 Molybdenite crystals from Råde, Norway,
illustrate hexagonal form and platy texture. This exceptional
specimen is about 6 cm across and belongs to the Natural
History Museum in Oslo. Crystals as large as 15 cm in diameter
have been reported from Råde (Photo by H. Stein).
713
commonly associated with quartz veins or as disseminated
blades or rosettes in felsic magmatic-metamorphic rocks
(Figure 5). The ability to directly date molybdenite gave
the economic geology community accurate and precise
ages for a wide array of ore deposits and gave the geologic
community direct ages for sulfide occurrences in the
Earth’s crust (Stein et al., 2001; cf. review in Stein, 2014).
Uniquely, molybdenite has very high Re/Os ratios on
crystallization (generally >106) and takes in essentially
no common (i.e., initial) Os. If common Os is present in
the sulfide-depositing fluid, it is accommodated by other
co-crystallizing sulfides. With vanishingly low common
Os (and therefore, no initial 187Os) in molybdenite, the
decay equation for calculating a molybdenite age is ele-
gant in its simplicity:
 
187
Osmeasured ¼ 187 Osinitial þ 187 Remeasured elt  1
ð3Þ
reduces to
 
187
Osmeasured ¼ 187 Remeasured elt  1 ð4Þ
Here the model age (t), reflecting the time of crystalliza-
tion, is the only unknown. The age is therefore readily
determined from a single sample of molybdenite. Lambda
(l) is the 187Re decay constant.
The 187Re–187Os radiometric clock held in the mineral
molybdenite was explored in earnest in the 1960s (Hirt
et al., 1963; Herr et al., 1967). Analytical issues and a
poorly determined 187Re decay constant stalled progress.
In the 1990s, early pioneering work in 187Re–187Os dating
of molybdenite was carried out by Du Andao, a Chinese
chemist, who early on realized the potential of the method
for this mineral (Du et al., 1995). Her lab quickly became a
venue for many Chinese geologists to simply acquire
dates from her for Chinese ore deposits, and a hailstorm
of papers reporting molybdenite ages ensued. These geol-
ogists depended on her meticulous work and laboratory
results. Also in the 1990s, another chemist, Katsuhiko
Suzuki, was independently exploring molybdenite as a
geochronometer in Japan. His lab became a focal point
for mixing geologists and geochemists to jointly tackle
geologic problems with Re–Os isotope geochemistry. In
the early 1990s, both Drs. Du and Suzuki were limited
by the analytical capability offered by ICP-MS
(Inductively Coupled Plasma Mass Spectrometry) prior
to the advent of multi-collector instruments.
Other groups were also limited by ICP-MS instrumen-
tation and by assessing the accuracy of 187Re–187Os
molybdenite ages through their agreement with K–Ar
ages (e.g., McCandless and Ruiz, 1993). It soon became
clear that primary ages in the ore-forming environment
were held by molybdenite and not by traditionally dated
ore-associated gangue minerals (e.g., alunite and sericite),
for which argon-based dating had been employed (Suzuki
et al., 1996; Watanabe and Stein, 2000; Stein et al., 2001;
Selby et al., 2002).
714 RHENIUM–OSMIUM GEOCHRONOLOGY: SULFIDES, SHALES, OILS, AND MANTLE

Rhenium–Osmium Geochronology: Sulfides, Shales, Oils, and Mantle, Figure 5 (a) Quartz-molybdenite veins in the Primos
rhyolite porphyry, Henderson Mo porphyry deposit, Colorado. Field of view is 10 cm across. (b) Banded quartz-molybdenite vein
cutting fine-grained biotite hornfels, Myszków Mo deposit, Poland. Thumb and forefinger to right, pencil tip to left for scale. Sample is
wet to enhance contrast. (c) Porphyritic granite with disseminated molybdenite and minor finely disseminated chalcopyrite (bottom
center) from the Myszków Mo deposit, Poland. Two drill pits (inside yellow circle) show how molybdenite is excavated for Re–Os
dating. Pencil tip for scale. Sample is wet to enhance contrast (Photos by H. Stein).

The ages achieved by early workers, nonetheless, were
encouraging. But a breakthrough in analytical technology
was needed. This was delivered with NTIMS (Negative
Thermal Ionization Mass Spectrometry), as two working
groups realized that Re and Os could be measured as neg-
ative ions (Creaser et al., 1991; Völkening et al., 1991).
This analytical breakthrough was immediately utilized to
successfully develop the 187Re–187Os chronometer in
molybdenite, achieving accurate ages at high precision
(Stein et al., 1997; Markey et al., 1998).
In addition to analytical challenges, a limiting factor in
putting the molybdenite chronometer to use was the accu-
rate determination of the 187Re decay constant and its
uncertainty. Ultimately, the 187Re decay constant of
Smoliar et al. (1996) became the established value and is
the most widely used today. More recent attempts to fur-
ther revise the 187Re decay constant by pairing up molyb-
denite Re–Os and zircon U–Pb ages from the literature
(Selby et al., 2007a) rely on two geologic assumptions:
(1) molybdenite and zircon formation are contemporane-
ous, and implicit to the first assumption, (2) there is only
one generation of molybdenite and one episode of zircon
formation at a locality. In hindsight, geologic reason tells
us this assumption is flawed; when multiple dates are
available, molybdenite (and zircon) generally shows a
range of ages at a single locality. The cleanest data sets
are from porphyry-style deposits, but these also may
have extended durations (e.g., Los Pelambres, Chile;
Stein, 2014). Therefore, the 187Re decay constant that ties
back to the meteorite community is best used; the com-
plexities of crustal rocks do not lend themselves well to
establishing decay constants.
Still other vexing issues lay ahead before 187Re–187Os
dating of molybdenite would become mainstream
geochronology. Recognition of parent-daughter
(187Re–187Os) decoupling within molybdenite crystals
presented a new challenge (Stein et al., 2001). Qualitative
trial-and-error studies with different grain sizes, different
crystals (chopped with a razor, peeled along cleavage
planes, or powdered by drilling), different occurrences,
and different milligram quantities of molybdenite revealed
that geologic sense, not quantity, of molybdenite should
drive the sampling. If part of a crystal is sampled, then
the whole crystal must be sampled to ensure that all of
the parent and daughter are included in the analysis. In
fact, acquisition of the molybdenite mineral separate in a
geologically meaningful context is the most critical step
for molybdenite dating (Stein et al., 2001; Stein, 2006,
2014). Unlike the approach for zircon U–Pb dating,
whole-rock crushing to acquire a vial of molybdenite
grains puts Re–Os dating at risk. If more than one genera-
tion of molybdenite is present, even if closely spaced in
time, this approach yields variable ages that cannot be well
reproduced. In the same way, if decoupling is present at
the scale of the sampling, ages will be variable and cannot
be well reproduced. It is the inability to replicate an age
from the same separate that signals decoupling. The
decoupling of 187Re–187Os in molybdenite is generally
greater in coarser-grained samples, in older high-Re
molybdenites that have accumulated significant radio-
genic 187Os, and in deformed molybdenites (Stein
et al., 2003). Simply put, the accumulated radiogenic Os
is desperate to get comfortable in the molybdenite struc-
ture and, at any opportunity, will migrate to crystal defects
or dislocations (Stein et al., 2003). The early empirical rec-
ognition of parent-daughter decoupling was later verified
through LA-ICP-MS of molybdenite (Stein et al., 2003;
Selby and Creaser, 2004). Thus, laser ablation technology
cannot be used for 187Re–187Os molybdenite dating (e.g.,
Košler et al., 2003). A subsequent atomic level study con-
firmed the suspected mobility of radiogenic Os within the
molybdenite structure (Takahashi et al., 2007).
A unique aspect of 187Re–187Os dating of molybdenite
is that this sulfide rarely forms overgrowths (Stein
et al., 2004; Aleinikoff et al., 2012). This is not the case
for other sulfides; for example, pyrite is notorious for
overgrowths. Without the complexity of overgrowths, this
makes working with molybdenite far simpler than
deciphering some U–Pb histories in zircon and monazite.
Another positive feature is that the molybdenite
 RHENIUM–OSMIUM GEOCHRONOLOGY: SULFIDES, SHALES, OILS, AND MANTLE
chronometer remains intact, even at granulite facies tem-
peratures (e.g., Bingen and Stein, 2003). Compromise of
the 187Re–187Os clock in molybdenite should not be asso-
ciated with temperature, but with chemical stability of sul-
fide (i.e., oxidation; cf. discussion in Stein, 2014).
In nature, however, there are always exceptions when
things at first seem simple. Though rare, a small popula-
tion of molybdenites began to show up with measurable
common Os. Not surprisingly, these molybdenites gener-
ally occur in the absence of other sulfides that accommo-
date Os, a platinum group element (PGE), in their
structures. Silicate minerals typically do not substitute
Os in their structures, so in such circumstances this left
molybdenite vulnerable to squeezing Os into its structure.
This reality led to the development of a mixed Re-double
Os spike, now widely in use, for analyzing molybdenite
(Markey et al., 2003). The concept is simple: a single spike
solution contains enriched 185Re–188Os–190Os. The 185Re
is used to determine the parent 187Re concentration in a
molybdenite sample, as there are only two isotopes of
Re. The 188Os in the spike solution is used to precisely
determine the radiogenic 187Os, just one amu (atomic
mass unit) away, and the 190Os is used to determine the
192
Os (the most abundant common Os mass). From these
measurements, the total Re and Os concentrations in the
sample can be reconstructed, and the dating is atomically
accurate. Further, the double Os spike eliminates weighing
errors for single Re and Os spike solutions and permits a
mass fractionation correction for Os (Markey et al., 2003).
A new technology requires a reference material that can
be accessed by multiple labs and used for interlaboratory
comparison. Early workers used a well-homogenized,
milled molybdenite powder from the Huanglongpu Mo
deposit in China to assess internal (within lab) and exter-
nal (between labs) repeatability of age results (Stein
et al., 1997; Selby and Creaser, 2001; Du et al., 2004).
From this, the concept of a molybdenite isochron was
derived, with the regression line defined by coupled vari-
ation in 187Re and 187Os concentrations and projecting to
an intercept of zero initial 187Os (Stein et al., 1997). Sub-
sequently, a milled molybdenite powder from the Hender-
son mine in Colorado was characterized as a reference
material (Markey et al., 2007). Molybdenite from this par-
ticular porphyry Mo deposit was specifically chosen for
three reasons: (1) undeformed, minimizing decoupling
issues, (2) relatively low Re, and (3) young age (27.656
 0.022 Ma), limiting the amount of 187Os introduced as
a possible lab contaminant. The Henderson molybdenite
was endorsed by NIST (RM#8599) and can be purchased
through NIST as a reference material for 187Re–187Os age
determinations. Importantly, RM#8599 is not a reference
material for Re and 187Os concentrations, only for the
accuracy of 187Re–187Os ages. The Re and 187Os concen-
trations vary, but are perfectly coupled such that the age is
highly reproducible (Markey et al., 2007).
While dating ore deposits was the original motivation
and a primary application for molybdenite dating, the min-
eral’s widespread occurrence as an accessory phase
715
provides additional opportunities. Molybdenite crystalli-
zation can be tied to specific tectonic events, yielding
timing, for example, of fluid-flow events during metamor-
phism (Stein, 2006), specific deformation events (Stein
and Bingen, 2002), and progressive and episodic exten-
sional tectonism (Zimmerman et al., 2008).
In sum, molybdenite presents the geologic community
with a simple, yet elegant, radiometric clock through
187
Re–187Os. What was novel turned mainstream with
NTIMS, the use of a double Os spike, and a certified refer-
ence material for the community. Depending on Re con-
centration in the molybdenite, the chronometer can
accommodate samples from Early Archean to well less
than 1 Ma. Potential decoupling can be overcome with
sensible sampling and thoughtful mineral separation. For
a detailed review of molybdenite and sulfide geochronol-
ogy, rich in examples, consult Stein (2014).
In this brief section, we have focused on the steps lead-
ing to breakthroughs and the hurdles in establishing the
187
Re–187Os chronometer for this unassuming, soft, pli-
able, and isotopically strong mineral.
LLHR (low-level highly radiogenic) sulfides
The LLHR term was introduced to describe samples with
molybdenite-like Os isotopic compositions but with much
lower Re concentrations, at the ppb or even ppt level
(Stein et al., 2000). That is, the common Os is low enough
relative to the radiogenic 187Os that single samples may
provide accurate and precise 187Re–187Os ages. The prin-
ciples are the same as described for molybdenite, with the
caveat that the blank correction may be important. The key
equation for dating is therefore Eq. 4 (above).
Arsenopyrite (FeAsS) and some pyrites (FeS2) may be
LLHR. Because arsenopyrite and pyrite are commonly
associated with Au, these two sulfides are sought out for
dating gold deposits when molybdenite is absent from
the sulfide assemblage. Like molybdenite, arsenopyrite
and pyrite may permit 187Re–187Os dating of ore deposits.
Further, pyrite is the most common sulfide in the earth’s
crust, so this mineral has additional value for Re–Os geo-
chronology in general. If sulfides, including arsenopyrite
and pyrite, have significant common Os, they can still be
dated using the traditional 187Re/188Os versus 187Os/188Os
isochron approach, but this requires analyses of multiple
contemporaneously formed samples. For further discus-
sion of LLHR samples, refer to Stein (2014). Several
key studies brought attention to arsenopyrite as a valuable
sulfide for Re–Os dating (Arne et al., 2001; Morelli
et al., 2005; Bierlein et al., 2006; Morelli et al., 2007,
2010). In addition, Morelli et al. (2010) showed that the
Re–Os systematics in arsenopyrite were intact, even at
temperature >400  C. Although unusual, even pyrrhotite
can be LLHR, yielding primary ages if the sample does
not include other sulfides (Demaiffe et al., 2013).
Dating shales: source rocks for oil
The Re–Os budget held in crustal rocks is dominated by
shales (Figure 1). Groundbreaking Re–Os work on dating
716 RHENIUM–OSMIUM GEOCHRONOLOGY: SULFIDES, SHALES, OILS, AND MANTLE
shales relied on whole-rock analyses of organic-rich
shales (e.g., Ravizza and Turekian, 1989; Creaser
et al., 2002). Subsequently, Hannah et al. (2004) made use
of syn-sedimentary pyrite with remarkable success on sam-
ples of Paleoproterozoic age (2,316  7 Ma). The field of
Re–Os geochronology quickly expanded to include dating
of the organic (hydrogenous) component of organic-rich
shales isolated from the clastic or detrital component of the
shale (e.g., Selby and Creaser, 2003, 2005a; Kendall
et al., 2009; Yang et al., 2009). Re and Os concentrations
in the organic component are easily in the ppb range. The
technique soon became well established as a tool to place
age pins in the stratigraphic column for geologic time scale
calibration. Previously, only U–Pb zircon ages on ash beds
had prevailed as absolute time markers in sedimentary sec-
tions. Absolute time is also a valuable commodity for corre-
lations where provincial species could not be used to link
widely separated paleogeographic terranes (e.g., Xu
et al., 2009).
The newfound ability to date shales led quickly to dat-
ing key units associated with major earth events; this can
help resolve debates over the reach of these events in time
and space – whether global or regional, abrupt or
protracted. For example, localities with shale records asso-
ciated with Neoproterozoic glaciation and the theory of a
frozen snowball earth could now be time-bracketed with
Re–Os dating of shales (e.g., Kendall et al., 2004, 2006;
Rooney et al., 2010, 2011; Kendall et al., 2013). Other
earth events such as biotic crises associated with mass
extinctions or major faunal turnovers provide excellent
venues for Re–Os dating, as widespread anoxia preserved
vast areas of organic material intact with their trace metal
record of toxic seawater (e.g., Turgeon and Creaser, 2008;
Selby et al., 2009; Georgiev et al., 2011; Xu et al., 2014).
Since organic-rich black shales are potential source
rocks for hydrocarbons, the petroleum industry became
immediately interested in using the method not only as a
measure of absolute time but as a correlation tool across
basins, continents, and hemispheres. Good isochron
regressions define initial 187Os/188Os ratios for paleo-
seawater, useful measures for what was entering, mixing
with, and characterizing the Re–Os budget of oceans on
a more global scale (see Figure 8). Simplistically summa-
rized, the marine geochemical record, including Re–Os
chemistry, is transferred to black shales on the seabed.
During periods in earth history when the whole of seawa-
ter may have been anoxic, including the photic zone (e.g.,
Grice et al., 2005), the connection between top and bottom
is profound. A more stratified ocean with variable oxygen-
ation presents a more complex architecture linking the
photic zone and seabed. This scenario holds for anoxic-
euxinic waters in lacustrine sediments as well. With ade-
quate ventilation and mixing, the seawater (lake water)
Os isotope ratio represents the balance of (1) continental
erosion (intimately linked to continental uplift) and
(2) hydrothermal input. Continental erosion at a particular
time in earth history reflects the kinds of rocks that are
dominating the erosional cycle, that is, what rivers are
delivering to seas. Hydrothermal input at a particular time
in earth history may be dominated by mafic magmatism
penetrating the seabed or terrestrial volcanic systems
belching aerosols and particulate matter that ultimately
mix at the seawater interface where they may incite algal
blooms or otherwise feed microorganisms in the photic
zone. While the notion of a global seawater Os isotopic
composition is a tidy concept, it soon became apparent
that restricted basins and isolated water masses, both hor-
izontally and vertically, both locally and widespread, have
the potential to harbor Os isotopic compositions that devi-
ate from an assigned global Os isotopic value. Neverthe-
less, the concept of a variable Os isotopic composition
for seawater through time works well, and the short resi-
dency time for Os in seawater gives us an increasingly
tuned record of past earth events at high resolution
(see Figure 8).
As discussed throughout this contribution, the Re–Os
data are governed by geologically responsible sampling.
Drill core samples are highly preferred to outcrop samples
to eliminate the possibility of disturbed isotope systemat-
ics related to weathering (Georgiev et al., 2012). At a min-
imum, flawed sampling at the outset compromises results
and, more seriously, may disguise potentially meaningful
results. Since we do not know the rate of sedimentation,
proper sampling of shale means that the smallest possible
interval, but representative, should be taken. After this
interval is powdered, an aliquot of several hundred milli-
grams is removed for the Re–Os analysis. Sampling too
large a stratigraphic interval risks capturing too large an
interval of time – this compromises isochroneity and risks
reducing the spread of the data points for a statistically
well-constrained isochron regression. Sampling too small
a stratigraphic interval risks capture of postdepositional,
small-scale isotopic exchange – for example, exchange
among organic particles and sulfides after pore waters
have been isolated from the overlying water column or
redistribution of sediment by burrowing organisms.
What does this mean for Re–Os shale chronology?
First, there is no threshold value for the amount of shale
to be acquired, pulverized, and homogenized; each inter-
val of core requires its own sampling strategy. Ideally, an
isochron is derived from 5 to 8 samples, taken from
20 to 30 cm of core (Figure 6). Sampling for an isochron
should not be carried out over many meters or tens
of meters of section. Generally speaking, for 20–30 strat-
igraphic centimeters of section and for ideal, well-
laminated black shales without bioturbation, <30 g of
pulverized shale sample (i.e., <1 cm stratigraphic span)
routinely provides good ages (Figure 7 and discussion in
caption). Again, these are generalized parameters, and
the sample stratagem depends on depositional rates, stabil-
ity of the initial Os ratio, and Re/Os ratios in the shale sam-
ples. Shales with lower Re require larger aliquots for the
analyses, but not necessarily larger bites of the strati-
graphic section for homogenization.
 RHENIUM–OSMIUM GEOCHRONOLOGY: SULFIDES, SHALES, OILS, AND MANTLE 717

Rhenium–Osmium Geochronology: Sulfides, Shales, Oils, and Mantle, Figure 6 Sampling shales is the most critical step for Re–Os
dating. Drill core showing organic-rich Jurassic black shale from Svalbard. The sampling strategy illustrated here is that used by
the AIRIE Program. Re–Os analysis of an aliquot from a shale sample acquired from within each green rectangle is very likely to yield
an accurate and precise isochron age. This particular interval will provide direct age constraints for the Boreal ammonite,
A. subkitchini.

Os isotope seawater curve
The Os isotopic composition of the oceans has varied
through time because of changes in the balance among
the various Os sources, especially inputs from continental
weathering (e.g., Peucker-Ehrenbrink and Ravizza, 2000)
and the several mechanisms for acquiring chondritic
Os. Organic-rich sedimentary rocks, especially black
shales, hold a record of ancient seawater Os isotopic com-
position. By far the best evidence that a sedimentary sec-
tion has remained closed to Os gain, loss, or isotopic
exchange is a suite of samples that form a precise
isochron. In the last two decades, since high-precision
2 m.y.; McArthur et al., 2000) and the comparatively
Re–Os geochronology became a reality, a growing
28,000 years, Georg
archive of initial 187Os/188Os ratios and corresponding
ages has been acquired from isochrons. Figure 8
shows a plot of seawater 187Os/188Os versus age; for com-
parison, the Sr isotope seawater curve (Howarth and
McArthur, 1997; McArthur et al., 2001; Shields, 2007)
is plotted on the same figure. The well-known Sr isotope
curve reflects four decades of effort; the Os isotope curve
is in its early stages, with much still to be developed. In
addition to these data, a number of studies have generated
calculated initial 187Os/188Os ratios, derived from Re–Os
isotopic ratios and a known (or assumed) age (e.g.,
Peucker-Ehrenbrink et al., 1995). Although a closed isoto-
pic system is not proven, many of these data are likely
valid, especially for young samples, and the effect of out-
liers tends to be swamped by the sheer volume of data.
Studies based solely on calculated initial 187Os/188Os
ratios have effectively illustrated, for example, the effect
of a bolide impact at the Cretaceous-Paleogene boundary
(Peucker-Ehrenbrink et al., 1995).
Figure 8 reveals a striking difference between the Sr
and Os isotope seawater curves: the Os curve is spiky,
with sharp highs and lows; the Sr curve is smooth,
with more subdued variations. This is a simple result of
the relatively long residence time for Sr in the oceans
(
short residence time for Os (
et al., 2013; still controversial, Paquay and Ravizza,
2012). Short-term effects, such as bolide impacts or inputs
from large igneous provinces, readily deflect the Os curve
but are washed out by the long residence and mixing of
Sr isotopes in seawater. A second critical difference is the
continual buildup of Re and thus radiogenic Os in the crust.
Neither element was mobilized in surface environments
under earth’s anoxic Archean atmosphere (Hannah
et al., 2004), but both have accumulated since then, seques-
tered in large amounts in organic-rich sedimentary rocks.
While both the Sr and Os seawater curves reflect variable
input from continental weathering, Os is strongly controlled
by the proportion of organic-rich sedimentary rocks
718 RHENIUM–OSMIUM GEOCHRONOLOGY: SULFIDES, SHALES, OILS, AND MANTLE

Rhenium–Osmium Geochronology: Sulfides, Shales, Oils, and Mantle, Figure 7 Sampling shales is the first critical step in Re–Os
dating. Three intervals from a drill core, Jurassic black shale, Svalbard. CT images on top are paired with corresponding photo of the
same drill core piece below. All samples are positioned stratigraphically up toward the top of the page. Cores are 5 cm in diameter.
(a) IDEAL for Re–Os dating. Beautifully laminated organic-rich black shale with lighter, less organic-rich, siltier intervals visible.
(b). SUITABLE for Re–Os dating. Laminated black shale with brachiopod shells concave up indicating settling in quiet water;
disseminated bright points are pyritized foraminifera indicating anoxic-euxinic conditions in the water column and bottom waters.
(c) LESS DESIRABLE for Re–Os dating. Organic-rich black shale shows a myriad of fossil burrows, a record of intense bioturbation.
As the organisms work their way into and out of the soft sediment bed, any variation in the osmium isotope composition is
stirred at the mm to cm scale. With slow sedimentation rates, this has the potential to introduce scatter into an isochron
(Samples collected and imaged by H. Stein and Ø. Hammer).

undergoing weathering (Georgiev et al., 2012), whereas the
Sr isotope balance is controlled by weathering of carbon-
ates and silicates (Ravizza, 2007). The two systems provide
complementary data for deducing causes of changes in
marine chemistry through time.
Hydrocarbons: what are we “dating”?
The proven concentration of Re and Os in organic matter
incites immediate interest in the potential for directly
“dating” hydrocarbons. Simple thought experiments,
however, draw immediate doubts: hydrocarbon systems
are by definition open systems. Hydrocarbons are gener-
ated by maturation of organic material in a source rock,
are expelled, migrate, and may eventually accumulate in
a porous reservoir rock with an adequate seal. At what
point might Re–Os systematics reveal useful information?
How can we know which processes are elucidated by
those systematics?
 RHENIUM–OSMIUM GEOCHRONOLOGY: SULFIDES, SHALES, OILS, AND MANTLE 719

Rhenium–Osmium Geochronology: Sulfides, Shales, Oils, and Mantle, Figure 8 Variations in the 187Os/188Os and 87Sr/86Sr ratios in
seawater through time. Inset shows the Phanerozoic with an expanded scale. The Os isotope curve shows only initial ratios from
isochrons (see discussion in text). The Phanerozoic Sr seawater curve is from Howarth and McArthur (1997) and McArthur et al. (2001);
the Precambrian Sr curve is from Shields (2007). The Os curve was compiled by the AIRIE Program from the literature and
unpublished data.

The first attempted Re–Os measurements on organic
matter focused on physically separated organic detritus
and associated pyrite in the Upper Jurassic Morrison for-
mation (Hannah et al., 2000; Hannah and Stein, 2003).
All samples yielded high, readily measured concentrations
of both Re and Os, with Os and especially Re enriched in
organic matter relative to pyrite. Although some useful
information could be extracted from closely spaced sam-
ples, an approach that subsequently proved useful for dat-
ing shales, most of the section was disturbed by open-
system behavior. The organic detritus served as a sink
for Re and Os dissolved in oxidizing groundwater that
flowed through the Morrison sandstones for tens of mil-
lions of years.
The first breakthrough was a study by Selby and
Creaser (2005b). They analyzed widely spaced samples
from the extensive oil sand deposits of Alberta, Canada.
A subset of their Re–Os data pointed to an age of 112 
5.3 Ma, interpreted as the petroleum emplacement age,
and ruling out models suggesting Late Cretaceous hydro-
carbon maturation and migration. The conclusions assume
that the deposits spanning more than 400 km formed from
the same source at the same time, a hypothesis supported
by distinctive organic geochemistry. A concern is that
9 outliers of the 26 analyses were excluded on the basis
of model initial Os ratios – statistically the same as remov-
ing any points not lying close to the isochron derived from
all data. In addition, the preferred age was partly con-
trolled by a single analysis (out of five) for one unusual
deposit. Nevertheless, their landmark effort constituted a
significant advance in analytical techniques and
demonstrated that useful age information could be
extracted from migrated hydrocarbons.
Subsequent work has documented Re–Os systematics
in several hydrocarbon systems, both marine (e.g., Finlay
et al., 2011; Lillis and Selby, 2013; Hannah et al., 2014)
and lacustrine (Cumming et al., 2014). Moreover, geolog-
ically reasonable ages for hydrocarbon maturation/expul-
sion tied to tectonism have been extracted from shales
(Stein et al., 2012). Comparison of asphaltene and maltene
fractions of individual crude oils (Selby et al., 2007b) and
hydrous pyrolysis experiments (artificial hydrocarbon
maturation; Rooney et al., 2012) both document minimal
differences in both 187Re/188Os and 187Os/188Os between
source rock and extracted oil and among hydrocarbon
fractions. Unfortunately, the oil generated by hydrous
pyrolysis did not contain measurable Re or Os and was
not able to fully replicate natural processes (Rooney
et al., 2012). Nevertheless, these studies affirm early work
that shows little or no disturbance of Re–Os systematics in
organic-rich sedimentary rocks upon hydrocarbon matura-
tion (Creaser et al., 2002 and many subsequent papers on
shale geochronology).
A major challenge for deriving age information from
oils is generating a spread in 187Re/188Os from different
components of a single oil. Without this, a statistically sig-
nificant linear regression cannot be achieved, and thus the
age and initial 187Os/188Os cannot be well determined.
Selby et al. (2007b) showed that Re and Os are concen-
trated in the asphaltene fraction, Mahdaoui et al. (2013)
further separated asphaltenes by sequential precipitation
into strongly and weakly polar fractions. Her results
720 RHENIUM–OSMIUM GEOCHRONOLOGY: SULFIDES, SHALES, OILS, AND MANTLE
further support observations that separation of lighter oils
from asphaltenes does not significantly fractionate Re/Os
or 187Os/188Os, but a method is still needed for inducing
a spread in Re/Os from a single oil sample. Notably, how-
ever, Ni and V concentrations peaked in mid-polarity frac-
tions, while Re and Os dropped continuously from most to
least polar. This argues against the commonly held
assumption that Re and Os, like Ni and V, are bound in
metalloporphyrins.
Re–Os analysis of oils is challenging and very much in
its infancy. The potential to place time constraints on mat-
uration, expulsion, and migration of hydrocarbons
remains a powerful driver of research.
Summary and challenges going forward
Re–Os geochronology can be used to date earth materials
(e.g., sulfides, shales, oils) previously undatable by other
isotopic methods. Of pivotal importance, these materials
are directly linked to natural resources, both minerals
and hydrocarbon, critical to human civilization.
For the immediate future, the challenge is no longer
analytical, but geologic. We need to put forward bright
and innovative thinking, with the basis for that thinking
no longer tied to traditional models, but pointed toward
new understanding of how earth’s resources are ultimately
generated. Re–Os offers more than geochronology.
Acquiring an age for an ore deposit, or shale, gives us
the last step in a long process. We have the data in hand
to start thinking about “how it came about.”
Questions for the future must probe deeper than they
have in the past. How can Re–Os be used to link the
nascent beginnings of resource formation to the end prod-
uct that is mined or drained from the earth’s crust for
profit? This is not esoteric information, but needed infor-
mation for responsible exploration in the next century.
What conditions, far and wide, deep and shallow, make
it possible to liberate metals and hydrocarbons? This is
part of the slippery question of source – where and how
does it start? Our thinking should not be limited to the lith-
osphere and hydrosphere, but must include atmospheric
processes. These three spheres collectively govern
resources. Not surprisingly, they also govern the timely
issue of climate change.
As a first next step and an easy one to take, reconciling
seemingly disparate and non-isochronous data is a good
place to start new thinking. There is no such thing as bad
data; there is only complexity in data sets that should chal-
lenge our imaginations, not prompt us to look only toward
the easily interpreted data sets. The barrage of analytical
data in the last decade should drive new and thoughtful
sampling with different questions in hand.
Bibliography
Aleinikoff, J. N., Creaser, R. A., Lowers, H. A., Magee, C. W., Jr.,
and Grauch, R. I., 2012. Multiple age components in individual
molybdenite grains. Chemical Geology, 300–301, 55–60.
Arne, D., Bierlein, F., Morgan, J. W., and Stein, H. J., 2001. Re–Os
dating of sulfides associated with gold mineralization in central
Victoria, Australia. Economic Geology, 96, 1455–1459.
Bierlein, F. P., Stein, H. J., Coira, B., and Reynolds, P., 2006. Timing
of gold and crustal evolution of the Paleozoic south central
Andes, NW Argentina – implications for the endowment of oro-
genic belts. Earth and Planetary Science Letters, 245, 702–721.
Bingen, B., and Stein, H., 2003. Molybdenite Re–Os dating of bio-
tite dehydration melting in the Rogaland high-temperature gran-
ulites, S Norway. Earth and Planetary Science Letters, 208,
181–195.
Birck, J.-L., Roy, B. M., and Capmas, F., 1997. Re–Os isotopic
measurements at the femtomole level in natural samples.
Geostandards Newsletter, 20, 19–27.
Brandon, A. D., Puchtel, I. S., Walker, R. J., Day, J. M. D., Irving,
A. J., and Taylor, L. A., 2012. Evolution of the Martian mantle
inferred from the 187Re–187Os isotope and highly siderophile
element abundances systematics of shergottite meteorites.
Geochimica et Cosmochimica Acta, 76, 206–235.
Brauns, C. M., 2001. A rapid, low-blank technique for the extrac-
tion of osmium from geological samples. Chemical Geology,
176, 379–384.
Cohen, A. S., and Waters, F. G., 1996. Separation of osmium from
geological materials solvent extraction for analysis by TIMS.
Analytica Chimica Acta, 332, 269–275.
Creaser, R. A., Papanastassiou, D. A., and Wasserburg, G. J., 1991.
Negative thermal ionization mass spectrometry of osmium, rhe-
nium and iridium. Geochimica et Cosmochimica Acta, 55,
397–401.
Creaser, R. A., Sannigrahi, P., Chacko, T., and Selby, D., 2002. Fur-
ther evaluation of the Re–Os geochronometer in organic-rich
sedimentary rocks: a test of hydrocarbon maturation effects in
the Exshaw formation, Western Canada Sedimentary Basin.
Geochimica et Cosmochimica Acta, 66, 3441–3452.
Cumming, V. M., Selby, D., Lillis, P. G., and Lewan, M. D., 2014.
Re–Os geochronology and Os isotope fingerprinting of petro-
leum sourced from a Type I lacustrine kerogen: insights from
the natural Green River petroleum system in the Uinta Basin
and hydrous pyrolysis experiments. Geochimica et
Cosmochimica Acta, 138, 32–56.
Demaiffe, D., Wiszniewska, J., Krzemińska, E., Williams, I. S.,
Stein, H., Brassinnes, S., Ohnenstetter, D., and Deloule, E.,
2013. A hidden alkaline and carbonatite province of early
Carboniferous age in NE Poland: zircon U–Pb and pyrrhotite
Re–Os geochronology. Journal of Geology, 121, 91–104.
Du, A., He, H., Yin, N., Zou, X., Sun, Y., Sun, D., Chen, S., and Qu,
W., 1995. A study of the rhenium-osmium geochronometry of
molybdenite. Acta Geologica Sinica, 8, 171–181 (in English).
Du, A., Wu, S., Sun, D., Wang, S., Qu, W., Markey, R., Stein, H.,
Morgan, J., and Malinovskiy, D., 2004. Preparation and certifi-
cation of Re–Os dating reference materials: molybdenites HLP
and JDC. Geostandards and Geoanalytical Research, 28,
41–52.
Faure, G., and Mensing, T. M., 2005. Isotopes: Principles and
Applications. Hoboken, NJ: Wiley.
Finlay, A. J., Selby, D., and Osborne, M. J., 2011. Re–Os geochro-
nology and fingerprinting of United Kingdom Atlantic margin
oil: temporal implications for regional petroleum systems. Geol-
ogy, 39, 475–478.
Georg, R., West, A., Vance, D., Newman, K., and Halliday, A.,
2013. Is the marine osmium isotope record a probe for CO2
release from sedimentary rocks? Earth and Planetary Science
Letters, 367, 28–38.
Georgiev, S., Stein, H. J., Hannah, J. L., Bingen, B., Weiss, H. M.,
and Piasecki, S., 2011. Hot acidic Late Permian seas stifle life
in record time. Earth and Planetary Science Letters, 310,
389–400.
 RHENIUM–OSMIUM GEOCHRONOLOGY: SULFIDES, SHALES, OILS, AND MANTLE
Georgiev, S., Stein, H. J., Hannah, J. L., Weiss, H. M., Bingen, B.,
Xu, G., Rein, E., Hatlø, V., Løseth, H., Nali, M., and Piasecki,
S., 2012. Chemical signals for oxidative weathering predict
Re–Os isochroneity in black shales, East Greenland. Chemical
Geology, 324–325, 108–121.
Gilchrist, R., 1932. A new determination of the atomic weight
of osmium. Bureau of Standards, Journal of Research, 9,
279–290.
Gramlich, J. W., Murphy, T. J., Garner, E. L., and Shields, W. R.,
1973. Absolute Isotopic abundance ratio and atomic weight of
a reference sample of Rhenium. Journal of Research of the
National Bureau of Standards A, 77A, 691–698.
Grice, K., Cao, C., Love, G. D., Böttcher, M. E., Twitchett, R. J.,
Grosjean, E., Summons, R. E., Turgeon, S. C., Dunning, W.,
and Jin, T., 2005. Photic zone euxinia during the Permian-
Triassic superanoxic event. Science, 307, 706–709.
Hannah, J. L., and Stein, H. J., 2003. Re–Os systematics in
syn-sedimentary/diagenetic pyrite: precise ages and Os cycling.
In Eliopoulos, D. G., et al. (eds.), Mineral Exploration and Sus-
tainable Development. Rotterdam: Millpress, pp. 81–84.
Hannah, J. L., and Stein, H. J., 2012. Re–Os geochemistry. In
Melezhik, V. A., Kump, L. R., Fallick, A. E., Strauss, H., Hanski,
E. J., Prave, A. R., and Lepland, A. (eds.), Reading the Archive of
Earth’s Oxygenation, volume 3: Global Events and the
Fennoscandian Arctic Russia – Drilling Early Earth Project.
Berlin/Heidelberg: Springer, pp. 1506–1514.
Hannah, J. L., Scherstén, A., and Morgan, J. W., 2000. What can the
Re–Os isotopic system tell us about diagenetic processes in con-
tinental environments? Quite a bit! Geological Society of Amer-
ica Abstracts with Programs, 32, A-346.
Hannah, J. L., Bekker, A., Stein, H. J., Markey, R. J., and Holland,
H. D., 2004. Primitive Os and 2316 Ma age for marine shale:
implications for paleoproterozoic glacial events and the rise of
atmospheric oxygen. Earth and Planetary Science Letters, 225,
43–52.
Hannah, J. L., Stein, H. J., Marolf, N., and Bingen, B., 2014. Cli-
matic instability and regional glacial advances in the late Edia-
caran: American Geophysical Union, December 15–19, San
Francisco, electronic abstract #31603.
Hassler, D. R., Peucker-Ehrenbrink, B., and Ravizza, G. E., 2000.
Rapid determination of Os isotopic composition by sparging
OsO4 into a magnetic-sector ICP-MS. Chemical Geology, 166,
1–14.
Herr, W., and Merz, E., 1955. Eine neue Methode zur
Alterbestimmung von Rhenium- Haltigen Mineralien Mittels
Neutronaktivierung. Zeitschrift für Naturforschung, Part
A Astrophysik, Physik und Physikalische Chemie, 10, 613–615.
Herr, W., Wölfle, E. P., and Kopp, E., 1967. Development and
recent applications of the Re/Os dating method. In Radioactive
Dating and Methods of Low-Level Counting. Vienna: Interna-
tional Atomic Energy Agency, pp. 499–508.
Hirt, B., Herr, W., and Hoffmeister, W., 1963. Age determination by
the rhenium-osmium method. In International Atomic Energy
Agency, Radioactive Dating: Proceedings of the Symposium on
Radioactive Dating, Athens, 19-23 November 1962, pp. 35–44.
Howarth, R. J., and McArthur, J. M., 1997. Statistics for strontium
isotope stratigraphy: a robust LOWESS fit to the marine
Sr-isotope curve for 0–206 Ma, with look-up table for the deriva-
tion of numerical age. Journal of Geology, 105, 441–456.
Kendall, B. S., Creaser, R. A., Ross, G. M., and Selby, D., 2004.
Constraints on the timing of marinoan “snowball Earth” glacia-
tion by 187Re–187Os dating of a neoproterozoic, post-glacial
black shale in Western Canada. Earth and Planetary Science Let-
ters, 222, 729–740.
Kendall, B., Creaser, R. A., and Selby, D., 2006. Re–Os geochro-
nology of postglacial black shales in Australia: constraints on
the timing of “sturtian” glaciation. Geology, 34, 729–732.
721
Kendall, B., Creaser, R. A., and Selby, D., 2009. 187Re–187Os geo-
chronology of Precambrian organic-rich sedimentary rocks.
Geological Society, London, Special Publications, 326, 85–107.
Kendall, B., van Acken, D., and Creaser, R. A., 2013. Depositional
age of the early paleoproterozoic klipputs member, nelani
formation (Ghaap Group, Transvaal Supergroup, South Africa)
and implications for low-level Re–Os geochronology and
paleoproterozoic global correlations. Precambrian Research,
237, 1–12.
Korzhinsky, M. A., Tkachenko, S. I., Shmulovich, K. I., Taran,
Y. A., and Steinberg, G. S., 2004. Discovery of a pure rhenium
mineral at Kudriavy Volcano. Nature, 369, 51–52.
Košler, J., Simonetti, A., Sylvester, P., Cox, R., Tubrett, M. N., and
Wilton, D., 2003. Laser ablation ICP-MS measurements of
Re/Os in molybdenites and implications for Re–Os geochronol-
ogy. Canadian Mineralogist, 41, 307–320.
Li, J., Liang, X.-R., Xi, J.-F., Suzuki, K., and Dong, Y.-H., 2010.
Simplified technique for the measurements of Re–Os isotope
by multicollector inductively coupled plasma mass spectrometry
(MC-ICP-MS). Geochemical Journal, 44, 73–80.
Lillis, P. G., and Selby, D., 2013. Evaluation of the rhenium–
osmium geochronometer in the Phosphoria petroleum system,
Bighorn Basin of Wyoming and Montana, USA. Geochimica
et Cosmochimica Acta, 118, 312–330.
Lindner, M., Leich, R. J., Russ, G. P., Bazan, J. M., and Borg, R. J.,
1989. Direct determination of the half-life of 187Re. Geochimica
et Cosmochimica Acta, 53, 1597–1606.
Luck, J. M., and Allègre, C. J., 1982. The study of molybdenites
through the 187Re–187Os chronometer. Earth and Planetary Sci-
ence Letters, 61, 291–296.
Ludwig, K., 2012. Isoplot/Ex version 3.75, A Geochronological
Toolkit for Microsoft Excel. Berkeley, CA: Berkeley Geochro-
nology Center Special Publication.
Lugmair, G. W., and Galer, S. J. G., 1992. Age and isotopic relation-
ships among the angrites Lewis Cliff 86010 and Angra dos Reis.
Geochimica et Cosmochimica Acta, 56, 1673–1694.
Mahdaoui, F., Reisberg, L., Michels, R., Hautevelle, Y., Poirier, Y.,
and Girard, J.-P., 2013. Effect of progressive precipitation of
petroleum asphaltenes on the Re–Os radioisotope system.
Chemical Geology, 358, 90–100.
Markey, R. J., Stein, H. J., and Morgan, J. W., 1998. Highly precise
Re–Os dating of molybdenite using alkaline fusion and NTIMS.
Talanta, 45, 935–946.
Markey, R. J., Hannah, J. L., Morgan, J. W., and Stein, H. J., 2003.
A double spike for osmium analysis of highly radiogenic sam-
ples. Chemical Geology, 200, 395–406.
Markey, R., Stein, H. J., Hannah, J. L., Zimmerman, A., Selby, D.,
and Creaser, R. A., 2007. Standardizing Re–Os geochronology:
a new molybdenite reference material (Henderson, USA) and the
stoichiometry of Os salts. Chemical Geology, 244, 74–87.
McArthur, J. M., Donovan, D. T., Thirlwall, M. F., Fouke, B. W.,
and Mattey, D., 2000. Strontium isotope profile of the early
Toarcian (Jurassic) anoxic event, the duration of ammonite
biozones, and belemnite palaeotemperatures. Earth and Plane-
tary Science Letters, 179, 269–285.
McArthur, J. M., Howarth, R. J., and Bailey, T. R., 2001. Strontium
isotope stratigraphy: LOWESS version 3. Best-fit line to the
marine Sr-isotope curve for 0 to 509 Ma and accompanying
look-up table for deriving numerical age. Journal of Geology,
109, 155–169.
McCandless, T. E., and Ruiz, J., 1993. Rhenium-osmium evidence
for regional mineralization in southwestern North America. Sci-
ence, 261, 1262–1266.
Miller, C. A., Peucker-Ehrenbrink, B., and Ball, L., 2009. Precise
determination of rhenium isotope composition by multi-
collector inductively-coupled plasma mass spectrometry. Jour-
nal of Analytical Atomic Spectrometry, 24, 1069–1078.
722 RHENIUM–OSMIUM GEOCHRONOLOGY: SULFIDES, SHALES, OILS, AND MANTLE
Morelli, R. M., Creaser, R. A., Selby, D., Kontak, D. J., and Horne,
R. J., 2005. Rhenium-osmium geochronology of arsenopyrite in
Meguma Group gold deposits, Meguma terrane, Nova Scotia,
Canada: evidence for multiple gold-mineralizing events. Eco-
nomic Geology, 100, 1229–1242.
Morelli, R., Creaser, R. A., Seltmann, R., Stuart, F. M., Selby, D.,
and Graupner, T., 2007. Age and source constraints for the giant
Muruntau gold deposit, Uzbekistan, from coupled Re–Os–He
isotopes in arsenopyrite. Geology, 35, 795–798.
Morelli, R. M., Bell, C. C., Creaser, R. A., and Simonetti, A., 2010.
Constraints on the genesis of gold mineralization at the
Homestake gold deposit, Black Hills, South Dakota from
rhenium-osmium sulfide geochronology. Mineralium Deposita,
45, 461–480.
Morgan, J. W., and Walker, R. J., 1989. Isotopic determinations of
rhenium and osmium in meteorites by using fusion, distillation
and ion-exchange separations. Analytica Chimica Acta, 222,
291–300.
Morgan, J. W., Horan, M. F., Walker, R. J., and Grossman, J. N.,
1995. Rhenium-osmium concentration and isotope systematics
in group IIAB iron meteorites. Geochimica et Cosmochimica
Acta, 59, 2331–2344.
Nozaki, T., Suzuki, K., Ravizza, G., Kimura, J.-I., and Chang, Q.,
2012. A method for rapid determination of Re and Os isotope com-
positions using ID-MC-ICP-MS combined with sparging method.
Geostandards and Geoanalytical Research, 36, 131–148.
Paquay, F. S., and Ravizza, G., 2012. Heterogeneous seawater
187
Os/188Os during Late Pleistocene glaciations. Earth and
Planetary Science Letters, 349–350, 126–138.
Peucker-Ehrenbrink, B., and Ravizza, G., 2000. The marine
osmium isotope record. Terra Nova, 12, 205–219.
Peucker-Ehrenbrink, B., Ravizza, G., and Hofmann, A. W., 1995.
The marine 187Os/186Os record of the past 80 million years.
Earth and Planetary Science Letters, 130, 155–167.
Platzner, I., 1999. Modern Isotope Ratio Mass Spectrometry,
2nd edn. Chichester, UK: Wiley.
Ravizza, G., 2007. Reconstructing the marine 187Os/188Os record
and the particulate flux of meteoritic osmium during the late Cre-
taceous. Geochimica et Cosmochimica Acta, 71, 1355–1369.
Ravizza, G., and Turekian, K. K., 1989. Application of the
187
Re–187Os system to black shale geochronometry.
Geochimica et Cosmochimica Acta, 53, 3257–3262.
Reisberg, L., and Lorand, J. P., 1995. Longevity of sub-continental
mantle lithosphere form osmium isotope systematics in orogenic
peridotite massifs. Nature, 376, 159–162.
Reisberg, L., Lorand, J. P., and Bedini, R. M., 2004. Reliability of
Os model ages in pervasively metasomatized continental mantle
lithosphere: a case study of Sidamo spinel peridotite xenoliths
(East African Rift, Ethiopia). Chemical Geology, 208, 119–140.
Rooney, A. D., Selby, D., Housay, J. P., and Renne, P. R., 2010. Re–
Os geochronology of a Mesoproterozoic sedimentary succes-
sion, Taoudeni basin, Mauritania: implications for basin-wide
correlations and Re–Os organic-rich sediments systematics.
Earth and Planetary Science Letters, 289, 486–496.
Rooney, A. D., Chew, D. M., and Selby, D., 2011. Re–Os geochro-
nology of the Neoproterozoic-Cambrian Dalradian Supergroup
of Scotland and Ireland: implications for Neoproterozic stratigra-
phy glaciations and Re–Os systematics. Precambrian Research,
185, 202–214.
Rooney, A. D., Selby, D., Lewan, M. D., Lillis, P. G., and Houzay,
J.-P., 2012. Evaluating Re–Os systematics in organic-rich sedi-
mentary rocks in response to petroleum generation using
hydrous pyrolysis experiments. Geochimica et Cosmochimica
Acta, 77, 275–291.
Roy-Barman, M., and Allègre, C. J., 1995. 187Os/186Os in oceanic
island basalts: tracing oceanic crust recycling in the mantle.
Earth and Planetary Science Letters, 129, 145–161.
Rudnick, R. L., and Walker, R. J., 2009. Interpreting ages from
Re–Os isotopes in peridotites. Lithos, 112S, 1083–1095.
Schoenberg, R., Nägler, T. F., and Kramers, J. D., 2000. Precise Os
isotope ratio and Re–Os isotope dilution measurements down to
the picogram level using multicollector inductively coupled
plasma mass spectrometry. International Journal of Mass Spec-
trometry, 197, 85–94.
Selby, D., and Creaser, R. A., 2001. Re–Os geochronology and sys-
tematics in molybdenite from the Endako porphyry molybde-
num deposit, British Columbia, Canada. Economic Geology,
96, 197–204.
Selby, D., and Creaser, R. A., 2003. Re–Os geochronology of
organic rich sediments: an evaluation of organic matter analysis
methods. Chemical Geology, 200, 225–240.
Selby, D., and Creaser, R. A., 2004. Macroscale NTIMS and micro-
scale LA-MC-ICP-MS Re–Os isotopic analysis of molybdenite:
testing spatial restrictions for reliable Re–Os age determinations,
and implications for the decoupling of Re and Os within molyb-
denite. Geochimica et Cosmochimica Acta, 68, 3897–3908.
Selby, D., and Creaser, R. A., 2005a. Direct radiometric dating
of the Devonian-Mississippian time-scale boundary using the
Re–Os black shale chronometer. Geology, 33, 545–548.
Selby, D., and Creaser, R. A., 2005b. Direct radiometric dating of
hydrocarbon deposits using rhenium-osmium isotopes. Science,
308, 1293–1295.
Selby, D., Creaser, R. A., Hart, C. J. R., Rombach, C. S., Thompson,
J. F. H., Smith, M. T., Bakke, A. A., and Goldfarb, R. J., 2002.
Absolute timing of sulfide and gold mineralization: a compari-
son of Re–Os molybdenite and Ar–Ar mica methods from the
Tintina Gold Belt, Alaska. Geology, 30, 791–794.
Selby, D., Creaser, R. A., Dewing, K., and Fowler, M., 2005. Eval-
uation of bitumen as a 187Re–187Os geochronometer for hydro-
carbon maturation and migration: a test case from the Polaris
MVT deposit, Canada. Earth and Planetary Science Letters,
235, 1–15.
Selby, D., Creaser, R. A., Stein, H. J., Markey, R. J., and Hannah,
J. L., 2007a. Assessment of the 187Re decay constant by cross
calibration of Re–Os molybdenite and U–Pb zircon chronome-
ters in magmatic ore systems. Geochimica et Cosmochimica
Acta, 71, 1999–2013.
Selby, D., Creaser, R. A., and Fowler, M. G., 2007b. Re–Os elemen-
tal and isotopic systematics in crude oils. Geochimica et
Cosmochimica Acta, 71, 378–386.
Selby, D., Mutterlose, J., and Condon, D. J., 2009. U–Pb and
Re–Os geochronology of the Aptian/Albian and Cenomanian/
Turonian stage boundaries: implications for timescale calibra-
tion, osmium isotope seawater composition and Re–Os system-
atics in organic-rich sediments. Chemical Geology, 265,
394–409.
Sen, I. S., and Peucker-Ehrenbrink, B., 2014. Determination of
osmium concentrations and 187Os/188Os of crude oils and source
rocks by coupling high-pressure, high-temperature digestion with
sparging OsO4 into a multicollector inductively coupled plasma
mass spectrometer. Analytical Chemistry, 86, 2982–2988.
Sharma, M., Papanastassiou, D. A., and Wasserburg, G. J., 1997.
The concentration and isotopic composition of osmium in the
oceans. Geochimica et Cosmochimica Acta, 61, 3287–3299.
Shen, J. J., Papanastassiou, D. A., and Wasserburg, G. J., 1996. Pre-
cise Re–Os determinations and systematics of iron meteorites.
Geochimica et Cosmochimica Acta, 60, 2887–2900.
Shields, G. A., 2007. A normalized seawater strontium isotope
curve: possible implications for Neoproterozoic-Cambrian
weathering rates and the further oxygenation of the Earth.
eEarth, 2, 35–42.
Shirey, S. B., and Walker, R. J., 1995. Carius tube digestion for low
blank rhenium-osmium analysis. Analytical Chemistry, 34,
2136–2141.
 RUBIDIUM–STRONTIUM DATING, HYDROTHERMAL EVENTS
Shirey, S. B., and Walker, R. J., 1998. The Re–Os isotope system in
cosmochemistry and high-temperature geochemistry. Annual
Review of Earth and Planetary Sciences, 26, 423–500.
Smoliar, M. I., Walker, R. J., and Morgan, J. W., 1996. Re–Os ages
of Group IIA, IIIA, IVA and IVB iron meteorites. Science, 271,
1099–1102.
Stein, H. J., 2006. Low-rhenium molybdenite by metamorphism in
northern Sweden: recognition, genesis, and global implications.
Lithos, 87, 300–327.
Stein, H. J., 2014. Dating and tracing the history of ore formation. In
Holland, H. D., and Turekian, K. K. (eds.), Treatise on Geo-
chemistry, 2nd edn. Oxford: Elsevier, Vol. 13, pp. 87–118.
Stein, H. J., and Bingen, B., 2002. 1.05–1.01 Ga Sveconorwegian
metamorphism and deformation of the supracrustal sequence at
Sæsvatn, south Norway: Re–Os dating of Cu–Mo mineral
occurrences. In Blundell, D., Neubauer, F., and von Quadt,
A. (eds.), The Timing and Location of Major Ore Deposits in
an Evolving Orogen. Geological Society, London, Special Pub-
lications, Vol. 204, pp. 319–335.
Stein, H. J., Markey, R. J., Morgan, J. W., Du, A., and Sun, Y., 1997.
Highly precise and accurate Re–Os ages for molybdenite from
the East Qinling molybdenum belt, Shaanxi Province, China.
Economic Geology, 92, 827–835.
Stein, H. J., Morgan, J. W., and Scherstén, A., 2000. Re–Os dating
of low-level highly-radiogenic (LLHR) sulfides: the Harnäs gold
deposit, southwest Sweden records continental scale tectonic
events. Economic Geology, 95, 1657–1671.
Stein, H. J., Markey, R. J., Morgan, J. W., Hannah, J. L., and
Scherstén, A., 2001. The remarkable Re–Os chronometer in
molybdenite: how and why it works. Terra Nova, 13, 479–486.
Stein, H., Scherstén, A., Hannah, J., and Markey, R., 2003.
Sub-grain scale decoupling of Re and 187Os and assessment of
laser ablation ICP-MS spot dating in molybdenite. Geochimica
et Cosmochimica Acta, 67, 3673–3686.
Stein, H. J., Hannah, J. L., Zimmerman, A., Markey, R., Sarkar,
S. C., and Pal, A. B., 2004. A 2.5 Ga porphyry Cu–Mo–Au
deposit at Malanjkhand, central India: implications for Late
Archean continental assembly. Precambrian Research, 134,
189–226.
Stein, H. J., Andrews, S. D., Hannah, J. L., Gaina, C., Whitham,
A. G., Yang, G., and Xu, G., 2012. Triassic mudstones
(Gråklint Beds) containing hydrocarbons with early Barremian
and Santonian ages, Jameson Land, East Greenland. In Ameri-
can Association of Petroleum Geologists Annual Conference
and Exhibition, Abstract #1236837.
Suzuki, K., Qi-Lu, Shimizu, H., and Masuda, A., 1992. Determina-
tion of osmium abundance in molybdenite mineral by isotope
dilution mass spectrometry with microwave digestion using potas-
sium dichromate as oxidizing agent. Analyst, 117, 1151–1156.
Suzuki, K., Qi-Lu, Shimizu, H., and Masuda, A., 1993. Reliable
Re–Os age for molybdenite. Geochimica et Cosmochimica Acta,
57, 1625–1628.
Suzuki, K., Shimizu, H., and Masuda, A., 1996. Re–Os dating of
molybdenites from ore deposits in Japan: implications for clo-
sure temperature of Re–Os system for molybdenite and cooling
history of molybdenum ore deposits. Geochimica et
Cosmochimica Acta, 60, 3151–3159.
Suzuki, K., Miyata, Y., and Kanazawa, N., 2004. Precise Re Isotope
ratio measurements by negative thermal Ionization mass Spec-
trometry (NTI-MS) using total evaporation technique. Interna-
tional Journal of Mass Spectrometry, 235, 97–101.
Takahashi, Y., Tomoya, U., Suzuki, K., Tanida, H., Terada, Y., and
Hattori, K. H., 2007. An atomic level study of rhenium and
radiogenic osmium in molybdenite. Geochimica et
Cosmochimica Acta, 71, 5180–5190.
Tessalina, S. G., Yudovskaya, M. A., Chaplygin, I. V., Birck, J.-L.,
and Capmas, F., 2008. Sources of unique rhenium enrichment in
723
fumaroles and sulphides at Kudryavy volcano. Geochimica et
Cosmochimica Acta, 72, 889–909.
Turgeon, S. C., and Creaser, R. A., 2008. Cretaceous oceanic anoxic
event 2 triggered by a massive magmatic episode. Nature, 454,
323–326.
Völkening, J., Walczyk, T., and Heumann, K. G., 1991. Osmium
isotope ratio determinations by negative thermal ionization mass
spectrometry. International Journal of Mass Spectrometry and
Ion Processes, 105, 147–159.
Walker, R. J., Morgan, J. W., Smoliar, M. I., Beary, E., Czamanske,
G. K., and Horan, M. F., 1997. Applications of the 190Pt–186Os
isotope system to geochemistry and cosmochemistry.
Geochimica et Cosmochimica Acta, 61, 4799–4808.
Watanabe, Y., and Stein, H. J., 2000. Re–Os ages for the Erdenet
and Tsagaan Suvarga porphyry Cu–Mo deposits, Mongolia,
and tectonic implications. Economic Geology, 95, 1537–1542.
Xu, G., Hannah, J. L., Stein, H. J., Bingen, B., Yang, G.,
Zimmerman, A., Weitschat, W., Mørk, A., and Weiss, H. M.,
2009. Re–Os geochronology of Arctic black shales to evaluate
the Anisian-Ladinian boundary and global faunal correlations.
Earth and Planetary Science Letters, 288, 581–587.
Xu, G., Hannah, J. L., Stein, H. J., Mørk, A., Vigran, J. O., Bingen,
B., Schutt, D., and Lundschien, B. A., 2014. Cause of Upper Tri-
assic climate crisis revealed by Re–Os geochemistry of Boreal
black shales. Palaeogeography, Palaeoclimatology,
Palaeoecology, 395, 222–232.
Yang, G., Hannah, J. L., Zimmerman, A., Stein, H. J., and Bekker,
A., 2009. Re–Os depositional age for Archean carbonaceous
slates from the southwestern Superior Province: challenges and
insights. Earth and Planetary Science Letters, 280, 83–92.
Yin, Q. Z., Jacobsen, S. B., Lee, C.-T., McDonough, W. F., Rudnick,
R. L., and Horn, I., 2001. A gravimetric K2OsCl6 standard:
application to precise and accurate Os spike calibration.
Geochimica et Cosmochimica Acta, 65, 2113–2127.
York, D., 1969. Least-squares fitting of a straight line with corre-
lated errors. Earth and Planetary Science Letters, 5, 320–324.
Zimmerman, A., Stein, H. J., Hannah, J. L., Tuttas, B., Yang, G., and
Beitscher, B., 2007. Instrumental mass fractionation overcome
by total evaporation. Geochimica et Cosmochimica Acta, 71,
A1176.
Zimmerman, A., Stein, H. J., Hannah, J. L., Kozelj, D., Bogdanov,
K., and Berza, T., 2008. Tectonic configuration of the Apusini-
Banat-Timok-Srednogorie belt, Balkans-South Carpathians,
constrained by high precision Re–Os molybdenite ages.
Mineralium Deposita, 43, 1–21.
RUBIDIUM–STRONTIUM DATING,
HYDROTHERMAL EVENTS
Shu’nichi Nakai
Earthquake Research Institute, The University of Tokyo,
Bunkyo-ku, Tokyo, Japan
Definition
Dating hydrothermal events related to the formation of ore
deposits by the Rb–Sr dating method
Historical background
Dating of a hydrothermal event related to ore formation
was conducted initially by dating the granite that was
724 RUBIDIUM–STRONTIUM DATING, HYDROTHERMAL EVENTS
regarded as the heat source for the hydrothermal activity
and assuming that ore deposition occurred contemporane-
ously with emplacement of the granite. Indirect dating of
alteration minerals such as sericite using K–Ar, Ar–Ar,
and Rb–Sr methods was also attempted. Results from
these studies, however, have been problematic because
of (1) complex associations between ore and alteration
minerals and (2) possible multiple hydrothermal events,
which can reset chronological information. Gangue min-
erals can be used as more direct targets of dating. Shep-
herd and Darbyshire (1981) obtained the Rb–Sr age of
392  5 Myr from fluid inclusions in quartz from the
Carrock Fell tungsten vein deposits, northern England.
However, the cause of the large variable Rb–Sr ratios
observed in their fluid inclusion samples remains uncer-
tain, which suggests mixing of different stages of
ore-forming fluids. These problems strongly demand
direct dating of ore minerals.
Rb–Sr dating of sphalerite from Mississippi
Valley-type ore deposits
Most long-lived radioactive nuclides, aside from Re, do
not enter into metal sulfides, which makes direct dating
of ore deposits difficult. Sphalerite from Mississippi
Valley-type (MVT) ore deposits was the first example of
successful dating of a sulfide mineral by Rb–Sr method.
Mississippi Valley-type deposits are carbonate-hosted
lead–zinc deposits. These deposits are regarded as the
product of large-scale fluid movements in the upper crust.
However, they are formed from warm basinal brines and
are not hydrothermal deposits in the strict sense. Direct
dating of an ore mineral, sphalerite, was attempted by
Nakai et al. (1990) and Nakai et al. (1993). In those ana-
lyses, sphalerite samples were crushed and leached in
purified water to separate fluid inclusions, which represent
solutions trapped at the time of ore-mineral formation.
Sphalerite solid residues contain Rb and Sr of 0.05–0.5
and 0.3–2 ppm, respectively, with variable 87Rb/86Sr of
0.2–2. Water from fluid inclusions trapped in these same
grains, however, contains high Sr with low and limited
87
Rb/86Sr of 0.05–0.26. Figure 1 shows a Rb–Sr isochron
plot of sphalerite residues and their fluid inclusions from
Coy Mine, Mascot–Jefferson City district, eastern Ten-
nessee, United States. The Rb–Sr isochron of sphalerite
solid residues yields an age of 377  29 Ma. The isochron
was regressed on data for solid residues without using
fluid inclusion data. The point for C15, which was more
deformed than the other samples, was not considered in
the age calculation. The resulting age ruled out any genetic
connection between mineralization and the Alleghenian
orogeny (330–250 Ma). Combining the Rb–Sr age of
sphalerites from the Immel Mine, it was concluded that
MVT deposits in the eastern United States were formed
by brine migration triggered by the earlier Acadian orog-
eny (380–350 Ma). Sphalerites from Pine Point district,
Canada (Nakai et al., 1993); from the West Hayden ore
0.722
0.720
377 ± 29 Ma
0.718
Sr/86Sr
0.716
87
0.714
C15
0.712
0.710
0 0.5 1 1.5 2
87
Rb/ Sr 86
Rubidium–Strontium Dating, Hydrothermal Events,
Figure 1 Rb–Sr isochron of sphalerites from Coy Mine,
Mascot–Jefferson City district, east Tennessee (Data were taken
from Nakai et al. (1990)). Data for sphalerite residues are shown
as filled circles, whereas data for fluid inclusions are shown as
open squares. The isochron was fitted to data for sphalerite
residues analyzed after removing solutions contained in fluid
inclusions, excluding data from C15 (see text).
body, Wisconsin, in the Upper Mississippi Valley district
(Brannon et al., 1992); from the Polaris Mine in the Cana-
dian Arctic Archipelago (Christensen et al., 1995a); and
from Blendevale, Western Australia (Christensen
et al., 1995b), were dated using this method.
Nakai et al. (1993) reported that sphalerites from Dan-
iel’s Harbour Mine, Newfoundland, Canada, and from
Monte Cristo Mine, Arkansas, United States, showed no
isochron. The samples used for dating might be mixtures
of sphalerites from different stages of ore deposition.
Nakai et al. (1993) reported that minute amounts of
inherited silicate inclusions might have disturbed the
Rb–Sr isotopic system in sphalerites from the Monte
Cristo Mine because the samples contained very low Sr.
Validity of Rb–Sr isochron age
The observed variable Rb–Sr ratios in sphalerite residues
engender some doubt about whether the Rb–Sr correla-
tions were formed by mixing of sphalerite and (1) a high
Rb–Sr phase, such as a silicate mineral, or (2) a low
Rb–Sr phase, such as a fluid inclusion.
Nakai et al. (1993) analyzed Rb and Sr isotopes in
K-feldspars from unmineralized carbonate rocks in the
Immel Mine. Their results showed that K-feldspar data
plotted above an isochron formed by sphalerite. They con-
cluded that mixing between K-feldspar and other
 RUBIDIUM–STRONTIUM DATING, HYDROTHERMAL EVENTS
0.722
0.720
0.718
Sr/86Sr
0.716
87
0.714
C15
0.712
0.710
0 1 2 3 4
1/[Sr]
Rubidium–Strontium Dating, Hydrothermal Events,
Figure 2 Plot of 87Sr/86Sr versus reciprocal Sr concentration
(1/[Sr]) for sphalerites from Coy Mine, Tennessee (Data are from
Nakai et al. (1990)). Symbols are the same as those in Figure 1.
Sphalerite residues do not form a straight line, ruling out the
possibility that Rb–Sr isochron was formed by variable degrees
of mixing with fluid inclusions.
components with low Rb–Sr ratio could not have formed
the observed sphalerite isochron.
Regarding the possible influence of fluid inclusions,
Pettke and Diamond (1996) examined the relation
between 87Sr/86Sr and reciprocal of Sr concentration for
samples showing Rb–Sr isochrons. Figure 2 presents data
for sphalerites from the Coy Mine using data from Nakai
et al. (1990). Pettke and Diamond found that this type of
plot shows no correlation indicative of mixing in the
observed isochrons except for the sphalerites from the
West Hayden ore body reported by Brannon et al. (1992).
It can be concluded that Rb–Sr isochrons were not mixing
lines between sphalerite residues and low Rb–Sr fluid
inclusions except in this case. In addition, Christensen
et al. (1995b) showed that Rb–Sr ratio in sphalerite resi-
dues from the Canning Basin was not controlled by Sr
concentration negating the influence of fluid inclusions.
They concluded that Rb incorporation into sulfide min-
erals during precipitation might be controlled kinetically.
Apart from MVT deposits, Shneider et al. (2007) com-
pared Rb–Sr and Re–Os ages of ore-stage Zn–Cu–Ge
sulfides, including sphalerite from the Kipushi base metal
deposit, Democratic Republic of Congo. They respec-
tively obtained concordant Rb–Sr and Re–Os ages of
451.1  6.0 and 450.5  3.4 Ma. The concordant ages fur-
ther clarified the geological significance of direct Rb–Sr
dating of sphalerites.
725
Rb–Sr dating of pyrite and other minerals
Apart from sphalerite, some other sulfide minerals have
been found to be suitable for Rb–Sr dating. Tretbar
et al. (2000) reported that Galkhaites [(Cs,Tl)(Hg,Cu,
Zn)6(As,Sb)4S12] from sedimentary-rock-hosted dissemi-
nated gold deposits known as Carlin-type gold deposits
in Nevada formed a Rb–Sr isochron. The mineral shows
high and variable 87Rb/86Sr ratios from 3,000 to 23,000.
Pyrite has also been found to be a suitable mineral for
Rb–Sr dating. Yang and Zhou (2001) dated pyrites from
Linglong gold deposit, Jiaodong Peninsula, China. Yang
and Zhou (2001) analyzed three sets of pure pyrite grains
from three hand specimens of 6  6  9 cm assuming that
subsamples from a single hand specimen are more likely
to have a uniform initial Sr isotope composition. The three
isochrons show concordant ages of 122  11, 122.7  3.3,
and 123.0  4.2 Ma. Wei et al. (2003) also applied Rb–Sr
dating to pyrite from the Liaodong gold province, China.
Summary and conclusions
Rb–Sr dating of sulfide minerals has been confirmed as
a robust method for dating hydrothermal events. Because
sulfide minerals contain only trace amounts of Rb and
Sr, careful treatment of analyses is necessary. Combining
Rb–Sr dating with other geochronometers such as Re–Os
dating of sulfide minerals and Sm–Nd dating of scheelites
is expected to shed light on details of ore formation pro-
cesses of hydrothermal deposits.
Bibliography
Brannon, J. C., Podsek, F. A., and McLimans, R. K., 1992. Alleghe-
nian age of upper Mississippi Valley zinc-lead deposit deter-
mined by Rb–Sr dating of sphalerite. Nature, 356, 509–511.
Christensen, J. N., Halliday, A. N., Leigh, K. E., Randell, R. N., and
Kesler, S. E., 1995a. Direct dating of sulfides by Rb–Sr: a critical
test using the Polaris Mississippi Valley-type Zn–Pb deposit.
Geochimica et Cosmochimica Acta, 59, 5191–5197.
Christensen, J. N., Halliday, A. N., Vearncombe, J. R., and Kesler,
S. E., 1995b. Testing models of large-scale crustal fluid flow
using direct dating of sulfides: Rb–Sr evidence for early
dewatering and formation of Mississippi Valley-type deposits,
Canning Basin, Australia. Economic Geology, 90, 877–884.
Nakai, S., Halliday, A. N., Kesler, S. E., and Jones, H. D., 1990.
Rb–Sr dating of sphalerites from Tennessee and the genesis of
Mississippi Valley type ore deposits. Nature, 346, 354–357.
Nakai, S., Halliday, A. N., Kesler, S. E., Jones, H. D., Kyle, J. R.,
and Lane, T. E., 1993. Rb–Sr dating of sphalerites from Missis-
sippi Valley-type (MVT) ore deposits. Geochimica et
Cosmochimica Acta, 57, 417–427.
Pettke, T., and Diamond, L. W., 1996. Rb–Sr dating of sphalerite
based on fluid inclusion – host mineral isochrones:
a clarification of why it works. Economic Geology, 91, 951–956.
Shepherd, T. J., and Darbyshire, D. P. F., 1981. Fluid inclusion Rb–
Sr isochrons for dating mineral deposits. Nature, 290, 578–579.
Shneider, J., Melcher, F., and Braus, M., 2007. Concordant ages for
the giant Kipushi base metal deposit (DR Congo) from direct
Rb–Sr and Re–Os dating of sulfides. Mineralium Deposita,
42, 791–797.
726 RUBIDIUM–STRONTIUM DATING, HYDROTHERMAL EVENTS

Tretbar, D. R., Arehart, G. B., and Christensen, J. N., 2000. Dating Cross-references
gold deposition in a Carlin-type gold deposit using Rb/Sr
methods on the mineral galkhaite. Geology, 28, 947–950. Crustal Sulfide Minerals (Re-Os)
Wei, J.-H., Liu, C.-Q., and Tang, H.-F., 2003. Rb–Sr and U–Pb iso- Rb–Sr Dating
topic systematic of pyrite and granite in Liaodong gold province, Rb–Sr Geochronology (Igneous Rocks)
North China: implication for the age and genesis of a gold Sedimentary Rocks (Rb-Sr Geochronology)
deposit. Geochemical Journal, 37, 567–577. Sm–Nd Dating
Yang, J.-H., and Zhou, X.-H., 2001. Rb–Sr, Sm–Nd and Pb isotope
systematic of pyrite: Implications for the age and genesis of lode
gold deposits. Geology, 29, 711–714.
 S
SEA LEVEL CHANGE (U-SERIES)
Christina Gallup
Department of Earth and Environmental Sciences,
University of Minnesota Duluth, Duluth, MN, USA
Synonyms
Quaternary sea level change; Sea level variations; U-series
disequilibrium dating; U-Th dating
Definitions
U-series dating: A method for determining the age of
material based on the changes in amounts of various iso-
topes within the decay chains of uranium-238 or
uranium-235 due to radioactive decay.
Sea level change: Changes in global mean ocean level
over time, usually associated with climatic changes.
Introduction
One of the most dramatic applications of uranium-series
dating has been the quantification of past changes in
late-Quaternary global sea level through U-Th dating of
fossil corals and speleothems (cave deposits). Large-scale
changes in Quaternary sea level result from changes in the
volume of ice on land associated with the ice ages. This is
because the water that formed the great ice sheets of past
glaciations came from the ocean. As sea level changes,
the location and elevation of coral growth and speleothem
formation in coastal caves also must change.
Corals make excellent indicators of sea level because
reef forming corals require sunlight and so must grow near
the sea surface. A typical coral reef has a reef crest; behind
the crest is the back reef or lagoon, and in front of the crest
is the fore reef that slopes down into deeper water. Reef-
crest corals make excellent sea level markers, such as
Acropora palmata in the Atlantic Ocean (Goreau, 1959),
J.W. Rink, J.W. Thompson (eds.), Encyclopedia of Scientific Dating Methods, DOI 10.1007/978-94-007-6304-3,
© Springer Science+Business Media Dordrecht 2015
which grows predominantly within a few meters of the
sea surface. In the Pacific Ocean, it is more common to
have an assemblage of corals that indicate growth on
a reef crest (Huston, 1985). When a coral reef becomes
exposed on land, either due to dropping sea level or tec-
tonic uplift, the reef becomes a terrace, with the reef crest
forming the lip of the terrace. Many fossil coral terraces
around the world have been sampled and dated with
U-series techniques, resulting in data that forms the
backbone of our record of Late-Quaternary sea level (see
references in Medina-Elizalde, 2013).
Speleothems also help to constrain past sea level
changes. Speleothems are cave deposits, many of which
formed from water flowing or dripping into vadose cavi-
ties. Caves often form in bedrock near sea level because
of the corrosive quality of the groundwater where fresh
water meets sea water. When these caves are exposed to
air and dripping water after sea level falls, speleothems
form. When sea level rises and floods the cave again,
speleothems stop growing. The growth of speleothems in
these types of caves thus means that sea level was below
the elevation of the cave at the time the speleothem grew.
Dating such speleothems with U-series methods provides
a history of periods during which sea level was below
a certain elevation (see references in Edwards et al., 2003).
Vertical movement caused by tectonic forces can cause
fossil corals and speleothems to be moved from their orig-
inal elevation. This can be an advantage, such as when tec-
tonically uplifting coastlines cause corals or speleothems
that grew below modern sea level to be exposed on land
and more easily sampled.
Obtaining sea level information from a fossil coral or
a speleothem deposit requires knowledge of the deposit’s
elevation and age. If the deposit is on a tectonically stable
substrate, then the measured elevation indicates the eleva-
tion of sea level when the deposit was formed. If the
deposit is on a tectonically active substrate, then one must
728 SEA LEVEL CHANGE (U-SERIES)
correct the measured elevation for the amount of tectonic
movement that occurred since the deposit was formed.
The tectonic uplift or subsidence rate is needed for this
correction and is usually determined by measuring the ele-
vation of a deposit that was formed at a known elevation in
the past. For sea level studies, the deposits formed during
the last interglacial period, when sea level was ~6 m higher
than present (Kopp et al., 2009), are often used to deter-
mine tectonic uplift or subsidence rates. Local sea level
can also vary due to changes in the shape of the earth
(called isostatic changes) due to changes in the weight of
the ice sheets and the oceans resulting from the ice ages.
U-Th dating can provide precise age determination of
fossil corals and speleothems. The precision in ages is
important not only because it determines the precision of
the sea level record but also because the sea level record
is used to test the current theory for the cause of the ice
ages, the Milankovitch Theory (Milankovitch, 1941). This
theory states that the ice ages wax and wane due to the
amount of solar insolation reaching 65
N latitude. When
summer insolation is low, ice will build up from year to
year, causing a glaciation. When summer insolation is
high, ice will melt, causing an interglacial period. The
changes in summer solar insolation at 65
N are caused
by changes in the orbit of the Earth around the sun. These
changes can be calculated with a precision of about 1 ka
(1,000 years) (Berger, 1978). To test the Milankovitch
Theory, it is necessary to date geologic records of the ice
ages to a similar or better precision.
Dating systematics and analytical methods
U-Th dating is based on the radioactive decay products of
238
U: 234U and 230Th. 234U has a half-life of 245,250 years
and 230Th has a half-life of 75,690 years (Cheng
et al., 2000). Uranium is weathered out of the continental
rocks, and because it is soluble, it is transported through
surface waters to the oceans. Thorium is also weathered
out of continental rocks, but because it has very low solu-
bility in natural waters, it has very low concentrations in
most water on Earth. When corals grow in the ocean, they
incorporate approximately 3 mg/g of U into their skeleton,
which is made of the mineral aragonite. When
speleothems crystallize from dripping water, they incorpo-
rate U also, though the concentration is much more vari-
able. Thorium produced by the decay of uranium is
essentially absent from corals and speleothems due to its
very low solubility, unless the sample has incorporated
clay-rich or organic-rich sediments, which can have sig-
nificant amounts of thorium in them. After the coral or
speleothem forms, the uranium in these deposits will
decay, producing 230Th. The amount of 230Th increases
at a predictable rate with time, providing a chronometer.
Ages calculated from the ingrowth of 230Th are called
230
Th ages, which can, in principle, be used to date mate-
rials as young as 3 years and as old as 600,000 years
(Edwards et al., 1987; Edwards, 1988). There is also
a chronometer based on the decay products of 235U
(present in all uranium in a ratio of 1 part 235U to ~138
parts 238U), which includes 231Pa (protactinium) with
a half-life of 31,000 years (Cheng et al., 1998). Protactin-
ium has very similar chemical properties to thorium, and
so the ingrowth of 231Pa in a sample provides
a chronometer. Ideally, the 230Th and 231Pa ages on the
same sample should be the same (often called concordant)
within error. 231Pa ages are not commonly available
because the measurements are difficult to make.
For fossil corals, the isotopes 234U and 238U are also
important. When uranium is weathered from the conti-
nents, the 234U atoms are dissolved more easily because
their chemical bonds have been damaged by the radioac-
tive decay process. This means that the oceans have
a higher activity ratio (atomic ratios normalized by the
ratio of their decay constants) of 234U to 238U than they
would have due to decay in a chemically closed environ-
ment. The 234U to 238U activity ratio due to the long-term
decay process is 1.000 (i.e., in secular radioactive
equilibrium), whereas the ratio in the ocean water is
1.147. The activity ratio of 234U to 238U is often
expressed as d234U in delta units (d234Usample ¼
[[234U/238Usample  234U/238Ustandard]  1]  1,000, where
234 238
U/ Usample and 234U/238Ustandard are expressed as
activity ratios and 234U/238Ustandard is equal to the secular
equilibrium value of 1.000). The modern ocean d234U
value is 147  1 per mil (%) (Stirling and Andersen,
2009). Evidence suggests that this ratio has remained rel-
atively constant during the last 400,000 years
(Henderson et al., 1993), consistent with the long time that
uranium resides in the ocean. The 230Th age of a fossil
coral can be used to calculate the initial d234U value.
How closely the value matches the modern ocean value
provides a measure of the coral’s quality of preservation.
Analytical methods for measuring uranium and tho-
rium isotopes have improved dramatically since the first
230
Th ages were measured in the 1950s (Barnes
et al., 1956). The first ages reported were measured using
alpha counting techniques, where the alpha particles given
off during radioactive decay were detected and counted.
A typical measurement of a 125,000-year-old (last inter-
glacial age) sample had a precision (2s) of
10,000–20,000 years in age and 20–40 per mil in d234U
value. Thermal ionization mass spectrometry (TIMS)
replaced alpha counting as the preferred method of analy-
sis in the late 1980s. TIMS methods involve measuring
atoms of the various uranium and thorium isotopes rather
than just those that decay during the measurement period
using heat to ionize the atoms, a sector magnetic analyzer
to separate the isotopes by mass, and a secondary electron
multiplier to measure the ions collected (Chen
et al., 1986). A typical measurement of a 125,000-year-
old sample had a precision of 500–2,000 years in age
and 1.5–8 per mil in d234U value. Magnetic sector induc-
tively coupled plasma mass spectrometry (MS-ICPMS)
became the preferred method of analysis in the late
1990s (Shen et al., 2002). This method differs from TIMS
in the manner in which the isotopes are ionized before
 SEA LEVEL CHANGE (U-SERIES)
being separated by the magnet. ICPMS uses a very hot gas
(plasma) to ionize the sample, which is introduced into the
plasma as a liquid nebulized into tiny droplets. The preci-
sion ranges from 300 to 1,000 years for a 125,000-year-
old sample and 1–3 per mil in d234U value. The greater
ionization efficiency of the plasma results in a higher per-
centage of ions reaching the collector and the increase in
precision relative to TIMS measurements. The sample size
in this technological progression decreased from grams to
10s of micrograms, and the time involved decreased from
weeks of counting decays to hours of counting atoms.
There is also the option of using large samples (grams)
to run samples on MS-ICPMS with very high intensities,
where the ions are measured on faraday cups instead of
a secondary electron multiplier, which results in precisions
of 100–200 years for 125,000-year-old samples and 0.2
per mil in d234U value (Stirling and Andersen, 2009;
Cheng et al., 2013).
Preservation and location
Fossil corals from all over the world have been collected
and analyzed. Initially, all data were considered exciting
and encouraging, as the timing of high sea level events
was found to match the predictions of the Milankovitch
Theory (Broecker et al., 1968). As more corals were ana-
lyzed and the precision of the measurements improved, it
became clear that many of the samples had been altered
such that their uranium and thorium isotopic composition
did not provide a perfect clock.
There are two main indicators of alteration in fossil
corals. One is the presence of calcite, which has the same
chemical formula (CaCO3), but a different arrangement of
atoms, than aragonite. Calcite is the more stable configu-
ration of atoms of the two minerals. Rain water percolat-
ing down into the fossil coral reef can cause the rock to
dissolve and reprecipitate, often resulting in a change from
aragonite to calcite. Calcite accommodates less uranium in
its structure than aragonite, so it is likely that recrystalliza-
tion causes a disruption of the uranium-series isotopes in
the fossil coral. Samples are often measured for calcite
content prior to uranium-series analysis to avoid measur-
ing altered samples. The second indicator is an initial
d234U value that is higher than the modern marine value.
Even with lower precision, early studies noted that many
samples did not have initial d234U values near the modern
value of about 150 but had values between 160 and
200, and a smaller number of samples had values well in
excess of 200. The trend toward higher initial d234U
values was noticed to be accompanied by a trend toward
higher 230Th ages in the late 1970s (Bender et al., 1979)
and later quantified (Gallup et al., 1994). For every
increase of 4 per mil in initial d234U value, the 230Th age
increased about 1,000 years in samples from Barbados,
making it clear that some process was mobilizing uranium
resulting in shifts in ages. Since this trend was quantified,
many studies have adopted a cutoff for how far above the
modern value the initial d234U value can be and still
729
provide acceptable ages. This cutoff, or criterion, has var-
ied from 0 (meaning that the initial d234U value must be
within error of the modern value; Gallup et al., 2002) to
10 per mil (Medina-Elizalde, 2013). As precision has
improved, the initial d234U criterion has become more use-
ful. Not all samples conform to this trend (e.g., data from
New Guinea); data without these trends are much harder
to evaluate. There are other methods to evaluate the qual-
ity of coral data, such as the uranium and thorium concen-
trations in the fossil coral and a holistic evaluation of the
data set to see if it makes sense in terms of the geologic
deposit.
Several workers noted that the observed trends in initial
d234U value and 230Th age are consistent with what might
happen if the decay process itself caused the mobilization
of uranium and thorium isotopes (Henderson and
O’Nions, 1995; Thompson et al., 2003). Most steps in
the decay chain between 238U and 230Th are by alpha
decay, where the equivalent of a helium atom is released
during decay. Each time an alpha decay occurs, there is
a physical movement called alpha recoil, where the atom
can be dislodged from its site in the mineral structure.
Thompson et al. (2003) used a mathematical model to
show that alpha-recoil processes can produce the types
of trends observed in Barbados corals and elsewhere. He
further used the model to develop model ages, where the
effects of this alteration can be corrected and produce
a corrected 230Th age. These model ages are very useful
for checking a data set to see if the samples from a given
deposit converge on a narrow age range when corrected,
as alteration tends to spread the 230Th ages out toward
higher values. The model ages are only appropriate where
the uranium and thorium isotopes display the type of trend
produced by alpha recoil.
Late-Quaternary sea level record from fossil corals
and speleothems
A recent compilation of three decades of work using ura-
nium series dating of fossil corals provides
a comprehensive documentation of sea level changes
(Medina-Elizalde, 2013). Much of the data for high sea
level comes from locations with no to slow tectonic uplift,
such as Barbados, Australia, the Bahamas, Hawaii, and
Henderson Island (in the Pacific ocean), while much of
the low sea level data comes from locations with high
uplift rates, such as New Guinea, or from submerged sam-
ples obtained from ocean drilling, such as from Barbados
and Tahiti. Figure 1a shows the data that has been col-
lected, regardless of signs of alteration, superimposed on
a record of deep-sea oxygen isotope changes as recorded
by shells of foraminifera (Lisiecki and Raymo, 2005).
Because ice sheets preferentially incorporate light oxygen
isotopes, the values of marine oxygen isotopes vary
mainly as a function of ice volume. Sea level in the Late
Quaternary also varies mainly due to changes in ice vol-
ume. Therefore, records of deep-sea oxygen isotopes and
fossil coral sea level should reflect similar fluctuations.
730 SEA LEVEL CHANGE (U-SERIES)

a 3

LR05 Stack (V-PDB, ‰)

0

Sea Level (m relative to

3.5
−40
4

present)
−80
4.5

−120
All Coral RSL Data 5
LR05 benthic δ18O stack
−160

b 3
0

LR05 Stack (V-PDB, ‰)

Sea Level (m relative

3.5
−40
to present)

4
−80
4.5
−120
δ234U = 137-157
δ234U = 146-148 5
−160
0 100 200 300 400
Age (ka)
Sea Level Change (U-Series), Figure 1 Comparison between sea level elevation determined from fossil coral data and the
stratigraphically continuous deep-sea oxygen isotope record (Lisiecki and Raymo, 2005) spanning the last 400 ka (Reprinted from
Medina-Elizalde (2013) with permission). Panel a contains all of the coral relative sea level (RSL) data in the compilation (n ¼ 897) (blue
squares). Panel b contains selected coral RSL data based on initial d234U values within the 137–157 % (red squares) and 146–148 %
(green squares) ranges (See Medina-Elizalde (2013) for the references to data in this figure).

It is clear from Figure 1a that there is quite a bit of scat-
ter and conflict in this graph of coral data. Figure 1b shows
two applications of the d234U criterion, one in red with
values accepted between 137 and 157 per mil and one in
green with a much narrower range of 146–148 per mil.
The strictest criterion reduces the number of data points
significantly, but conflicting data points remain. One con-
flict occurs during the last interglacial high sea stand
(about 120,000–128,000 years ago) where multiple data
sets suggest that sea level was higher than present,
whereas data sets from New Guinea for this same period
suggest a much lower sea level. The source of this discrep-
ancy is unclear. The prevalence of data showing high sea
level during the last interglacial period suggests that sea
level did remain above the present value for most of the
period. Another area of conflict is the many different sea
level elevations reported for the high sea stand 80,000
years ago.
Despite these issues, the coral data generally follow the
marine oxygen isotope record fairly well, with a few nota-
ble exceptions. The coral record suggests a much higher
sea level around 80,000 years ago, a much longer last
interglacial which remains high until roughly 115,000
years ago, and a much higher sea level 170,000 and
200,000 years ago. Model ages (Thompson et al., 2003)
from the data that meet the 137 to 157 per mil initial
d234U criterion do reduce the scatter in Figure 1b (see
Figure 3b in Medina-Elizalde, 2013), but do not remove
the conflicting sea level data or change the relationship
with the marine oxygen isotope record. The reasons for
the scatter and conflicts in the data even after it has been
screened for its initial d234U value and modeled to correct
for alpha-recoil effects are likely many, including isostatic
changes, uncertainties and/or inaccuracies in tectonic cor-
rections, uncertainties in the water depth at which corals
grow, and dating uncertainties and/or inaccuracies.
Medina-Elizalde (2013) performed a statistical analysis
on the coral data that meet the 137–157 per mil initial
d234U criterion to form a best fit record of sea level
through time with a 95 % confidence level based on uncer-
tainties in the sea level and the measured age. Figure 2a
shows the results of this analysis superimposed on the data
it is based on, and Figure 2b shows the results
superimposed on two, deep-sea oxygen isotope records
(Waelbroeck et al., 2002; Lisiecki and Raymo, 2005)
and one oxygen isotope record from the Red Sea
(Rohling et al., 2009). The coral sea level curve agrees
broadly with the other sea level records. However, the
coral record (a) shows higher sea level about 80,000 years
ago, (b) does not show a drop between the last interglacial
 SEA LEVEL CHANGE (U-SERIES) 731

Age (ka)
a 0 40 80 120 160 200

Sea Level (m relative to present)

0

−40

−80

−120
δ234U = 137-157‰
−160

b 3
Sea Level (m relative to present)

Benthic δ18O stack (V-PDB, ‰)

3.5

−40
4

−80
4.5

−120 Red Sea (Rohling et al., 2009, 2010) 5

Waelbroeck et al. (2002)
Lisiecki and Raymo (2005)

−160 5.5
0 40 80 120 160 200
Age (ka)
Sea Level Change (U-Series), Figure 2 Monte Carlo 95 % confidence intervals (red curve) based on sea level benchmarks calculated
from samples having initial d234U values within the137–157 % range. The Monte Carlo analysis is based on 1,000 runs and takes into
account a sea level uncertainty of 6 m (1s) and age uncertainties at a 2s level (Reprinted from Medina-Elizalde (2013) with
permission). The original coral RSL data used in this statistical analysis are also plotted (a, blue circles). The continuous sea level proxies
are shown for comparison in (b) with references based on oxygen-isotope data as indicated in the figure.

125,000 years ago and the next high sea level event at
about 100,000 years, (c) shows a high sea level about
140,000 years ago that is not present in the other records
(though this is based on very sparse data), and (d) shows
sea level much higher between 160,000 and 200,000 years
ago (though this is based on very sparse data).
Edwards et al. (2003) assembled sea level data for the
last 210,000 years (Figure 3) based on speleothems and
fossil corals that had an initial d234U within error of the
modern seawater value and, where available, concordant
230
Th and 231Pa ages. The data in Figure 3 are connected
by a line that is the best estimate (unlike the statistical
method of Medina-Elizalde) of sea level based on the
available data. For the sea level rise since the last glacial
maximum 20,000 years ago, the line is based on drill core
data from Barbados (not shown). For the period between
20,000 and 70,000 years ago, the line is based on an anal-
ysis by Chappell (2002) of 230Th dates and coral terrace
elevations from New Guinea. The speleothem data is
shown with gray symbols accompanied by a downward
facing arrow, indicating that the speleothem constrains
sea level to have been below the plotted elevation. The
speleothem data are generally consistent with the coral
data, though the data at 200,000 years ago are only barely
consistent within error. The estimated sea level records
from Figures 2 and 3 are very similar in overall shape
and timing. There are some differences, as Figure 2 has
a higher estimate for sea level 80,000 years ago and
140,000 years ago and suggests a more detailed and vari-
able sea level record between 80,000 and 130,000 years
ago.
Finally, Figure 3 makes it possible to evaluate how the
uranium-series record of sea level compares with the
timing of the orbital cycles that drive the glacial cycles
according to the Milankovitch Theory. The peaks in the
top curve in Figure 3 are times when summers are warm
in the northern high latitudes and when sea level should
be high due to low glacial activity. While it is true that
periods with high sea level almost always coincide with
times of high 65
N summer insolation, it is also true that
periods of high insolation do not necessarily result in high
sea level. There are fewer low sea level data points,
732 SEA LEVEL CHANGE (U-SERIES)

Sea Level Change (U-Series), Figure 3 Direct sea level data (210,000 years to present) from corals (solid symbols) and speleothems
(gray symbols). The gray symbols represent maximum possible sea level based on speleothem growth. The marine oxygen isotope
stages are indicated by numbers, where even numbers are glacial periods and odd numbers are interglacial periods. The line
segments above the last interglacial high stand (125,000 years ago) are different estimates of the timing and duration of peak last
interglacial sea level from corals and from direct dating of the marine oxygen isotope record. The placements of the line segments
have no significance in terms of sea level elevation. The curve between 65,000 and 35,000 years is from Chappell (2002). Note that the
curve disagrees with a 231Pa – 230Th concordant point at about 50,000 years. This indicates that either the curve or the point are in
error, or that the curve does not capture the full range of sea level variability in this time range. July summer solar insolation at
65
north latitude is depicted in the top panel (Berger, 1978). The vertical gray bars indicate times of high insolation and are intended
as aids in comparing the curves.

but what data there are show that low sea levels coincide
with periods of low 65
N summer insolation. Thus, the
uranium-series sea level data generally support the
Milankovitch Theory for the ice ages. However, there
are many unexplained relationships between the orbital
cycles and the sea level data, such as the observation that
higher 65
N summer insolation does not necessarily cor-
respond to higher sea level and the observation that high
sea level periods often continue well into periods of
declining 65
N summer insolation.
Bibliography
Barnes, J. W., Lang, E. J., and Potratz, H. A., 1956. The ratio of
ionium to uranium in coral limestone. Science, 124, 175–176.
Bender, M. L., Fairbanks, R. G., Taylor, F. W., Matthews, R. K.,
Goddard, J. G., and Broecker, W. S., 1979. Uranium-series dat-
ing of the Pleistocene reef tracts of Barbados, West-Indies. Geo-
logical Society of America Bulletin, 90, 577–594.
Berger, A. L., 1978. Long-term vatiations of caloric insolation
resulting from the earth’s orbital elements. Quaternary
Research, 9, 139–167.
Broecker, W. S., Thurber, D. L., Goddard, J., Ku, T. K., Matthews,
R. K., and Mesolella, K. J., 1968. Milankovitch hypothesis
supported by precise dating of coral reefs and deep sea sedi-
ments. Science, 159, 297–300.
Chappell, J., 2002. Sea level changes forced ice breakouts in the
Last Glacial cycle: new results from coral terraces. Quaternary
Science Reviews, 21, 1229–1240.
Chen, J. H., Edwards, R. L., and Wasserburg, G. J., 1986. 238U, 234U
and 232Th in seawater. Earth and Planetary Science Letters, 80,
241–251.
Cheng, H., Edwards, R. L., Murrell, M. T., and Goldstein, S., 1998.
The systematics of uranium-thorium-protactinium dating.
Geochimica et Cosmochimica Acta, 62, 3437–3452.
Cheng, H., Edwards, R. L., Hoff, J., Gallup, C. D., Richards, D. A.,
and Asmerom, Y., 2000. The half-lives of uranium-234 and
thorium-230. Chemical Geology, 169, 17–33.
Cheng, H., Edwards, R. L., Shen, C.-C., Polyak, V. J., Asmerom, Y.,
Woodhead, J., Hellstrome, J., Wang, Y., Kong, X., Spötl, C.,
 SEAWATER Sr CURVE
Wang, X., and Alexander, E. C., Jr., 2013. Improvements in
230
Th dating, 230Th and 234U half-life values, and U–Th isotopic
measurements by multi-collector inductively coupled plasma
mass spectrometry. Earth and Planetary Science Letters,
371–372, 82–91.
Edwards, R. L., 1988. High Precision Thorium-230 Ages of Corals
and the Timing of Sea Level Fluctuations in the Late Quaternary.
Doctoral dissertation, California Institute of Technology.
Edwards, R. L., Chen, J. H., and Wasserburg, G. J., 1987. 238U,
234
U, 230Th, 232Th systematics and the precise measurement of
time over the past 500,000 years. Earth and Planetary Science
Letters, 81, 175–192.
Edwards, R. L., Cutler, K. B., Cheng, H., and Gallup, C. D., 2003.
Geochemical evidence for quaternary sea level. In Elderfield,
H. (ed.), Oceans and Marine Geochemistry. In Holland, H. D.,
and Turekian, K. K. (eds.), Treatise on Geochemistry. Amster-
dam: Elsevier. Vol. 6, pp. 343–364.
Gallup, C. D., Edwards, R. L., and Johnson, R. G., 1994. The timing
of high sea levels over the past 200,000 years. Science, 263,
796–800.
Gallup, C. D., Cheng, H., Taylor, F. W., and Edwards, R. L., 2002.
Direct determination of the timing of sea level change during
Termination II. Science, 295, 310–313.
Goreau, T. F., 1959. The ecology of Jamaican reefs. I. Species com-
position and zonation. Ecology, 40, 67–90.
Henderson, G. M., and O'Nions, R. K., 1995. 234U/238U ratios in
quaternary planktonic foraminifera. Geochimica et
Cosmochimica Acta, 59, 4685–4694.
Henderson, G. M., Cohen, A. S., and O'Nions, R. K., 1993.
234 238
U/ U ratios and 230Th ages for Hateruma Atoll corals:
implications for coral diagenesis and seawater 234U/238U ratios.
Earth and Planetary Science Letters, 115, 65–73.
Huston, M. A., 1985. Patterns of species diversity on coral reefs.
Annual Review of Ecology and Systematics, 16, 149–177.
Kopp, R. E., Simons, F. J., Mitrovica, J. X., Maloof, A. C., and
Oppenheimer, M., 2009. Probabilistic assessment of sea level
during the last interglacial stage. Nature, 462, 863–867.
Lisiecki, L. E., and Raymo, M. E., 2005. A Pliocene–Pleistocene
stack of 57 globally distributed benthic delta O-18 records.
Paleoceanography, 20, 1–7.
Medina-Elizalde, M., 2013. A global compilation of coral sea-level
benchmarks: implications and new challenges. Earth and Plan-
etary Science Letters, 362, 310–318.
Milankovitch, M., 1941. The Canon of the Ice Ages. Belgrade:
Royal Serbian Academy.
Rohling, E. J., Grant, K., Bolshaw, M., Roberts, A. P., Siddall, M.,
Hemleben, C., and Kucera, M., 2009. Antarctic temperature
and global sea level closely coupled over the past five glacial
cycles. Nature Geoscience, 2, 500–504.
Shen, C.-C., Edwards, R. L., Cheng, H., Dorale, J. A., Thomas,
R. T., Moran, S. B., Weinstein, S., and Edmonds, H. N., 2002.
Uranium and thorium isotopic and concentration measurements
by magnetic sector inductively coupled plasma mass spectrome-
try. Chemical Geology, 185, 165–178.
Stirling, C. H., and Andersen, M. B., 2009. Uranium-series dating
of fossil coral reefs: extending the sea-level record beyond the
last glacial cycle. Earth and Planetary Science Letters, 284,
269–283.
Thompson, W. G., Spiegelman, M. W., Goldstein, S. L., and Speed,
R. C., 2003. An open-system model for U-series age determina-
tions of fossil corals. Earth and Planetary Science Letters, 210,
365–381.
Waelbroeck, C., Labeyrie, L., Michel, E., Duplessy, J. C.,
McManus, J. F., Lambeck, K., Balbon, E., and Labracherie,
M., 2002. Sea-level and deep water temperature changes derived
from benthic foraminifera isotopic records. Quaternary Science
Reviews, 21, 295–305.
733
Cross-references
Carbonates, Marine Carbonates (U-Series)
Carbonates, Speleothem Climatic (U-Series)
Mass Spectrometry
Thermal Ionization Mass Spectrometer (TIMS)
U-Series Dating
SEAWATER Sr CURVE
Galen P. Halverson and Lucie Hubert-Théou
Department of Earth and Planetary Sciences/Geotop,
McGill University, Montréal, QC, Canada
Definition
The seawater Sr curve is a proxy record of the 87Sr/86Sr
value of seawater through time, as reconstructed from
well-preserved minerals precipitated from seawater.
Introduction
Prior to the development of the analytical technique to
measure precise strontium isotope ratios, Wickman
(1948) predicted that the ratio of radiogenic to
unradiogenic strontium (87Sr/86Sr), as measured in car-
bonates and evaporate minerals precipitated from seawa-
ter, might be used as a tool to date ancient marine
sediments. Wickman’s rationale was straightforward,
even if, upon hindsight, partially incorrect and incom-
plete: the gradual concentration of Rb in the continental
crust and decay of radiogenic 87Rb would result in mono-
tonically increasing 87Sr/86Sr values of the crust and hence
a steady rise in seawater strontium isotope composition
over geological time. He presciently noted that due to
the effective separation of Sr and Rb in carbonates, in par-
ticular mollusk shells, limestone was the ideal target for
determining ancient seawater Sr isotope ratios. Although
87
Sr/86Sr is now measured on a wide variety of minerals,
well-preserved foraminifera tests and brachiopod shells
are among the prime targets for Sr isotope chemostra-
tigraphy of Phanerozoic sediments (Veizer et al., 1999;
McArthur et al., 2012).
Burke et al. (1982) published the first comprehensive
record of Phanerozoic 87Sr/86Sr, which revealed multiple
peaks and troughs that correspond in part to major plate
tectonic events. This entry firmly established the applica-
tion of strontium isotope chemostratigraphy as
a geochronological tool with wide applications in Earth
history. The history of the development of the seawater
Sr curve and the fundamentals of its application are briefly
reviewed here. For a thorough treatment of strontium iso-
tope stratigraphy and its geochronological applications,
the reader is referred to the recent review by McArthur
et al. (2012).
734 SEAWATER Sr CURVE

Development and application of the method
With the development of techniques to separate strontium
and analyze its isotopic ratios by solid source mass spec-
trometry, Wickman’s (1948) hypothesis that seawater
87
Sr/86Sr changed through time was put to the test in the
1950s and 1960s. Early results suggested that due to
a combination of lower-than-predicted variation in Rb/Sr
ratios in rock-forming minerals, poor age control on sam-
ples, and as yet limited analytical precision, strontium iso-
topes might not live up to their promise as an effective tool
in dating ancient marine sediments (Gast, 1955). How-
ever, geochemists did not give up, and subsequent applica-
tion of the technique highlighted the potential to unravel
past geological processes, including growth of the conti-
nental crust, using the strontium isotope system (e.g.,
Hurley et al., 1962; Hedge and Walthall, 1963).
Another early target of the Sr isotope method was in
characterization of seawater (e.g., Compston and Pidgeon,
1962; Hedge and Walthall, 1963; Faure et al., 1965). This
work demonstrated that the strontium isotope composition
of the present oceans is effectively uniform (0.7092) due
to the long residence time of strontium in the oceans
(3–5  106 years) versus the effective mixing time of
the oceans (103 years). Faure et al. (1965) further argued
that seawater composition at any given time was the result
of a mixture of Sr from several different reservoirs with
strongly contrasting strontium isotope ratios, most notably
mafic versus felsic continental crust and marine carbon-
ates. Armstrong (1971) proposed that weathering related
to continental glaciation might also exert a strong influ-
ence on seawater Sr isotope compositions.
Hedge and Walthall (1963) published the first dataset
on ancient carbonates and argued that these limited data
supported a roughly linear evolution in marine 87Sr/86Sr
over geological time, consistent with Wickman’s (1948)
hypothesis. However, Peterman et al. (1970) published
the first Phanerozoic seawater strontium isotope curve,
which, although at a low resolution (by modern standards)
and limited by poor age constraints, showed that 87Sr/86Sr
both rose and decreased during the Phanerozoic, with
a notable trough across the Jurassic–Cretaceous boundary.
Veizer and Compston (1974) subsequently upgraded the
Phanerozoic record, taking into account diagenetic effects
on 87Sr/86Sr in marine carbonates and applying the now
canonical principle that the lowest 87Sr/86Sr values for
rocks of any given time interval most reliably record coe-
val seawater composition. This compilation confirmed the
late Jurassic–Cretaceous 87Sr/86Sr trough and also
revealed a prominent Cenozoic rise in 87Sr/86Sr (Figure 1).
A decade later, Burke et al. (1982) published a much
higher-resolution Phanerozoic seawater 87Sr/86Sr compi-
lation that reproduced the broad patterns documented by
Veizer and Compston (1974) but also demonstrated
a more complex history of seawater strontium, with

Seawater Sr Curve, Figure 1 The seawater 87Sr/86Sr curve for the past 1.0 billion years. The Ordovician–Present portion of the curve
is the LOWESS fit from McArthur et al. (2012), updated from McArthur et al. (2001). The Neoproterozoic–Cambrian portion of the
curve (dashed line) is much more poorly calibrated and of lower resolution (Adapted from Shields and Veizer (2002) and Halverson
et al. (2007, 2010)).
 SEAWATER Sr CURVE
Seawater Sr Curve, Figure 2 A basic box model illustrating the
marine Sr cycle with present strontium isotope compositions for
fluxes and reservoirs. The main sources of sweater Sr are the
seafloor hydrothermal flux and the continental weathering flux.
The sizes of the arrows scale to the relative size of the associated
flux into and out of the ocean (from Banner, 2004). Note that the
flux from diagenesis of marine sediments (dominated by
carbonates) is effectively negligible because it releases
strontium with nearly the same composition as the seawater
reservoir (Kump, 1989). Strontium is removed from seawater as
a trace component of marine carbonates (and to a minor extent
sulfates).
multiple peaks, troughs, and intervals of rapid change.
Burke et al. (1982) interpreted the fluctuations in seawater
87
Sr/86Sr in terms of the competing influence of
unradiogenic (0.7035) strontium derived from mid-
ocean ridges (i.e., hydrothermal) and radiogenic strontium
(0.720 at present) derived from weathering of the conti-
nents (Figure 2), with carbonate-derived (weathered)
strontium acting as a buffer rather than a driver of change.
These fluctuations, they argued, corresponded to tectonic
events and reorganization of the continents during the
Phanerozoic. This compilation also implicitly demon-
strated a much improved potential for using strontium iso-
topes in marine carbonates as a dating tool, in particular
during intervals of rapidly evolving 87Sr/86Sr, such as the
Cambro-Ordovician, Permian, and Jurassic (McArthur
et al., 2001). DePaolo and Ingram (1985) later showed
how high-precision Sr isotope chemostratigraphy could
be used as a more effective dating tool than paleomagne-
tism and stable isotope chemostratigraphy during intervals
such as the past 35 m.y., where 87Sr/86Sr has been rising
steadily and rapidly, presumably due in large part to the
Alpine–Himalayan orogeny (Edmond, 1992) and/or Ant-
arctic glaciation (Zachos et al., 1999).
Subsequent work on the seawater strontium isotope
curve has involved focused efforts to produce high-
resolution records through specific time intervals (e.g.,
De Paolo and Embry; Koepnick et al., 1990; Paytan
et al., 1993; Shields et al., 2003), comprehensive Phaner-
ozoic compilations from well-constrained and well-
preserved samples (McArthur, 1994; Veizer et al., 1999),
and Precambrian compilations (Shields and Veizer,
2002), most specifically for the Neoproterozoic
735
(1,000–541 Ma), which shows changes comparable to
the Phanerozoic (Shields, 2007; Halverson et al., 2007;
Figure 1). Strontium isotope chemostratigraphy has
proven particularly valuable in the Neoproterozoic, where
lack of biostratigraphic constraints and consequent poor
age control necessitate alternative dating methods.
Sample selection and diagenetic considerations
As Wickman (1948) noted, the best materials for stron-
tium isotope chemostratigraphy are minerals that precipi-
tated from seawater. Many different materials have been
explored and analyzed as proxies for seawater composi-
tion, including both aragonite and calcite bulk rock, shells,
otoliths (fish ear bones), other biogenic and inorganic
CaCO3 precipitates, phosphatic material such as
conodonts, and sulfates, namely, evaporite minerals
(e.g., CaSO4) and barite (BaSO4). Due to the relatively
high Sr concentration in primary aragonite (up to 1 wt%),
originally aragonitic tests and shells (and to a lesser extent
early marine cements) may be good targets for Sr isotope
analyses. However, primary aragonite is unstable in the dia-
genetic realm and rare in the geological record. Further-
more, the type of materials available for strontium isotope
stratigraphy is heavily dependent on geological age.
Calcitic foraminifera tests recovered from deep sea cores
have proven to be well suited for the Cenozoic record
(McArthur et al., 2012), whereas brachiopod shells (Veizer
et al., 1999) and belemnite guards (Jones et al., 1994)
are among the most robust proxies for the Mesozoic and
Paleozoic eras.
The paucity to absence of biogenic carbonates in Cam-
brian and older strata necessitates the use of bulk rocks for
strontium isotope analyses. Conveniently, seawater and
calcite contain only trace amounts of Rb; unfortunately,
carbonate sediments also typically contain an acid insolu-
ble (detrital) component, resulting in a degree of ingrowth
of 87Sr from detrital 87Rb, mostly from clays. Coupled
with loss of Sr during diagenesis, this can result in signif-
icantly elevated 87Sr/86Sr values in the carbonate fraction,
even in carefully processed carbonate samples. Tech-
niques are available for correcting for radiogenic 87Sr
ingrowth (e.g., McArthur, 1994); however, where possi-
ble, low-Rb carbonates should be targeted for Sr isotope
stratigraphy. Analyzing multiple samples from the same
or closely spaced stratigraphic horizons is also useful in
determining the pattern of alteration of 87Sr/86Sr during
diagenesis (Banner and Hanson, 1990; McArthur et al.,
2012) and constraining the most likely seawater composi-
tion, which is typically assumed to be less than or equal to
the lowest 87Sr/86Sr value.
The crystal lattice of dolomite, a common secondary
carbonate mineral, does not accommodate the relatively
large Sr2+ cation. Hence, dolostones, which comprise
a significant fraction of the stratigraphic record, have spar-
ingly low Sr concentrations (typically less than 100 ppm),
making them highly prone to isotopic overprinting and in
most cases unsuitable for strontium isotope stratigraphy.
736 SEAWATER Sr CURVE
Other carbonate minerals, such as ankerite and siderite,
typically form authigenically and hence are also poor
proxies for seawater Sr isotope composite.
Analytical techniques
Sample preparation and analyses for Sr isotopes on car-
bonate minerals (and other seawater precipitates) involve
four steps: sample selection, sample cleaning and dissolu-
tion, chemical separation, and measurement of isotope
ratios by mass spectrometry. Because only samples with
minimal Rb contents consistently yield reliable data, only
Sr isotope ratios (and not precise Sr or Rb concentrations)
are typically measured for Sr isotope stratigraphy,
hence eliminating the requirement to add an isotopic tracer
solution, unless a radiogenic ingrowth correction is
required:
1. Sample selection is performed based on a combination
of petrographic and geochemical prescreening. Those
samples showing the least recrystallization and
overprinting by secondary cements, relatively high
Sr/Rb ratios (or in the absence of Rb data, high Sr/Ca
ratios), and low detrital component are best suited for
Sr isotope analyses.
2. Sample dissolution is usually performed in a dilute acid
such as weak acetic acid, which minimizes partial dis-
solution of the non-carbonate fraction. In order to fur-
ther reduce incorporation of detrital strontium, it is
common to pre-leach sample powders in ammonium
acetate, which removes loosely bound Sr ions (Bailey
et al., 2000). A prewash with methanol (McArthur
et al., 2006) is also effective in removing clays from
a sample powder, reducing the potential for leaching
of detrital strontium during sample dissolution.
3. Dissolved samples are passed through an ion exchange
resin (e.g., Eichrom™ Sr-spec resin) to separate Sr
from other ions (see also entry on Chromatography),
most importantly Ca and Rb, which can greatly affect
Sr isotope measurements.
4. Traditionally, Sr isotope analyses have been performed
by solid source, thermal ionization mass spectrometry
(TIMS; see also entry on Mass Spectrometry). Mea-
sured 87Sr/86Sr values are corrected against an interna-
tional standard, usually NIST (NBS) 987, with an
accepted 87Sr/86Sr value of 0.710248. Reproducibility
is commonly as good as 0.000005 (2s). Strontium
isotope ratios may also be measured in solution or by
laser ablation on a multi-collector inductively coupled
plasma mass spectrometer (MC-ICP-MS). These tech-
niques are faster, and laser ablation has the advantage
that it does not require time-consuming chemical sepa-
ration. However, they also yield less precise results
than TIMS, although reproducibility as low as
0.00001 (2s) has been achieved in solution mode
(e.g., Balcaen et al., 2005). Similar precision and accu-
racy has not yet been realized by laser ablation MC-
ICP-MS, but the technique is increasingly applied to
high spatial resolution analyses of media such as fish
otoliths, bones, and teeth where the high precision pro-
vided by TIMS analysis is not required (Copeland
et al., 2010).
The strontium isotope seawater curve
(1.0–0 billion years)
Since Burke et al. (1982) published their record of Phaner-
ozoic seawater 87Sr/86Sr, a great deal of work has gone
both into developing high-resolution records for specific
intervals of the Phanerozoic (e.g., Koepnick et al., 1990;
Figure 3) and in producing compilations of the highest-
quality and best dated samples (Veizer et al., 1999;
McArthur et al., 2001). The most comprehensive compila-
tion of Phanerozoic 87Sr/86Sr data is found in the database
of McArthur et al. (2001), which has been progressively
updated (e.g., McArthur and Howarth, 2004; McArthur
et al., 2012), and now comprises over 4,000 data points
from some 50 sources (McArthur et al., 2012). The con-
struction of such a record requires the application of age
models to specific data sets, the integration of these data
sets, and a technique for fitting a curve to the data, which
can then be used to calculate an age and appropriate error
(see Howarth and McArthur, 1997, for detailed descrip-
tion of data treatment and analysis for the Phanerozoic
seawater Sr curve shown in Figure 1). For the relatively
well-dated Phanerozoic record, age models are reasonably
robust and there exist high-resolution records, resulting in
typical 95 % confidence intervals about the best-fit curve
of 0.00001–0.00004 (McArthur et al., 2012).
The most complete database for Precambrian strontium
isotopes is from Shields and Veizer (2002). However, the
age constraints on Precambrian samples are deficient,
and the quality of data is lower and more difficult to verify
than for the Phanerozoic. Here, we include only data from
1.0 billion years onward because the pre-1.0 Ga record is
poor and shows only limited change, in contrast with the
1,000–485 Ma Sr isotope record, which has been exten-
sively studied and shows significant variability. Because
the resolution and quality of the data from this time inter-
val is low, rigorous fitting techniques as used for the Phan-
erozoic record are not yet applicable.
The Proterozoic–Phanerozoic strontium isotope seawa-
ter curve (Figure 1) reveals many striking features. The
first-order trend is one of increasing 87Sr/86Sr, which
results from the progressive decay of 87Rb (Wickman,
1948; Shields, 2007). The second-order pattern is one of
peaks at present and c. 500 Ma ago and troughs at about
180 and 1,000 Ma ago, which may reflect the cycle of
supercontinental breakup and assembly (Halverson et al.,
2007). Superimposed on this pattern are prominent fluctu-
ations, some by as much as 0.001, over intervals of a few
to 25 Ma, many of which closely correspond to tectonic
events.
The main parameters controlling the seawater Sr iso-
tope composition are the magnitude of the seafloor hydro-
thermal flux (both mid-ocean ridge volcanism and off-axis
alteration of the seafloor) and continental weathering and
 SEAWATER Sr CURVE 737

their respective 87Sr/86Sr values. The composition of the
seafloor flux (0.7037; Banner, 2004) is approximately
fixed by mantle composition, whereas the 87Sr/86Sr of
continental weathering is presently 0.712 (Figure 2)
but presumably varied considerably through time.
Because the residence time of Sr in the oceans is 5 Ma
and there are no plausible mechanisms to increase or
decrease the flux into or out of the oceans rapidly, changes
in 87Sr/86Sr are steady state (i.e., gradual; see also entry on
Marine Isotope Stratigraphy). In principle, changes in any
one of the four major influx parameters (Figure 2) can
drive a change in seawater 87Sr/86Sr, but due to the large
isotopic difference between the hydrothermal and conti-
nental weathering sources, many researchers invoke
changes in the relative contribution of these two fluxes
to explain changes in seawater 87Sr/86Sr. On the other
hand, although the continental flux is buffered by the
weathering of marine carbonates, the huge variations
between different rivers (Gaillardet et al., 1999) suggest
that the continental weathering flux should have changed
substantially through time as a result of paleogeographic
evolution of the continents and paleoclimate.
Use as a geochronological tool
Given a reference seawater 87Sr/86Sr curve, dating
a marine rock or sediment of poorly constrained age may
be accomplished by comparing its strontium isotope com-
position to a reference curve. However, the effectiveness
of this tool is contingent upon the age and available age
constraints on that sample, whether or not there are stron-
tium isotope data from overlying or underlying strata, and
the robustness of the curve. For example, a marine carbon-
ate sample that is known only to be Jurassic or early Cre-
taceous in age and has an 87Sr/86Sr value of 0.70720 could
equally be Pliensbachian (190.8–182.7 Ma), Toarcian
(182.7–174.1 Ma), Bajocian (170.3–168.3 Ma), or
Berriasian (145.0–139.4 Ma) in age (Figure 3A). How-
ever, if the stage of the sample is known through biostra-
tigraphy or other chronostratigraphic constraints, then
the age can be specified more precisely (Figure 3B), with

Seawater Sr Curve, Figure 3 A. The LOWESS fit Sr isotope curve for the time interval 200–140 Ma ago (Jurassic stages in blue and
gray: Aal Aalenian, Bj Bajocian, Bt Bathonian, Cal Callovian, Kimm Kimmeridgian. Cretaceous Berriasian stage is in green), from
McArthur et al. (2012). The red line is the LOWESS fit, and the blue lines mark the upper and lower 95 % confidence intervals. Dashed
line at 87Sr/86Sr ¼ 0.70720 demonstrates the nonuniqueness of this value during the Jurassic, where it occurs during four distinct
stages. B. Inset is a zoom-in of the interval from 170 to 169 Ma. If a sample with a Sr isotope composition of 0.70720 is known to be
Bajocian in age (e.g., based on biostratigraphy or other chemostratigraphic data), then an age can be determined by comparison with
the reference curve. In this case, a value of 0.70720 yields an age of 169.60  0.04 Ma (note that the error does not factor in
uncertainty related to the calibration of the time scale).
738 SEAWATER Sr CURVE
the precision being a measure of the confidence interval
for the Sr isotope curve at that age (Figure 3) and the
uncertainty in the age model upon which the reference
Sr isotope curve is constructed (McArthur et al., 2012).
In the absence of additional age constraints that elimi-
nate some of the possible matches between the sample
and the reference curve, it helps to match both the absolute
value and stratigraphic (i.e., temporal) trends. For exam-
ple, if multiple samples from different stratigraphic levels
are analyzed from the section containing the original sam-
ple of interest, then these can be used to determine the tra-
jectory in 87Sr/86Sr. If the samples show a trend of
decreasing 87Sr/86Sr up-section (i.e., in progressively
younger rocks), then this pattern eliminates either
a Toarcian or Berriasian age, both of which correspond
to intervals of increasing 87Sr/86Sr (Figure 3A). Whereas
the slope of the declining trends (d[87Sr/86Sr]/dt) differs
between the Pliensbachian and Bajocian, it would be vir-
tually impossible to use the distinction alone in a poorly
dated section of strata to discriminate between these two
ages. Rather, a more complete 87Sr/86Sr profile that cap-
tured, for example, the late Pliensbachian trough
(0.707075) or the continued decline to 87Sr/86Sr values
lower than 0.707075 in the Bathonian (Figure 3) would
be required. Alternatively, additional chemostratigraphic
data such as carbonate carbon isotopes or sulfate sulfur
isotopes (see entry on Marine Isotope Stratigraphy) or
magnetic polarity data may provide sufficient constraints
to narrow down the appropriate age for the 0.70720
sample.
Other applications
The seawater strontium isotope curve has other useful
applications in stratigraphy. By comparing the Sr isotope
trend of a given succession of strata to the reference curve,
sediment accumulation rates can be reconstructed. This
application is most useful where the rate of change in
87
Sr/86Sr is high and of limited to no use where the rate
of change is low or effectively nil (McArthur et al.,
2012). Strontium isotopes are also useful for detecting
and quantifying breaks in the stratigraphic record.
Because seawater Sr isotope composition evolves gradu-
ally, any abrupt shifts within strata must record either
a depositional hiatus (diastem) or an erosional disconfor-
mity. Again, this application works best where the rate of
change in 87Sr/86Sr is high, in which case the offset in
87
Sr/86Sr across a stratigraphic boundary can be used to
approximate the amount of time missing at that strati-
graphic level.
Summary and conclusions
Strontium isotope stratigraphy has important applications
as a chronostratigraphic tool. The basis for this dating
method is the construction of a reference seawater stron-
tium curve against which undated or poorly dated samples
can be compared. The ability to date a given sedimentary
sample or interval of strata depends on several factors:
(1) the reliability of that sample as a proxy for seawater
composition and the protocol by which it is analyzed,
(2) independent constraints on the approximate age of that
sample, and (3) the precision of both the age model and the
Sr isotope data upon which the appropriate reference sea-
water Sr isotope curve has been constructed. A high-
precision Sr isotope curve now exists for the interval from
the Ordovician through the present (McArthur et al.,
2012), and a lower-resolution and as yet poorly age-
calibrated record for the Neoproterozoic–Cambrian is
available (Figure 1). The reliability of strontium isotope
stratigraphy as a dating tool decreases with increasing
age; however, its importance increases with older rocks
because other dating methods are rare or less reliable.
For example, in the absence of a robust biostratigraphy
for Precambrian strata, strontium isotope stratigraphy will
play an important role in defining the geological time scale
for these older rocks.
Bibliography
Armstrong, R. L., 1971. Glacial erosion and the variable isotopic
composition of strontium in sea water. Nature, 230, 132–133.
Balcaen, L., De Schrijver, I., Moens, L., and Vanhaecke, F., 2005.
Determination of the 87Sr/86Sr isotope ratio in USGS silicate ref-
erence materials by multi-collector ICP-mass spectrometry.
International Journal of Mass Spectrometry, 242, 251–255.
Banner, J. L., 2004. Radiogenic isotopes: systematics and applica-
tions to earth surface processes and chemical stratigraphy.
Earth-Science Reviews, 65, 141–194.
Banner, J. L., and Hanson, G. N., 1990. Calculation of simultaneous
isotopic and trace element variations during water-rock interac-
tion with applications to carbonate diagenesis. Geochimica et
Cosmochimica Acta, 54, 3123–3137.
Burke, W. M., Denison, R. E., Hetherington, E. A., Koepnik, R. B.,
Nelson, M., and Omo, J., 1982. Variation of seawater 87Sr/86Sr
throughout Phanerozoic time. Geology, 10, 516–519.
Compston, W., and Pidgeon, R. T., 1962. Rb-Sr dating shales by the
total rock method. Journal of Geophysical Research, 67,
3493–3502.
Copeland, S. R., Sponheimer, M., Lee-Thorp, J. A., le Roux, P. J.,
de Ruiter, D. J., and Richard, M. P., 2010. Strontium isotope
ratios in fossil teeth from South Africa: assessing laser ablation
MC-ICP-MS analysis and the extent of diagenesis. Journal of
Archaeological Science, 37, 1437–1446.
DePaolo, D. J., and Ingram, B. L., 1985. High-resolution stratigra-
phy with strontium isotopes. Science, 227, 938–941.
Edmond, J. M., 1992. Himalayan tectonics, weathering processes,
and the strontium isotope record in marine limestones. Science,
258, 1594–1597.
Faure, G., Hurley, P. M., and Powell, J. L., 1965. The isotopic com-
position of strontium in surface water from the North Atlantic
Ocean. Geochimica et Cosmochimica Acta, 29, 209–220.
Gaillardet, J., Dupré, B., Louvat, P., and Allègre, C. J., 1999. Global
silicate weathering and CO2 consumption rates deduced from the
chemistry of large rivers. Chemical Geology, 159, 3–30.
Gast, P. W., 1955. Abundance of Sr87 during geological time. Bulle-
tin of the Geological Society of America Bulletin, 66,
1449–1454.
Halverson, G. P., Dudas, F. O., Maloof, A. C., and Bowring, S. A.,
2007. Evolution of the 87Sr/86Sr composition of Neoproterozoic
seawater. Palaeogeography Palaeoclimatology Palaeoecology,
256, 103–129.
 SECONDARY ION MASS SPECTROMETRY (SIMS)
Halverson, G. P., Wade, B. P., Hurtgen, M. T., and Barovich, K.,
2010. Neoproterozoic chemostratigraphy. Precambrian
Research, 182, 337–350.
Hedge, C. E., and Walthall, F. E., 1963. Radiogenic strontium-87 as
an index of geological processes. Science, 140, 1214–1217.
Howarth, R. J., and McArthur, J. M., 1997. Statistics for strontium
isotope stratigraphy: a robust LOWESS fit to the marine
Sr-isotope curve for 0 to 206 Ma, with look-up table for deriva-
tion of numeric age. Journal of Geology, 105, 441–456.
Hurley, P. M., Hughes, H., Faure, G., Fairbairn, H. W., and
Pinson, W. H., 1962. Radiogenic strontium-87 model of conti-
nent formation. Journal of Geophysical Research, 67,
5315–5334.
Koepnick, R. B., Denison, R. E., Burke, W. H., Hetherington, E. A.,
and Dahl, D. A., 1990. Construction of the Triassic and Jurassic
Portion of the Phanerozoic curve of seawater 87Sr/86Sr. Chemical
Geology, 80, 327–349.
Kump, L. R., 1989. Alternative modeling approaches to the geo-
chemical cycles of carbon, sulfur, and strontium isotopes. Amer-
ican Journal of Science, 289, 290–410.
McArthur, J. M., 1994. Recent trends in strontium isotope stratigra-
phy. Terra Nova, 6, 331–358.
McArthur, J. M., Howarth, R. J., and Bailey, T. R., 2001. Strontium
isotope stratigraphy: LOWESS version 3: best fit marine Sr-
isotope curve for 0–509 Ma and accompanying look-up table
deriving numerical age. Journal of Geology, 109, 155–170.
McArthur, J. M., Rio, D., Castradori, D., Bailey, T. R., Thirlwall,
M., and Houghton, S., 2006. A revised Pliocene record for
marine 87Sr/86Sr used to date an interglacial event recorded in
the Cockbur Island Formation, Antarctic Peninsula.
Palaeogeography Palaeoclimatology Palaeoecology, 242,
126–136.
McArthur, J., Howarth, R. J., and Shields, G. A., 2012. Strontium
isotope stratigraphy. In The Geological Time Scale. Elsevier,
Chap. 7, pp. 127–144.
Paytan, A., Kastner, M., Martin, E., MacDougall, J., and Herbert, T.,
1993. Marine barite as a monitor of seawater strontium isotope
composition. Nature, 366, 445–449.
Peterman, Z. E., Hedge, C. E., and Tourtelot, H. A., 1970. Isotopic
composition of strontium in sea water throughout Phanerozoic
time. Geochemical et Cosmochimica Acta, 34, 105–120.
Shields, G. A., 2007. A normalized seawater strontium isotope
curve: possible implications for Neoproterozoic-Cambrian
weathering rates and the further oxygenation of the Earth.
eEarth, 2, 35–42.
Shields, G., and Veizer, J., 2002. Precambrian marine carbonate iso-
tope database: version 1.1. Geochemistry Geophysics
Geosystems, 3, 1, doi:10.1029/2001GC000266.
Veizer, J., and Compston, W., 1974. 87Sr/86Sr composition of sea-
water during the Phanerozoic. Geochimica et Cosmochimica
Acta, 38, 1461–1484.
Veizer, J., Buhl, D., Diner, A., Ebneth, S., Podlaha, O. G.,
Bruckschen, P., Jasper, T., Korte, C., Schaaf, M., Ala, D., and
Azmy, K., 1997. Strontium isotope stratigraphy: potential reso-
lution and event correlation. Palaeogeography Palaeocli-
matology Palaeoecology, 132, 65–67.
Veizer, J., Ala, D., Azmy, K., Bruckschen, P., Buhl, D., Bruhn, F.,
Carden, G., Diener, A., Ebneth, S., Godderis, Y., Jasper, T.,
Korte, C., Pawellek, F., Podlaha, O., and Strauss, H., 1999.
87
Sr/86Sr, d13C and d18O evolution of Phanerozoic seawater.
Chemical Geology, 161, 59–88.
Wickman, F. E., 1948. Isotope ratios: a clue to the age of certain
marine sediments. Journal of Geology, 56, 61–66.
Zachos, J. C., Opdyke, B. N., Quinn, T. M., Jones, C. E., and
Halliday, A. N., 1999. Early Cenozoic glaciation, Antarctic
weathering, and seawater 87Sr/86Sr: is there a link? Chemical
Geology, 161, 165–180.
739
SECONDARY ION MASS SPECTROMETRY (SIMS)
Trevor R. Ireland
Research School of Earth Sciences, The Australian
National University, Canberra, ACT, Australia
Synonyms
Focused ion beam mass spectrometry; Ion microprobe
mass spectrometry; Ion microscope mass spectrometry
Definition
Mass Spectrometer. Instrument for causing separation of
ions according to mass.
Introduction
SIMS is a technique for generating ion beams from solid
targets. A focused ion beam is used to sputter (¼ ablate)
the target mineral. The primary ion beam is typically quite
energetic (10–20 keV) and composed of chemically reac-
tive species (O, Cs) that enhance secondary ion production
in electropositive elements (metals) and electronegative
species (non metals). SIMS is applied through instruments
such as the ion microprobe or ion microscope. For U-Pb
geochronology, a negative oxygen ion beam (O or
O2) is used to generate U+ and Pb+ ions. Molecular spe-
cies UO+, UO2+, ThO+, etc., are also commonly mea-
sured. Sputtering results in complicated mass spectra that
include atomic, simple molecule, and complex molecule
species.
Instrumentation
In order to distinguish between the relevant Pb+ and
U(On)+ species, large mass spectrometers are frequently
used. The first mass spectrometer used for this purpose
was the SHRIMP (Sensitive High-Resolution Ion Micro-
probe), which had a magnet turning radius of 100 cm
and a total beam path length of approximately 700 cm,
allowing both high mass resolution and high sensitivity
(Ireland et al., 2008). SHRIMP utilized a double focusing
mass analyzer incorporating magnetic and electrostatic
sectors, the electrostatic sector compensating for velocity
dispersion in the magnet. CAMECA instruments also pro-
duce a similarly sized instrument (IMS 1280) based on
their ion microscope range, which can be used for
geochronology.
U–Pb geochronology
Studies suggest that instrumental mass fractionation of Pb
isotopes is low (0.2 % per amu) allowing use of
uncorrected Pb isotope ratios. Therefore, zircons older
than 1,000 billion years can be dated using directly mea-
sured 207Pb/206Pb ratios. Younger zircons have insuffi-
cient 207Pb for 207Pb/206Pb dating, but a novel
calibration technique allows the use of 206Pb/238U ratios
for geochronology (Compston et al., 1984).
740 SEDIMENT MIXING RATE, 210Pb AND 234Th
The Pb+/U+ ratio is fractionated from the true Pb/U
ratio because of the differences in ionization yields
between U and Pb. Furthermore, the Pb+/U+ can be quite
variable in an analytical session. Based on a model of local
thermodynamic equilibrium, the measured Pb+/U+ is
found to covary with uranium-uranium oxide speciation.
Pb+/U+ versus UO+/U+ is frequently used to monitor the
instrumental fractionation and allow measurement of cali-
brated 206Pb/238U ratios to a level of 1 % reproducibility.
Mineral analysis
The first mineral used for ion microprobe analysis was zir-
con (Compston et al., 1984). The beneficial characteristics
of zircon are that it contains elevated concentrations of
U (100–500 mg/g), but the crystal structure strongly
excludes Pb. Radiogenic Pb produced by the decay of
U is strongly retained in zircon during even high-grade
metamorphism. Zircons can be strongly zoned with
domains representing inherited material, as well as
igneous and metamorphic overgrowths. Ion microprobe
analysis allows in situ analyses of growth zones that
can be characterized through electron imagery
(cathodoluminescence) as well as optical imagery
(transmitted and reflected light) to optimize petrographic
context and avoid areas complicated by inclusions or frac-
turing (Hanchar and Hoskin, 2003). Other U-bearing min-
erals including monazite, allanite, titanite, and apatite
have been dated.
Precision of SIMS U-Pb analyses can be limited by the
small amount of material analyzed. A typical analysis con-
sumes only a few nanograms of material. For a 100-Ma-
old zircon with 100 mg/g U, only 3 fg of 206Pb are included
in an analysis, many orders of magnitude below detection
limits of other techniques. Precision can be further limited
for materials with low uranium concentrations or young
minerals where radiogenic Pb in-growth is limited.
Ion microprobe analysis has been applied to zircons
younger than a million years and to zircons formed at the
start of the solar system, 4.56 billion years ago.
Conclusion
SIMS analysis allows a reliable means of dating discrete
domains in U-Th-Pb-bearing minerals that may have com-
plex histories. In situ analysis allows retention of the pet-
rographic context and is one of the most powerful
attributes of the technique. However, because the amount
of material sampled is extremely small (few nanograms),
precision can be limited, especially in young minerals or
those with low uranium concentrations.
Bibliography
Compston, W., Williams, I. S., and Meyer, C., 1984. U-Pb geochro-
nology of zircon from lunar breccia 73217 using a sensitive high
mass-resolution ion microprobe. Journal of Geophysical
Research, 89(Supplement), B525–B534.
Hanchar, J. M., and Hoskin, P. W. O., 2003. Zircon. Reviews in Min-
eralogy and Geochemistry, 53, 500.
Ireland, T. R., Clement, S., Compston, W., Foster, J. J., Holden, P.,
Jenkins, B., Lanc, P., Schram, N., and Williams, I. S., 2008.
The development of SHRIMP. Australian Journal of Earth Sci-
ences, 55, 937–954.
Cross-references
Mass Spectrometry
Uranium–Lead Dating
Uranium–Lead, Zircon
Zircon
SEDIMENT MIXING RATE, 210Pb AND 234Th
Joseph M. Smoak
Department of Environmental Science, Policy and
Geography, University of South Florida, St. Petersburg,
FL, USA
Definition
Sediment mixing – the movement or redistribution of
deposited sediments within a sediment column by physi-
cal and/or biological means such that the sediment is
moved from its original sequence or position of
deposition.
Introduction
Two members of the naturally occurring uranium and tho-
rium decay series, 210Pb (half-life 22.3 years) and 234Th
(half-life 24.1 days), may occur in excess of their parent
activity at depth within the sediment column of aquatic
systems either from sediment accumulation or downward
mixing of sediment particles. The downward mixing of
sediment particles is produced by physical processes such
as current or wave action and biological processes such as
burrowing organisms. Being aware of and understanding
sediment mixing is important for several reasons. The
most basic of these reasons relates to using sediments to
interpret the past. Sediments are deposited in sequential
order and thus provide a temporal record of surface pro-
cesses and a historical record of material input. This infor-
mation can have a variety of uses from understanding past
climate or environmental changes to reconstruction of the
pollution or contaminant history of a drainage basin.
Interpreting the sequential sediment record or layers is
sometimes described by the analogy of reading the layers
like they are pages in a book. Now imagine the pages in
the book have been rearranged. This rearrangement alters
or smears the sediment record which can influence the
preservation of physical sedimentary structures
(Nittrouer and Sternberg, 1981). How much the sediment
record is altered depends on the intensity, depth, and
nature of the sediment mixing in relation to the sediment
accumulation rate (Smoak and Patchineelam, 1999). Sed-
iment mixing also influences degradation and preserva-
tion of biogenic components, pore-water concentrations
 SEDIMENT MIXING RATE, 210Pb AND 234Th
of dissolved chemical species within the sediment mixed
layer (Schink and Guinasso, 1977; Berner, 1980; Aller,
1982; Alperin et al., 1999), microbial activity (Yingst
and Rhoads, 1980), and fate and transport of contaminants
(Bothner et al., 1980; Officer and Lynch, 1989).
In marine systems, both 210Pb and 234Th are continually
produced within the overlying water column by their
radionuclide parents, while 210Pb is also supplied via
atmospheric fallout. In freshwater systems, 234Th may
not be present, and 210Pb supply may be dominated by
the atmospheric fallout due the low abundance of the
radionuclide parents within the water column. In either
case once in the water column, 210Pb and 234Th interact
with particles and are rapidly removed from dissolved to
particulate phase. Due to the rapid adsorption onto parti-
cles, these elements along with other elements that rapidly
become particle bound are referred to as “particle-
reactive” elements. Once in the particulate phase, the par-
ticles to which they are attached may settle from the water
column and be deposited at the sediment/water interface.
The occurrence of excess 210Pb or 234Th in the sediments
deeper than would be produced by sediment accumulation
alone can be used to characterize sediment mixing. While
physical and biological mixing are separate processes,
they may both occur within the same sediment column.
However, most systems can be characterized as mixed
by either dominantly physical or biological processes
depending on the preservation of primary sedimentary
structure. Physical mixing by waves, tides, and currents
is generally event driven and potentially resets the sedi-
ment record within the mixing zone (Sanford, 1992) while
preserving sedimentary structure below the mixing zone.
Biological mixing, referred to as bioturbation, is generally
a steady-state process that can destroy the sedimentary
structure (Dellapenna et al., 1998) depending on the
mixing rate compared to accumulation rate. The radioac-
tivity profiles of 210Pb or 234Th produced by sediment
mixing are often used to calculate diffusion coefficients
based on Fick’s law of diffusion. The diffusion coefficient
is defined as the rate at which the variance of particle loca-
tion changes with time, where the variance is a measure of
particle spread and is proportional to the squared velocity
of the diffusing particle (Crank, 1975). The coefficient
units are length squared per unit time typically expressed
as cm2-yr1.
Most attempts to model sediment mixing have been
used to evaluate biological mixing; however, the same
models have been applied to physical mixing as well.
The coefficients produced by biological mixing are
referred to as biodiffusion coefficients although they are
actually eddy diffusion coefficients. Biological sediment
mixing is not truly a diffusive process; however, in many
cases, biological mixing appears diffusive-like when
many nondiffusive events are integrated over a short time
period compared to the time scale of the radioactive tracer
(Aller and Dodge, 1974; Robbins et al., 1979; Boudreau,
1986; Wheatcroft et al., 1990; Maire et al., 2008). If
mixing is not rapid compared to the time scale of the
741
tracer, tracer-dependent artifacts can appear (Reed
et al., 2007; Lecroart et al., 2010). For a diffusion model
to accurately represent the physical process of sediment
mixing, sediment particles must be moved in a random
manner over short distances. Large step length in the
advection of sediment produced by conveyor belt-type
sediment feeders may violate these assumptions.
Nonrandom movement produced by selective feeding
due to size, shape, texture, or composition may also vio-
late these assumptions (Smoak and Patchineelam, 1999).
A one-dimensional biodiffusion model is the most com-
monly applied model due to its simplicity. Even though
the assumptions may be violated and the model does not
include realistic biological complexity, an accurate empir-
ical model is produced in most cases. The fact that the
model is used despite these issues has been referred to
by Meysmann et al. (2003) as the “biodiffusion paradox.”
Other more complex and realistic models have been devel-
oped for different types of reworking using nonlocal trans-
port models (Boudreau and Imboden, 1987; Francois
et al., 1997, 2001; Meysman et al., 2003; Maire
et al., 2007). For a complete review of methods to quantify
bioturbation rates, refer to Maire et al. (2008).
Sediment mixing model
Despite the fact that assumptions are often violated, sedi-
ment mixing is generally well described by a simple
one-dimensional diffusion model based on Fick’s law of
diffusion. This is the most commonly used model to calcu-
late diffusion coefficients. Activity (A; dpm-g1) of the
tracer in the sediment column will be affected by the sed-
iment mixing (D; cm2-yr1), sediment accumulation (S;
cm-yr1), radioactive decay (l; yr1), and production (P;
dpm-g1-y1) of these radionuclides supported by their
parents. All of these variables influence the observed
tracer profile in the sediment column and are all included
in the steady-state equation Eq. (1) presented by Guinasso
and Schink (1975) where z is depth below the sediment-
water interface.
@2A @A
D S  lA þ P ¼ 0 ð1Þ
@z2 @z
Using the excess activity allows the production term to
be removed, and diffusion coefficients can be obtained
from the following steady-state equation Eq. (2):
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi !
S  S 2 þ 4Dl
Az ¼ A0 exp z ð2Þ
2D
In Figure 1 Eq. (2) was used to create an idealized
plot showing the activity of the tracer as a percent of
the initial activity versus depth. In this plot the sediment
accumulation rate is 2 mm-yr1, diffusion coefficient is
35 cm2-yr1, and the mixed depth is 10 cm. The longer-
lived tracer (i.e., 210Pb) has a nearly vertical profile within
the mixed layer, while the shorter-lived tracer (i.e., 234Th)
exhibits a strong gradient. In this case the 234Th profile is
742 SEDIMENT MIXING RATE, 210Pb AND 234Th
% of Initial Activity
0 1
0
10
depth (cm)
15
20
25
30
Sediment Mixing Rate, Pb and 234Th, Figure 1 Idealized plot showing the activity of the tracer as a percent of the initial activity
210
versus depth (cm). Sediment accumulation rate is 2 mm-yr1, diffusion coefficient is 35 cm2-yr1, and the mixed depth is 10 cm.
dominated by mixing, and a simplified version of
the equation Eq. (3) can be used to calculate the
diffusion coefficient. This approach is commonly applied
within the mixed zone by fitting a regression line to the
activity profile of the tracer and using the gradient to cal-
culate the diffusion coefficient (D).
" #2
z Alperin, M. J., Martens, C. S., Albert, D. B., Suayah, I. B.,
D ¼ l A  ð3Þ
ln 0 =Az
In cases were accumulation and mixing rates are much
slower, 210Pb can be used in the same way 234Th was used
above to calculate diffusion coefficients.
Conclusion
Sediment mixing, either physically and/or biologically
induced, results in the alteration of the original location
of sediment particles. This sediment mixing, in turn, influ-
ences the degree of preservation of the sediment record
and biogenic components, microbial activity, and fate
and transport of contaminants. The most common
approach to modeling sediment mixing is to use
a one-dimensional diffusion-type model. The naturally
occurring radionuclides 234Th and 210Pb are well-suited
tracers to estimate the diffusion mixing coefficient in these
models. When mixing is relatively rapid, the shorter time-
scale tracer (234Th) is the most suitable. When sediment
accumulation and mixing rates are slower, the longer
time-scale tracer (210Pb) becomes more useful. While
these models do not include all the complexity actually
involved in sediment mixing, the models along with these
uranium-thorium series tracers do create an accurate
empirical representation of sediment mixing.
10 100
Pb-210
Th-234
Bibliography
Aller, R. C., 1982. The effects of macrobenthos on chemical proper-
ties of marine sediment and overlying water. In McCall, P. L.,
and Tevesz, M. S. J. (eds.), Animal-Sediment Relations. New
York: Plenum Press, pp. 53–102.
Aller, R. C., and Dodge, R. E., 1974. Animal-sediment relations in
a tropical lagoon, Discovery Bay, Jamaica. Journal of Marine
Research, 32, 209–232.
Benninger, L. K., Blair, N. E., and Jahnke, R. A., 1999. Benthic
fluxes and porewater concentration profiles of dissolved organic
carbon in sediments from the north Carolina continental slope.
Geochimica et Cosmochimica Acta, 63, 427–448.
Berner, R. A., 1980. Early Diagenesis: A Theoretical Approach.
Princeton: Princeton University Press, p. 241.
Bothner, M. H., Jahnke, R. A., Peterson, M. L., and Carpenter, R.,
1980. Rate of mercury loss from contaminated estuarine sedi-
ments. Geochimica et Cosmochimica Acta, 44, 273–285.
Boudreau, B., 1986. Mathematics of tracer mixing in sediment. II.
Non-local mixing and biological conveyor-belt phenomena.
American Journal of Science, 286, 199–238.
Bourdreau, B. P., and Imboden, D. M., 1987. Mathematics of tracer
mixing in sediments: III the theory of nonlocal mixing within
sediments. American Journal of Science, 287, 693–719.
Crank, J., 1975. The Mathematics of Diffusion. Oxford: Oxford
Press, London, p. 424.
Dellapenna, T. M., Kuehl, S. A., and Schaffner, L. C., 1998. Sea-bed
mixing and particle residence times in biologically and physi-
cally dominated estuarine systems: a comparison of lower Ches-
apeake Bay and the York River subestuary. Estuarine, Coastal
and Shelf Science, 46, 777–795.
Francois, F., Poggiale, J. C., Durbec, J. P., and Stora, G., 1997.
A new approach for the modeling of sediment reworking
induced by macrobenthic community. Acta Biotheoretica, 45,
295–319.
Francois, F., Poggiale, J. C., Durbec, J. P., and Stora, G., 2001.
A new model of bioturbation for a functional approach to sedi-
ment reworking resulting from macrobenthic communities. In
Aller, J. Y., Woodin, S. A., and Aller, R. C. (eds.), Organism-
 SEDIMENT, ESR
Sediment Interactions. Columbia: University of South Carolina
Press, pp. 73–86.
Guinasso, N. L., Jr., and Schink, D. R., 1975. Quantitative estimates
of biological mixing rates in abyssal sediments. Journal of Geo-
physical Research, 80, 3032–3043.
Lecroart, P., Maire, O., Schmidt, S., Gremare, A., Anschutz, P., and
Meysman, F. J. R., 2010. Bioturbation, short-lived radioisotopes,
and the tracer-dependence of biodiffusion coefficients.
Geochimica et Cosmochimica Acta, 74, 6049–6063.
Maire, O., Duchene, J.-C., Gremare, A., Malyuga, V. S., and
Meysman, F. J. R., 2007. A comparison of sediment reworking
rates by the surface deposit-feeder bivalve Abra ovate during
summertime and wintertime, with a comparison between two
models of sediment reworking. Journal of Experimental Marine
Biology and Ecology, 343, 21–36.
Maire, O., Lecroat, P., Meysman, F., Rosenberg, R., Duchene, J. C.,
and Gremare, A., 2008. Quantification of sediment reworking rates
in bioturbation research: a review. Aquatic Biology, 2, 219–238.
Meysman, F. J. R., Boudreau, B. P., and Middleburg, J. J., 2003. Rela-
tions between local, nonlocal, discrete and continuous models of
bioturbation. Journal of Marine Research, 61, 391–410.
Nittrouer, C. A., and Sternberg, R. W., 1981. The formation of
sedimentary strata in an allochthonous shelf environment: the
Washington continental shelf. Marine Geology, 31, 297–316.
Officer, C. B., and Lynch, D. R., 1989. Bioturbation, sedimentation
and sediment-water exchanges. Estuarine, Coastal and Shelf
Science, 28, 1–12.
Reed, D. C., Doudreau, B. P., and Huang, K., 2007. Transient tracer
dynamics in a lattice-automation model of bioturbation. Journal
of Marine Research, 65, 813–833.
Robbins, J. A., McCall, P. L., Fisher, J. B., and Krezoski, J. R.,
1979. Effect of deposit feeders on migration of 137Cs in lake sed-
iments. Earth and Planetary Science Letters, 42, 277–287.
Sanford, L. P., 1992. New sedimentation, resuspension, and burial.
Limnology and Oceanography, 37, 1164–1178.
Schink, D. R., and Guinasso, N. L., 1977. Effects of Bioturbation on
sediment-seawater interaction. Marine Geology, 23, 133–154.
Smoak, J. M., and Patchineelam, S. R., 1999. Sediment mixing and
accumulation in a mangrove ecosystem: evidence from 210Pb,
234
Th and 7Be. Mangroves and Salt Marshes, 3, 17–27.
Wheatcroft, R. A., Jumard, P. A., Smith, C. R., and Nowell,
A. R. M., 1990. A mechanistic view of the particulate
biodiffusion coefficient: step lengths, rest period and transport
directions. Journal of Marine Research, 48, 177–207.
Yingst, J. Y., and Rhoads, D. C., 1980. The role of bioturbation in
the enhancement of microbial turnover rates in marine sediment.
In Tenore, K. R., and Coul, B. C. (eds.), Marine Benthic Dynam-
ics. Columbia: University of South Carolina Press, pp. 407–422.
Cross-references
210
Pb Dating
U-Series Dating
SEDIMENT, ESR
Helene Tissoux
DGR/GAT, Bureau de Recherches Géologiques et
Minières (BRGM), Orléans, France
Synonyms
ESR dating of sediment; ESR dating of sedimentary
quartz
743
Introduction
As quartz is the most common mineral forming the rocks
of the earth surface, its dating is of potential value. Elec-
tron spin resonance (ESR) dating of sedimentary quartz
has many applications in geology and archaeology. The
first studies on ESR dating of sedimentary quartz were
done by McMorris in the early 1970s and then by
Yokoyama et al. (1985) and Tanaka et al. (1985) and sum-
marized by and Ikeya (1993). During the past two
decades, our knowledge about the ESR dating of sedimen-
tary quartz has improved, and this method can now be rou-
tinely used for dating alluvial sedimentary quartz deposits.
Paramagnetic centers used for dating of quartz
The silicon atom, Si, is inside four oxygen atoms forming
a tetrahedral SiO4. In case of silica, or quartz (SiO2), each
oxygen atom is bonded with two Si atoms to form a three-
dimensional network in a crystalline form. Impurity-
associated defects gives ESR signals associated with
impurities such as substitutional tetravalent ions (Ge4+
and Ti4+) and trivalent ion (A13+) as well as interstitial
monovalent cations, M+ (H+, U+, and Na+) (see Figure 1;
Weil, 1984; Ikeya, 1993). These paramagnetic centers,
radiosensitive and with a high thermal stability, can be
used as dosimeters, though the use of the Ge4+ center
remains unclear (see below).
Formation of Al paramagnetic centers
In the SiO2 lattice, trivalent aluminum ion A13+ is easily
substituted at a Si4+ site. The principle of ionic charge neu-
trality in an ionic crystal requires a charge-compensating
monovalent cation interstitial. In this case, a monovalent
cation M+ such as Li+, Na+, and H+ at the interstitial site
is associated with A13+ as a charge compensator and gives
rise to a [AlO4/M+]
center. When the charge-
compensated [AlO4/M+] traps a hole created by ionizing
radiation at room temperature on one of the two long-
bonded oxygen neighbors, the cation M+ diffuses away
leading to [A1O4/h]
, where h is a hole trapped in
a nonbonding 2p-orbital of oxygen located adjacent to
the substitutional aluminum (O’Brien, 1955; Hitt and
Martin, 1983 in Ikeya, 1993).
Formation of Ti and Ge paramagnetic centers
Ge4+ or Ti4+ substitutes for a Si4+ site in the SiO2 lattice,
but its electron affinity (ionization potential) is consider-
ably larger, resulting in trapping of an electron created
by ionizing irradiation to form [GeO4/TiO4-e] in the
SiO2 lattice. The total charge of the center is 1, and thus
Ge3+ or Ti3+ at a Si4+ site in an SiO2 structure has
a negative effective charge, attracting an interstitial mono-
valent cation M+ such as H+, Li+, or Na+, to form stable
[GeO4/M+]
or [TiO4/M+]
at room temperature
(Andersen et al., 1974 in Ikeya, 1993).
744 SEDIMENT, ESR
Sediment, ESR, Figure 1 Impurity-associated defects present in the SiO2 lattice.
Paramagnetic center features
Al center is commonly employed for dating as it is
a radiosensitive center with a high saturation dose and it
is one of the most thermally stable radiation-induced cen-
ters in SiO2. The ESR signal related to the aluminum can
be detected in all natural quartz and used for dating sedi-
mentary of quartz grains deposited during all the Plio-
Pleistocene period (McMorris, 1971; Yokoyama et al.,
1985; Toyoda and Ikeya, 1991; Falguères et al., 1991;
Laurent et al., 1998; Toyoda and Falguères, 2003; Lin
et al., 2006).
Ti–Li and Ti–H center can be detected in almost all
sedimentary quartz. The saturation of the Ti–H center is
rapidly reached (Beerten et al., 2006; Tissoux et al.,
2007, 2008). This implies that Ti–H center can be used
only for the dating of samples with low palaeodoses
(Upper Pleistocene deposits, or older in case of very low
annual dose). Ti–Li center could be used for dating sedi-
ments of Lower Pleistocene age (Rink et al., 2007; Liu
et al., 2010).
It is possible that the Ge center could be employed over
the same range of ages as the OSL dating method. How-
ever, the ESR Ge signal is rarely observed in the natural
samples, although growing with small artificial irradiation
doses, suggesting that it is not stable over geological time
scales (Rink, 1997).
The Ge center can be detected by ESR at room temper-
ature, whereas Al and Ti center signals are observed at low
temperature (Figure 2).
Bleaching
When dating sedimentary quartz, the geological or archae-
ological events have to be dated, not the quartz growth.
Thus, a zeroing of the quartz geochronometer is necessary
before the time of deposition of the sedimentary sequence.
This zeroing is also named “bleaching.” The bleaching
occurs when the energy brought to the quartz grain is suffi-
cient to release the electron trapped in the defects
(activation energy). This energy can be brought by pres-
sure, light exposure, or heat, and the exposure has to be long
enough for a maximal release of the trapped electron. Oth-
erwise, a residual dose is inherited from the primary history
of the quartz, leading to an overestimation of the dose
acquired during the depositional history of the quartz grain.
Regarding marine, alluvial, or aeolian sediments, an
optical bleaching by sunlight of the light-sensitive para-
magnetic centers of quartz occurs during the transporta-
tion of the quartz grain by water or by wind, before
deposition in the sedimentary sequence (Figure 3).
Many authors investigated the behavior of some para-
magnetic centers of quartz when they are exposed to natu-
ral of artificial light (Brumby and Yoshida, 1994; Walther
and Zilles, 1994; Yoshida, 1996; Tanaka et al., 1997;
Toyoda et al., 2000; Voinchet et al., 2003; Tissoux
et al., 2007, 2012).
They observed that Al centers of quartz cannot be
totally bleached by exposure to the sunlight. An ESR
residual dose, related to Al centers, is observed in quartz
after a maximal optical bleaching and is generated by the
presence of deep aluminum traps in the crystal, which can-
not be reset by solar light. This residual dose must be
determined for all the sedimentary quartz when ESR geo-
chronology studies are performed. Bleaching experiments
indicated that several weeks of exposure were necessary to
obtain a maximal bleaching of the Al center (plateau
value); the Ge center can be totally bleached in 3 h of
exposure by simulated solar light; the Ti–Li center can
be totally bleached after a time of exposure similar to the
 SEDIMENT, ESR 745

Sediment, ESR, Figure 2 Al and Ti center measured samples measured with a JEOL PX-2300 ESR spectrometer. Temperature 80 K;
microwave power 5 mW; width of the magnetic field 10 mT; modulation amplitude 0.1 mT; time constant 0.03 s.; frequency 100 kHz
(After Tissoux et al., 2008, modified).

Sediment, ESR, Figure 3 Optical bleaching history of Ti and Al centers in sedimentary quartz.
746 SEDIMENT, ESR
one of Al center, whereas Ti–H center needs less than half
the time to be totally bleached.
ESR dating of optically bleached fluvial quartz
deposits
The ESR dating of alluvial quartz is based on the hypoth-
esis of ESR trap resetting by sunlight during the transport
phase in the river prior to the sediment burial. The calcu-
lated age does not correspond to the quartz crystallization
but to the time elapsed since the quartz deposition in the
alluvial formation.
The exposure time necessary for a maximal bleaching
during transport by the river seems to be a very long expo-
sure in natural conditions. However, studies of modern sed-
iments deposited in sandy bars show that the maximum
bleaching is effectively reached for Al center at few kilome-
ters from the river’s spring (Voinchet et al., 2007). Liu and
Grün (2011) conclude that tumbling due to fluvial transport
may take a part in the resetting of the ESR centers. ESR dat-
ing of fluvial deposits has been then successfully applied
using Al center of quartz (Yokoyama et al., 1985; Laurent
et al., 1994; Voinchet et al., 2003, 2004). “Absolute” chro-
nologies of alluvial systems could then be established with
strong implications in archaeology (Bahain et al., 2007;
Voinchet et al., 2010; Despriée et al., 2011).
ESR dating of optically bleached aeolian quartz
deposits
Signal resetting by sunlight prior to burial is a crucial
assumption in electron spin resonance (ESR) dating of
sediments. This resetting process was successfully tested
for Ti–Li and Ti–H signals of quartz from
a contemporary desert surface deposit, demonstrating the
potential of using the Ti–Li and the Ti–H centers in quartz
for dating fossil sediments from similar environments
(Beerten and Stesmans, 2005). On the contrary, optical
bleaching experiments showed that sunlight does not
completely bleach the ESR Al signal of quartz grains in
aeolian sand from Australia (Tanaka et al., 1995) and in
loess of northern Europe (Tissoux et al., 2010).
ESR dating of optically bleached marine quartz
deposits
Tanaka et al. (1997) and Tissoux et al. (2008) applied ESR
dating on known-age marine terrace sediments collected
in Japan using Ti signal and Ti and Al signals, respec-
tively. In both cases, it was apparent that the signal in the
samples studied here had not received complete
pre-burial bleaching. On the contrary, ESR dating of
quartz from fossil beach in Nice, France (Falguères
et al., 1988), and from fossil estuarine in the Somme Val-
ley, France (Laurent et al., 1998), provided data in agree-
ment with the expected ages. Burdette et al. (2013)
showed that using the Ti–Li signal from sands from
stranded ancient coastal deposits in Florida were in
concordance with increasing geomorphic age and had
ages older than OSL saturation indicated. Moreover, bio-
stratigraphic age constraints in one sample were consistent
with the ESR optical ages.
ESR dating of glacial deposits
Together with sunlight bleaching, such grinding mecha-
nisms could be an effective zeroing mechanism in ESR
dating of glacially derived sediments. As such, the
Ge-related impurity center in quartz was applied for the
dating of Pleistocene glaciations in Central Asia (Zhou
et al., 2002; Yi et al., 2002; Zhao et al., 2006, 2010).
However, some fundamental points remain unclear as
mentioned here above.
Summary
ESR applied on sedimentary quartz is a powerful dating
method when applied on quartz extracted from fluvial,
coastal, or desert or sediments deposited during the Plio-
Pleistocene period.
Bibliography
Bahain, J.-J., Falguères, C., Laurent, M., Voinchet, P., Dolo, J.-M.,
Antoine, P., and Tuffreau, A., 2007. ESR chronology of the
Somme River Terrace system and first human settlements in
Northern France. Quaternary Geochronology, 2, 356–362.
Beerten, K., and Stesmans, A., 2005. Single quartz grain ESR dat-
ing of a contemporary desert surface deposit, Eastern Desert,
Egypt. Quaternary Science Reviews, 24, 223–231.
Beerten, K., Lomax, J., Clémer, K., Stesmans, A., and Radtke, U.,
2006. On the use of Ti centres for estimating burial ages of Pleis-
tocene sedimentary quartz: multiple grain data from Australia.
Quaternary Geochronology, 1, 151–158.
Brumby, S., and Yoshida, H., 1994. An investigation of the effect of
sunlight on the ESR spectra of quartz centres: implications for
dating. Quaternary Science Reviews, 13, 615–618.
Burdette, K., Rink, W. J., Mallinson, D., Means, G., and Parham, P.,
2013. Electron spin resonance optical dating of marine, estua-
rine, and aeolian sediments in Florida, USA. Quaternary
Research, 79, 66–74.
Despriée, J., Voinchet, P., Tissoux, H., Bahain, J. J., Falguères, C.,
Courcimault, G., Dépont, J., Moncel, M.-H., Robin, S.,
Arzarello, M., Sala, R., Marquer, L., Messager, E., Puaud, S.,
and Abdessadok, S., 2011. Lower and Middle Pleistocene
human settlements recorded in fluvial deposits of the middle
Loire River Basin, Centre Region, France. Quaternary Science
Reviews, 30(11–12), 1474–1485.
Falguères, C., Yokoyama, Y., and Quaegebeur, J. P., 1988.
Datations de sédiments quaternaires par la méthode de résonance
de spin électronique (ESR). L’Anthropologie, 92, 723–726.
Falguères, C., Yokoyama, Y., and Miallier, D., 1991. Stability of
some centres in quartz. Nuclear Tracks and Radiation Measure-
ments, 18, 155–161.
Ikeya, M., 1993. New Applications of Electron Spin Resonance:
Dating, Dosimetry and Microscopy. Singapore: World Scien-
tific, 500 pp.
Laurent, M., Falguères, C., Bahain, J.-J., and Yokoyama, Y., 1994.
Géochronologie du système de terrasses fluvialtiles quaternairs
du bassin de la Somme par datation RPE sur quartz,
 SEDIMENT, ESR
déséquilibres des familles de l’uanium et magnétostratigrapie.
Comptes-Rendus de l’Académie des Sciences, II, 318, 521–526.
Laurent, M., Falguères, C., Bahain, J. J., Rousseau, L., and Van
Vliet Lanoe, B., 1998. ESR dating of quartz extracted from Qua-
ternary and Neogene sediments: method, potential and actual
limits. Quaternary Science Reviews, 17, 1057–1061.
Lin, M., Yin, G.-M., Ding, Y., Cui, Y., Chen, K., Wu, C., and Xu, L.,
2006. Reliability study on ESR dating of the aluminum center in
quartz. Radiation Measurements, 41, 1045–1049.
Liu, C. R., and Grün, R., 2011. Fluvio-mechanical resetting of the Al
and Ti centres in quartz. Radiation Measurements, 46, 1038–1042.
Liu, C. R., Yin, G.-M., Gao, L., Bahain, J.-J., Li, J. P., Lin, M., and
Chen, S. M., 2010. ESR dating of Pleistocene archaeological
localities of the Nihewan Basin, North China-Preliminary
results. Quaternary Geochronology, 5, 385–390.
McMorris, D. W., 1971. Impurity color centers in quartz and trapped
electron dating Electron Spin Resonance, thermoluminescence
studies. Journal of Geophysical Research, 76, 7875–7887.
Rink, W. J., 1997. Electron Spin Resonance (ESR) Dating and ESR
Applications in Quaternary Science and Archaeometry. Radia-
tion Measurements, 27, 975–1025.
Rink, W. J., Bartoll, J., Schwarcz, H. P., Shane, P., and Ber-Yosef, O.,
2007. Testing the reliability of ESR dating of optically exposed
buried quartz sediments. Radiation Measurements, 4, 1618–1626.
Tanaka, T., Sawada, S., and Ito, T., 1985. ESR dating of Late Pleis-
tocene near-shore and terrace sands in Southern Kanto, Japan. In
Ikeya, M., and Miki, T. (eds.), ESR Dating and Dosimetry.
Tokyo: Ionics, pp. 275–280.
Tanaka, K., Machette, M. N., Crone, A. J., and Bowman, J. R.,
1995. ESR dating of aeolian sand near Tennant Creek, Northern
Territory, Australia. Quaternary Science Reviews, 14, 285–293.
Tanaka, K., Hataya, R., Spooner, N. A., Questiaux, D. G., Saito, Y.,
and Hashimoto, T., 1997. Dating of marine terrace sediments by
ESR, TL and OSL methods and their applicabilities. Quaternary
Science Reviews, 16, 257–264.
Tissoux, H., Falguères, C., Voinchet, P., Toyoda, S., Bahain, J. J., and
Despriée, J., 2007. Potential use of Ti-center in ESR dating of flu-
vial sediment. Quaternary Geochronology, 2(1–4), 367–372.
Tissoux, H., Toyoda, S., Falguères, C., Voinchet, P., Takada, M.,
Bahain, J. J., and Despriée, J., 2008. ESR dating of quartz from
two different Pleistocene deposits using Al and Ti-centers.
Geochronometria, 30, 23–31.
Tissoux, H., Valladas, H., Voinchet, P., Reyss, J. L., Mercier, N.,
Falguères, C., Bahain, J. J., Zöller, L., Antoine, P., and Rousseau,
D. D., 2010. OSL and ESR studies of aeolian quartz sediments
from the Upper Pleistocene loess sequence of Nussloch
(Germany). Quaternary Geochronology, 5(2–3), 131–136.
Tissoux, H., Voinchet, P., Lacquement, F., Prognon, F., Moreno, D.,
Falguères, C., Bahain, J. J., and Toyoda, S., 2012. Investigation
on non-optically bleachable components of ESR aluminium sig-
nal in quartz. Radiation Measurements, 47(9), 894–899.
Toyoda, S. 1992. Production and Decay Characteristics of Para-
magnetic Defects in Quartz: Applications to ESR Dating. PhD
dissertation, Toyonaka, Osaka, Osaka University.
Toyoda, S., and Falguères, C., 2003. The method to represent the
ESR intensity of the aluminium hole center in quartz for the pur-
pose of dating. Advances in ESR Applications, 20, 7–10.
Toyoda, S., and Ikeya, M., 1991. Thermal stabilities of paramag-
netic defect and impurity centres in quartz: basis for ESR dating
of thermal history. Geochemistry Journal, 25, 437–445.
Toyoda, S., Goff, F., Ikeda, S., and Ikeya, M., 1995. ESR dating of
quartz phenocrysts in the El Cajete and Battleship Rock Mem-
bers of Valles Rhyolite, Valles Caldera, New Mexico. Journal
of Volcanology and Geothermal Research, 67, 29–40.
747
Toyoda, S., Voinchet, P., Falguères, C., Dolo, J. M., and Laurent,
M., 2000. Bleaching of ESR signal by the sunlight:
a laboratory experiment for establishing the ESR dating of sedi-
ments. Applied Radiation and Isotopes, 52, 1357–1362.
Voinchet, P., Falguères, C., Laurent, M., Toyoda, S., Bahain, J. J.,
and Dolo, J. M., 2003. Artificial optical bleaching of the alumin-
ium centre in quartz implications to ESR dating of sediments.
Quaternary Science Reviews, 22, 1335–1338.
Voinchet, P., Bahain, J. J., Falguères, C., Laurent, M., Dolo, J. M.,
Despriée, J., Gageonnet, R., and Chaussé, C., 2004. ESR dating
of quartz extracted from Quaternary sediments: application to
fluvial terraces systems of Northern France. Quaternaire, 15,
135–141.
Voinchet, P., Falguères, C., Tissoux, H., Bahain, J.-J., Despriée, J.,
and Pirouelle, F., 2007. ESR dating of fluvial quartz: estimate
of the minimal distance transport required for getting
a maximum optical bleaching. Quaternary Geochronology, 2,
363–366.
Voinchet, P., Despriée, J., Tissoux, H., Falguères, C., Bahain, J. J.,
Gageonnet, R., and Dolo, J.-M., 2010. ESR chronology of allu-
vial deposits and first human settlements of the middle Loire
basin (Region Centre, France). Quaternary Geochronology, 5,
381–384.
Walther, R., and Zilles, D., 1994. ESR studies on bleached sedimen-
tary quartz. Quaternary Geochronology, 13, 611–614.
Weil, J. A., 1984. A review of electron spin spectroscopy and its
application to the study of paramagnetic defect in crystalline
quartz. Physics and Chemistry of Minerals, 10, 149–165.
Yi, C. L., Li, X. Z., and Qu, J. J., 2002. Quaternary glaciation of
Puruogangri the largest modern ice field in Tibet. Quaternary
International, 97/98, 111–121.
Yin, G., Lin, M., Yanchun, L., Jianping, L., and Fei, H., 2007. Pre-
liminary ESR dating results on loess samples from the loess–
paleosol sequence at Luochuan, Central Loess Plateau, China.
Quaternary Geochronology, 2(1–4), 381–385.
Yokoyama, Y., Falguères, C., and Quaegebeur, J. P., 1985. ESR dat-
ing of quartz from quaternary sediments: first attempt. Nuclear
Tracks and Radiation Measurements, 10, 921–928.
Yoshida, H., 1996. Quaternary Dating Studies Using ESR Signals,
with Emphasis on Shell, Coral, Tooth Enamel and Quartz. PhD
thesis, Canberra, Australian National University.
Zhao, J. D., Zhou, S. Z., He, Y. Q., Ye, Y. G., and Liu, S. Y., 2006.
ESR dating of glacial tills and glaciations in the Urumqi River
headwaters, Tianshan Mountains, China. Quaternary Interna-
tional, 144, 61–67.
Zhao, J. D., Song, Y. G., King, J. W., Liu, S. Y., Wang, J., and Wu,
M., 2010. Glacial geomorphology and glacial history of the
Muzart River valley, Tianshan range, China. Quaternary Science
Reviews, 29, 1453–1463.
Zhou, S. Z., Jiao, K. Q., Zhao, J. D., Zhang, S. Q., Cui, J. X., and
Xu, L. B., 2002. The geomorphology of Urümqi River valley
and the uplift of Tianshan mountain in the Quaternary. Science
in China (Series D), 45, 961–968.
Cross-references
Electron Spin Resonance (ESR) Dating of Coral
Electron Spin Resonance (ESR) Dating of Fossil Tooth Enamel
Electron Spin Resonance (ESR) Dating, General Principles
Electron Spin Resonance Spectrometer
Quartz
Radiation and Radioactivity
Radiation Defect
Radiation Dose Rate
748 SEDIMENTARY ROCKS (Rb-Sr GEOCHRONOLOGY)
SEDIMENTARY ROCKS (Rb-Sr GEOCHRONOLOGY)
Tod Waight
Department of Geosciences and Natural Resource
Management, Geology Section, University of
Copenhagen, Copenhagen, Denmark
Synonyms
Clay dating; Glauconite dating; Rb-Sr dating; Shale chro-
nology; Sr dating; Sr geochronology
Definition
Application of the radioactive decay of 87Rb to 87Sr to
determining the age of deposition of sedimentary rocks.
Characteristics
Absolute dating of sedimentary rocks is of interest to
obtain depositional ages and to provide a quantitative link
between the relative ages provided by fossil successions
(biostratigraphy) and the geological time scale, and
improve our understanding of the evolution of life through
the fossil record. However, application of the Rb-Sr sys-
tem to dating of sedimentary rocks has had variable results
and has often proved to be problematic.
The Rb-Sr geochronological system is based on the
beta decay of 87Rb to 87Sr with a half-life of 48.8 Ga,
corresponding to a decay constant (l) of 1.42  1011
a1, making the decay system potentially able to provide
absolute age information for geological events throughout
most of the Earth’s history. The general decay formula for
the Sr isotope system is
87 87 87
Sr Sr Rb lt
86 Sr
¼ 86 Sr
þ 86 Sr
e 1
ð0Þ ðiÞ
where the subscripts (0) and (i) refer to the measured pre-
sent-day and initial Sr isotope composition, respectively.
Age determination with the Rb-Sr system typically
involves construction of an isochron using minerals or
rocks with as wide a range in Rb-Sr ratios as possible.
Rb and Sr have chemical characteristics similar to K and
Ca, respectively; therefore, Rb will tend to be concen-
trated in K-rich minerals and Sr will be concentrated in
Ca-rich minerals. Examples of minerals in sedimentary
systems with high-K contents and therefore high Rb/Sr
are micas, alkali feldspar, and various clay minerals,
whereas minerals such as carbonates and phosphates
(e.g., calcite and apatite) are Ca-rich and therefore have
low Rb/Sr. With time the 87Rb in the minerals will decay
to 87Sr, such that high Rb/Sr minerals will become pro-
gressively more radiogenic (higher 87Sr/86Sr) with time,
whereas the amount of ingrowth of 87Sr in the Rb-poor
minerals will be lower or, in the case of Rb-free minerals,
absent. The measured 87Sr/86Sr of the different minerals or
rocks phases will then define a line, an isochron, with
a slope that is proportional to the age of the rock. The inter-
section of the isochron with the Y-axis, or alternatively
analysis of a Rb-free phase or an assumption of an initial
isotopic composition, provides the initial Sr isotopic com-
position of the system 87Sr/86Sr(i).
The age of the rock can then be calculated using the
equation:
1
t ¼ lnð1 þ slopeÞ
l
In order to obtain meaningful age information from an
isochron, three fundamental assumptions must be fulfilled:
(a) The minerals/rocks must have all formed in equilib-
rium with each other.
(b) The minerals/rocks must have all formed from
a reservoir of homogeneous isotopic composition
(87Sr/86Sr(i)).
(c) The minerals/rocks have been a closed system (i.e.,
not been subject to thermal or chemical disturbance)
subsequent to formation.
There are three main groups of sedimentary rocks: clas-
tic (transported), carbonate (biological), and chemical.
Chemical sediments form by precipitation of minerals
out of water (e.g., evaporates, carbonates, phosphates,
and travertine in caves). Biological carbonate sediments
(peat and coal are also biological) are formed from the
accumulation of skeletal remains of organisms. The min-
erals formed in chemical sediments and in shells and skel-
etons are normally calcium-rich (e.g., calcite, apatite) and
show a limited range in Rb/Sr. Thus, they are not suitable
for Rb-Sr geochronology, although some can be dated
using the U-series dating method, and will not be
discussed further here. Siliciclastic sediments such as
sandstones and mudstones form at relatively low tempera-
tures at the Earth’s surface as a consequence of erosion,
transportation, and deposition of fragments and minerals
from preexisting rocks dominated by silicate minerals.
These sediments then undergo burial and diagenesis to
become cemented into sedimentary rocks. Minerals that
are unstable at surficial pressures and temperatures will
tend to progressively break down during sedimentary pro-
cesses to form new mineral phases, for example, feldspars
will undergo hydrolysis to form clays (e.g., illite, smectite,
kaolinite), many of which are K-rich and thus theoretically
provide potential for dating using Rb-Sr.
The main difficulties encountered when applying the
Rb-Sr system to dating of sedimentary rocks reflect uncer-
tainties in the assumption that the sedimentary compo-
nents had homogeneous isotopic compositions at the
time of deposition and that the system has remained closed
since. Most sandstones and clays consist of some mix of
detrital phases and new authigenic minerals. Detrital min-
erals are derived from the eroding source region and have
been transported unchanged or partially modified, and
these grains will retain a Sr isotope signature reflecting
the amount of ingrowth of 87Sr from the decay of 87Rb
since their source region cooled through the closure tem-
perature for the mineral in question. As the source regions
 SEDIMENTARY ROCKS (Rb-Sr GEOCHRONOLOGY)
for any particular region can comprise many different rock
types of different ages and origins, this detrital component
is potentially heterogeneous. Authigenic minerals grow
during weathering, transport, or deposition and may there-
fore be in isotopic equilibrium with their transport
medium (e.g., water). Because sedimentary processes take
place at low temperatures, significantly below the closure
temperatures for Sr isotope diffusion in the constituent
minerals, the isotopic systematics are not reset to equilib-
rium. Therefore, the fundamental assumption that all min-
erals in a sediment at the time of deposition or early
diagenesis formed in equilibrium with each other and from
a reservoir of homogenous isotopic composition
(87Sr/86Sr(i)) is unlikely to hold true.
Rb-Sr dating of clays
To avoid these complications, workers trying to date
siliciclastic sediments have focused on fine-grained
authigenic minerals such as clays that grow during deposi-
tion and early diagenesis and are so fine grained that they
can equilibrate with the surrounding water. The assumption
is then that these phases form in isotopic equilibrium and
thus fulfill the criteria for creating an isochron. Subsequent
variations in Sr isotopic composition in various clay fractions
reflect their variations in Rb/Sr and ingrowth with time and
can theoretically be used to construct an isochron. Two point
isochrons may also be constructed if constraints on the initial
Sr isotopic composition of the sediment can be made using
the Sr isotope evolution curve for seawater, i.e. if an approx-
imate age of the sediment is known beforehand or from ana-
lyses of interbedded carbonate or calcareous fossils.
Complications in the use of fine-grained clays for Rb-
Sr geochronology exist because detrital components
(including some clays) may be present even in the finest
grain size fraction of sediments, and not all clays
Sedimentary Rocks (Rb-Sr Geochronology), Figure 1 Rb-Sr isochron for a representative clay sample from the Lontova claystones
studied by Gorokhov et al. (1994) (sample 77NC80 < 0.1 mm fraction). Clays were leached with 1 N ammonium acetate, and a bulk
sample, leachate, and residue were analyzed.
749
equilibrate with the surrounding water. Therefore, detailed
X-ray diffraction studies of the clays are necessary to dis-
criminate between detrital components and authigenic
clays that have grown during diagenesis and low-
temperature metamorphism. Furthermore, the ability of
the clay minerals to equilibrate with the surrounding water
also makes them susceptible to later disturbances by post-
diagenetic interaction with pore waters which violates the
remaining fundamental premise of isochron geochronol-
ogy; that the minerals have remained a closed system since
formation. It has also been shown that clay minerals can
contain at least two different reservoirs of Sr with different
origins and isotopic compositions, a loosely bound reser-
voir that can be released by leaching in acid or ammonium
acetate and that represents the environment during deposi-
tion and clay crystallization, and a reservoir that is part of
the crystal structure (e.g., Clauer et al., 1982).
The care and level of characterization that is required,
and the complications involved, in obtaining geologically
meaningful age data from clays in sediments are well illus-
trated in the study of Gorokhov et al. (1994). They separated
the <2 mm clay size fraction from a series of sediment sam-
ples from the Cambrian–Precambrian boundary in northern
Estonia and Russia and then further separated these
according to grain size. The samples were characterized
by XRD to determine their illite crystallinity index by inves-
tigating the width of the (001) peak at half its height. A low
index indicates a well-defined sharp peak and is indicative
of well-crystallized illite which formed under relatively
high-temperature conditions, whereas a high index indi-
cates poorly ordered, less regular illite formed at lower tem-
peratures (Dickin, 2005). The clay fractions were then
leached to produce 3-point isochrons (leachate, residue,
and bulk) for each clay size fraction (Figure 1).
The results from all clay samples and size fractions
yielded a large scatter of ages but more importantly also
750 SEDIMENTARY ROCKS (Rb-Sr GEOCHRONOLOGY)
Sedimentary Rocks (Rb-Sr Geochronology), Figure 2 Compilation of apparent Rb-Sr ages for clay fractions from the Lontova
claystones (see Figure 1) against size of the clay fraction analyzed, showing a clear positive correlation between grain size and age
(From Gorokhov et al. (1994)).
revealed a trend of increasing age with increasing grain
size (Figure 2). In addition, the coarser-grained fractions
were also characterized by XRD signatures that suggest
crystallization at higher temperature, and thus,
a significant detrital component was considered to be pre-
sent in these fractions, subsequently resulting in older
ages. Most of the other coarser-grained clay fractions were
considered to be mixtures between detrital and authigenic
components, and thus, the isochrons do not represent geo-
logically meaningful ages but are instead mixing arrays.
Nevertheless, when the data are combined in terms of
grain size and crystallinity, some potentially meaningful
age information can be obtained. The well-crystallized
fine-grained (<0.2 mm) clays define an isochron with an
age of 533  8 Ma which Gorokhov et al. (1994) estimate
to be the best estimate of the age of sedimentation,
whereas fine-grained disordered clays yielded a younger
age (430–480 Ma) that may be related to later disturbance
and interactions with subsurface waters.
Rb-Sr dating of glauconite
A more promising approach to dating sediments using
both the Rb-Sr and K-Ar systems has been to analyze
authigenic glauconite, a distinctive green to black K-Fe
mica that is a member of the glaucony family of minerals
and forms at the sediment–water interface in shallow
marine environments characterized by low energy and
low rate of deposition. Sediments with high abundances
of glauconite are commonly called greensands due to the
characteristic color of the mineral. Glauconite originates
as a result of biological activity such as ingestion and
excretion of sediments. These fecal pellets, composed pri-
marily of clay as well as potential detrital and organic
material, become progressively dissolved and
recrystallized on the seafloor and during burial in
a process termed glauconitization. In order for glauconite
to be of use in either Rb-Sr or K-Ar geochronology, it must
be first assumed that it formed in equilibrium with
a homogeneous reservoir (e.g., seawater) and has
remained a closed system since. Detailed studies on mod-
ern glauconite have provided important insights into the
glauconization process and show that it occurs as the
result of a relatively slow, two-stage process (Clauer
et al., 1992; Stille and Clauer, 1994). Clauer et al. (1992)
investigated modern glauconites from the Guinea and Cal-
ifornia coasts. The investigated samples were leached in
acid prior to analysis to remove contaminants such as
detrital clays and silicates, and carbonates. Sr isotope ana-
lyses of these leachates are similar to modern seawater and
indicate that the source of this Sr is mostly marine. The
isotopic compositions of the leached glauconites were
then used to create 2-point isochrons, using the composi-
tion of modern seawater (87Sr/86Sr ¼ 0.7092) as the initial
composition. If there is no difference between the glauco-
nite and seawater (i.e., an age of 0 Ma), then the glauconite
is in equilibrium with seawater and could thus with burial
and time yield meaningful geochronological information
about the age of deposition. Strontium isotopic composi-
tions of leached glauconite that are more radiogenic than
seawater will result in positive present-day “ages” and
suggest the presence of Sr in the glauconite inherited from
detrital sources. These ages are of course geologically
meaningless as the samples are effectively zero age. With
burial and time, these rocks would develop isotopic com-
positions that would yield meaningless ages older than the
age of deposition. Clauer et al. (1992) discovered a wide
variation in apparent ages for the investigated glauconites
and that the apparent age decreased with increasing
K content, reaching a zero age at K contents in excess of
7 wt% (Figure 3).
Clauer et al. (1992) therefore proposed that the initial
stages of glauconite formation primarily involve recrystal-
lization in equilibrium with progenitor materials and sur-
rounding clays and are thus dominated by Sr inherited
from the sedimentary source region. Isotopic analysis of
 SEDIMENTARY ROCKS (Rb-Sr GEOCHRONOLOGY) 751

Sedimentary Rocks (Rb-Sr Geochronology), Figure 3 Compilation of model Rb-Sr ages calculated assuming initial
87
Sr/86Sr ¼ modern seawater against K2O content for glauconites (circles) and coexisting clays (triangles) from modern sediment
samples. The variation indicates that glauconites with <7 wt% K2O contain a significant component of Sr derived from detrital
sources (From Clauer et al. (1992)).

Sedimentary Rocks (Rb-Sr Geochronology), Figure 4 Rb-Sr isochron for glauconites (circles) separated from sandstones and
siltstones from southeastern France. K-Ar analyses on the same material yield an identical isochron age of 97.3  0.4 Ma, and the age
is interpreted as representing the time of deposition (From Rousset et al. (2004)).

these immature glauconites will not yield geologically
meaningful ages. Later stages of glauconite growth are
characterized by increasing K contents (>4 %) and
involve chemical exchange with surrounding seawater.
At equilibrium, this effectively resets the system to
a homogeneous initial isotopic composition that is the
same as that of the surrounding seawater. The glauconite
then fulfills the first assumption of Rb-Sr geochronology
as it formed from a homogeneous reservoir. With burial
and isolation, it will begin to accumulate 87Sr and thus
potentially preserve the age of deposition and early dia-
genesis if there is sufficient spread in the Rb/Sr ratios of
various glauconite fractions or when an assumed seawater
composition or Rb-free phase such as apatite or calcite is
used to constrain the initial composition. Careful chemical
and mineralogical characterization of glauconite is there-
fore critical to ascertaining the validity of glauconite
Rb-Sr ages.
An example of successful Rb-Sr glauconite dating is
given by Rousset et al. (2004). Several carefully sepa-
rated, cleaned, and well-characterized fractions of glauco-
nite from Cretaceous sediments in France were
investigated. The glauconites have K2O ranging from
6.1 to 7.4 wt% suggesting they are mature and well equil-
ibrated based on the studies of Clauer et al. (1992)
described above and yield enough variation in Rb/Sr to
define an isochron with an age of 97.9  3.5 Ma (Figure 4).
The initial Sr isotope ratio of 0.70736 calculated from the
752 SEDIMENTS, TERRESTRIAL (PALEOMAGNETISM)
isochron is in agreement with the composition of seawater
at that time and calcareous fossils from the same sedimen-
tary succession. A K-Ar age on the same material that
is identical within error confirms the validity of the
Rb-Sr age and indicates that no post-depositional alter-
ation or recrystallization affected the sediments as this
would likely have affected the two isotope systems
differently.
Conclusions
Dating the age of deposition or early diagenesis of sedi-
mentary rocks is highly problematic, but in certain
instances, meaningful depositional ages can be deter-
mined. The constituent minerals in most chemical and bio-
logical sediments do not have enough range in Rb/Sr to
produce measurable differences in 87Sr/86Sr with time.
Clastic rocks on the other hand comprise a mixture of
detrital grains with variable isotopic compositions
inherited from their source regions and authigenic min-
erals that grew during burial and therefore cannot be
assumed to have been in isotopic equilibrium at the time
of deposition. To avoid these problems, age information
can be derived from Rb-Sr isotopic analyses of authigenic
minerals such as clays and glauconite. However, such
studies are complex and require careful characterization
of the material prior to analysis in order to assess the
robustness and significance of the geochronological
results. In particular, for clay analysis, detailed mineralog-
ical analyses are required to identify different generations
of clays developed during various diagenetic and post-
diagenetic processes and to avoid detrital minerals that
may still be present in the fine-grained clay fraction. Stud-
ies of modern glauconites have shown that only mature
glauconites with high K2O contents appear to be in equi-
librium with the surrounding seawater and thus fulfill the
requirement of having formed from a homogeneous isoto-
pic source.
Bibliography
Clauer, N., Hoffert, M., and Karpoff, A.-M., 1982. The Rb-Sr iso-
tope system as an index of origin and diagenetic evolution of
southern pacific red clays. Geochimica et Cosmochimica Acta,
46, 2659–2664.
Clauer, N., Keppens, E., and Stille, P., 1992. Sr isotopic constraints
on the process of glauconitization. Geology, 20, 133–136.
Dickin, A. P., 2005. Radiogenic isotope geology, 2nd edn. Cam-
bridge: Cambridge University Press.
Garorkhov, I. M., Clauer, N., Turenko, T. L., Melnikov, N. N.,
Kutyavin, E. P., Pirrus, E., and Basakov, A. V., 1994. Rb-Sr sys-
tematics of Vendian-Cambrian claystones from the East Euro-
pean platform: implications for a multi-stage illite evolution.
Chemical Geology, 112, 71–89.
Rousset, D., Leclerc, S., Clauer, N., Lancelot, J., Cathelineau, M.,
and Aranyossy, J.-F., 2004. Age and origin of Albian
glauconites and associated clay minerals inferred from
a detailed geochemical analysis. Journal of Sedimentary
Research, 74, 631–642.
Stille, P., and Clauer, N., 1994. The process of glauconitization:
chemical and isotopic evidence. Contributions to Mineralogy
and Petrology, 117, 253–262.
Cross-references
Apatite
Ar–Ar and K–Ar Dating
Biostratigraphy
Carbonates, Marine Carbonates (U-Series)
Clays and Glauconites (K–Ar/Ar–Ar)
Feldspars
Minerals (40Ar–39Ar)
SEDIMENTS, TERRESTRIAL (PALEOMAGNETISM)
Wout Krijgsman1 and Gillian Turner2
1
Department of Earth Sciences, University of Utrecht,
Utrecht, The Netherlands
2
School of Chemical and Physical Sciences, Victoria
University of Wellington, Wellington, New Zealand
Synonyms
Magnetic polarity stratigraphy; Magnetostratigraphy
Definitions
Geomagnetic north/south poles: the points where the axis
of the geocentric magnetic dipole that best fits the geo-
magnetic field intersects the surface of the earth. The south
(north) pole of the best fitting dipole is actually in the
northern (southern) hemisphere during Normal polarity,
and is, by convention, called the north (south) geomag-
netic pole.
Virtual geomagnetic poles (VGP’s): the geomagnetic
poles corresponding to the geocentric dipole that would
produce an observed palaeomagnetic direction. VGP’s
are calculated from the declination and inclination of a
paleomagnetic measurement and the latitude and longi-
tude of the site from which the measurement was obtained.
Magnetostratigraphy: the application of the well-
known principles of stratigraphy to the observed reversal
pattern of the geomagnetic polarity recorded in rock
sequences.
Paleosecular variation: the gradual changes in direction
and intensity of the geomagnetic field observed at all loca-
tions on the Earth’s surface on timescales of years to
millennia.
Introduction
Paleomagnetism has become a standard discipline in dat-
ing the terrestrial record as it can be used to date and cor-
relate sedimentary successions from all over the globe. It
uses the well-documented variations in polarity, direction,
and intensity of the Earth’s magnetic field that have been
independently dated by radiometric, astronomical, or
other dating techniques. The geomagnetic polarity time-
scale (GPTS) is a “bar code” of alternating normal and
reversed polarity intervals with characteristic lengths
(Figure 1). Sedimentary successions can be magnetostrati-
graphically dated by correlating their individual polarity
 SEDIMENTS, TERRESTRIAL (PALEOMAGNETISM) 753

Sediments, Terrestrial (Paleomagnetism), Figure 1 Schematic representation of the geomagnetic field of a geocentric axial dipole
(“bar magnet”). During normal polarity of the field, the average magnetic north pole is at the geographic north pole, and a compass
points northwards. During normal polarity, the inclination is positive (downward directed) in the northern hemisphere and negative
(upward directed) in the southern hemisphere. Conversely, during reversed polarity, a compass needle points to the south; the
inclination is negative in the northern hemisphere and positive in the southern hemisphere. In the GPTS, periods of normal and
reversed polarity are represented by black- and white-colored intervals, respectively.

patterns to the GPTS, with each recognized reversal pro-
viding an age tie point. The duration of individual polarity
intervals varies from 20,000 years to several Myr, so that
this technique will mainly work for long (>1 Myr) and
continuous successions. The most recent polarity reversal
(Brunhes/Matuyama) took place 780,000 years ago.
Since then several “excursions,” thought to be failed or
aborted reversals, have been recorded against
a background of continuous low amplitude changes in
direction and intensity, known as the geomagnetic secular
variation. The combination of continuous records from
sedimentary cores with data from archaeomagnetic studies
has resulted in the construction of regional secular varia-
tion master curves and models. Holocene lake sediments
can be dated by correlating their direction and, in some
cases, their intensity records to these secular variation
master curves. Terrestrial sediments are generally more
difficult to date than marine sediments because the conti-
nuity of the depositional record is more easily disrupted
by erosional or nondepositional processes and because
biostratigraphic information (the fossil record) is often
scarce.
The paleomagnetic signal of terrestrial sediments
The magnetic field at any point on the Earth’s surface is
a vector that is described by three parameters: its strength
(F), declination (D) – the angle, measured eastwards,
between geographical north and the horizontal component
of the field – and inclination (I), the angle of the field
below the horizontal (Figure 2). Ferrimagnetic grains that
are deposited by wind and water experience a torque and
tend to align with the ambient magnetic field as long as
they are in the water column or in the soft water-saturated
topmost layer of the sediment (Figure 3). These grains will
become mechanically locked by dewatering and compac-
tion and will preserve the ambient magnetic field direction
as depositional remanent magnetization (DRM).
Compactional processes, however, may also affect the ori-
entation of elongated magnetic grains and result in
a shallowing of the inclination (e.g., Tauxe, 2010). The
paleomagnetic signal of terrestrial sediments may also be
affected by secondary components of magnetization. Sec-
ondary magnetic minerals may grow chemically within
the sediment, and their magnetization will become locked
when they grow through a critical grain size, after which
their remanence may be stable over billions of years as
chemical remanent magnetization (CRM). These geo-
chemical processes may result in the (partial) removal of
the original paleomagnetic signal and in a secondary com-
ponent that may be significantly younger than the deposi-
tion of the sediment. Such processes may offset the
position of a reversal boundary by a time of up to tens of
thousands of years. In most sediments, however, the offset
is small and the paleomagnetic signal may be considered
as reliable and near-depositional. Inclination shallowing
does not occur in CRM because chemical growth con-
tinues after dewatering.
In terrestrial sediments, early diagenesis is widespread
and occurs in virtually every sedimentary environment.
Depending on the role of organic matter, existing
magnetic minerals may dissolve, iron may be mobilized,
and it may precipitate elsewhere. Suitable environments
for paleomagnetic dating are generally those with
a sufficiently high sedimentation rate, a significant detrital
input, and a predominantly oxic environment. Sediments
that are deposited under oxic conditions typically contain
the iron oxides magnetite (Fe3O4) and/or hematite
(a-Fe2O3) that are both magnetically stable, especially if
they are fine-grained. These conditions are common in
many alluvial, fluvial, and deltaic depositional environ-
ments. Sediments that are deposited under anoxic or
dysoxic conditions may contain iron sulfides like greigite
(Fe3S4) and pyrrhotite (Fe7S8). Pyrrhotite is generally
754 SEDIMENTS, TERRESTRIAL (PALEOMAGNETISM)

Sediments, Terrestrial (Paleomagnetism), Figure 2 (a) The magnetic field at any point on the Earth’s surface is a vector (F) which
possesses a component in the horizontal plane called the horizontal component (H) which makes an angle (D) with the geographical
meridian. The declination (D) is an angle from north measured eastward ranging from 0
to 360
. The inclination (I) is the angle made
by the magnetic vector with the horizontal. By convention, it is positive if the magnetic field vector points below the horizontal and
negative if it points above. (b) and (c) The standard method for presentation and analysis of paleomagnetic demagnetization results
is called the Zijderveld diagram (Zijderveld, 1967). The endpoint of the vector measured after each demagnetization step is projected
both onto the horizontal plane (closed symbols) and onto the vertical plane (open symbols). Difference vectors (lines between data
points) then show the magnetization removed between steps.

interface, and authigenic components of secondary origin

Sediments, Terrestrial (Paleomagnetism), Figure 3 Illustration
of how rocks acquire a depositional remanent magnetization.
When magnetic particles are eroded from a source rock and are
deposited as a sediment, the particles can align with the
geomagnetic field to give rise to a depositional remanent
magnetization (DRM). If the sediment is remobilized (e.g., by
bioturbation), high water contents can allow the magnetic
particles to realign with the geomagnetic field via the torque
exerted on such particles by the field (From Roberts and Turner
(2013), after Tauxe and Yamazaki (2007)).
considered to form during late diagenesis and therefore
not suitable for paleomagnetic studies. Greigite can form
authigenically any time after deposition, which compli-
cates the interpretation of its paleomagnetic records
(Roberts et al., 2011). Recent studies, however, show that
two particular populations of greigite may provide reliable
records of the ancient geomagnetic field variations,
biogenic components of primary origin generated by
magnetotactic bacteria living near the sediment/water
that formed by early diagenetic processes (Vasiliev et al.,
2008). Greigite is especially common in restricted lacus-
trine and brackish water (e.g., Black Sea) environments.
The paleomagnetic signal in terrestrial sediments is
extremely small compared to that of volcanic rocks and
lavas and strongly depends on the magnetic carriers. Mag-
netite-bearing sediments usually yield a much stronger
paleomagnetic signal than hematite- or greigite-bearing
sediments. Rocks with relatively fine grain sizes (clay,
marls, silts, fine sands) usually provide the best results,
because smaller magnetic particles are generally more
stable.
Sampling and laboratory techniques for
magnetostratigraphy
Paleomagnetic samples of rocks and consolidated sedi-
mentary outcrops are commonly collected in the field
using a portable electrical or gasoline-powered drill with
a water-cooled diamond drill bit. Rocks can also be drilled
with compressed air, e.g., if the cooling water dissolves
the unconsolidated rocks (e.g., Dupont-Nivet and
Krijgsman, 2012). The drilling direction (azimuth and
dip) of the cores is measured (and marked) in the field
by a magnetic compass and a clinometer before extraction
from the outcrop. Alternatively, samples can be collected
as oriented blocks which need to be drilled in the labora-
tory later. Standard paleomagnetic cores have a diameter
of 2.5 cm and are at least 2 cm long. For unlithified
deposits, oriented samples are often obtained by pushing
PVC sample boxes into the soft sediment and orienting
them before extraction.
 SEDIMENTS, TERRESTRIAL (PALEOMAGNETISM)
Ideally, the ancient geomagnetic field can be
reconstructed from its record in rocks during the geologi-
cal past. The primary paleomagnetic signal, i.e., originat-
ing from near the time of deposition, however, is often
contaminated with chemical overprints and “viscous”
remanent components, referred to as VRM. Such a VRM
may result from partial realignment of “soft” magnetic
domains in the present-day field or from low-temperature
oxidation of magnetic minerals. Weathered rocks will
most certainly yield a paleomagnetic signal overprinted
in the present-day field. Overprints can generally be
removed through magnetic “cleaning” by demagnetiza-
tion techniques.
The standard demagnetization procedure is to progres-
sively remove the remanence of the samples by applying
stepwise increasing temperatures or alternating fields
(AF) in a magnetically shielded environment. After each
demagnetization step, the remaining remanence is mea-
sured on either a spinner or a cryogenic magnetometer.
Most rocks contain several components of magnetization,
which are demagnetized within different temperature (or
AF) ranges. If these components are formed at different
times, they often have recorded the paleomagnetic field
with a different direction. In the ideal case, the decay of
the remanence will be linear within separate demagnetiza-
tion ranges, so that direction and intensity of individual
components can be reliably determined. The stepwise
demagnetization results are commonly displayed in so-
called Zijderveld diagrams or Z-plots (Figure 2b, c). Their
advantage over stereographic projections is that they
enable resolution of the relative intensity and directions
of the various linear components. The direction is usually
determined by principal component analysis and is
documented by declination, inclination, and maximum
angular deviation (MAD).
Magnetostratigraphic dating of terrestrial
sediments
There are several prerequisites for the successful applica-
tion of magnetostratigraphy as a dating technique for ter-
restrial sedimentary sequences (Opdyke and Channell,
1996; Langereis et al., 2010). First, the originally recorded
geomagnetic field has to be retrieved with sufficient accu-
racy and resolution to enable the successive intervals of
normal and reversed polarity to be faithfully determined.
Second, one needs to have some approximate (biostrati-
graphic or radiometric) age control. Finally, the studied
sequences must represent a sufficiently long period to
reveal a characteristic pattern of polarity intervals (gener-
ally >1 Myr during the Cenozoic) that can be unambigu-
ously correlated to the standard GPTS. Naturally,
hiatuses and major changes in sedimentation rate may
obscure this “barcode” pattern, requiring careful field
observations, while sampling resolution must ideally be
several times higher than the shortest duration of (sub)
chrons (20–30 kyr). Sampling of multiple sections,
755
covering the same stratigraphic interval, increases the reli-
ability of the age determination.
Often, these prerequisites are met, enabling
magnetostratigraphy to be used as a high-resolution
age control, which has several fundamental advantages
above other dating techniques (Opdyke and Channell,
1996; Langereis et al., 2010). Unlike most biostrati-
graphic events, reversals of the geomagnetic field are
globally synchronous events, within sampling resolu-
tion, allowing precise correlations between different
parts of the world and between different environments.
A successful correlation of the recorded polarity pattern
to the GPTS provides accurate ages of every recognized
reversal boundary, allowing estimation of rates and rates
of change (Figure 4). This enables, for example,
detailed continental-marine correlations and hence the
determination of synchrony or diachrony of regional or
even global paleoclimatic or tectonic events. Magnetos-
tratigraphy can also be used to study the paleogeo-
graphic evolution and paleoclimatic history of different
regions over long periods.
The latest development in dating the terrestrial record
comes from orbital tuning of observed sedimentary
cycles to cyclic changes in Earth’s orbit around the
Sun and its axial tilt. In the “astronomically tuned polar-
ity timescale,” each reversal boundary – or any other
geological boundary for that matter, e.g., biostrati-
graphic datum levels or stage and epoch boundaries –
is dated individually. The astronomical polarity time-
scale (APTS) has provided a breakthrough in dating of
the geological record (Hilgen et al., 1997) and has sig-
nificantly increased our understanding of, for example,
the (paleo)climate system, since astronomical tuning
relies on deciphering and understanding environmental
changes driven by climate change which in turn is
orbitally forced. Some parts of the geological time-
scale – e.g., the Late Triassic and the Middle Miocene –
are largely based on astronomically tuned polarity rever-
sals documented in terrestrial sediments (Kent and
Olsen, 1999; Abdul Aziz et al., 2000).
Geomagnetic excursions
There is growing evidence from dynamo theory and the
paleomagnetic record that for every successful polarity
reversal, there have been a number of unsuccessful
attempts of the geodynamo to reverse. Theory suggests
that this is due to the longer magnetic time constant of
the inner core: a reversal is initiated in the liquid outer
core, but the inner core often fails to reverse before the ini-
tiating anomaly has reverted to the original polarity. In the
paleomagnetic record, an excursion is seen as a significant
deviation of the direction of magnetization – the virtual
geomagnetic pole moving more than 45
from the rotation
axis and a drop in intensity to maybe 20 % or 30 % of its
value during steady polarity. Many excursions are
observed worldwide and therefore make good global strat-
igraphic markers. In the Brunhes these include the Mono
756 SEDIMENTS, TERRESTRIAL (PALEOMAGNETISM)

Age
OCS GPTS (Ma)
VT-III
10.5

VT-lll
VTO C5n.2n-3

C5n.2n
79

VTO
VT-ll
68
C5r.1r 11.0
LY
C5r.1n

C5r.2r-1

VT-ll
VT-l
59
C5r.2r

55 55

51

C5r.2n 11.5

C5r.3r
VT-l

ORE

12.0
C5An.1n

30
C5An.1r

C5An.2n
ORE

CM
19

12.5
C5Ar.1r
CM

Age (CK95)
13 12 11 10
160 OB 20 C5Ar.1n
Sratigraphic Level (m)

C5Ar.2r
120
8 C5Ar.2n
80 10
4 C5Ar.3r
40
OB

0 0m 13.0

Sediments, Terrestrial (Paleomagnetism), Figure 4 Cyclostratigraphic and magnetostratigraphic correlations between subsections
and correlation of the terrestrial Orera composite section (OCS) to the GPTS. The OCS is located in the continental Calatayud Basin
of NE Spain. Solid lines between OCS and GPTS connect (interpretative) concomitant polarity boundaries. The inset diagram
illustrates the changes in accumulation rates through time (After Abdul Aziz et al. (2000)).
 SEDIMENTS, TERRESTRIAL (PALEOMAGNETISM)
Lake (33 ka), Laschamp (41 ka), Blake (120 ka), Iceland
Basin (180 ka), Pringle Falls (210 ka), Big Lost
(570 ka), and Stage 17/Delta (670 ka) (e.g., Roberts and
Turner, 2013).
Paleosecular variation curves of Holocene lake
sediments
Secular variation refers to the gradual changes in the geo-
magnetic field that occur at all locations on the globe on
timescales of years to millennia and are due to changes
in the internal geodynamo. At any given location, the
direction may vary by several tens of degrees and the
intensity by a factor of two or more over a few centuries.
In general the variations are not regular but consist of char-
acteristic time-varying signals in each of the three compo-
nents. Spherical harmonic analysis of historical
observations made over the past 400 years shows that all
the Gauss coefficients vary with time indicating that both
the dipole and non-dipole field contribute to the secular
variation (Jackson et al., 2000).
Several prerequisites are of crucial importance for
paleosecular variation studies. Sediments should be fine-
grained and contain single-domain detrital or biogenic mag-
netic minerals in sufficient concentration to yield a stable
magnetization that can be measured with accuracy. They
should have accumulated in a quiet environment, free from
currents that might otherwise upset the alignment of mag-
netized grains as they fall, settle, and are locked into posi-
tion in the sediment (Figure 3). In many cases, the
direction of such a detrital remanent magnetization
(DRM) is an accurate record of the paleomagnetic field
direction; however, sometimes the inclination appears to
be shallowed during the depositional process, resulting in
erroneously low average inclinations. Techniques are avail-
able for detecting and estimating a correction for such incli-
nation error (e.g., Tauxe, 2010).
Lake and shallow marine sediments typically accumu-
late at rates of between 0.5 and 5 m per millennium, and
so the DRM of such sediment sequences can, in favorable
circumstances, provide a continuous high-resolution
record of prehistoric (or paleo-) secular variation (Tauxe,
2010). The intensity of a DRM depends on the intensity
of the paleomagnetic field and a complex function of the
properties of the magnetic grains carrying the remanence.
In order to obtain an estimate of paleointensity,
a normalizing factor must be found that gives an empirical
estimate of this function. In principle this should be
a laboratory-grown remanence that affects the same spec-
trum of magnetic grains as the DRM, that is, it has
a similar magnitude and coercivity spectrum. In practice
normalization using anhysteritic remanent magnetization
(ARM) is often found to yield a good proxy for relative
paleointensity in sequences where there is little variation
in magnetic mineralogy or grain size (Levi and Banerjee,
1976; Tauxe and Yamazaki, 2007).
757
Coring and sampling for secular variation studies
When coring unconsolidated subaqueous sediments for
paleomagnetism, the most important considerations are
to core vertically, to avoid rotation of the core barrel which
might upset the declination record, and to avoid distur-
bance of the sediment due to compaction or vibration. Pis-
ton corers, operated from boats or small ships, are the
usual choice. The Mackereth corer (Mackereth, 1958),
made of nonmagnetic materials, is operated pneumatically
from cylinders of compressed air and collects 6, 9, or even
12 m long cores in a single drive. Once cut into 1.0 or 1.5 m
sections and split into halves, cores are sampled either in
discrete, typically 2.0–2.5 cm plastic cubes, or in continu-
ous U-channels of 2  2 cm cross-section and up to 1.5 m
in length. Many laboratories now employ pass-through
cryogenic magnetometers to measure the NRM of
U-channel samples continuously; some also have in-line
automated alternating field demagnetization instrumenta-
tion (Turner et al., 2007). A downside of such automation,
however, is that the measured signal is a convolution of
the magnetization and the sampling profile of the sensors
and deconvolution should be performed (e.g., Jackson
et al., 2010). Measurement of discrete samples is more
time-consuming but circumvents the convolution problem.
Clearly there is a trade-off between time and resolution:
with very long marine cores, the efficiency of automated
measurements on U-channels wins, but with Holocene sec-
ular variation records from perhaps 6 m long cores, the
extra resolution often makes discrete samples preferable.
Secular variation “master curves” are now available cov-
ering the Holocene period in several parts of the world. Typ-
ically these have been compiled by correlating and
combining the records of replicate cores within lakes and
then combining such records from different lakes. Where
lakes are more than about 100 km apart, the directional
records are reduced to a common location, usually via
a virtual geomagnetic pole transformation (e.g., Roberts
and Turner, 2013). Age control is typically provided by
radiocarbon dating, pollen assemblages, tephra identifica-
tion, etc. The stacking of several records also allows the esti-
mation of statistical confidence in the records. An example is
the UK master curve of declination and inclination of Turner
and Thompson (1981, 1982) and later European records in
which many of the same features are recognizable, modified
increasingly with distance from the United Kingdom.
Confidence in such lake sediment records is enhanced
by their correlation with archaeomagnetic records from
the same region (Figure 5). Archaeomagnetic data derived
from the thermoremanent magnetization (TRM) imparted
to an artifact (pottery, brick, fireplace, cooking stone, etc.)
the last time it cooled through the Curie temperature of its
constituent magnetic minerals. Age control may be avail-
able directly from the historical context or may be pro-
vided by radiocarbon dating of associated organic
material. Although archaeomagnetic records lack the
758 SEDIMENTS, TERRESTRIAL (PALEOMAGNETISM)
Sediments, Terrestrial (Paleomagnetism),
Figure 5 Paleomagnetic secular variation records from Europe
for the past 4,000 years. Declination and inclination: black data
points with error bars are archaeomagnetic data averaged over
160-year windows, calculated at 50-year intervals, with 95 %
confidence intervals, from Bucur (1994) and Gallet et al. (2002).
Red curves are the UK lake sediment records of Turner and
Thompson (1981, 1982), transformed to the latitude and
longitude of Paris. Paleointensity: archaeomagnetic intensities
from Spain, France, Norway, and Denmark, referred to the
latitude of Paris from Gómez-Paccard et al. (2008). The red curve
and shading represent the mean and 95 % confidence. In all
three plots, the blue and yellow curves are calculated from the
regional and global models, SCHA3k and CALS3k of Pavón-
Carrasco et al. (2009) and Korte et al. (2009), respectively (After
Roberts and Turner (2013)).
continuity of sedimentary ones, the TRM process leads to
a stronger remanence. Further, there is little or no smooth-
ing or averaging in either the magnetization or measure-
ment process, so confidence in the direction is usually
very good, and in addition an estimate of absolute
paleointensity can often be made from a TRM.
Databases, models, and dating tools
Various compilations of lake sediment and
archaeomagnetic secular variation data are available,
many in online databases. Most of these have been com-
piled with the aim of producing mathematical models of
the geomagnetic field, and the datasets included all fulfill
agreed quality criteria. The overall distribution of avail-
able data is heavily biased towards the northern and Atlan-
tic hemispheres, and relatively few archaeomagnetic data
are older than 3 ka.
Various models have been constructed using these data.
The CALS suite of models (e.g., Korte et al., 2011) are
continuous time-varying models of the global field based
on spherical harmonic expansion of the magnetic potential
in space and cubic B-splines in time. The most recent of
these, CALS10k, covers the entire Holocene to degree
and order 10. As is to be expected, such models are most
reliable where the data coverage is best, for example,
Europe and North America. In such regions, they are use-
ful dating tools: a model master curve can be calculated for
any given location (latitude and longitude) and observed
declination, inclination, and/or intensity matched, either
by eye or using formal statistical procedures. In other
regions, where data are currently sparse, the reliability of
global models and their potential for dating is poorer and
will improve as more data accumulate.
An alternative approach is to model the high-density
data at the regional level: to produce regional models,
using only a restricted dataset. Whereas spherical har-
monic analysis is a well-validated technique for use in
a spherical reference frame such as the Earth, methods
for producing models on restricted areas of the globe are
less well developed. One method is spherical cap har-
monic analysis (Haines, 1985; Thébault et al., 2006),
which has been used to produce models for Europe,
SCHA.DIF.3k and SCHA.DIF.8k (Pavón-Carrasco et al.,
2010). Although the method is still controversial, these
European models reproduce the data as well as or better
than existing global models based on spherical harmonic
analysis (Figure 6).
Pavon-Carassco et al. (2011) have gone one step fur-
ther – they have produced a Matlab dating tool, whereby
input of a paleomagnetic declination, inclination, and ide-
ally intensity, together with sampling location, results in
an estimate of age, based on comparison with a PSV mas-
ter curve for the location that is calculated from regional
and global models. The next decade will almost certainly
see both improvements in global models and the appear-
ance of PSV-based dating tools for many more regions
of the globe.
Summary/outlook
Paleomagnetic records of polarity reversals and secular
variation can provide valuable age control for geological
 SEDIMENTS, TERRESTRIAL (PALEOMAGNETISM) 759

Sediments, Terrestrial (Paleomagnetism), Figure 6 Black: Paleomagnetic secular variation “master curve” for the United Kingdom,
compiled from ten 6 m long sediment core records (four from Loch Lomond, three from Windermere, three from Llyn Geirionydd)
with age control provided by suites of radiocarbon dates on three of the cores (From Turner and Thompson (1981, 1982)). Red:
Modeled declination and inclination variations at the latitude and longitude of Loch Lomond calculated from the CALS10k model,
which is based on spherical harmonic analysis of globally distributed secular variation records (Korte et al., 2011). Green: Modeled
declination and inclination variations at the location of Loch Lomond calculated from the SCHA3k and SCHA8k models, which are
based on spherical cap harmonic analysis of European secular variation records (Pavón-Carrasco et al., 2009, 2010). In each case, the
uncertainty bars and bands signify 95 % confidence limits.

and archaeological materials, particularly sedimentary Bibliography

sequences. The geomagnetic polarity timescale is applica-
ble worldwide. Secular variation is coherent over regions
of continental scale. The use of regional master curves of
declination, inclination, and sometimes intensity is gradu-
ally being supplanted by mathematical models at regional
and global scale from which dating tools can be created.
The current GPTS offers high-resolution age control
for the past 160 Ma, where it is constructed from seafloor-
magnetic anomaly profiles, complemented by astronomi-
cal and radiometric (40Ar/39Ar; U-Pb) dating of
terrestrial rocks. Research is now focused on excursions
and older parts of the GPTS. Secular variation models will
be improved as more data is obtained to fill in the gaps,
particularly in the southern hemisphere and Pacific
regions.
Abdul Aziz, H., Hilgen, F., Krijgsman, W., Sanz, E., and Calvo,
J. P., 2000. Astronomical forcing of sedimentary cycles in the
middle to late Miocene continental Calatayud Basin (NE Spain),
Earth Planet. Science Letters, 177, 9–22.
Bucur, I. 1994. The direction of the terrestrial magnetic field in
France during the last 21 centuries. Recent progress. Physics of
the Earth and Planetary Interiors, 87, 95–109.
Dupont-Nivet, G., and Krijgsman, W., 2012. Magnetostratigraphic
methods and applications. In Busby, C., and Azor, A. (eds.), Tec-
tonics of Sedimentary Basins: Recent Advances. Hoboken:
Blackwell Publishing Ltd, pp. 80–94.
Gallet, Y., Genevey, A., and Le Goff, M., 2002. Three millennia of
directional variation of the Earth’s magnetic field in western
Europe as revealed by archaeological artefacts. Physics of the
Earth and Planetary Interiors, 131, 81–89.
Gómez-Paccard, M., Chauvin, A., Lanos, P., and Thiriot, J., 2008.
New archeointensity data from Spain and the geomagnetic
760 SINGLE-CRYSTAL LASER FUSION
dipole moment in western Europe over the past 2000 years. Jour-
nal of Geophysical Research, 113, B09103, doi:10.1029/
2008JB005582.
Haines, G. V., 1985. Spherical cap harmonic analysis. Journal of
Geophysical Research, 90, 2583–2591.
Hilgen, F. J., Krijgsman, W., Langereis, C. G., and Lourens, L. J.,
1997. Breakthrough made in dating of the geological record.
EOS, Transactions American Geophysical Union, 78, 285–288.
Jackson, A., Jonkers, A. R. T., and Walker, M. R., 2000. Four cen-
turies of geomagnetic secular variation from historical records.
Philosophical Transactions of the Royal Society of London A,
358, 957–990.
Jackson, M., Bowles, J. A., Lascu, I., and Sølheid, P., 2010.
Deconvolution of u channel magnetometer data: experimental
study of accuracy, resolution, and stability of different inversion
methods. Geochemistry, Geophysics, Geosystems, 11, Q07Y10,
doi:10.1029/2009GC002991.
Kent, D. V., and Olsen, P. E., 1999. Astronomically tuned geomag-
netic polarity timescale for the Late Triassic. Journal of Geo-
physical Research-Solid Earth, 104, 12831–12841.
Korte, M., Donadini, F., and Constable, C. G., 2009. Geomagnetic
field for 0–3 ka: 2. A new series of time-varying models. Geo-
chemistry, Geophysics, Geosystems, 10, Q06008, doi:10.1029/
2008GC002297.
Korte, M., Constable, C. G., Donadini, F., and Holme, R., 2011.
Reconstructing the Holocene geomagnetic field. Earth and
Planetary Science Letters, 312, 497–505, doi:10.1016/j.
epsl.2011.10.031.
Langereis, C. G., Krijgsman, W., Muttoni, G., and Menning, M.,
2010. Magnetostratigraphy – concepts, definitions, and applica-
tion, Newslett. Stratigraphy, 43, 207–233, doi:10.1127/0078-
0421/2010/0043-0207.
Levi, S., and Banerjee, S. K., 1976. On the possibility of obtaining
relative paleointensities from lake sediments. Earth and Plane-
tary Science Letters, 29, 219–226.
Mackereth, F. J. H., 1958. A portable core sampler for lake deposits.
Limnology and Oceanography, 3, 181–191.
Opdyke, N. D., and Channell, J. E. T., 1996. Magnetic Stratigraphy.
San Diego: Academic Press, p. 346.
Pavón-Carrasco, F. J., Osete, M. L., Torta, J. M., and Gaya-Piqué,
L. M., 2009. A regional archaeomagnetic model for Europe for
the last 3000 years, SCHA.DIF.3K: applications to
archaeomagnetic dating. Geochemistry, Geophysics,
Geosystems, 10, Q03013, doi:10.1029/2008GC002244.
Pavón-Carrasco, F. J., Osete, M. L., and Torta, J. M., 2010. Regional
modeling of the geomagnetic field in Europe from 6000 BC to
1000 BC. Geochemistry, Geophysics, Geosystems, 11, Q11008,
doi:10.1029/2010GC003197.
Pavón-Carrasco, F. J., Rodriguez-Gonzalez, J., Osete, M. L., and
Torta, J. M., 2011. A Matlab tool for archaeomagnetic dating.
Journal of Archaeological Science, 38, 408–419, doi:10.1016/
j.jas.2010.09.021.
Roberts, A. P., and Turner, G. M., 2013. Geomagnetic excursions and
secular variations. In Elias, S. A. (ed.), The Encyclopedia of Qua-
ternary Science. Amsterdam: Elsevier, Vol. 1, pp. 705–720.
Roberts, A. P., Chang, L., Rowan, C. J., Horng, C.-S., and Florindo,
F., 2011. Magnetic properties of sedimentary greigite (Fe3S4):
an update. Reviews of Geophysics, 49(1).
Tauxe, L., 2010. Essentials of Paleomagnetism. Berkeley: Univer-
sity of California Press.
Tauxe, L., and Yamazaki, T., 2007. Paleointensities. In Schubert, G.
(ed.), Treatise on Geophysics. Oxford: Elsevier, pp. 509–564,
Geomagnetism 5.
Thébault, E., Schott, J. J., and Mandea, M., 2006. Revised spherical
cap harmonic analysis (R-SCHA): validation and properties.
Journal of Geophysical Research, 111, B01102, doi:10.1029/
2005JB003836.
Turner, G. M., and Thompson, R., 1981. Lake sediment record of
the geomagnetic secular variation in Britain during Holocene
times. Geophysical Journal of the Royal Astronomical Society,
65, 703–725.
Turner, G. M., and Thompson, R., 1982. Detransformation of the
British geomagnetic secular variation record for Holocene times.
Geophysical Journal of the Royal Astronomical Society, 70,
789–792.
Turner, G. M., Rasson, J. L., and Reeves, C. V., 2007. Observation
and measurement techniques. In Schubert, G. (ed.), Treatise on
Greophysics. Amsterdam: Elsevier, Vol. 5, pp. 93–146.
Vasiliev, I., Franke, C., Meeldijk, J. D., Dekkers, M. J., Langereis,
C. G., and Krijgsman, W., 2008. Putative greigite magnetofossils
from the Pliocene epoch. Nature Geoscience, 1, 782–786,
doi:10.1038/ngeo335.
Zijderveld, J. D. A., 1967. Demagnetisation of rocks: analysis of
results. In Collinson, D., Creer, K., and Runcorn, S. (eds.),
Methods in Paleomagnetism. New York: Elsevier, pp. 254–286.
Cross-references
Archaeomagnetic Dating
Continental Drift (Paleomagnetism)
Geomagnetism
Magnetic Anomalies
Magnetic Chronology
Magnetostratigraphic Dating
SINGLE-CRYSTAL LASER FUSION
James K. W. Lee
Department of Geological Sciences and Geological
Engineering, Queen’s University, Kingston, ON, Canada
Department of Earth and Planetary Sciences, Macquarie
University, Sydney, NSW, Australia
Synonyms
Laser spot dating; Single-crystal step-heating; Single-
crystal total fusion
Definition
An analytical technique in Ar–Ar (40Ar/39Ar) dating in
which a laser is used to heat single crystals of minerals
in order to extract argon isotopes for measurement in
a mass spectrometer.
Introduction and modes of operation
Improvements in the sensitivity of argon mass spectrome-
ters have enabled the measurement of argon isotopes
released from the single crystals of K-bearing minerals.
In order to release the argon from the crystal structure, in
an environment with minimal contamination from the
atmosphere and experimental conditions, lasers have been
employed in combination with low-volume extraction
lines and sample chambers. The earliest use of lasers in
single-crystal Ar–Ar dating was by Megrue (1967), who
employed a pulsed ruby laser to analyze meteorite sam-
ples. York et al. (1981) documented the first use of
 SINGLE-CRYSTAL LASER FUSION 761

Single-Crystal Laser Fusion, Figure 1 40Ar/39Ar laser system. The laser (in this case, a Lexel Ar-ion laser) sits horizontally on the
bench. Mineral grains are placed inside the cylindrical steel sample chamber (see Figure 2) on the right-hand side of the photograph
on the stage of the binocular microscope. A CCD camera is aligned along the optical axis of the microscope and laser beam so that
mineral grains can be viewed “live” on the television screen (at the left) as they are heated. The laser power is controlled by the
rectangular console (foreground, left) and the sample chamber can be moved precisely in small increments in both the X and
Y directions using the joystick controller (foreground, center).

a continuous-wave laser in a thermochronological study of
metamorphic rocks and showed that such lasers could be
used to progressively step-heat a single crystal to obtain
an apparent age spectrum, analogous to that obtained by
heating a much larger aliquot of crystals in a furnace. Fur-
ther information about the step-heating technique and the
interpretation of Ar–Ar age spectra can be found in the
section “K–Ar and Ar–Ar Dating.” Since then, the use
of single-crystal laser dating techniques in Ar–Ar dating
has become widespread. Figure 1 shows a typical laser
dating system and Figure 2 shows a close-up of the sample
chamber containing mineral grains to be analyzed.
Single-crystal 40Ar/39Ar laser dating can be
performed in three modes: (1) step-heating, (2) total
fusion, and (3) spot dating.
Laser step-heating
In laser step-heating, a large-diameter, defocused laser
beam is used to progressively heat single crystals in
a series of steps of increasing power (i.e., temperature)
until the crystal has completely melted and fused into
a glass bead. At each step, the argon isotopes released by
the heating are analyzed in a mass spectrometer to obtain
an 40Ar/39Ar age for that step. A defocused beam is essen-
tial in order to uniformly heat the entire crystal in each
step. The complete sequence of steps (from low to high
temperature) yields an 40Ar/39Ar apparent age spectrum.
This technique can be used to elucidate potential age gra-
dients within crystals from rocks which may have cooled
slowly or been partially reset due to metamorphism or
local reheating. One of the earliest geological applications
is given by York et al. (1981). Lee et al. (1990) used both
laser step-heating and laser spot dating (described below)
to elucidate age profiles from rocks in a contact metamor-
phic zone. Laser step-heating is now one of the most com-
mon modes of single-grain laser dating. Typical types of
lasers currently used for this mode of dating are continu-
ous-wave lasers that emit in the infrared (IR) and visible
wavelengths, such as CO2 and Ar-ion lasers.
Laser total fusion
In single-crystal total fusions, a laser is used to completely
fuse a single mineral crystal in one step using a focused
or slightly defocused beam. The released argon isotopes
are analyzed in a mass spectrometer to obtain an 40Ar/39Ar
age for the crystal. This age is identical to the integrated age
that would be obtained from an 40Ar/39Ar apparent
age spectrum. Because this mode produces only a single
age per crystal, it is most typically used in dating studies
in which there may only be a limited amount of sample or
in detrital or provenance studies of crystals which may have
come from a variety of source rocks or regions which
therefore reflect rocks of different ages. The data are com-
monly displayed in the form of a probability density dia-
gram of total fusion age versus relative probability
(i.e., frequency) (see K–Ar and Ar–Ar Dating).
A thorough overview summarizing a variety of 40Ar/39Ar
detrital studies is given by Hodges et al. (2005). Typical
762 SINGLE-CRYSTAL LASER FUSION

Single-Crystal Laser Fusion, Figure 2 Laser dating sample chamber. Single crystals of the samples to be analyzed are placed in small
holes drilled into a metal (e.g., Al or Cu) planchet (bottom, center). The planchet is then placed inside a steel sample chamber with
a viewport that is transparent to the laser beam radiation, and the chamber is evacuated under ultrahigh vacuum. In this system, the
laser beam comes in horizontally from the upper right of the photo and is deflected downwards onto the mineral grains in the
planchet by a circular mirror angled at 45
(top, center).

types of lasers currently used for this mode of dating are
continuous-wave or pulsed lasers that emit in the infrared
(IR), visible, and ultraviolet (UV) wavelengths, such as
CO2, Ar-ion, and Nd:YAG lasers.
Laser spot dating
In laser spot dating, a finely focused laser beam is used to
ablate small volumes of sample within single crystals. The
argon isotopes released by the ablation are analyzed to
yield an age for that spot. Samples are usually prepared
in the form of a single crystal with a flat surface perpendic-
ular to the laser beam (e.g., as a polished thick section)
such that the laser can be used effectively as an age micro-
probe, in which laser spot traverses can be conducted from
the core to the rim of a single crystal. Age contour (or
“chrontour”) maps across an entire crystal can be
constructed (Figure 3) if a suitable number of sampling
points can be obtained, e.g., Phillips and Onstott (1988),
Lee et al. (1990), and Kramar et al. (2001). Typical laser
beam diameters can produce spots as small as 5–10 mm
in diameter, though typical applications employ larger
(e.g., 20–50 mm) spots in order to reduce analytical uncer-
tainties. This mode of dating is used to directly ascertain
potential age gradients within crystals. A variant of this
mode involves rastering the laser beam over a large area
of the crystal (e.g., 100  100 mm) or drilling a line of
spots in the form of a “trench” in order to obtain enough
argon to measure in a single analysis with reasonable pre-
cision, e.g., Kelley and Wartho (2000). This variant is
most commonly employed in Ar diffusion studies or in
depth profiling. Typical types of lasers currently used for
this mode of dating are pulsed lasers that emit in the IR
and UV wavelengths, such as Nd:YAG, frequency-
quadrupled Nd:YAG, and excimer lasers.
Advantages and limitations
40
Ar/39Ar single-crystal laser dating has a number of
advantages. Single-crystal analysis enables the unambigu-
ous measurement of the distribution of Ar isotopes within
a crystal, i.e., it eliminates the potential for the mixing of
different Ar reservoirs from various sources which may
exist in analyzing a mineral separate composed of tens or
hundreds of mineral grains. In addition, the effects of con-
tamination from other minerals or phases that may be pre-
sent in larger aliquots of minerals can be reduced. Another
advantage is that the amount of material that is required for
analysis is significantly reduced, which may be particu-
larly important in the analysis of trace or rare minerals.
The reduced sample size also greatly simplifies and
reduces the time required to prepare samples for analysis
and also minimizes the volume of sample that must be arti-
ficially irradiated and safely stored.
This technique also has some limitations. The reduced
sample size means that volumes of Ar derived from laser
heating are proportionally smaller; this means that suffi-
ciently old, K-rich, or large crystals should be dated to pro-
vide ages with acceptable precision. Alternatively,
modified techniques such as rastering or trenching
(discussed above) should be used. The smaller gas vol-
umes generally mean that age uncertainties on single-
crystal laser ages are somewhat larger than those derived
from analyzing larger aliquots of sample. For laser
 SKELETAL REMAINS (14C)
Single-Crystal Laser Fusion, Figure 3 Age contour or
“chrontour” map of a single amphibole grain with an age
gradient ranging from ~2,300 Ma in the grain core to ~2,700 Ma
along the grain rim. Chrontours were constructed from 22
individual laser spot dates obtained within the grain (Modified
after Lee et al. (1990)).
step-heating, it is important that the laser radiation couples
well with the mineral structure to ensure uniform heating
of the crystal; for example, although Nd:YAG lasers have
been successfully used to step-heat mafic minerals such as
biotite and amphibole, they are known to have difficulty
coupling with optically transparent minerals such as K-feld-
spar. Care must also be taken when performing laser spot
dating analyses on single crystals which are inhomoge-
neous, e.g., those with exsolution lamellae or Ar reservoirs,
as the resultant ages may reflect the ages of the different
phases/reservoirs or a combined mixture of their ages.
Summary
Single-crystal laser dating techniques are now commonly
employed in Ar–Ar dating studies. Depending on the type
of application, either pulsed lasers or continuous-wave
lasers may be used. Continuous-wave lasers are typically
used for single-crystal step-heating and total fusions; in this
case, the laser beam radiation must couple well with the
mineral structures to be heated (i.e., laser energy needs to
be efficiently absorbed by the mineral), and CO2 lasers are
most commonly used in current applications because the
wavelength of the radiation appears to couple efficiently
with all oxygen-bearing compounds. Pulsed lasers are
763
typically used for laser spot dating; UV wavelengths are
commonly employed because of the improved efficiency
in ablating a large variety of silicate minerals. Numerous
advantages of single-crystal dating have led to its wide-
spread application in a variety of Ar–Ar dating studies.
Bibliography
Hodges, K. V., Ruhl, K. W., Wobus, C. W., and Pringle, M. S.,
2005. 40Ar/39Ar thermochronology of detrital minerals. Reviews
in Mineralogy and Geochemistry, 58, 239–257.
Kelley, S. P., and Wartho, J. A., 2000. Rapid kimberlite ascent and
the significance of Ar-Ar ages in xenolith phlogopites. Science,
289, 609–611.
Kelley, S. P., Arnaud, N. O., and Turner, S. P., 1994. High spatial
resolution 40Ar/39Ar investigations using an ultra-violet laser
probe extraction technique. Geochimica Cosmochimica Acta,
58, 3519–3525.
Kramar, N., Cosca, M. A., and Hunziker, J. C., 2001. Heterogenous
40
Ar* distributions in naturally deformed muscovite: in situ UV-
laser ablation evidence for microstructurally controlled
intragrain diffusion. Earth and Planetary Science Letters,
192(3), 377–388.
Lee, J. K. W., Onstott, T. C., and Hanes, J. A., 1990. An 40Ar/39Ar
investigation of the contact effects of a dyke intrusion,
Kapuskasing Structural Zone, Ontario: a comparison of laser
microprobe and furnace extraction techniques. Contributions to
Mineralogy and Petrology, 105, 87–105.
Megrue, G. H., 1967. Isotopic analysis of rare gases with a laser
microprobe. Science, 157, 1555–1556.
Phillips, D., and Onstott, T. C., 1988. Argon isotopic zoning in man-
tle phlogopite. Geology, 16, 542–546.
York, D., Hall, C. M., Yanase, Y., Hanes, J. A., and Kenyon, W. J.,
1981. 40Ar/39Ar dating of terrestrial minerals with a continuous
laser. Geophysical Research Letters, 8, 1136–1138.
Cross-references
Ar–Ar and K–Ar Dating
Kimberlites (K-Ar/Ar-Ar)
Metamorphic Terranes (K–Ar/40Ar/39Ar)
Minerals (40Ar–39Ar)
Noble Gas Mass Spectrometer
Sedimentary Rocks (Rb-Sr Geochronology)
Thermochronology, Meteorites
Volcanogenic Sedimentary Rocks (K/Ar, 40Ar/39Ar)
SKELETAL REMAINS (14C)
Gregory W. L. Hodgins
Accelerator Mass Spectrometry Laboratory, Department
of Physics, School of Anthropology, University of
Arizona, Tucson, AZ, USA
Definition
Bone and teeth are complex tissues and thus radiocarbon
dating them is also complex. They contain carbon in both
organic and inorganic mineral phases. The organic phase
is primarily collagen. It is a fibrous extracellular structural
protein. The mineral phase is bioapatite. Bioapatite is
a biologically synthesized hydroxylapatite that is
764 SKELETAL REMAINS (14C)
predominantly calcium phosphate within which carbonate
ions are substituted.
Biomineralized tissues tend to survive better than soft
tissues after death. Nevertheless, they undergo a variety
of structural and chemical changes caused by physical,
chemical, and biological agents. Post mortem processes
that alter tissue structure and chemistry are termed diagen-
esis. Diagenetic reactions lead to both the loss of endoge-
nous carbon and the uptake of exogenous carbon. One of
the main challenges in radiocarbon dating skeletal mate-
rials is reliably removing the exogenous carbon that has
accumulated within the sample over time. Many different
sample preparation protocols have been developed to cope
with different sources and levels of contamination. Each
protocol has its strengths and weaknesses and no single
protocol is suitable for all sub-fossil skeletal remains.
The origin of bone dating goes right to the earliest pub-
lications on the 14C method (Arnold and Libby, 1951).
The earliest dates were derived from total bone carbon.
The reliability of these early measurements was almost
immediately questioned because bone dates frequently
contradicted dates obtained from stratigraphically associ-
ated wood and charcoal. This focused attention on deter-
mining whether fractions of bone could be identified that
would provide reliable dates, and this objective has pro-
pelled bone dating research up until the present day.
Chemical preparation methods have tended to focus
either on the organic fraction, principally collagen, or the
inorganic mineral fraction, the bioapatite. The following
discussion deals first with collagen dating. It begins with
a description of the structure of the protein itself, followed
by a presentation of the five common methods used for
collagen extraction from bones and teeth. The parameters
that define collagen purity are presented, and this is
followed by a brief consideration of burnt bone dating.
Protocols for dating bioapatite, the mineral phase of bone,
are then discussed. This entry concludes with a review of
a special case of bone mineral phased dating, radiocarbon
measurement from cremated or “calcined” bone.
Collagen in bone and teeth
Most radiocarbon dates on bone, teeth, tusk, and antler are
derived from the radiocarbon content of collagen in the
organic phase. This is because, in contrast with bone min-
eral, collagen does not readily exchange carbon with its
surrounding environment.
Collagens are in fact a family of proteins (Nelson and
Cox, 2009). The major type, Type I collagen, is the pri-
mary structural protein in bone. It comprises approxi-
mately 20 % of dry bone mass. The collagen content of
tooth enamel is lower than bone, typically about 3 % by
mass. The collagen content of tooth dentin is similar to
that of bone.
The genes that encode collagen are highly conserved
both between collagen types and across species. Organ-
isms on widely separated branches of the evolutionary tree
have collagen genes with similar structure. Consequently,
collagen purification protocols can be applied without
modification to a wide range of animal species.
A single collagen molecule is composed of three poly-
peptides hydrogen bonded in a rigid triple-helical rod.
A small number of covalent bonds within the triple helix
stabilize the structure. Each polypeptide is approximately
350 amino acids in length, with a molecular weight of
100,000 Da. The amino acid composition is unusual in
that every third amino acid is glycine, frequently followed
by proline and hydroxyproline. Hydroxyproline is rarely
encountered in other proteins and so can function as a
marker molecule for collagen.
Individual collagen tripeptides are bundled in
a staggered array, and these arrays are ordered into higher
structures and filaments. Cross-links between molecules
accumulate within the filaments over time. Hydroxylapatite
crystals form within the arrays and the fibrous nature of
these biomineralized structures give bones, teeth, and ant-
lers rigidity and compressive and tensile strength. Colla-
gens are also present in non-mineralized tissues, arranged
into both ordered and unordered fibers and sheets. The
accumulation of cross-links within and between collagen
molecules slowly alters the mechanical properties of
collagen-based tissues during life.
Archaeological collagen
Collagen can survive in archaeological bone for extremely
long periods, and survival beyond the 50,000 year radio-
carbon time frame is not uncommon. The mechanisms of
decay are diverse (Collins et al., 2002). Typically, collagen
is progressively lost from archaeological bone during dia-
genesis. Bones retaining less than 1 % collagen by mass
are usually too decayed to be reliably dated. Depending
upon the burial environment, hydrolytic reactions can
fragment the polypeptides decreasing collagen molecular
weight and lowering collagen melting temperature. Con-
densation reactions that cross-link collagen molecules to
other biomolecules can also occur, thereby covalently
linking contaminants into the collagen matrix. These
changes alter the behavior of the collagen fraction during
purification and can reduce collagen yield.
Methods for extracting collagen from bones and
teeth for radiocarbon dating
A variety of methods exist for extracting collagen from
bones and teeth. Considerations that affect the decision
about what method to use include the quantity of bone
available, its estimated age, its condition, whether or not
other isotopic parameters are to be determined, equipment
availability, cost, and other factors. Five methods will be
discussed: (1) total collagen purification by the Longin
and the Modified Longin method, (2) collagen ultrafiltra-
tion, (3) total amino acid purification by ion-exchange
chromatography, (4) single amino acid purification
including hydroxyproline extraction, and (5) the ninhy-
drin method.
 SKELETAL REMAINS (14C)
It is important to keep in mind that although the more
elaborate preparation protocols seem to promise more reli-
able dates, this is not necessarily the case. Theoretically,
they more precisely exclude contaminants that infiltrated
the skeletal remains during burial, yet the paradox is that
more elaborate protocols require more extensive labora-
tory handling and thus are prone to higher levels of
laboratory-induced contamination. It is very important to
carry out the experimental controls so contamination
levels are clearly defined.
The longin method
Berger et al. (1964) first proposed a collagen isolation
method based upon the suspension of bone powder in dilute
strong acids such as 0.1 N HCl. Acidification causes the
bone mineral to decompose. The mineral carbon is released
as carbon dioxide gas; the mineral salts dissolve into the
acidic solution, as do soluble organic molecules, and an
insoluble collagen-rich organic fraction remains. Contami-
nants can be washed out of the insoluble fraction without
substantially reducing collagen yield. The isolated “colla-
gen” fraction is quite crude, often contaminated with humic
or fulvic acids, but radiocarbon measurements of it were
a clear improvement over whole bone measurements.
Longin (1971) proposed a simple but powerful modifi-
cation to this early collagen purification protocol. It utilized
the acid extraction of collagen in 0.5 N HCl, but followed
this with a gelatinization step that rendered the collagen into
soluble gelatin. Heating an aqueous suspension of collagen
to temperature of greater than 65
C causes the triple helices
in the individual collagen molecules to unwind, producing
randomly coiled, soluble polypeptide monomers. Intramo-
lecular and intermolecular cross-links prevent the polypep-
tides from completely dissociating. Upon cooling, the
solution forms a gel. Triple-helical collagen molecules do
not reform, but a high degree of hydrogen bonding links
monomers together in a random manner. Depending upon
the concentration, the randomly coiled collagen peptides
either stay in solution or at higher concentrations form
a gel. Gelatin is the common name for this solubilized col-
lagen and denotes both the material and its form.
The rationale of the Longin method is that it is unlikely
that contaminants absorbed into the bone during burial
would undergo the same change in solubility that occurs
with collagen as a result of heating in aqueous solution.
The initial acid demineralization step allows the removal
of acid-soluble contaminants from the insoluble collagen
matrix by simple washing steps, and then gelatinization
changes the game so that any remaining insoluble contam-
inants can be removed from the collagen solution by either
centrifugation or filtration.
The Modified Longin method adds a NaOH extraction
step after the initial acid demineralization step. If the bone
is contaminated with soil-derived organic matter that is
insoluble in acidic solution, washes in high pH solutions
can extract these from the collagen matrix. As atmospheric
carbon dioxide dissolves readily in high pH solutions, the
765
base wash must be followed by another acid wash to
ensure that any absorbed atmospheric carbon dioxide is
removed before the gelatinization steps.
The Modified Longin method is commonly referred to as
an “ABA plus gelatinization” method where ABA ¼ acid/
base/acid. It has the advantage that it requires a minimum
of equipment and minimal sample handling, but achieves
a high degree of purification. The acid–base–acid washing
steps can be automated using readily available chromatog-
raphy equipment, and when carried out in continuous-flow
systems, there is an added advantage that extracted contam-
inants are completely removed from the system during
extraction. Finally, the gelatinized collagen can be isolated
from near-neutral pH solutions by freeze-drying. This dried
collagen solid is easy to handle, store, and archive.
Ultrafiltration
This method is a modification of the Longin method. First
suggested by Brown et al. (1988), it exploits the large size
of collagen protein (>100 kDa for intact protein mono-
mers) and hypothesizes that contaminants are likely to be
much smaller molecules. The method involves isolating
collagen by the Longin or Modified Longin method,
followed by ultrafiltration using 30 kDa molecular weight
cutoff membranes. The retained fraction, containing high
molecular weight collagen molecules, is retained and dated,
while the pass-through fraction containing low molecular
weight contaminants is discarded. Extensive washing of
the ultrafiltration membranes have been found necessary.
Higham et al. (2006) suggest using the method in contexts
in which bones are several half-lives old, and small
amounts of contamination have a large effect on the
resulting radiocarbon date.
Collagen total amino acid dating
Ho, Marcus, and Berger (1969) first proposed a method to
date bone by isolating and dating collagen-derived amino
acids. Collagen was first isolated from bone by 1.0 N HCl
hydrolysis of bone mineral. The proteinaceous residue
was then hydrolyzed in 6 N HCl at 100
C for 24 h. The
amino-acid-containing residue was then purified by
ion-exchange chromatography. After contaminants were
washed through the column, total amino acids were eluted
and subsequently combusted for radiocarbon measure-
ment. This approach was also developed by Gillespie
et al. (1984). Stafford et al. (1987) published a variant
using a mixed-bed ion-exchange resin called XAD-2.
The use of ion-exchange chromatography exposed a risk
that the ion-exchange columns themselves bleed off car-
bon that contaminated the sample; care was required to
minimize it and control it at the measurement phase.
Hydroxyproline dating
In parallel with the development of total collagen, amino
acid dating was the targeting of specific amino acids.
Hydroxyproline, though uncommon in most proteins, is
approximately 8–12 % of collagen. Its rarity made it a good
766 SKELETAL REMAINS (14C)
marker for collagen and several investigators focused on
developing methods for its isolation. Stafford
et al. (1982), Gillespie et al. (1984), and others have
published hydroxyproline isolation protocols. Until
recently, these methods have had relatively little impact
on bone dating, probably due to their complexity, concerns
about column bleed, and the fact that 90 % of collagen car-
bon is discarded during hydroxyproline purification.
Recently, new hydroxyproline purification methods
have been developed (McCullagh et al., 2010; Marom
et al., 2012). The development of preparative scale HPLC
systems, new ion-exchange resins with low bleed col-
umns, and improved techniques for the measurement of
small radiocarbon samples has made this possible.
The ninhydrin method
Ninhydrin is a reagent that reacts with primary and sec-
ondary amines. In the presence of a solution of amino
acids, ninhydrin condenses with the amino group nitrogen
and ultimately releases the carboxyl carbon of the amino
acid as carbon dioxide gas. A blue-colored ninhydrin
chromophore is also a product of the reaction. Nelson
(1991) formulated an elegant two-step protocol for
extracting collagen carbon using the ninhydrin reaction,
and it has been utilized extensively by Tisnérat-Laborde
et al. (2003). In the first step, crude collagen extracts are
treated with ninhydrin in a slightly acidic buffered solu-
tion. A small amount of carbon dioxide is produced from
the terminal amino acid of each collagen polypeptide
and any free amino acids or other contaminants in the
preparation that contain primary or secondary amines.
The first pulse of carbon dioxide gas is discarded. The
ninhydrin-treated collagen polypeptides are recovered
and then hydrolyzed to their constituent amino acids in
hot 6 N HCl. When this preparation is treated a second
time with ninhydrin, the resulting carbon dioxide gas is
recovered for radiocarbon dating.
Part of the power of the method lies in the selectivity of
ninhydrin for collagen amino acids. Burial contaminants
comprised of lipids, carbohydrates, humic acids, and other
organic compounds are unreactive. Another advantage
comes from the phase change that occurs in the
carbon of interest during the reaction. The ninhydrin reac-
tion takes place in a solution, but the carbon of interest
bubbles out as a gas so unwanted contaminants remain
behind.
The main disadvantage of the method is inefficiency. At
best, only one tenth of the collagen carbon is available for
radiocarbon measurement. There are further complexities
if stable isotope data are required, and these were investi-
gated by Keeling et al. (1999).
The inefficiency of the ninhydrin, hydroxyproline puri-
fication, and other elaborate protocols that by their nature
only recover a small percentage of collagen carbon pre-
sents a paradox: Only bones in a good state of preservation
contain sufficient collagen so that elaborate purifications
are possible, yet if bone preservation is good, then elabo-
rate protocols are probably not needed.
Collagen quality indicators
Collagen structure has been elucidated both biochemi-
cally and genetically in a broad range of species. A
archaeological collagen more difficult to characterize.
Nevertheless, several basic parameters are used to define
it. Solubility characteristics, carbon content, nitrogen
content, carbon to nitrogen ratio, carbon d13C and nitro-
gen d15N values, glycine content, the presence of
hydroxyproline, and other parameters have been used
to identify collagen in extracts of archaeological bone.
Moreover, these parameters can, to some extent, be used
to assess the purity of the preparation (van Klinken,
1999). This is in contrast to the short list of parameters
that defines the other two most commonly dated organic
materials, charcoal and wood cellulose. The identifica-
tion of contaminants within these materials is much more
difficult because the only parameters available are car-
bon content, which is not particularly enlightening
because the principle component of potential contami-
nants is also likely to be carbon, and d13C.
Burnt bone
Burnt bone and teeth present several challenges for radio-
carbon dating. The degree of charring affects the survival
of organic material and also changes its chemistry. Conse-
quently, protocols for collagen purification are not reliable
and the collagen quality indicators such as carbon and
nitrogen content cannot be applied. For these reasons,
burnt bone is best avoided.
Carbonate dating
The crystals of bioapatite, though predominantly calcium
phosphate, contain carbonate substitutions to the extent
that the mineral phase of bone contains about 5 % of total
bone carbon. The carbonate content of bioapatite is about
0.5–1.0 % by weight. The crystal morphology varies
within and between bones and teeth and also changes dur-
ing life, after death, and during burial.
The earliest bone radiocarbon measurements were car-
ried out on total bone carbon and thus included mineral
carbon. The variability of such bone dates compared to
associated charcoal made it clear that groundwater infiltra-
tion during burial could deposit secondary calcite, con-
taminate the mineral fraction, and thus alter the
radiocarbon date. Haynes (1968) demonstrated that acetic
acid was capable of removing secondary calcite, and this
treatment was adopted within protocols designed to
extract stable isotope information from bone and tooth
mineral (Koch et al., 1997). In spite of this advance and
subsequent attempts to refine it (Haas and Banewicz,
1980), radiocarbon dates on mineral carbon persistently
generate dates offset from associated charcoal or other
organic material. These findings have limited the use of
 SKELETAL REMAINS (14C)
bone mineral in dating studies. Nevertheless Zazzo and
Saliège (2011) have recently called for renewed research
on apatite dating because in some environments, typically
hot and arid regions, apatite carbon is all that survives.
Calcined bone dating
When bone is burnt above approximately 650
C, as can
be encountered during cremation, its mineral structure is
altered. The bone fragment, the organic fraction, is
completely lost, and the color of the interior becomes pure
white. The majority of the carbonate is lost, and apatite
crystal size and degree of order (crystallinity) increase
(Stiner et al., 1995). The bone becomes “calcined.” Cal-
cined bone is initially very friable after burning but
undergoes recrystallization, after which the bones become
structurally stable and resistant to environmental carbon-
ate exchange reactions.
Lanting et al. (2001) first demonstrated that the radio-
carbon content of calcined bone generally matched the
radiocarbon content of associated organic material and
concluded that traces of carbonate carbon survived the
high-temperature decomposition. Subsequently, calcined
bones were shown to give identical radiocarbon dates to
stratigraphically related organic material in a variety of
burial circumstances (DeMulder et al., 2007; Starkovich
et al., 2013). The question of why this is the case moti-
vated considerable research (e.g., Van Strydonck
et al., 2005). Subsequently, Van Strydonck et al. (2010)
and Hüls et al. (2010) demonstrated that at least some of
the carbon within calcined bone originates from the fuel
used during cremation. This opened up the possibility that
radiocarbon dates on calcined bones might be altered if
old wood was utilized during cremation. Evidence of this
effect can be seen in Olsen et al. (2008) in which cremated
remains were older than associated dendrochronologically
dated wood.
Summary and conclusions
Skeletal remains contain two carbon pools, an organic
phase that is primarily collagen protein and an inorganic
mineral phase containing calcium carbonate. The organic
phase, primarily collagen, is the preferred target for radio-
carbon dating because it does not readily incorporate car-
bon from the burial environment, whereas secondary
carbonates from the burial environment readily precipitate
within the mineral phase. Diagenesis, the physical and
chemical reactions that alter skeletal remains during burial
and decay, differs according to the particular chemistry of
the burial environment. Thus, the preservation state of
a particular bone is difficult to predict. A variety of purifi-
cation protocols have been developed to isolate collagen
protein or subcomponents of it.
Although historically skeletal remains were considered
a riskier material for radiocarbon dating than cellulosic
materials, their inherent complexity translates into
increased information content. Advances in understanding
767
how dateable components in archaeological bone and
teeth survive, and how they can be extracted, have made
them important materials for radiocarbon dating.
Bibliography
Arnold, J. R., and Libby, W. F., 1951. Radiocarbon dates. Science,
New Series, 113(2927), 111–120.
Berger, R., Horney, A. G., and Libby, W. F., 1964. Radiocarbon dat-
ing of bone and shell from their organic components. Science,
144, 999–1001.
Brown, T. A., Nelson, D. E., Vogel, J. S., and Southon, J. R., 1988.
Improved collagen extraction by modified Longin method.
Radiocarbon, 30(2), 171–177.
Collins, M. J., Nielsen-Marsh, C. M., Hiller, J., Smith, C. I., Rob-
erts, J. P., Prigodich, R. V., Weiss, T. J., Csapò, J., Millard,
A. R., and Turner-Walker, G., 2002. The survival of organic mat-
ter in bone: a review. Archaeometry, 44(3), 383–394.
DeMulder, G., van Strydonck, M., Boudin, M., Leclercq, W.,
Paridaens, N., and Warmenbol, E., 2007. Re-evaluation of
the Late Bronze Age and Early Iron Age chronology of the
Western Belgian urnfields based on14C dating of cremated
bones. Radiocarbon, 49(2), 499–514.
Gillespie, R. A., Hedges, R. E. M., and Wand, J. O., 1984. Radiocar-
bon dating of bone by accelerator mass spectrometry. Journal of
Archaeological Science, 11, 165–170.
Haas, H., and Banewicz, J., 1980. Radiocarbon dating of bone apa-
tite using thermal release of CO2. Radiocarbon, 22(2), 537–544.
Haynes, C. V., 1968. Radiocarbon analysis of inorganic carbon of
fossil bone and enamel. Science, 161, 687–688.
Higham, T., Bronk Ramsey, C., Karavanić, I., Smith, F. H., and
Trinkaus, E., 2006. Revised direct radiocarbon dating of the
Vindija G1 Upper Paleolithic Neanderthals. Proceedings of the
National Academy of Sciences of the United States of America,
103(3), 553–557.
Ho, T. Y., Marcus, L. F., and Berger, R., 1969. Radiocarbon dating
of petroleum-impregnated bone from Tar Pits at Rancho La Brea.
California Science, 164(3883), 1051–1952.
Hüls, C. M., Erlenkeuser, H., Nadeau, M.-J., Grootes, P. M., and
Andersen, N., 2010. Experimental study of the origin of cre-
mated bone apatite carbon. Radiocarbon, 52(2–3), 587–599.
Keeling, C. I., Nelson, D. E., and Slessor, K. N., 1999. Stable carbon
isotope measurements of the carboxyl carbons in bone collagen.
Archaeometry, 41, 151–164.
Koch, P. L., Tuross, N., and Fogel, M. L. 1997. The effects of sam-
ple treatment and diagenesis on the isotopic integrity of carbon-
ate in biogenic hydroxylapatite. Journal of Archaeological
Science, 24(5), 417–429.
Lanting, J. N., Aerts-Bijma, A. T., and van der Plicht, J., 2001. Dat-
ing of cremated bones. Radiocarbon, 43(2A), 249–254.
Longin, R., 1971. New method of collagen extraction for radiocar-
bon dating. Nature, 230, 241–242.
Marom, A., McCullagh, J. S. O., Higham, T. F. G., Sinitsyn, A. A.,
and Hedges, R. E. M., 2012. Single amino acid radiocarbon dat-
ing of Upper Paleolithic modern humans. Proceedings of the
National Academy of Sciences of the United States of America,
109(18), 6878–6881.
McCullagh, J. S. O., Marom, A., and Hedges, R. E. M., 2010.
Radiocarbon dating of individual amino acids from archaeolog-
ical bone collagen. Radiocarbon, 52(2–3), 620–634.
Nelson, D. E., 1991. A new method for carbon isotope analysis of
protein. Science, 251(4993), 552–554.
Nelson, D. L., and Cox, M. M., 2009. Lehninger Principles of Bio-
chemistry, 5th edn. New York: W.H. Freeman.
768 Sm–Nd DATING
Olsen, J., Heinemeier, J., Bennike, P., Krause, C., Hornstrup, K. M.,
and Thrane, H., 2008. Characterization and blind testing of
radiocarbon dating of cremated bone. Journal of Archaeological
Science, 35, 791–800.
Stafford Jr., T. W., Duhamel, R. C., Haynes Jr., C. V., and Brendel,
K., 1982. Isolation of proline and hydroxyproline from fossil
bone. Life Sciences, 31, 931–938.
Stafford Jr., T. W., Jull, A. J. T., Brendel, K., Duhamel, R. C., and
Donahue, D., 1987. Radiocarbon, 29(1), 24–44.
Starkovich, B. M., Hodgins, G. W. L., Voyatzis, M. E., and Gilman
Romano, D., 2013. Dating gods: radiocarbon dates from the
Sanctuary of Zeus on Mt. Lykaion (Arcadia, Greece). Radiocar-
bon, 55(2–3), 501–513.
Stiner, M. C., Kuhn, S. L., Weiner, S., and Bar-Yosef, O., 1995. Dif-
ferential burning, recrystallization, and fragmentation of archae-
ological bone. Journal of Archaeological Science, 22, 223–237.
Tisnérat-Laborde, N., Valladas, H., Kaltnecker, E., and Arnold, M.,
2003. AMS radiocarbon dating of bones at LSCE. Radiocarbon,
45(3), 409–419.
van Klinken, G. J., 1999. Bone collagen quality indicators for
palaeodietary and radiocarbon measurements. Journal of
Archaeological Science, 26, 687–695.
van Strydonck, M., Boudin, M., Hoefkens, M., and De Mulder, G.,
2005. 14C-dating of cremated bones, why does it work? Lunula,
13, 3–10.
van Strydonck, M., Boudin, M., and De Mulder, G., 2010. The car-
bon origin of structural carbonate in bone apatite of cremated
bone. Radiocarbon, 52(2–3), 578–586.
Zazzo, A., and Saliège, J.-F., 2011. Radiocarbon dating of biologi-
cal apatites: a review. Palaeogeography, Palaeoclimatology,
Palaeoecology, 310, 52–61.
Sm–Nd DATING
Richard W. Carlson
Department of Terrestrial Magnetism, Carnegie Institution
of Washington, Washington, DC, USA
Definition
Samarium–neodymium dating uses the radioactive decay
of 147Sm to 143Nd and 146Sm to 142Nd to measure the age
of rocks and minerals and to trace and date geochemical
processes responsible for the chemical differentiation of
the Earth and the terrestrial planets and planetesimals.
Introduction
Introduced to the geochemical and cosmochemical com-
munities in the early 1970s, the Sm–Nd radioactive decay
systems have grown into essential tools in geochronology
and particularly for the tracing of geochemical processes.
A key strength of Sm–Nd dating is that it involves two
rare earth elements (REE). The behavior of the REE dur-
ing geologic processes is relatively well understood
because, with the exception of Eu and Ce under some con-
ditions, they all have the same ionic charge of +3 and their
ionic radius changes monotonically as a function of
atomic number. As a result, their fractionation from one
another during silicate rock melting and crystallization is
predictable. In addition, the low solubility of REE in water
leads them to be relatively immobile during rock
weathering and to precipitate rapidly out of seawater once
delivered to the oceans by rivers. For this reason, the Nd
isotopic composition of sediments precipitated from sea-
water, such as manganese nodules (Piepgras et al.,
1979), serves as an excellent tracer for seawater circula-
tion (Jones et al., 2008).
Because Sm and Nd behave similarly in most geologic
processes, are not strongly affected by weathering pro-
cesses, and diffuse through minerals relatively slowly,
the Sm–Nd dating system holds promise of being more
resistant to resetting than other dating systems and thus
is capable of providing accurate initial formation ages
for rocks with complicated metamorphic histories. This
promise, in part, has been fulfilled. Forty years of applica-
tion of the Sm–Nd system, however, has provided a better
appreciation that not even Sm and Nd are immune to dis-
turbance by routine geologic processes. While this can be
a problem if one is trying to obtain an accurate age for an
igneous rock subjected to later metamorphism, the growth
of a metamorphic mineral like garnet, one of the few min-
erals that strongly fractionates Sm from Nd, allows the
Sm–Nd system to be an extremely useful tool for deter-
mining the time of peak metamorphism in a garnet-
bearing rock. Applications of the Sm–Nd dating system
range widely and touch on many fundamental questions
of solid Earth and planetary science. Examples include
the following: the time of Moon formation; new estimates
for the bulk composition of the Earth; the nature and age of
magma ocean differentiation on Earth, Mars, and the
Moon; the age of the oldest terrestrial crust; the growth
rate of the continental crust; age dating of mafic/ultramafic
rocks; the deformation history of high-grade metamorphic
belts; and the patterns of ocean circulation.
Once adequate isotope ratio precisions could be
obtained to resolve the radiogenic contributions to 143Nd
from the decay of 147Sm, the system could be applied to
add a temporal component to the story of REE fraction-
ation by geologic processes that had been developed pri-
marily in the proceeding decade. As the precision of
isotope ratio measurements improved further, the even
smaller variations in 142Nd due to the decay of the much
less abundant, and now extinct, 146Sm came into use for
studies of early solar system and planetary processes.
The nuclear basics
The REE provide an excellent example of the odd–even
effect in nuclear stability. Due to the extra nuclear stability
contributed by spin pairing of nucleons, an even number
of either neutrons or protons in the nucleus adds to nuclear
stability, whereas an odd number of either nucleon
decreases nuclear stability. All REE with an even number
of protons (Z) have at least four stable or long-lived iso-
topes, most, including Nd (Z ¼ 60) and Sm (Z ¼ 62), have
seven (Figure 1). The odd-Z REE have between zero (pro-
methium (Z ¼ 61) that does not exist in nature for this
 Sm–Nd DATING 769

Sm–Nd DATING, Figure 1 Portion of the chart of the nuclides in the Sm–Nd mass region. Numbers along the x-and y-axes indicate
the number of neutrons and protons, respectively, in each nucleaus with the numbers in the boxes showing the combination of
neutrons and protons that establish the mass of each nuclide. Blue boxes identify stable isotopes. Green boxes are isotopes with half-
lives ranging from a few days to 108 years. Brown boxes are nuclides with half-lives ranging from minutes to days. White boxes are
nuclides with half-lives of less than a minute. The black arrows show the path of s-process nucleosynthesis. The blue arrows show one
path for r-process nucleosynthesis of Nd and the green arrows show one path to create 146Sm via p-process nucleosynthesis.

reason) and two stable isotopes. Sm and Nd also benefit
from having isotopes with 82 neutrons, one of the nuclear
“magic numbers” (Mayer, 1950) where there is additional
stability in the nuclear shell structure model. The effect is
apparent in both 142Nd (Z ¼ 60, N ¼ 82) which is the most
abundant of the stable Nd isotopes though also lightest in
mass, but particularly in 144Sm (Z ¼ 62, N ¼ 82) that is
a stable isotope, whereas the next three higher mass Sm
isotopes are radioactive with half-lives ranging from 340
days (145Sm) to geologically useful times of 68 Ma
(146Sm) and 106 Ga (147Sm).
The isotopic compositions of Sm and Nd are listed in
Table 1. Samarium-147 constitutes just slightly less than
15 atom % of Sm and so is a relatively abundant isotope.
In contrast, although 146Sm was present when planet for-
mation began in the early solar system, it no longer exists
in nature because its decay half-life is only 1.5 % of the
age of the Earth. Neodymium presents the complication
that during the development of the Sm–Nd system, two
quite different values were used for mass fractionation
during analysis that results in very different measured iso-
topic compositions for the same Nd (Table 1). In
a common technique for Nd isotopic analysis, the sample
Nd is loaded onto an electrically heated filament placed
in the vacuum of a mass spectrometer (see Thermal Ioni-
zation Mass Spectrometer (TIMS)). At temperatures
between 1,200
C and 1,800
C, the Nd is both evaporated
and ionized from the filament. The ionized Nd is then
accelerated by an electric field into the mass spectrometer
where a large magnet separates the isotopes according to
their mass. During evaporation/ionization, the lighter
isotopes evaporate preferentially, so the Nd entering the
mass spectrometer starts out enriched in light over heavy
isotopes compared to the average isotopic composition
of the Nd loaded onto the filament. As the analysis pro-
ceeds, the heavier Nd isotopes become progressively more
abundant. The magnitude of the mass fractionation is on
the order of 0.1 % to 2 % per atomic mass unit depending
on the measurement technique used. The fractionation
would limit isotope ratio determinations to similar preci-
sion were it not for the ability to monitor and correct for
the fractionation during measurement. The degree of frac-
tionation is determined by comparing the measured ratio
of two stable isotopes against some assumed standard
value for that ratio. The measured fractionation is then
applied to the other isotope ratios using some model of
its mass dependency (Carlson, 2013). For Nd, the two
approaches used assume either 146Nd/144Nd ¼ 0.7219
(O’Nions et al., 1977) or 150Nd/142Nd ¼ 0.2096 (DePaolo
and Wasserburg, 1976b) to measure and correct for instru-
mental fractionation. While the former value produces Nd
isotope ratios similar to the values obtained during the first
Nd isotope measurements that used the NdO + ion and
assumed 148NdO/144NdO ¼ 0.242436 (Lugmair et al.,
1975b), the latter results in 143Nd/144Nd ratios that are
lower by 0.153 % for exactly the same Nd. Care must be
taken to compare only data measured using the same mass
fractionation correction value or correct the data to the
same mass fractionation ratio, in order to accurately com-
pare data from these two data reduction schemes. Fortu-
nately, the majority of data now reported uses a single
fractionation correction value, most commonly
770 Sm–Nd DATING

Sm–Nd DATING, Table 1 Isotopic composition of Sm and Nd

144 147 148 149 150 152 154
Sm Sm Sm Sm Sm Sm Sm
152
Ratio to Sm 0.114973 0.56081 0.420430 0.516833 0.275984 1.0 0.850791
Atomic % 0.030746 0.149957 0.112418 0.138201 0.073796 0.267394 0.227486
Atomic weight 143.91200 146.91489 147.91482 148.91718 149.91727 151.91973 153.92221
Atomic weight Sm 150.3653 g/mole
142 143 144 145 146 148 150
Nd Nd Nd Nd Nd Nd Nd
Ratio to 144Nd 1.141842 0.512630 1.0 0.348395 0.7219 0.241583 0.236460
Atomic % 0.271685 0.121973 0.237936 0.082896 0.171766 0.057481 0.056262
Atomic weight 141.90772 142.90981 143.91008 144.91257 145.91311 147.91689 149.92089
Atomic weight Nd 144.2395 g/mole
Ratio to 144Nd 1.138334 0.511845 1.0 0.348927 0.724095 0.243044 0.238595
Atomic % 0.270720 0.121728 0.237821 0.082982 0.172205 0.057801 0.056743
Atomic weight Nd 144.2470 g/mole

Sm and first-row Nd data from Qin et al. (2011). These Nd data are corrected for instrumental mass fractionation to 146Nd/144Nd ¼ 0.7219.
The second row of Nd data is for the same material, but mass fractionation corrected to 150Nd/142Nd ¼ 0.2096 (DePaolo and Wasserburg,
1976b). Atomic masses from Emsley (1991)

146
Nd/144Nd ¼ 0.7219. Given this issue and also to more
clearly express the small variations in 142Nd/144Nd and
143
Nd/144Nd ratios in nature, the isotope ratios are com-
monly reported as “epsilon” where
    
e143 Nd ¼ 143
Nd=144 Nd = 143 Nd=144 Nd
SA
and
  
e142 Nd ¼ 142
Nd=144 Nd = 142 Nd=144 Nd
SA
where “SA” is sample, “Chon” is average chondrite
(Bouvier et al., 2008), and “Std” is a terrestrial Nd isotope
standard, commonly JNdi (Tanaka et al., 2000). Given the
very small variations in 142Nd/144Nd, e142Nd is sometimes
replaced with m142Nd, which is calculated as shown in
Eq. 2 except with a multiplier of 106 instead of 104. Given
the isotopic compositions and atomic weights for Sm and
Nd listed in Table 1, one can transform the Sm/Nd weight
ratio (ppm/ppm) into 147Sm/144Nd atomic ratio by
multiplying the weight ratio by 0.6046 when data
are reported relative to 146Nd/144Nd ¼ 0.7219. The multi-
plier is 0.6049 when Nd data are reported relative to
150
Nd/142Nd ¼ 0.2096, another indication of the need
for care when comparing results reported from different
laboratories that use different mass fractionation correc-
tions for Nd.
As with all elements significantly greater in mass than
iron, the REE primarily are made in stellar interiors by
adding neutrons to preexisting seed nuclei (Burbidge
et al., 1957). This occurs through three routes depending
on the neutron density of the stellar site of nucleosynthesis
(Truran and Heger, 2003). In the latter stages of the life of
low- to moderate-mass stars (asymptotic giant branch or
AGB stars) when all the hydrogen has been consumed
by nuclear fusion in the core of the star, the star contracts
and the core density rises until fusion of He begins. As
the He is “burned” away to make carbon and oxygen,
fusion moves into less dense shells surrounding the core
 1  104 where mixing of material from outside the stellar interior
Chon leads to reactions such as the capture of an alpha particle
ð1Þ (a 4He nucleus composed of two protons and two neu-
trons) by 13C to produce 16O plus a neutron. These neu-
trons can then be captured by heavier nuclei, which
  cause them to move to the neutron-rich side of the stable
 1  104
Std mixture of neutron and protons, the so-called island of b-
ð2Þ decay stability (Figure 1). The more neutron-rich
a nucleus becomes, the more likely it is to decay by emit-
ting an electron (b-particle) from its nucleus to transform
a neutron into a proton and thus move up in mass along
the island of b-decay stability. If the neutron addition
occurs more slowly than the decay, the nucleus moves in
stair-step fashion up the stable portion of the nuclear chart
(Figure 1). This manner of creating heavier elements is
known as slow or s-process nucleosynthesis (Gallino
et al., 1997; Busso et al., 2004). For example, the sequen-
tial addition of neutrons to 142Nd moves the Nd up in mass
through the stable Nd isotopes to 146Nd. Addition of
another neutron produces 147Nd, an unstable nuclei with
a half-life of 11 days. If no additional neutron comes in
during the stability period of 147Nd, it decays to 147Sm,
which can capture a neutron to become 148Sm, and so con-
tinues to move up mass and Z in the chart of the nuclides
(Figure 1). Because of the relatively short half-life of
147
Nd and particularly the 1.7-h half-life of 149Nd, slow
neutron addition to Nd has trouble jumping over 147Nd
to make stable 148Nd and a particularly difficult time get-
ting past 149Nd to create stable 150Nd.
This is where the second style of nucleosynthesis
comes in. If the neutron density gets high enough, for
example, in the interior of an exploding star called
 Sm–Nd DATING
a supernova, neutron capture occurs more rapidly than b-
decay, so the seed nuclei can be driven to large neutron
excesses. Eventually, the half-lives of the neutron-rich iso-
topes become short enough that, even at these high neu-
tron capture rates, they decay back toward stable nuclei.
This more rapid or r-process nucleosynthesis (Cowan
and Sneden, 2004) is the primary production mechanism
for the neutron-rich isotopes of Nd and Sm. The propor-
tion of each Nd isotope created by s-process and r-process
nucleosynthetic paths thus changes with mass, with the
higher mass Nd isotopes being dominated by r-process
production (Arlandini et al., 1999).
The final nucleosynthetic process, the p-process, is
responsible for creating the proton-rich isotopes (Meyer
and Zinner, 2006). In the p-process, energetic photons or
neutrinos in a supernova cause the transformation of
a neutron into a proton. The unstable proton-rich isotope
then decays by either alpha emission or electron capture
until it reaches the island of nuclear stability. Both 144Sm
and 146Sm are made only by the p-process as neither
can be made by the s-process because the s-process path
takes 146Nd to 147Sm and both are blocked by stable iso-
topes of Nd from being produced by the r-process.
The elements now in our solar system consist of
a mixture of the elements produced by, and blown off
from, individual stars either by mass outflows from AGB
stars or from the ejecta of supernova explosions. The
importance of this as far as Sm–Nd dating is concerned
is that 142Nd cannot be produced by the r-process because
b-decay of neutron-rich isotopes will stop at stable 142Ce,
which then blocks further decay to 142Nd. If the stellar
contributions to Nd in the solar nebula were not perfectly
mixed when planet formation began, then variable propor-
tions of s-process and r-process stellar ejecta would cause
potentially large variations in the 142Nd/144Nd ratio in the
solar system that are not due to the decay of 146Sm. Nucle-
osynthetic heterogeneity in the solar nebula is apparent in
comparisons of meteorite and terrestrial Nd data
(Andreasen and Sharma, 2007; Carlson et al., 2007), so
this cause of isotopic variation in Nd must be taken into
account particularly when applying the 146Sm–142Nd
dating system. The 147Sm–143Nd system, however, is
much less sensitive to this problem, in part because the
magnitude of the 147Sm decay contribution to 143Nd is
much larger than the 146Sm contribution to 142Nd. In addi-
tion, because 143Nd is made by a combination of s-process
and r-process nucleosynthesis, its abundance is less sensi-
tive to imperfect mixtures of the two.
Another nuclear feature of the Sm–Nd system that must
be appreciated when the Sm–Nd dating technique is
applied to extraterrestrial objects is that some Nd and Sm
isotopes, particularly 149Sm, have large thermal neutron
capture cross sections. When planetary surfaces are
exposed to the high-energy particles, mostly protons,
known as cosmic rays, the collision of cosmic rays with
atoms in the upper meters of the planet/planetesimal can
knock neutrons out of the nucleus of the impacted atom.
At first, the emitted neutrons will be energetic
771
(epithermal), but collisions with other nearby atoms slow
them to thermal energies that allow them to be captured
by atoms with large thermal neutron capture cross sec-
tions, such as 149Sm (Lingenfelter et al., 1972). On the
Moon, the 149Sm/152Sm ratio can be reduced by more
than 0.4 % through neutron capture, raising the
150
Sm/152Sm ratio by about 0.8 % in the process
(Rankenburg et al., 2006; Boyet and Carlson, 2007;
Hidaka and Yoneda, 2007). In fact, at least some of the
techniques that allowed the development of the Sm–Nd
dating system originated from efforts to measure the cos-
mic ray exposure history of the lunar surface using neu-
tron capture by Sm, and particularly Gd, which has even
higher thermal neutron capture cross sections (Eugster
et al., 1970; Lugmair and Marti, 1971).
With the exception of very minor neutron capture
effects in Nd (Rankenburg et al., 2006), neutron capture
would have minimal affect on Sm–Nd dating were it not
for the fact that the precision needed in Sm/Nd ratio deter-
mination is such that the most commonly used technique
for both Sm and Nd concentration measurement is isotope
dilution. In isotope dilution, a known amount of a highly
purified isotope of the element is added to a known weight
of sample. The degree to which the ratio of the “spike” to
some stable isotope of that element in the spike plus
sample mixture is changed allows the calculation of the
concentration of the element in the sample. Spiking
low-abundance isotopes requires less of the often very
expensive spike per sample in order to make a precisely
measureable shift in isotopic composition, but in the case
of Sm, one also aims to avoid potential interference with
Nd. For this reason, 149Sm is the most commonly used
spike for Sm measurement by isotope dilution. Conse-
quently, one must know the sample’s 149Sm/152Sm ratio
in order to calculate an accurate concentration via this
technique. Half a percent depletion in 149Sm due to neu-
tron capture will result in a calculated Sm concentration
that is 0.5 % low if the neutron-induced depletion in
149
Sm is not taken into account. On Earth, where the atmo-
sphere and magnetic field shield the surface from the
majority of cosmic rays, the 149Sm/152Sm ratio is constant,
so neutron capture is not a significant concern for terres-
trial Sm–Nd dating.
Sm–Nd decay systematics
Both parents 146Sm and 147Sm decay by emitting
an alpha particle from their nucleus to end up as
daughters 142Nd and 143Nd, respectively. The standard
radioactive decay equation for 147Sm–143Nd dating is
     
143
Nd=144 Nd ¼ 143 Nd=144 Nd þ 147 Sm=144 Nd
p 0 p
 lt 
 e 1
ð3Þ
where the subscript “p” refers to present day and “0” to the
value at t ¼ 0 where t is time. Lambda is the decay
772
Sm–Nd DATING, Figure 2 147Sm–143Nd and 146Sm–142Nd isochrons for two eucritic meteorites illustrating the need for
high-precision isotopic ratio measurements in order to obtain precise isochron slopes. The data from Notsu et al. (1973a) and
Lugmair et al. (1975a) are both for the Juvinas eucrite, while the data from Boyet et al. (2010) are for the related eucrite Binda.
constant for 147Sm. For two or more materials formed at
the same time and with the same initial Nd isotopic com-
position, for example, different minerals rapidly crystal-
lized from a homogeneous lava, plotting the two
measureable quantities in this equation – (143Nd/144Nd)p
and (147Sm/144Nd)p – against one another will give
a line whose slope is equal to elt-1 and the y-intercept is
the 143Nd/144Nd at the time of formation of the minerals.
The line so obtained is called an isochron (Figure 2).
Begemann et al. (2001) summarize the measurements
of the half-life of 147Sm by radioactive counting.
Beginning around 1960, the determinations began to clus-
ter around a mean of 106  109 years, which was the value
adopted in one of the early Sm–Nd dating studies
(Lugmair et al., 1975b) and has been in use ever since.
This translates into a decay constant of 6.54  1012
year1. The value was cross-compared with U–Pb dating
of the Stillwater igneous complex by Nunes (1981) who
produced a zircon U–Pb age of 2,713  3 Ma in compar-
ison with the Sm–Nd isochron age of 2,701  8 Ma
(DePaolo and Wasserburg, 1979) that changes to 2,706
Ma using the 146Nd/144Nd ¼ 0.7219 fractionation correc-
tion. The overlapping Sm–Nd and U–Pb ages indicate
that the 147Sm decay constant currently in use produces
ages within 0.5 % of those determined using the U decay
system where considerable effort has gone into determin-
ing precise decay constants (Begemann et al., 2001). With
a 106 billion year (Ga) half-life, the abundance of 147Sm
has decreased by only 3 % over the 4.568 Ga history of
the solar system.
Until recently, the half-life used for 146Sm was 103  5
million years (Ma) (Friedman et al., 1966), but a more
recent experiment derives a significantly shorter half-life
of 68  7 Ma (Kinoshita et al., 2012). As either half-life
is significantly shorter than the age of the Earth, 146Sm
no longer exists in measureable quantities in natural
Sm–Nd DATING
materials. Thus, one cannot use the standard radioactive
decay equation (3) in order to calculate ages from
the 146Sm–142Nd system. Instead, like all dating systems
based on extinct radionuclides, ages from the
146
Sm–142Nd system utilize equations of the form
   
142
Nd=144 Nd Dt ¼ 142 Nd=144 Nd þ146 Sm=144 Nd  elDt
0
ð4Þ
Equation 4 can be expanded as
     
142
Nd=144 Nd Dt ¼ 142 Nd=144 Nd þ 146 Sm=144 Sm
 0

 144 Sm=144 Nd elDt
ð5Þ
Plotting the two measureable parameters in this
equation – (142Nd/144Nd)Dt and (144Sm/144Nd) – against
one another for a group of minerals formed at the same
time and with the same (142Nd/144Nd)0 will give a slope
equal to the 146Sm/144Sm ratio at the time the minerals
formed and a y-intercept equal to (142Nd/144Nd)0. Chro-
nologies derived from extinct nuclides thus do not give
absolute ages relative to the present day, as in
147
Sm–143Nd, but instead give the abundance ratio of
146
Sm to a stable Sm isotope, 144Sm, at the time the mate-
rial formed. If one rock gives a slope of 0.008 and another
a slope of 0.004, then the age difference between the rocks
is one half-life of 146Sm, or 68 Ma.
These age differences can be transformed into absolute
ages if one knows the 146Sm/144Sm in the solar system at
some well-dated time. For example, if one can measure
an accurate 147Sm–143Nd, or more commonly U–Pb, age
for a given sample, then the 146Sm/144Sm ratio determined
for the same sample can serve as a time marker for the
 Sm–Nd DATING
declining abundance of 146Sm in the early solar system.
This approach is documented in Carlson and Boyet (2009).
Using this approach, Boyet et al. (2010) derive a best esti-
mate for the initial 146Sm/144Sm in the solar system of
0.0084  0.005 at 4.568 Ga. This value, however, was
determined using the 103 Ma half-life of 146Sm. Using
instead the 68 Ma half-life, the solar system initial
146
Sm/144Sm calculated from the data considered by
Boyet et al. (2010) is 0.0094 (Kinoshita et al., 2012). Of
concern, and a topic of continuing research, is the question
of whether the 146Sm/144Sm ratio was homogeneous in the
solar nebula at the beginning of planet formation. If 146Sm
and other short-lived, recently synthesized radionuclides
were injected into the nebula just before the start of solar
system formation, then there may not have been enough
time for them to have been homogeneously mixed into
the nebula prior to the start of planet(esimal) formation.
If not, then different 146Sm/144Sm ratios determined for
different rocks could reflect, in part, the heterogeneous
distribution of 146Sm in the nebula. Sorting out how much
of the variability in the relative abundances of 142Nd in
a given rock is due to the possibly heterogeneous initial
abundance of 146Sm, or due to different times of rock for-
mation, is not straightforward. The best way to answer
this question is to produce a series of 146Sm–142Nd
“ages” for various ancient solar system materials and com-
pare the 146Sm–142Nd ages with absolute ages determined
by long-lived radiometric systems such as 147Sm–143Nd,
U–Pb, Lu–Hf, and Rb–Sr. If the absolute ages
indicate the same time span as the relative ages from
146
Sm–142Nd, then the argument is strong that 146Sm
was homogeneously distributed in the nebula at the start
of planet formation.
At this point, the reader may well be asking “Why
bother?” with extinct radionuclide geochronology. The
answer is twofold. First, the age precision provided by iso-
chrons is proportional, among other things, to the decay
constant, so short-lived radioactive decay schemes have
the potential of producing much more precise ages than
can be obtained from long-lived radiometric systems. Sec-
ond, because the parent nuclide becomes extinct early in
Earth history, radionuclide decay schemes are less sensi-
tive to modification of parent/daughter ratios that may
occur due to metamorphic events occurring long after rock
formation and thus can “see through” the later events to
provide information on the time of initial rock formation
(O’Neil et al., 2012).
The geochemistry and cosmochemistry of
Sm and Nd
Sm and Nd, and all the other REE, belong to the group of
elements known as “refractory lithophile” elements. The
“refractory” part means that both elements have high con-
densation temperatures (Nd ¼ 1,602
C, Sm ¼ 1,590
C;
Lodders, 2003) from a gas of solar composition. As the
gaseous solar nebula was cooling after the formation of
the Sun, both Sm and Nd were among the first elements
773
to condense into solid grains. Because they have similar
condensation temperatures, Sm and Nd should not be dra-
matically fractionated from one another by condensation
or evaporation processes. The “lithophile” part means that
both elements are much more soluble in silicate than in
iron (siderophile) or sulfide (chalcophile) melts/solids.
As a consequence, a process involving metal–silicate sep-
aration, for example, core formation, should not fraction-
ate Sm from Nd as both will stay behind in the silicate
mantle. The result of this refractory lithophile behavior
of the REE is that most types of primitive meteorites,
whose chemistry is controlled largely by volatility and
separation of metal from silicate, show a very narrow
range of REE relative abundances and a very small range
(approx. 3 %) in Sm/Nd ratios (Jacobsen and
Wasserburg, 1980; Boyet and Carlson, 2007; Bouvier
et al., 2008). For this reason, primitive meteorites
often are used to model the refractory lithophile element
abundances of the bulk silicate Earth, that part of Earth
excluding the core (McDonough and Sun, 1995; Palme
and O’Neill, 2003). The current best estimate for the aver-
age 147Sm/144Nd ratio of primitive meteorites, and hence
the solar system, is 0.1960  0.0004 with
a corresponding 143Nd/144Nd ¼ 0.512630  11 (Bouvier
et al., 2008).
The most effective natural processes capable of separat-
ing Sm from Nd are partial melting and fractionation crys-
tallization. The ability of minerals to fractionate a given
element is expressed by their partition coefficient, which
is the ratio of the concentration of an element in
a mineral compared to the concentration of the same ele-
ment in a melt with which the mineral is in chemical equi-
librium. Table 2 lists the partition coefficients for Sm and
Nd in various minerals. Most minerals have low partition
coefficients for REE, meaning that the REE tend to be
“incompatible” in most silicate minerals. As a result, they
selectively partition into the melt during partial melting
and concentrate in the residual liquid during fractional
crystallization. There are exceptions in that some minerals
concentrate REE, but these minerals tend to be relatively
rare trace phases, with the exception of apatite that is
a common minor phase in many rock types. Apatite and
plagioclase prefer Nd over Sm (D(Nd) > D(Sm)), but they
are exceptions as most minerals fractionate Sm and Nd in
the opposite sense. The most extreme in this sense is gar-
net, where the Sm partition coefficient is at least three
times higher than for Nd, but can go even higher
depending on the composition of the garnet (Table 2).
For most other minerals, however, the difference between
Sm and Nd partition coefficients is substantially less than
a factor of two. The partition coefficients listed in Table 2
provide only an estimate of the values likely to operate in
igneous differentiation as partition coefficients are func-
tions of temperature, pressure, and the composition of
both the mineral and melt. A particularly thorough discus-
sion of partition coefficients and the crystal chemical
parameters that determine them is given in Wood and
Blundy (2003).
774 Sm–Nd DATING

Sm–Nd DATING, Table 2 Melt–solid partition coefficients for where D is the bulk distribution coefficient of the solid
Sm and Nd in various minerals
assemblage of minerals, which is calculated from the
Mineral/basaltic melt D(Sm) D(Nd) References sum of the individual mineral distribution coefficients
weighted by their modal abundance in the rock:
Olivine 0.00055 0.00022 1
Clinopyroxene 0.462 0.277 2 D ¼ SX M DM ð7Þ
Garnet 1.1 0.363 2
Garnet 0.18 0.03 3 where DM is the distribution coefficient for mineral “M”
Clinopyroxene 0.23 0.14 3 and XM is the modal abundance of that mineral in the solid
Orthopyroxene 0.0056 0.015 3 rock. For a typical fertile mantle peridotite composed of
Plagioclase 0.098 0.123 4
Amphibole 0.28 0.25 5 60 % olivine, 20 % clinopyroxene, 10 % orthopyroxene,
Phlogopite 0.014 0.012 5 and 10 % garnet, the bulk D for Sm is 0.2 and for Nd is
Apatite 12 14 6 0.093. Estimates of fertile mantle Sm and Nd concentra-
Zircon 0.38 0.2 6 tions are 0.406 and 1.25 ppm, respectively (McDonough
and Sun, 1995). Using equation 6, a 1 % melt of the mantle
The partition coefficient, D, is defined as the ratio of the concentra- will have 1.92 ppm Sm and 12.2 ppm Nd for
tion of the element in the solid compared to melt. References are 1,
Beattie (1994); 2, Hauri et al. (1994); 3, Green et al. (2000); 4, a 147Sm/144Nd ¼ 0.095. At 20 % melting, the melt will
Bindeman and Davis (2000); 5, LaTourette et al. (1995); 6, Irving have 1.12 ppm Sm and 4.55 ppm Nd for
(1978) a 147Sm/144Nd ¼ 0.149. Partial melting of the mantle thus
will produce magmas with a fairly small range in Sm/Nd
ratio that is always lower than the Sm/Nd ratio of the
source rock. In contrast, because so much more Nd than
The similar partition coefficients for Sm and Nd in most Sm is partitioned into the melt during melting, the
minerals lead to relatively small variability in Sm/Nd ratio residue can develop quite high Sm/Nd ratios. In the
in most natural materials. For example, given the plagio- 20 % melting case described above, the 147Sm/144Nd of
clase and pyroxene partition coefficients listed in the residue is 0.323. The fractionation of Sm from Nd dur-
Table 2, a rock containing only plagioclase and pyroxene ing partial melting is the basis for perhaps the most signif-
with a whole rock 147Sm/144Nd ¼ 0.196 (chondritic) icant use of Sm–Nd dating where variations in
143
will have 147Sm/144Nd ratios of 0.156 and 0.322 in the Nd/144Nd of igneous rocks are used to infer the time
plagioclase and clinopyroxene, respectively. If the when their mantle source regions experienced some frac-
rock is 4.568 Ga old and today has a whole rock tionation event that caused their composition to deviate
143
Nd/144Nd ¼ 0.51263 (chondritic), then, using Eq. 3, from an estimate of the fertile mantle (see Model Ages
the rock formed with 143Nd/144Nd ¼ 0.506686. Today, (Sm–Nd)).
the plagioclase and pyroxene have The examples above illustrate one of the weaknesses of
143
Nd/144Nd ¼ 0.511417 and 0.516451, respectively, the Sm–Nd dating system, the combination of small par-
a range of just under 1 % in isotopic composition. ent–daughter fractionation and the long half-life of
147
Using Eq. 5 with a modern-day whole Sm conspire to create very limited variation in
143
rock 142Nd/144Nd ¼ 1.141842, solar system initial Nd/144Nd. With typical modern measurement uncer-
146
Sm/144Sm ¼ 0.0094 at 4.568 Ga, and tainties of 0.1 % on Sm/Nd ratio and 20 ppm
144
Sm/147Sm ¼ 0.20501, the rock modeled above formed for 143Nd/144Nd ratio, the igneous rock data modeled
with 142Nd/144Nd ¼ 1.1414573. Today, the plagioclase above produce an isochron with an age uncertainty of
and pyroxene have 142Nd/144Nd ¼ 1.1417579 and 13 Ma. For the 146Sm/142Nd system, where typical mea-
1.1420778, respectively, a range of 0.028 %. If the rock surement precisions on 142Nd/144Nd are  5 ppm, the data
instead formed at 1 Ga, then the present-day 143Nd/144Nd above provide an age precise to 2 Ma. These age preci-
ratios of its plagioclase and pyroxene would be 0.512368 sions are roughly an order of magnitude larger than the
and 0.513456, respectively, a range of 0.2 %. In a 1 Ga most precise ages currently being obtained with U–Pb in
rock, the plagioclase, whole rock, and pyroxene would zircon (Schoene et al., 2006; Amelin, 2008). Where Sm–
have essentially identical 142Nd/144Nd ratios because the Nd becomes important as a dating tool, however, is that
rock formed when 146Sm/144Sm ¼ 1.5  1018; hence, not all rocks contain minerals with high enough U–Pb
146 ratios to make them amenable for U–Pb dating. Examples
Sm was, in essence, extinct at 1 Ga.
Sm and Nd also can be fractionated by partial melting. include some types of meteorites, particularly those from
In the simple case where a source rock melts to a given Mars, most lunar rocks, and terrestrial Mg-rich rocks
degree (F) and the melt is then extracted from the such as komatiites, basalts, layered mafic intrusions, and
remaining solid, the concentration of a trace element in peridotites. In addition, garnet’s unique ability to strongly
the liquid (CL) compared to its concentration in the concentrate Sm over Nd, with 147Sm/144Nd ratios up to
starting solid (CO) is given by the batch melting equation over 4 (Pollington and Baxter, 2011), has turned garnet
into a particularly useful mineral for Sm–Nd
CL ¼ CO =ðF þ D  FDÞ ð6Þ geochronology.
 Sm–Nd DATING
The high isotope ratio precisions needed to provide use-
ful chronological precisions were the primary factor that
delayed the development of the Sm–Nd system as
a geochronological tool. The first inroads into the use
of Sm–Nd dating were two papers from the group at the
University of Tokyo who reported 147Sm–143Nd
data for an eclogite inclusion in the Roberts Victor kimber-
lite of South Africa (Notsu et al., 1973b) and both
147
Sm–143Nd and 146Sm–142Nd data for the Juvinas
eucritic meteorite (Notsu et al., 1973a). In both cases, the
Nd isotopic composition for all phases measured was
unresolved within the uncertainties of ~0.1–0.2 %
(Figure 2), but the work showed the potential of Sm–Nd
dating if isotope ratio precisions could be improved. In
1975, Lugmair et al. (1975b) used 147Sm–143Nd with Nd
isotope ratio precisions of 80 ppm to determine an age of
3.70  0.07 Ga for lunar mare basalt 75075. Perhaps more
important, this work also started the use of the Nd model
age approach to understand the time of chemical differen-
tiation in the source region of a volcanic rock. Lugmair
et al. (1975a) then attempted to find evidence of 146Sm
decay in the Juvinas meteorite (Figure 2), but even with
precision on 142Nd/144Nd improved to 60 ppm, they could
not resolve variability in 142Nd/144Nd in the mineral sepa-
rates. Convincing evidence for live 146Sm in the early
solar nebula had to wait until Lugmair and Marti (1977)
improved the 142Nd/144Nd precision to 20 ppm and
resolved variability in 142Nd/144Nd in mineral separates
from the Angra dos Reis meteorite. With these precision
improvements, the use, and diversity of uses, of the Sm–
Nd dating system grew rapidly (DePaolo and Wasserburg,
1976b; Lugmair et al., 1976; O’Nions et al., 1979; Allègre
and Othman, 1980; DePaolo, 1980; Allègre et al., 1983;
DePaolo, 1988).
Sm–Nd dating
While U–Pb dating of zircon (see Zircon) has reached
a level of sophistication and precision in age determina-
tion for rocks that contain zircon that cannot be bettered
by other dating systems, the Sm–Nd system has proven
particularly useful for age dating of zircon-free rocks,
which includes the majority of extraterrestrial rocks (see
Meteorites, Rubidium–Strontium, and Samarium–Neo-
dymium Chronology) and Mg-rich rocks including
basalts, komatiites, and peridotites. Sm–Nd dating has
also been applied to date altered/metamorphosed rocks
due to its greater stability through these processes com-
pared to, for example, Rb–Sr and K–Ar dating.
The Sm–Nd system has been heavily used in lunar geo-
chronology because most lunar rocks do not contain zir-
con, and the Moon is highly depleted in volatile
elements with the result that many lunar rocks have too
low a concentration of volatile Rb to be amenable to pre-
cise Rb–Sr dating and such low concentrations of volatile
Pb as to be extremely sensitive to contamination, thereby
making U–Pb chronological measurements an analytical
challenge. The first high-precision Sm–Nd age was
775
determined on a lunar basalt (Lugmair et al., 1975b). An
excellent summary of the application of Sm–Nd to under-
standing the history of lunar volcanism is given by
Nyquist and Shih (1992). Sm–Nd chronology has been
particularly important in defining the age range of the
lunar highlands crust as most of the highlands crust con-
sists of rocks highly dominated by plagioclase with vari-
able amounts of pyroxene and/or olivine as their
main mineral constituents. An example of a recent appli-
cation of 146Sm, 147Sm–142Nd, and 143Nd dating of what
is perhaps the oldest lunar rock is found in Borg et al.
(2011). For much the same reasons, Sm–Nd has been an
essential dating tool for the Martian samples delivered to
Earth as meteorites of the SNC – shergottite, nahklite,
and chassignite clan (Nyquist et al., 2001).
Another area where Sm–Nd dating has made a major
contribution is in determining the timing of high-grade
metamorphic events. This is enabled largely because of
the extreme fractionation of Sm and Nd typical of garnet.
For this reason, eclogites, the high-pressure form of basalt
that is composed of pyroxene and garnet, are a common
target for Sm–Nd dating (Mork et al., 1988; Chavagnac
and Jahn, 1996). As with all minerals, Sm and Nd diffuse
through garnet at high temperature, but the diffusion is
slow enough that garnet begins to faithfully maintain its
record of Sm decay at temperatures below about 600
C
(Mezger et al., 1992), allowing garnet to serve as
a useful thermochronometer for high-pressure metamor-
phism (Ducea et al., 2003; Pollington and Baxter, 2010;
see Luminescence, Earthquake and Tectonic Activity).
Perhaps the most common application of Sm–Nd dat-
ing has been on ultramafic rocks on Earth as these gener-
ally have too low K, Rb, and U concentrations for dating
by other methods. Good examples of this approach
include layered mafic cumulate bodies such as the Stillwa-
ter (DePaolo and Wasserburg, 1979), Muskox (Stewart
and DePaolo, 1992) and Great Dyke (Mukasa et al.,
1998), ophiolites (Jacobsen and Wasserburg, 1979;
McCulloch et al., 1981; Shaw et al., 1987; Li et al.,
1997), and particularly komatiites (Hamilton et al., 1977;
Jahn et al., 1982; Machado et al., 1986; Puchtel et al.,
2009).
The use of Sm–Nd dating to determine the age of
komatiites opens the discussion of the benefits/risks of
isochrons constructed from compositionally distinct
whole rocks as opposed to isochrons created by separating
the constituent minerals from a single rock. In most cases,
komatiite Sm–Nd ages involve whole rock isochrons.
There are two reasons for turning to whole rock isochrons.
First, many old mafic/ultramafic rocks have been
subjected to sufficient alteration/metamorphism that little
of their original igneous mineralogy remains. In the case
of komatiites, the original pyroxenes and olivines often
have been transformed into amphiboles and serpentines
by the addition of water during metamorphic reactions
upon their burial in the deep crust. In these cases, the
Sm–Nd ratios of the metamorphic minerals will not be
the same as the original igneous minerals and hence will
776 Sm–Nd DATING
Sm–Nd DATING, Figure 3 Sm–Nd isochron diagram illustrating
the potential and pitfall of whole rock isochrons. In this model,
a magma is created by melting the mantle at 2.7 Ga. A portion of
this magma undergoes crystal fractionation to create a series of
compositionally distinct rocks with different Sm–Nd ratios, but,
initially, the same 143Nd/144Nd ratio (blue dots along horizontal
line). After 2.7 billion years of evolution, these compositionally
distinct rocks are measured today and give the blue points near
the top of the figure that define a line whose slope gives an age
of 2.7 Ga. The other example takes a portion of the 2.7 Ga
magma and lets it mix, at 2.7 Ga, with a 3.8 Ga granite that has
lower Sm/Nd and 143Nd/144Nd ratios than the mantle-derived
magma. This creates a group of mixed rocks shown by the
green squares in the sloped line at the bottom of the graph. After
2.7 Ga of evolution, the rocks formed from the mixed magma
create a steep-sloped line whose slope corresponds to 3.8 Ga,
not 2.7 Ga.
not define an isochron that gives the igneous age of the
rock. Second, heating of minerals causes their constituent
elements to move, diffuse, through the crystal lattice.
Burial of rocks in the deep crust can raise their temperature
by hundreds of degrees. At these temperatures, fast-
diffusing elements can move millimeter distances over bil-
lions of years. The diffusional movement will serve to mix
the Nd in nearby minerals, erasing, or lessening, the Nd
isotopic difference caused by the decay of 147Sm in the
minerals due to their different Sm/Nd ratios. By moving
to the many centimeter-size hand-sample scale of whole
rocks, the consequences of this diffusional resetting of
the Sm–Nd clock will be minimized.
Equation 3, however, requires that all the samples used
to define an isochron must have started with the same initial
143
Nd/144Nd ratio in order for that line to provide an accu-
rate age. While the expectation that all minerals crystalliz-
ing from a magma will inherit the same 143Nd/144Nd from
the magma usually, but not always (Prevec et al., 2005),
is met, the limited Sm–Nd fractionation during partial melt-
ing and fractional crystallization often requires including
a wide range of rock compositions on a Sm–Nd whole
rock isochron in order to obtain enough spread in Sm/Nd
ratio to define a precise isochron. If all the rocks used to
construct the isochron indeed started with the same initial
143
Nd/144Nd, and formed at the same time, then the whole
rock isochron will give the correct age (Figure 3).
Magmas moving from source to eruption, however,
often interact with the surrounding rock. If the surround-
ing rock is isotopically distinct from the magma, then
mixing between the two can create hybrid magmas that
have isotopic compositions between the two end-members
as shown in Figure 3. Allowing these mixtures to evolve
through 147Sm decay with their different Sm/Nd and
initial 143Nd/144Nd ratios to the present day will produce
a linear array of data that gives a meaningless age some-
where between the age of magmatism and that of the wall
rock contaminant. This problem was first highlighted
when whole rock Sm–Nd dating of komatiites from
Kambalda, Australia, gave an age 500 Ma older than their
stratigraphic age constrained by U–Pb geochronology of
over- and underlying rock strata (Chauvel et al., 1985).
Although presenting a major obstacle for Sm–Nd whole
rock dating, using Sm–Nd systematics to detect and
understand the process of crustal contamination has been
an important application of the Sm–Nd system in rocks
where the age can be constrained by other means (Carlson
et al., 1981; DePaolo, 1981; Stewart and DePaolo, 1992;
Lambert et al., 1994).
Sm–Nd tracing of geologic processes
Because the processes that most strongly fractionate Sm
from Nd are partial melting and crystal fractionation, the
Sm–Nd system is particularly useful for determining the
time at which a rock, or its source materials, had its Sm/
Nd ratio changed from that of some model for mantle Nd
isotope evolution. Figure 4 shows an example of how
Nd isotope measurements can be used to understand the
source history of igneous rocks. In this model, the
143
Nd/144Nd of the mantle is evolving with a Sm/Nd ratio
equal to chondritic, often called the “chondritic uniform
reservoir,” or CHUR. If a portion of CHUR melts to pro-
duce a melt and residue, the different Sm/Nd ratios of
these complementary components cause their Nd isotope
evolution to diverge from the CHUR Nd isotope evolution
line. If nothing else happens to either melt or residue, the
age of this differentiation event can be calculated from
the measured 143Nd/144Nd and Sm/Nd ratios of either melt
or residue using the equation
    
TCHUR ¼ 1=l  ln 143 Nd=144 Nd  143 Nd=144 Nd =
   M   CHUR
147
Sm=144 Nd  147 Sm=144 Nd
M CHUR
where “M” stands for the measured values. While the
model age approach is straightforward mathematically,
interpretation of the age obtained depends on the number
 Sm–Nd DATING 777

Sm–Nd DATING, Figure 4 Model time evolution in 143Nd/144Nd (left) and transformed to e143Nd (right). The thick black line in each
plot shows the ingrowth of 143Nd in a material that has chondritic Sm/Nd ratio (CHUR). In this example, a melting event takes place at
3 Ga that creates a melt with low Sm/Nd ratio (CC) and a residue (DM) with high Sm/Nd ratio. The different Sm/Nd ratios of the two
rocks formed in this event cause their 143Nd/144Nd to diverge from the CHUR line. Measuring the 143Nd/144Nd and Sm/Nd ratio of the
samples of CC and DM today allows reconstruction of their Nd isotope evolution lines back in time. The slope of the isotope evolution
line is proportional to Sm/Nd ratio. The intersection of these lines with the CHUR evolution line defines the chondritic model age
(TCHUR). In this model, a second melting event occurs to CC at 1 Ga to create a new rock with lower Sm/Nd ratio (CM). Measuring the
Sm/Nd ratio and 143Nd/144Nd of this sample today allows the calculation of the green line. The extrapolation of this line to the CHUR
and DM evolution lines provides two different model ages. In this case, both are substantially older than the 1 Ga melting event and
show that CM was derived by reworking/remelting of CC, not as a new melt from the material evolving with either CHUR or DM
Sm/Nd ratio.

of differentiation steps involved in producing any given
rock sample and particularly on the Nd isotope evolution
path chosen for the model primitive reservoir. Early appli-
cations of Sm–Nd model ages usually referred to
a primitive mantle evolution characterized by
a chondritic Sm/Nd ratio. As Nd data became more abun-
dant, mantle-derived rocks through time defined a mantle
evolution characterized by higher than chondritic
143
Nd/144Nd (Shirey and Hanson, 1986). Interpretations
of this result ranged from assuming that the Sm/Nd ratio
of the mantle had been raised due to extraction of the
low Sm/Nd ratio continental crust (DePaolo et al., 1991)
to the possibility that the mantle never had a chondritic
Sm/Nd ratio (Boyet and Carlson, 2005). Whichever expla-
nation is correct, Nd model ages now are mostly
calculated relative to a “depleted mantle” instead
of CHUR evolution. There is some variability in the
values used for the Sm/Nd evolution parameters in calcu-
lating depleted mantle model ages, but a good compro-
mise, and a well reasoned argument for the choice of
values, is that presented by DePaolo et al. (1991). These
authors suggest replacing (143Nd/144Nd)CHUR with
(143Nd/144Nd)DM ¼ 0.51307 and (147Sm/144Nd)CHUR with
(147Sm/144Nd)DM ¼ 0.2105 for calculating Nd model
ages using equation 8 where “DM” is depleted mantle.
While uncertainties of this nature in the Nd evolution of
the “model” in Sm–Nd model age dating have
a significant effect on the ages calculated, they do not
detract from the usefulness of such “mantle separation
ages” in helping to distinguish whether a given rock was
newly added to the continental crust via melting of the
mantle or instead reflects reworking/remelting of older
crust (see entry on Model Ages (Sm–Nd)). This type of
information has been particularly useful in constructing
maps showing crustal age terranes that define the borders
of new crustal additions.
Conclusion
Forty years of application of the Sm–Nd dating system
has established it as a crucial component in the geochro-
nology/geochemistry toolbox. While hard-pressed to
provide dating precision comparable to techniques such
as U–Pb in zircon, the Sm–Nd systems can be applied
widely across a variety of rock types. This characteristic
has led it to be used extensively in dating mafic and
ultramafic rocks, be they from Earth or elsewhere. In
addition, Sm–Nd has become a useful chronometer for
high-pressure and high-temperature metamorphic
events. Among its many uses, perhaps the most com-
mon is its ability to illuminate the history of crust–man-
tle differentiation through model age approaches that
follow rocks from an early origin in a hypothetical
undifferentiated bulk planet to the differentiated crusts
and mantles that we find on Earth, Moon, Mars, and
asteroids today.
778 Sm–Nd DATING

Bibliography Carlson, R. W., Boyet, M., and Horan, M., 2007. Chondrite barium,
neodymium, and samarium isotopic heterogeneity and early
Allègre, C. J., and Othman, D. B., 1980. Nd-Sr isotopic relationship
earth differentiation. Science, 316, 1175–1178.
in granitoid rocks and continental crust development: a chemical
Chauvel, C., Dupre, B., and Jenner, G. A., 1985. The Sm-Nd age of
approach to orogenesis. Nature, 286, 335–342.
Kambalda volcanics is 500 Ma too old! Earth and Planetary Sci-
Allègre, C. J., Hart, S. R., and Minster, J.-F., 1983. Chemical struc-
ence Letters, 74, 315–324.
ture and evolution of the mantle and continents determined by
Chavagnac, V., and Jahn, B.-M., 1996. Coesite-bearing eclogites
inversion of Nd and Sr isotopic data, II. Numerical experiments
from the Bixiling Complex, Dabie Mountains, China: Sm-Nd
and discussion. Earth and Planetary Science Letters, 66,
ages, geochemical characteristics and tectonic implications.
191–213.
Chemical Geology, 133, 29–51.
Amelin, Y., 2008. U-Pb ages of angrites. Geochimica et
Cowan, J. J., and Sneden, C., 2004. Advances in r-process nucleo-
Cosmochimica Acta, 72, 221–232.
synthesis. In McWilliam, A., and Rauch, M. (eds.), Origin and
Andreasen, R., and Sharma, M., 2007. Mixing and homogenization
Evolution of the Elements. Cambridge: Cambridge University
in the early solar system: clues from Sr, Ba, Nd, and Sm isotopes
Press, pp. 27–42.
in meteorites. The Astrophysical Journal, 665, 874–883.
DePaolo, D. J., 1980. Sources of continental crust: neodymium iso-
Arlandini, C., Kappeler, F., Wisshak, K., Gallino, R., Lugaro, M.,
tope evidence from the Sierra Nevada and Peninsular ranges.
et al., 1999. Neutron capture in low-mass asymptotic giant
Science, 209, 684–687.
branch stars: cross sections and abundance signatures. The
DePaolo, D. J., 1981. Trace element and isotopic effects of com-
Astrophysical Journal, 525, 886–900.
bined wallrock assimilation and fractional crystallization. Earth
Beattie, P., 1994. Systematics and energetics of trace-element
and Planetary Science Letters, 53, 189–202.
partitioning between olivine and silicate melts: implications for
DePaolo, D. J., 1988. Neodymium Isotope Geochemistry: An Intro-
the nature of mineral/melt partitioning. Chemical Geology, 117,
duction. Heidelberg: Springer, p. 187.
57–71.
DePaolo, D. J., and Wasserburg, G. J., 1976a. Inferences about
Begemann, F., Ludwig, K. R., Lugmair, G. W., Min, K., Nyquist,
magma sources and mantle structure from variations of
L. E., et al., 2001. Call for an improved set of decay constants 143
Nd/144Nd. Geophysical Research Letters, 3, 743–746.
for geochronological use. Geochimica et Cosmochimica Acta,
DePaolo, D. J., and Wasserburg, G. J., 1976b. Nd isotopic variations
65, 111–121.
and petrogenetic models. Geophysical Research Letters, 3,
Bindeman, I. N., and Davis, A. M., 2000. Trace element partitioning
249–252.
between plagioclase and melt: investigation of dopant influence
DePaolo, D. J., and Wasserburg, G. J., 1979. Sm-Nd age of the Still-
on partition behavior. Geochimica et Cosmochimica Acta, 64,
water complex and the mantle evolution curve for neodymium.
2863–2878.
Geochimica et Cosmochimica Acta, 43, 999–1008.
Borg, L. E., Connelly, J. N., Boyet, M., and Carlson, R. W., 2011.
DePaolo, D. J., Linn, A. M., and Schubert, G., 1991. The continen-
Chronological evidence that the Moon is either young or did
tal crustal age distribution: methods of determining mantle sepa-
not have a global magma ocean. Nature, 477, 70–73.
ration ages from Sm-Nd isotopic data and application to the
Bouvier, A., Vervoort, J. D., and Patchett, P. J., 2008. The Lu-Hf and
Southwestern United States. Journal of Geophysical Research,
Sm-Nd isotopic composition of CHUR: constraints from
96, 2071–2088.
unequilibrated chondrites and implications for the bulk composi-
Ducea, M. N., Ganguly, J., Rosenberg, E. J., Patchett, P. J., Cheng,
tion of terrestrial planets. Earth and Planetary Science Letters,
W., and Isachsen, C., 2003. Sm-Nd dating of spatially controlled
273, 48–57.
domains of garnet single crystals: a new method of high-
Boyet, M., and Carlson, R. W., 2005. 142Nd evidence for early
temperature thermochronology. Earth and Planetary Science
(>4.53 Ga) global differentiation of the silicate Earth. Science,
Letters, 213, 31–42.
309, 576–581.
Emsley, J., 1991. The Elements. Oxford: Clarendon, p. 251.
Boyet, M., and Carlson, R. W., 2007. A highly depleted moon or
Eugster, O., Tera, F., Burnett, D. S., and Wasserburg, G. J., 1970.
a non-magma ocean origin for the lunar crust? Earth and Plane-
The isotopic composition of Gd and the neutron capture effects
tary Science Letters, 262, 505–516.
in samples from Apollo 11. Earth and Planetary Science Letters,
Boyet, M., Carlson, R. W., and Horan, M. F., 2010. Old Sm-Nd ages
8, 20–30.
for cumulate eucrites and redetermination of the solar system ini-
Friedman, A. M., Milsted, J., Metta, D., Henderson, D., Lerner, J.,
tial 146Sm/144Sm ratio. Earth and Planetary Science Letters,
et al., 1966. Alpha decay half-lives of 148Gd, 150Gd and 146Sm.
291, 172–181.
Radiochimica Acta, 5, 192–194.
Burbidge, E. M., Burbidge, G. R., Fowler, W. A., and Hoyle, F.,
Gallino, R., Busso, M., and Lugaro, M., 1997. Neutron capture
1957. Synthesis of the elements in stars. Reviews of Modern
nucleosynthesis in AGB stars. In Bernatowicz, T. J., and Zinner,
Physics, 29, 547–654.
E. K. (eds.), Astrophysical Implications of the Laboratory Study
Busso, M., Straniero, O., Gallino, R., and Abia, C., 2004. s-
of Presolar Materials. Woodbury, MN: The American Institute
processing in AGB stars and the composition of carbon stars.
of Physics, pp. 115–153.
In McWilliam, A., and Rauch, M. (eds.), Origin and Evolution
Green, T. H., Blundy, J. D., Adam, J., and Yaxley, G. M., 2000.
of the Elements. Cambridge: Cambridge University Press,
SIMS determination of trace element partition coefficients
pp. 67–84.
between garnet, clinopyroxene and hydrous basaltic liquids at
Carlson, R. W., 2013. Thermal ionization mass spectrometry. In
2–7.5 GPa and 1080-1200
C. Lithos, 53, 165–187.
McDonough, W. F., Holland, H. D., and Turekian, K. K. (eds.),
Hamilton, P. J., O’Nions, R. K., and Evensen, N. M., 1977. Sm-Nd
Treatise on Geochemistry. Amsterdam: Elsevier, pp. 337–354.
dating of Archean basic and ultrabasic volcanics. Earth and
Carlson, R. W., and Boyet, M., 2009. Short-lived radionuclides as
Planetary Science Letters, 36, 263–268.
monitors of early crust-mantle differentiation on the terrestrial
Hauri, E. H., Wagner, T. P., and Grove, T. L., 1994. Experimental
planets. Earth and Planetary Science Letters, 279, 147–156.
and natural partitioning of Th, U, Pb and other trace elements
Carlson, R. W., Lugmair, G. W., and Macdougall, J. D., 1981.
between garnet, clinopyroxene and basaltic melts. Chemical
Columbia River volcanism: the question of mantle heterogeneity
Geology, 117, 149–166.
or crustal contamination. Geochimica et Cosmochimica Acta,
Hidaka, H., and Yoneda, S., 2007. Sm and Gd isotopic shifts of
45, 2483–2499.
Apollo 16 and 17 drill stem samples and their implications for
 Sm–Nd DATING
regolith history. Geochimica et Cosmochimica Acta, 71,
1074–1086.
Irving, A. J., 1978. A review of experimental studies of crystal/liq-
uid trace element partitioning. Geochimica et Cosmochimica
Acta, 42, 743–770.
Jacobsen, S. B., and Wasserburg, G. J., 1979. Nd and Sr isotopic
study of the Bay of Islands ophiolite complex and the evolution
of the source of midocean ridge basalts. Journal of Geophysical
Research, 84, 7429–7445.
Jacobsen, S. B., and Wasserburg, G. J., 1980. Sm-Nd isotopic evo-
lution of chondrites. Earth and Planetary Science Letters, 50,
139–155.
Jahn, B.-M., Gruau, G., and Glickson, A. Y., 1982. Komatiites of
the Onverwacht Group, South Africa: REE geochemistry, Sm-
Nd age and mantle evolution. Contributions to Mineralogy and
Petrology, 80, 25–40.
Jones, K. M., Khatiwala, S. P., Goldstein, S. L., Hemming, S. R.,
and van der Flierdt, T., 2008. Modeling the distribution of Nd
isotopes in the oceans using an ocean general circulation model.
Earth and Planetary Science Letters, 272, 610–619.
Kinoshita, N., Paul, M., Kashiv, Y., Collon, P., Deibel, C. M., et al.,
2012. A shorter 142Nd half-life measured and implications for
146
Sm-142Nd chronology in the solar system. Science, 335,
1614–1617.
Lambert, D. D., Walker, R. J., Morgan, J. W., Shirey, S. B., Carlson,
R. W., et al., 1994. Re-Os and Sm-Nd isotopic evidence for the
petrogenesis of the Stillwater Complex, Montana. Journal of
Petrology, 35, 1717–1753.
LaTourette, T., Hervig, R. L., and Holloway, J. R., 1995. Trace ele-
ment partitioning between amphibole, phlogopite, and basanite
melt. Earth and Planetary Science Letters, 135, 13–30.
Li, X., Zhao, J., McCulloch, M. T., and Zhou, G. F., 1997. Geo-
chemical and Sm-Nd isotopic study of Neoproterozoic
ophiolites from southeastern China: petrogenesis and tectonic
interpretations. Precambrian Research, 81, 129–144.
Lingenfelter, R. E., Canfield, E. H., and Hampel, V. E., 1972. The
lunar neutron flux revisited. Earth and Planetary Science Let-
ters, 16, 335–369.
Lodders, K., 2003. Solar system abundances and condensation tem-
peratures of the elements. Astrophysical Journal, 591, 1220–1247.
Lugmair, G. W., and Marti, K., 1971. Neutron capture effects in
lunar gadolinium and the irradiation histories of some lunar
rocks. Earth and Planetary Science Letters, 13, 32–42.
Lugmair, G. W., and Marti, K., 1977. Sm-Nd-Pu timepieces in the
Angra dos Reis meteorite. Earth and Planetary Science Letters,
35, 273–284.
Lugmair, G. W., Scheinin, N. B., and Marti, K., 1975a. Search for
extinct 146Sm, 1. The isotopic abundance of 142Nd in the Juvinas
meteorite. Earth and Planetary Science Letters, 27, 79–84.
Lugmair, G. W., Scheinin, N. B., and Marti, K., 1975b. Sm-Nd age
and history of Apollo 17 basalt 75075: evidence for early differ-
entiation of the lunar interior. In Proceedings of the 6th Lunar
Science Conference, pp. 1419–1429.
Lugmair, G. W., Marti, K., Kurtz, J. P., and Scheinin, N. B., 1976.
History and genesis of lunar troctolite 76535 or: how old is
old? In Proceedings Lunar Science Conference 7th,
pp. 2009–2033.
Machado, N., Brooks, C., and Hart, S. R., 1986. Determination of
initial 87Sr/86Sr and 143Nd/144Nd in primary minerals from mafic
and ultramafic rocks: experimental procedure and implications
for the isotopic characteristics of the Archean mantle under the
Abitibi greenstone belt, Canada. Geochimica et Cosmochimica
Acta, 50, 2335–2348.
Mayer, M. G., 1950. Nuclear configurations in the spin-orbit cou-
pling model. I. Empirical evidence. Physical Review, 78, 16–21.
McCulloch, M. T., Gregory, R. T., Wasserburg, G. J., and Taylor,
H. P., Jr., 1981. Sm-Nd, Rb-Sr, and 18O/16O isotopic systematics
779
in an oceanic crustal section: evidence from the Samail ophiolite.
Journal of Geophysical Research, 86, 2721–2735.
McDonough, W. F., and Sun, S.-s., 1995. The composition of the
Earth. Chemical Geology, 120, 223–253.
Meyer, B. S., 1997. Supernova nucleosynthesis. In Bernatowicz,
T. J., and Zinner, E. K. (eds.), Astrophysical Implications of the
Laboratory Study of Presolar Materials. Woodbury, MN: The
American Institute of Physics, pp. 155–178.
Meyer, B. S., and Zinner, E., 2006. Nucleosynthesis. In Lauretta,
D. S., and McSween, H. Y. (eds.), Meteorites and the Early Solar
System II. Tucson, AZ: The University of Arizona Press,
pp. 69–108.
Mezger, K., Essene, E. J., and Halliday, A. N., 1992. Closure tem-
peratures of the Sm-Nd system in metamorphic garnets. Earth
and Planetary Science Letters, 113, 397–409.
Mork, M. B. E., Kullerud, K., and Stabel, A., 1988. Sm-Nd dating of
Seve eclogites, Norrbotten Sweden – evidence for early Caledo-
nian (505 Ma) subduction. Contributions to Mineralogy and
Petrology, 99, 344–351.
Mukasa, S. B., Wilson, A. H., and Carlson, R. W., 1998.
A multielement geochronologic study of the Great Dyke, Zimba-
bwe: significance of the robust and reset ages. Earth and Plane-
tary Science Letters, 164, 353–369.
Notsu, K., Mabuchi, H., Yoshioka, O., Matsuda, J., and Ozima, M.,
1973a. Evidence of the extinct nuclide 146Sm in “Juvinas”
achondrite. Earth and Planetary Science Letters, 19, 29–36.
Notsu, K., Mabuchi, H., Yoshioka, O., and Ozima, M., 1973b.
147
Sm-143Nd dating of a Roberts Victor eclogite. Geochemical
Journal, 7, 51–54.
Nunes, P. D., 1981. The age of the Stillwater complex –
a comparison of U-Pb zircon and Sm-Nd isochron systematics.
Geochimica et Cosmochimica Acta, 45, 1961–1963.
Nyquist, L. E., and Shih, C. Y., 1992. The isotopic record of lunar
volcanism. Geochimica et Cosmochimica Acta, 56, 2213–2234.
Nyquist, L. E., Bogard, D. D., Shih, C.-Y., Greshake, A., Stoffler,
D., and Eugster, O., 2001. Ages and geologic histories of Mar-
tian meteorites. Space Science Reviews, 96, 105–164.
O’Neil, J., Carlson, R. W., Paquette, J.-L., and Francis, D., 2012.
Formation age and metamorphic history of the Nuvvuagittuq
greenstone belt. Precambrian Research, 220–221, 23–44.
O’Nions, R. K., Hamilton, P. J., and Evensen, N. M., 1977. Varia-
tions in 143Nd/144Nd and 87Sr/86Sr in oceanic basalts. Earth
and Planetary Science Letters, 34, 13–22.
O’Nions, R. K., Carter, S. R., Evensen, N. M., and Hamilton, P. J.,
1979. Geochemical and cosmochemical applications of Nd iso-
tope analysis. Annual Review of Earth and Planetary Sciences,
7, 11–38.
Palme, H., and O’Neill, H. S. C., 2003. Cosmochemical estimates of
mantle composition. In Carlson, R. W., Holland, H. D., and
Turekian, K. K. (eds.), The Mantle and Core. Amsterdam:
Elsevier. Treatise on Geochemistry, Vol. 2, pp. 1–38.
Piepgras, D. J., Wasserburg, G. J., and Dasch, E. J., 1979. The iso-
topic composition of Nd in different ocean masses. Earth and
Planetary Science Letters, 45, 223–236.
Pollington, A. D., and Baxter, E. F., 2010. High resolution Sm–Nd
garnet geochronology reveals the uneven pace of tectonome-
tamorphic processes. Earth and Planetary Science Letters,
293, 63–71.
Pollington, A. D., and Baxter, E. F., 2011. High precision
microsampling and preparation of zoned garnet porphyroblasts
for Sm-Nd geochronology. Chemical Geology, 281, 270–282.
Prevec, S. A., Ashwal, L. D., and Mkaza, M. S., 2005. Mineral dis-
equilibrium in the Merensky Reef, western Bushveld Complex,
South Africa: new Sm-Nd isotopic evidence. Contributions to
Mineralogy and Petrology, 149, 306–315.
Puchtel, I. S., Walker, R. J., Brandon, A. D., and Nisbit, E. G., 2009.
Pt-Re-Os and Sm–Nd isotope and HSE and REE systematics of
780 STELLAR CHRONOLOGY
the 2.7 Ga Belingwe and Abitibi komatiites. Geochimica et
Cosmochimica Acta, 73, 6367–6389.
Qin, L., Carlson, R. W., and Alexander, C. M. O. D., 2011. Corre-
lated nucleosynthetic isotopic variability in Cr, Sr, Ba, Sm, Nd
and Hf in Murchison and QUE 97008. Geochimica et
Cosmochimica Acta, 75, 7806–7828.
Rankenburg, K., Brandon, A. D., and Neal, C. R., 2006. Neodym-
ium isotope evidence for a chondritic composition of the Moon.
Science, 312, 1369–1372.
Schoene, B., Crowley, J. L., Condon, D. J., Schmitz, M. D., and
Bowring, S. A., 2006. Reassessing the uranium decay constants
for geochronology using ID-TIMS U–Pb data. Geochimica et
Cosmochimica Acta, 70, 426–445.
Shaw, H. F., Chen, J. H., Saleeby, J. B., and Wasserburg, G. J., 1987.
Nd-Sr-Pb systematics and age of the Kings River ophiolite, Cal-
ifornia: implications for depleted mantle evolution. Contribu-
tions to Mineralogy and Petrology, 96, 281–290.
Shirey, S. B., and Hanson, G. N., 1986. Mantle heterogeneity and
crustal recycling in Archean granite-greenstone belts: evidence
from Nd isotopes and trace elements in the Rainy Lake area,
Ontario. Geochimica et Cosmochimica Acta, 50, 2631–2651.
Stewart, B. W., and DePaolo, D. J., 1992. Diffusive isotopic con-
tamination of mafic magma by coexisting silicic liquid in the
Muskox intrusion. Science, 255, 708–711.
Tanaka, T., Togashi, S., Kamioka, H., Amakawa, H., Kagami, H.,
et al., 2000. JNdi: a neodymium isotopic reference in consis-
tency with La Jolla neodymium. Chemical Geology, 168,
279–281.
Truran, J. W. J., and Heger, A., 2003. Origin of the elements. In
Davis, A. M., Holland, H. D., and Turekian, K. K. (eds.), Trea-
tise on Geochemistry. Amsterdam: Elsevier, pp. 1–16.
Wood, B. J., and Blundy, J. D., 2003. Trace element partitioning
under crustal and upper mantle conditions: the influences of
ionic radius, cation charge, pressure and temperature. In Carlson,
R. W., Holland, H. D., and Turekian, K. K. (eds.), Treatise on
Geochemistry: The Mantle and Core. Amsterdam: Elsevier,
pp. 395–424.
Cross-references
Luminescence, Earthquake and Tectonic Activity
Meteorites, Rubidium–Strontium, and Samarium–Neodymium
Chronology
Model Ages (Sm–Nd)
Thermal Ionization Mass Spectrometer (TIMS)
Zircon
STELLAR CHRONOLOGY
Trevor R. Ireland
Research School of Earth Sciences, The Australian
National University, Canberra, ACT, Australia
Synonyms
Nuclear cosmochronology; Nucleocosmochronology
Definition
Nucleocosmochronology. Use of radiogenic isotope ratios
for determining ages of nucleosynthesis and other frac-
tionations occurring in the galaxy and stars.
Introduction
Stars reflect the ongoing evolution of the galaxy as
nuclidic abundances of metals (elements from carbon up
in terms of proton number) are formed from the fusion
reactions transforming light elements in to heavy. At the
Fe abundance peak, fusion no longer releases energy,
and elements heavier than iron are the result of two reac-
tion pathways involving the addition of neutrons. The
s-process involves the slow addition of neutrons, where
slow indicates that the chances of beta decay is more likely
than the chance of adding another neutron. The s-process
therefore keeps the nucleosynthetic pathway close to the
stable nuclides we see around us. The s-process stops at
209
Bi, this being the heaviest stable nuclide. The site of
the s-process is likely dominated by asymptotic giant
branch (AGB) stars, also known as red giants, that are
burning He into heavier nuclides releasing abundant
neutrons in the process.
The r-process denotes rapid addition of neutrons, so
that many neutrons are added before a nuclide has
a chance to decay. In the r-process, neutrons are added to
quasi-stabilize nuclides far off the pathway of stability;
once the neutron source is extinguished, the nuclides beta
decay back to the canonical stable nuclides. The r-process
therefore leads to much heavier nuclides being formed as
the neutrons in the nucleus decay to protons. Uranium
and thorium isotopes are therefore formed in this way.
The astrophysical site of the r-process is likely to be super-
nova explosions resulting from core collapse as the den-
sity in the stellar core cannot resist gravitational
contraction forming abundant neutrons in the process.
These explosions take place over time periods of a few
seconds with secondary nucleosynthesis (related to decay)
on the order of months to years depending on the nuclide.
In terms of galactic chemical evolution, it is widely held
that massive star formation was more prevalent in the
early solar system, and hence r-process contributions are
dominated by early galaxy supernovae explosions.
Chronological implication
It follows that nucleosynthesis itself could be dated if
the initial abundances were known and the products
could be measured. Theoretical models of r-process nucle-
osynthesis can be used to establish the abundances of
uranium and thorium isotopes originating from that
source. A classic case is that of the 235U/238U of our
solar system first related in detail in the classic study of
Willy Fowler and Fred Hoyle. The benefit of simply using
uranium is that there is no inherent fractionation between
the isotopes because of chemical processing. The isotope
ratio is taken to simply reflect the decay of 235U (half-
life of 700 Myr) relative to 238U (half-life of 4.4 Gyr).
The r-process production (P) ratio 235U/238U is based
on the contributions of not only A ¼ 235 and A ¼ 238
but also their respective a-decay progenitors up to
A ¼ 255 where spontaneous fission alone occurs; P235/
P238 can be affected by details in r-process
 STELLAR CHRONOLOGY
nucleosynthesis models. Fowler and Hoyle (1960) used
a P235/P238U ratio of 1.65 in their calculations. The pre-
sent day 235U/238U of our solar system is 0.00725, and the
initial 235U/238U of the solar system is therefore 0.320 for
the age of the solar system of 4.567 Gyr (Amelin et al.,
2010). It follows then that
N235 =N238 ¼ P235 =P238 eðl235l238Þt
and the time t to evolve from a primitive r-process contri-
bution to the time of the formation of our solar system is
2.0 billion years.
The Th/U ratio can also be used as a chronometer
because of the difference in half-lives of 238U and 232Th.
Here the calculation will also be affected by the potential
for chemical fractionation of Th from U, which for the pur-
poses of this calculation will be ignored. For the present
Earth, the Th/U abundance is 3.6, and therefore 2.3 for
the early solar system (Anders and Grevesse, 1989). The
r-process production ratio is around 1.7. Following
a similar formalism to that above for uranium, the decay
time is 2.8 billion years. Thus, the actinide abundances
in the early solar system can be accommodated by injec-
tion of nucleosynthetic material that has been allowed to
decay for 2–3 billion years.
However, the actinide abundances are being
maintained in an equilibrium state balancing the nuclear
decay with augmentation from new stellar explosions.
Such a scenario is now supported additionally by the pres-
ence of short-lived radionuclides, many with half-lives
less than 10 Myr such as the r-process nuclide 182Hf. In
accounting for this ongoing production and decay, Fowler
and Hoyle derived an age of 15 (+5/3) Gyr for the age of
the galaxy, a remarkably prescient prediction.
Chronlogy from stellar abundances
For dating stellar systems other than our own, the ability to
perform this kind of analysis is severely limited by our
ability to measure the abundances of the nuclides in opti-
cal spectra. Isotopic analysis of heavy nuclides such as
781
uranium is not possible. In fact detection of uranium is
very difficult because of the coincidence of a CN (cya-
nide) line with U(II). Thorium is more promising to reflect
age. The Th/Eu ratio can be used as a proxy for age
because Eu is predominantly r-process and is a stable
nuclide. The Th/Eu abundance will fall with time
at a known rate because of the decay of 232Th. As an
example, the Th/Eu of the star CS 22892–052 is 0.219,
well below the r-process ratio of 0.463, consistent with
a simple radioactive decay age of 15.2  3.7 Gyr.
These calculations assume simple single-stage decay
models. Such a scenario is unlikely with a variety of stellar
inputs likely making up the galactic abundances in any
region of the galaxy.
Presolar grains are recovered from primitive meteorites,
and these grains are believed to have condensed around
stars that were extant before our solar system formed.
Most of these appear to be derived from AGB stars with
characteristic isotopic compositions from that source.
Within single grains, U and Th are observed reflecting
the ongoing r-process history of our galaxy.
Conclusion
The abundances of the radiogenic elements U and Th
reflect ongoing processing in the galaxy. The abundances
of these nuclides reflect billions of years of decay coupled
with ongoing augmentation from supernovae.
Bibliography
Amelin, Y., Kaltenbach, A., Iizuka, T., Stirling, C. H., Ireland, T. R.,
Petaev, M. I., and Jacobsen, S. B., 2010. U-Pb chronology of the
solar system’s oldest solids with variable 238U/235U. Earth and
Planetary Science Letters, 300, 343–350.
Anders, E., and Grevesse, N., 1989. Abundances of the elements –
meteoritic and solar. Geochimica et Cosmochimica Acta, 53,
197–214.
Cowan, J. J., McWilliam, A., Sneden, C., and Burris, D. L., 1997. The
thorium cosmochronometer in CS 22892–052: estimates of the
age of the galaxy. The Astrophysical Journal, 480, 246–254.
Fowler, W. A., and Hoyle, F., 1960. Nuclear cosmochronology.
Annals of Physics, 10, 280–302.

TEPHROCHRONOLOGY
David J. Lowe1 and Brent Alloway2
1
School of Science, Faculty of Science and Engineering,
Earth and Ocean Sciences, The University of Waikato,
Hamilton, New Zealand
2 clastic (fragmental) products of a volcanic eruption (Greek
School of Geography, Environment and Earth Sciences,
Victoria University of Wellington, Wellington,
New Zealand
Synonyms
Chronostratigraphy; Stratigraphic correlation
tephra; Tephrostratigraphy
Definitions
Tephra. All the explosively erupted, unconsolidated pyro-
clastic products of a volcanic eruption.
Cryptotephra. Tephra-derived glass-shard and/or crystal
concentration preserved in sediments or soils/paleosols
but not visible as a layer to the naked eye.
Tephrostratigraphy. Study of sequences of tephra layers or
cryptotephras (and associated deposits) and their litholo-
gies, spatial distribution, stratigraphic relationships, and
relative and numerical ages. Involves defining, describing,
characterizing, and dating tephra layers or cryptotephra
deposits in the field and laboratory as a basis for their
correlation.
Tephrochronometry. Obtaining a numerical age or calen-
drical date for a tephra or cryptotephra deposit.
Tephrochronology sensu stricto. The use of primary tephra
layers or cryptotephras as isochrons to connect and syn-
chronize depositional sequences, or soils, and to transfer
relative or numerical ages to the sequences or soils using
lithostratigraphic, compositional, and other data
pertaining to the tephras/cryptotephras.
J.W. Rink, J.W. Thompson (eds.), Encyclopedia of Scientific Dating Methods, DOI 10.1007/978-94-007-6304-3,
© Springer Science+Business Media Dordrecht 2015
T
Tephrochronology sensu lato. All aspects of tephra/
cryptotephra studies and their application.
Introduction and definitions
Tephras are the explosively erupted, unconsolidated pyro-
tephra, “ashes”) (Lowe, 2011). Typically they comprise
volcanic glass (including shards, pumice, and scoriae or
cinders), rock (lithic) fragments, and crystals (mineral
grains), which are erupted through the atmosphere and
deposited on the land, the seafloor, or ice caps relatively
using quickly – usually in a matter of hours or days according
to eruption duration (Lowe, 2011; Stevenson et al.,
2012). A tephra layer deposited from a powerful eruption,
and not reworked, consequently forms a widespread, thin
blanket on the surface of the Earth that has effectively the
same age – an isochron – wherever it occurs. The term
“tephra” encompasses all grain sizes: ash (grains <2 mm
in diameter), lapilli (64 to 2 mm), or blocks (angular) or
bombs (rounded) (>64 mm) (White and Houghton,
2006). Diminutive, distal tephras that are not visible as
layers in the field are called cryptotephras (Greek kryptein,
“to hide”). Cryptotephras comprise concentrations of
ash-sized glass shards or crystals, or both, usually
<150 mm in diameter, preserved in sediments (including
ice), in cave deposits, or in soils or paleosols (Lowe,
2011; Lane et al., 2014).
Tephrochronology may be defined as the use of primary
tephra layers or cryptotephra deposits as isochronous beds
to connect and synchronize depositional sequences and to
transfer relative or numerical ages (or dates) to such
sequences using stratigraphic information together with
lithological, compositional, and geochronological data
obtained for the tephras or cryptotephras.
Tephrochronology is thus a method for correlating and
dating geological, paleoecological, paleoclimatic, or
784 TEPHROCHRONOLOGY

Tephrochronology, Figure 1 Stratigraphic sequence of four Holocene tephra deposits of ash and lapilli grade, and buried soil
horizons, near Mt Tarawera, New Zealand. Exposure is ~2 m high. From top, Tarawera tephra (mainly Rotomahana Mud member)
(erupted 10 June 1886); shower-bedded Kaharoa tephra (AD 1314
12), characterized by abundant biotite; Taupo tephra
(AD 232
10) (note that the gray and very dark-brown soil horizons evident in the upper part of the nonwelded Taupo deposit, Taupo
ignimbrite, are the result of podzolization, an acidic soil-leaching process that occurred here prior to the soil’s burial by the Kaharoa
eruptives); and Whakatane tephra (5,526
145 cal. year BP) (Ages from Lowe et al., 2013) (Photo: Megan Balks).

archaeological sequences or events, or soils, using charac-
terized tephras or cryptotephras as chronostratigraphic
marker beds (Alloway et al., 2013).
Tephrochronology is undertaken typically in three
steps. Firstly, tephra deposits are correlated from place to
place in the field using their lithostratigraphic and relative
age relationships along with intrinsic physical properties,
associations with other deposits, and spatial distribution,
i.e., tephrostratigraphy (Figure 1). Fieldwork is carried
out at a range of scales because tephra deposits become
thinner, and constituent components become smaller,
from proximal to distal locations. A potential complica-
tion is tephra remobilization, which is discussed later. Sec-
ondly, the components of the deposits are characterized or
“fingerprinted” using mineralogical or geochemical
analytical methods, often both, in the laboratory to aid
their correlation. Such laboratory characterization is
essential in cryptotephra studies. Thirdly, when a numeri-
cal age or date for a tephra or cryptotephra deposit is
obtained, i.e., tephrochronometry, that age or date can be
transferred from one site to the next by comparing the
compositional characteristics or “fingerprints” of its com-
ponents with those of equivalent deposits elsewhere.
These three steps – (i) mapping tephras and determining
their lithostratigraphic relationships, (ii) characterizing
tephras/cryptotephras in the laboratory, and (iii) dating
them – enable tephrochronology to function as a unique
stratigraphic and geoscientific dating tool and are
described below in detail after a subsection on volcanolog-
ical applications.
 TEPHROCHRONOLOGY 785

Tephrochronology, Figure 2 (a) Road section in the vicinity of Lago Blanco, Chaitén sector (42 S), southern Chile. The stratigraphy at
this section reveals (from top) a prominent rhyolitic (c. 76 % SiO2) pumiceous lapilli deposit (Cha-1 correlative) dated at c. 8,700 14C
year BP and sourced from an ancestral Chaitén volcano, closely overlying a rhyolitic (c. 71 % SiO2) surge and fall couplet informally
named Lepué tephra. This lower rhyolitic eruptive couplet, dated at c. 9,800 14C year BP, is sourced from the nearby Michinmahuida
Volcanic Complex (MVC). The buried soil horizon (“andic paleosol”) marks the hiatus between the Lepué and Cha-1 eruptions.
(b) Low-angle cross-bedding and crosscutting relationship of the surge deposit across its co-eruptive fall deposit, Lepué tephra.
(c) The Michinmahuida-sourced eruptive couplet is closely underlain by a widespread and distinctive layer of banded high-Si rhyolite
breccia (indicated by arrows) that likely represents the products of an explosion of a pre-Cha-1 lava dome (ancestral Chaitén volcano).
Such sections are particularly important in terms of recognizing eruptives with different field expression and composition, as well as
for unraveling the complex and variable histories of closely situated (adjacent) eruptive centers (Photos: Brent Alloway).

Volcanological applications of tephrochronology
As well as providing isochrons for geological, archaeolog-
ical, and paleoenvironmental applications, tephrochro-
nological studies involving mapping, characterization,
and dating are critical for developing a comprehensive
record of past explosive eruptions and recurrence rates
from which time-space relationships of volcanism can be
established and volcanic hazards evaluated (e.g., Óladóttir
et al., 2008; Kuehn and Negrini, 2010). Knowledge of the
distribution and thickness relationships of tephra or
cryptotephra deposits enables magma volumes to be cal-
culated (Wilson et al., 2009; Ponomareva et al., 2013a).
Near to volcanic centers, deposits may include lavas as
well as an array of different types of pyroclastic deposits
(fall, surge, flow and density current, and explosion brec-
cias; Figure 2), and detailed studies of their stratigraphic
intercalation and physical, mineralogical, and geochemi-
cal properties provide insight into volcanic history and
petrogenesis (e.g., Ponomareva et al., 2004; Smith
et al., 2005; Turner et al., 2011a; Cioni et al., 2014). Infor-
mation about the release of volatiles into the atmosphere
and associated climatic impacts can also be gleaned from
studies on tephras (e.g., Zdanowicz et al., 1999; Alloway
et al., 2013; Sigl et al., 2013).
786 TEPHROCHRONOLOGY

Santa Barbara Puerto Cardenas

Section 9 Section 1
S 42 51’ 49.2” S 43⬚ 09’ 49.3”
W 72 47’ 12.2” W 72⬚ 25’ 44.8”
(m)
7 2008
(m) Chaiten
2
6
Cor-3B
5 Cor-3A

1
unnamed-2
4
Cor-2
Cor-1
3
(Lepué
tephra)
0
Till 2
unnamed-1

1
Yan-1
0
Till
Bedrock

Tephrochronology, Figure 3 Stratigraphic columns and corresponding section photos detailing the tephrostratigraphy of deposits
at two sites, Santa Barbara and Puerto Cardenas, in the Chaitén sector (42 S) of southern Chile. The tephra notation is from Naranjo
and Stern (2004). A key tephra of interest is the prominent decimeter-thick gray tephra layer (Cor-1; now more recently and informally
referred to as Lepué tephra) that contains conspicuous accretionary lapilli throughout. This tephra is the product of a large-scale
phreatomagmatic eruption sourced from the glaciated Michinmahuida Volcanic Complex (MVC) dated at c. 9,800 14C year BP. Tephra
from the 2008 eruption of Chaitén volcano occurs on the present-day ground surface. Other strongly pedogenically weathered
tephra beds represented within these sections have yet to be geochemically characterized and are probably variably sourced from
other nearby volcanoes (Puyuhuapi, Melimoyu, Yanteles, and Corcovado). The tephra sequences at both localities rest on glacial till
with a minimum age of c. 14,000 14C year BP (Photos: Brent Alloway).

Mapping and correlating tephras
The most successful approaches in the field include the
so-called hand-over-hand method whereby relatively
thick sequences of tephras, typically meter to decimeter
scale, at proximal or medial locations, are documented
and traced from one outcrop or section to the next using
distinctive physical properties in combination with their
stratigraphic associations (Figure 3). Physical properties
include color, bedding characteristics, or particle-specific
features such as pumice/scoria, lithic, and crystal
componentry, the presence of accretionary lapilli, or
marker mineral grains (crystals) such as biotite identifi-
able via a hand lens. Distinctive marker beds provide a
useful stratigraphic starting point in unraveling the com-
plexities of geological sequences seen in road cuttings or
natural exposures, and detailed lithostratigraphic columns
are constructed to provide the basis essential for correlat-
ing from one site to the next (Figure 3). The nature of bur-
ied soil horizons, loess, or other deposits associated with
tephra layers, including (for example) fossils or other
attendant paleoecological information such as pollen
assemblages (e.g., Newnham and Lowe, 1999; Housley
et al., 2013; Westgate et al., 2013a) or archaeological rela-
tionships (Feibel, 1999; Mullen, 2012; Riede and
Thastrup, 2013), may provide additional contextual
information helpful for affecting correlation (Figure 4).
High-resolution tephrostratigraphic records in Iceland
were constructed by Streeter and Dugmore (2013) using
digital photography to obtain thousands of stratigraphic
measurements of multiple tephra layers intercalated with
sediments at a resolution of 1 mm.
One potential difficulty in tephrostratigraphy arises
where tephra remobilization has occurred or where glass
shards or crystals have been disseminated or dispersed in
sediments or soils, and hence, primary isochron positions
are not clear (Alloway et al., 2004a; Pyne-O’Donnell,
2011; Housley et al., 2012). Remobilization of tephra
resulting from widespread landscape instability in combi-
nation with climatic factors (e.g., Manville and Wilson,
2004) can remobilize large volumes of tephra for decades
in the aftermath of a paroxysmal eruption (e.g., following
the 1991 Pinatubo eruption; Torres et al., 2004).
Remobilization of tephra can also be triggered by human
activities (Swindles et al., 2013). Such reworked or dis-
seminated tephra-derived constituents, including glass
shards, form diachronous rather than isochronous sur-
faces, and hence, their use as stratigraphic tie points is
compromised unless remobilization is very localized or
near contemporaneous with the primary depositional
event. The non-reworked part of a tephra deposit provides
 TEPHROCHRONOLOGY 787

Tephrochronology, Figure 4 Stratigraphy associated with hominin stone artifacts and vertebrate fossil remains from the So’a Basin,
central Flores, Indonesia (unpublished data of B. V. Alloway, A. Brumm, G. van den Bergh, and R. Setiawan; after O’Sullivan et al. (2001)
and Brumm et al. (2010)). The in situ stone artifacts occur immediately beneath a nonwelded (unconsolidated) ignimbrite deposit
(Wolo Sege ignimbrite, WSI) that can readily be correlated and distinguished from other enveloping silicic tephra interbeds across the
So’a Basin on the basis of its unique depositional architecture as well as its major-element glass-shard composition. The WSI, dated at
1.02
0.02 Ma by 40Ar/39Ar (Brumm et al., 2010), provides important evidence for the presence of hominins on Flores for over
1 million years. (a) The base of WSI at the Wolo Sege section showing distinctive surge-like subunits containing accretionary lapilli
and bedding consistent with the density segregation of entrained particles. Note the in situ bifacial flake in the paleosol immediately
beneath the base of the WSI (indicated by arrow). (b) Close-up photo of the in situ bifacial flake identified at Wolo Sege. (c) In situ
flaked core stone identified at the Matago section on the paleo-ground surface buried by the WSI. (d) Close-up photo showing the
detail of the flaked core stone from the Matago section (Photos a–c, Brent Alloway; Photo d, Adam Brumm).

an isochron of maximum age (the date of the tephra erup-
tion and primary deposition), but any reworked compo-
nents are younger (Lowe, 2011).
Field correlation methods are increasingly limited as
the tephra layers become thinner and finer with increasing
distance away from source (e.g., Figure 5), and the tephras
may become more irregularly distributed (e.g., Turney and
Lowe, 2001; Lawson et al., 2012). The tephras typically
lose diagnostic physical features (such as internal bed-
ding) in subaerial sequences, and thin, separate beds can
become mixed together by soil-forming processes includ-
ing bioturbation or by cryoturbation in periodically or
perennially frozen landscapes (Sanborne et al., 2006;
Lowe and Tonkin, 2010; Dugmore and Newton, 2012).
Cores taken from lake sediments and peat bogs potentially
provide a reliable means to record thin and/or fine-grained
distal tephra layers, or at proximal or medial locations
where tephras tend to be thicker and may have more
restricted dispersal patterns around source vents – for
example, in basaltic volcanic fields (Shane and Hoverd,
2002). These lake or bog depositional environments allow
thin tephras or glass-shard concentrations (cryptotephras)
to be preserved amidst organic-bearing sediments
(Figure 6) (de Fontaine et al., 2007; Smith et al., 2013),
notwithstanding potential taphonomic difficulties posed
by tephra reworking or dispersion (Lowe, 2011;
788 TEPHROCHRONOLOGY
Tephrochronology, Figure 5 A distal occurrence of fine-
grained Mazama tephra, 1–2 cm in thickness and composed
mainly of glass shards, shows up as the pale “break” (where the
person is touching) in an alluvium-soil sequence on the
lowermost terrace alongside the North Saskatchewan River in
Edmonton, Alberta, Canada. Multiple buried soils formed on
intermittent flood deposits are evident in the section, with the
high sedimentation (up-building) rate contributing to the rapid
burial and preservation of the thin, fine-grained Mazama tephra
as an inconspicuous layer. The Mazama tephra, dated at
7,627
150 cal. year BP (Zdanowicz et al., 1999), provides a key
isochronous time plane (chronostratigraphic marker bed) for
Holocene archaeological and paleoenvironmental studies in the
USA and southwestern Canada. Located ~1,450 km northeast of
its source (Crater Lake, Oregon), the tephra was first identified
here by Westgate et al. (1969) (Photo: Maria Lowe).
Liu et al., 2014). Marine cores, especially from locations
downwind of eruption centers, may also contain detailed
records of tephra layers or cryptotephras intercalated
with sediments (e.g., Guðmundsdóttir et al., 2012;
Abbott et al., 2013; Ponomareva et al., 2013b; Austin
et al., 2014). Possible reworking of tephras (by currents,
ice rafting, bioturbation) adds complexity in some cases
(Brendryen et al., 2010; Larsen et al., 2014; Todd
et al., 2014).
Together with those from lakes and bogs, many of the
records of marine tephra deposits can thus document pat-
terns of explosive volcanism in time and space and integrate
the stratigraphic interfingering of eruptives from multiple
volcanic sources (Shane et al., 2006; Swindles et al., 2011;
Ponomareva et al., 2013a, Ponomareva et al., 2013b).
Tephrochronology, Figure 6 Close-up of a near-basal part of a
lake sediment core from Lake Okoroire, New Zealand, showing
(in middle of photo) 3-cm-thick Rerewhakaaitu tephra, erupted
17,496
462 cal. year BP from Mt Tarawera in the Okataina
Volcanic Centre (Lowe et al., 2013; see Figure 8 below). Many tiny
black specks in the pale layer represent free biotite crystals
diagnostic of this (and some other) tephras derived from Mt
Tarawera. Different sediment colors above and below the tephra
reflect a climatically driven change from shrubland/grassland to
forest in the region soon after the eruption (Newnham
et al., 2003; Alloway et al., 2007). An indistinct uncorrelated
tephra lies ~2 cm above the Rerewhakaaitu tephra. Scale marks
in centimeters (Photo: David Lowe).
Such compilations are often more comprehensive than
those obtainable near volcanic centers because of burial or
erosion of eruptives at such proximal locations, but an
important caveat is that proximal deposits are typically more
compositionally variable than distal deposits (e.g., Smith
et al., 2005; Shane et al., 2008a). On this basis, distal
deposits usually have a more restricted compositional range
than otherwise might be generated during a particular erup-
tive episode.
Distal tephras and cryptotephra deposits
Submillimeter-scale studies have involved the develop-
ment of new approaches to enable cryptotephras, and very
thin distal tephras, to be mapped, albeit discontinuously,
across landscapes using a range of methods firstly to
detect and isolate glass-shard or crystal concentrations
(e.g., Hall and Pilcher, 2002; Gehrels et al., 2006; Swin-
dles et al., 2010; Lane et al., 2014) and secondly to
 TEPHROCHRONOLOGY 789

correlate them with known (previously characterized) Tephrochronology, Table 1 Analytical methods (excluding
deposits using geochemical analyses of glass shards, melt geochronology) used to characterize glass or minerals in
inclusions (glass) within crystals, or crystals (Matsu’ura tephras to facilitate their correlation (After Lowe, 2011)
et al., 2011; Swindles et al., 2013) as described below. Tephra components and properties Main methods of analysisa
Cryptotephras have been discovered in peat, lake, marine,
aeolian, or frozen sediments or ice, in deposits in caves or Glass shardsb
rock shelters, and in soils or paleosols. They (and thin vis- Major and minor elementsc Electron microprobe
ible tephras) have been detected using ground-penetrating Trace elementsc including rare earths LA- or SN-ICP-MS,
radar, magnetic susceptibility and remanent magnetiza- INAA, SIMS
Sr, Nd, and Pb isotopes LA-ICP-MS, multi-
tion, X-radiography, X-ray fluorescence (including use collector NMS
of scanners), spectrophotometry, micro-petrography, enu- Shard morphology Optical microscope, SEM
meration of glass shards, and measurements of total Melt inclusions (glass)
organic carbon and loss on ignition (Turney and Lowe, Major and minor elements Electron microprobe
2001; Gehrels et al., 2008; Lawson et al., 2012). These Trace elements SIMS, LA-ICP-MS
novel cryptotephra studies, although not without problems Ferromagnesian silicate minerals
Assemblages or marker minerals Petrographic microscope,
and limitations (e.g., Davies et al., 2007; Payne and XRD
Gehrels, 2010; Swindles et al., 2011), have documented Pyroxenes, amphiboles, olivine, or Electron microprobe
tephra-fall occurrences at submillimeter scales at distances biotite crystals/phenocrysts
of hundreds to some thousands of kilometers from source, Crystal geometry Optical microscope, SEM
greatly extending known geographical limits (e.g., Payne Apatite, zircon
d
et al., 2008; Davies et al., 2010; Balascio et al., 2011; Apatite or zircon crystals/phenocrysts Electron microprobe,
Davies et al., 2012; Cullen et al., 2014). Pyne-O’Donnell TIMS-TEA
Crystal geometry Optical microscope, SEM
et al. (2012) showed that “ultra-distal” cryptotephras Fe-Ti oxides
derived from Holocene eruptions in Alaska and the Pacific Major and minor elements Electron microprobe
Northwest occur up to ~7,000 km from source in eastern- Eruption temperatures and oxygen Electron microprobe
most North America and one tephra, the Alaskan White fugacities
River Ash (eastern lobe, erupted around AD 860), occurs Feldsparsd
over ~7,000 km away in Ireland (Jensen et al., 2014). Lane Plagioclase, anorthoclase, or sanidine Electron microprobe
crystals/phenocrysts
et al. (2013) identified the c. 75,000-year-old Youngest
Toba Tuff tephra as a cryptotephra deposit in Lake Malawi a
LA- or SN-ICP-MS laser ablation or solution nebulization induc-
sediments in east Africa, >7,000 km west of the source tively coupled plasma mass spectrometry, INAA instrumental neu-
volcano in Sumatra. Cryptotephra associated with the tron activation analysis, SIMS secondary ionization mass
eruption of the Campanian Ignimbrite in Italy c. 40,000 spectrometry (ion probe), NMS nuclide mass spectrometry, SEM
calendar (cal.) year BP was identified across much of scanning electron microscopy, XRD X-ray diffraction, TIMS-TEA
thermal ionization mass spectrometry-trace element analysis
central Europe and the Mediterranean area by b
May also include glass coats (selvedges) or matrix glass in pumice
Lowe et al. (2012). As noted earlier with regard to visible clasts
tephra deposits, cryptotephra studies can help also in elu- c
Major elements expressed as oxides usually are defined as
cidating the eruptive history of volcanoes (e.g., Shane >1 wt%, minor element oxides as 0.1–1 wt%, and trace elements
et al., 2013). as <0.1 wt% or <1,000 parts per million (ppm) of the element
Ice cores provide detailed and valuable records of vol- (not oxides)
d
canism (Davies et al., 2010; Dunbar and Kurbatov, 2011; Analyses of these minerals generally are less useful for correlating
individual tephras
Abbott and Davies, 2012; Coulter et al., 2012). Until
recently, analytical limitations of geochemical techniques
have hindered adequate characterization of ultrafine glass
particles (<5 mm) and identification of their eruptive out in conjunction with lithostratigraphic and chronologi-
source. Similarly, sulfate records in ice cores may not cal criteria to obtain cogent correlations (Alloway
always act as suitable proxies for the occurrence of tephra et al., 2013).
or cryptotephra deposits composed of (typically sparse)
glass shards in the cores (Davies et al., 2010; Abbott and Ferromagnesian silicate mineral assemblages
Davies, 2012). A common method is to use a petrographic microscope to
identify ferromagnesian or mafic mineralogical assem-
blages where such minerals are abundant, usually at prox-
Characterizing tephras and cryptotephras in the imal or medial sites. These minerals can be extracted using
laboratory a magnetic separator together with nontoxic heavy liquids
The characterization, or “fingerprinting,” of mineral and such as sodium polytungstate. With stratigraphic and age
glass components of tephras or cryptotephras can be constraints, the relative abundances of ferromagnesian
undertaken using a range of analytical methods minerals may allow a source volcano to be identified.
(Table 1). Such characterization is almost always carried For example, in the Yukon Territory of Canada,
790 TEPHROCHRONOLOGY
Preece et al. (2000, 2011a) showed that tephras erupted
from the Aleutian arc-Alaska Peninsula, so-called Type
I beds, had low crystal contents comprising mainly pyrox-
ene (and mainly bubble-wall shards), whereas tephras
derived from the Wrangell volcanic field and Hayes vol-
cano, Type II beds, generally had high crystal contents
comprising mainly hornblende (and mostly pumiceous
shards). In New Zealand, orthopyroxene (mainly hyper-
sthene) is dominant in tephras erupted from the Taupo Vol-
canic Centre since c. 30,000 cal. year BP, whereas biotite,
hornblende, cummingtonite, or orthopyroxene predomi-
nate in tephras erupted from the Okataina Volcanic Centre
over the same period (Lowe et al., 2008).
In some cases an individual tephra can be identified
using distinctive, diagnostic marker minerals, such as
aegirine and aenigmatite in the Tuhua Tephra that were
erupted from peralkaline Mayor Island volcano
c. 7,000 cal. year BP in New Zealand. However, the
absence of diagnostic minerals does not necessarily negate
correlation because minerals such as olivine, biotite, and
hypersthene can be dissolved comparatively quickly in
some very acid peat bogs (within 700 years; Hodder
et al., 1991) or in soil-forming environments (Lowe,
1986; Churchman and Lowe, 2012). Ferromagnesian min-
erals and Fe-Ti oxides also tend to be sparse or absent at
distal localities, dropping out from proximal or medial
ash clouds earlier because of their high density (Juvigné
and Porter, 1985).
Although of limited application, the crystal geometry
(crystal width) of apatite (a phosphate-group mineral)
was shown to be useful, along with apatite trace-element
data, for differentiating beds in a study of Late Ordovician
K-bentonites in the USA by Sell and Samson (2011). Sim-
ilarly, Donoghue et al. (1991) showed that two distinct
crystal geometries of olivine (skeletal, non-skeletal) were
useful in correlating some andesitic tephras in New
Zealand.
Major-element analysis of glass shards, melt
inclusions, and minerals
The electron microprobe enables a range of tephra compo-
nents to be analyzed for major elements on a grain-by-
grain basis: individual glass shards, glassy coatings on
crystals, melt inclusions (glass preserved within crystals
including quartz, pyroxenes, amphiboles), pumice frag-
ments, and loose crystals or phenocrysts including various
ferromagnesian silicate minerals, apatite, zircon, and
Fe-Ti oxides such as titanomagnetite (Table 1). The main
advantage of the microprobe and other single-grain tech-
niques is that they allow mixed or heterogeneous
populations to be identified. Because volcanic glass is
amorphous, its analysis by microprobe needs especially
careful sample preparation and mounting, the use of
appropriate standards, and optimum probe-operating con-
ditions to derive accurate and robust data (Kuehn
et al., 2011; Hall and Hayward, 2014; Pearce et al.,
2014; Suzuki et al., 2014).
Usually a defocused (e.g., 10–20 mm) or rastered beam
is deployed to minimize the mobilization of Na, although
protocols that use narrower beam diameters (3–5 mm)
without loss of Na have been recently developed for ana-
lyzing fine-grained glass shards, as occur in distal or
ultra-distal tephras or cryptotephras or glasses with
microlites (Hayward, 2012; Pyne-O’Donnell et al., 2012;
Hall and Hayward, 2014). Glass microprobe analyses are
usually normalized (summed to 100 %, most, but not all,
of the deficit being attributable to water) to enable useful
comparisons of analyses (Shane, 2000; Lowe, 2011;
Westgate et al., 2013a). Using stratigraphic or age con-
straints, and usually also a knowledge of geochemical
affinities of potential source rocks/deposits and spatial dis-
tribution patterns, the microprobe analysis of glass may
allow tephra source volcanoes to be identified, and indi-
vidual eruptives may also be correlated in some cases
where compositions through time have changed from
one eruptive event to the next. Complexities arise where
glass analyses show heterogeneity, a consequence poten-
tially of (i) mingling or mixing of separate batches of
magmas that were tapped simultaneously or sequentially
as eruptions proceeded (e.g., Tryon et al., 2010; Turner
et al., 2011b), (ii) major-element evolution by fractional
crystallization or crustal assimilation (Bogaard and
Schminke, 1985; Ukstins Peate et al., 2008; Óladóttir
et al., 2012), (iii) blending of thin tephras in soil-forming
or cryogenic environments (Lowe, 1986; Eden
et al., 2001; Dugmore and Newton, 2012), or (iv) post-
depositional dissemination of glass shards in peat, lake,
or marine sediments (Gehrels et al., 2006; Allan
et al., 2008; Abbott et al., 2013). Heterogeneity arising
from magmatic or volcanic eruption processes accompa-
nied by changes in wind direction warns of the difficulty
of characterizing (thus fingerprinting) tephra beds using
a limited set of distal samples from restricted dispersal sec-
tors (Shane et al., 2008a). Another problem can arise with
analyses of melt inclusions where these do not reflect the
full compositional range of glass shards or matrix glass
(e.g., Shane et al., 2008b; Chesner and Luhr, 2010; Allan
et al., 2013) or they show a pattern different from that
associated with matrix glass analyses (e.g., Kilgour
et al., 2013). Once recognized, such compositional vari-
ability can enhance correlation by providing additional
fingerprinting criteria, but sampling from the full spatial
and temporal range of deposits of an eruptive sequence
is required (Alloway et al., 2013).
The correlation of andesitic or basaltic tephras using
glass chemistry generally can be complicated for various
reasons including the multiplicity of units, the paucity of
suitable glass for microprobing (shards may contain
micro-inclusions and can be highly vesicular), susceptibil-
ity of glass to weathering, and wide compositional ranges
and potential heterogeneity as noted previously (Moebis
et al., 2011; Shane and Zawalna-Geer, 2011; Óladóttir
et al., 2012). Platz et al. (2007) provided a procedure to
evaluate glass compositional variability arising from the
inadvertent microanalysis of plagioclase microlites in
 TEPHROCHRONOLOGY
shards using a wide beam, but the recent use of narrower
beam diameters is helping to obviate this problem
(Hayward, 2012; Pearce et al., 2014).
Analyses by microprobe of ferromagnesian silicate
minerals, such as biotite (Shane et al., 2003; Smith
et al., 2011) and cummingtonite (Matsu’ura et al., 2012),
apatite (a phosphate-group mineral) (Sell and Samson,
2011), and Fe-Ti oxides such as titanomagnetite and
ilmenite (Shane, 1998; Preece et al., 2011b; Marcaida
et al., 2014), have also been used for tephra fingerprinting.
In some cases, and after taking into account compositional
zoning within crystals, trace elements such as Mg, Cl, Mn,
Fe, Ce, and Y identified in apatite crystals and phenocrysts
provide a means of distinguishing or matching tephras
(Sell and Samson, 2011). The eruption temperature and
oxygen fugacity (oxidation state of magma) of rhyolitic
tephras – estimated using single-grain EMPA of Fe-Ti
oxide pairs of titanomagnetite and ilmenite – provide
another way to correlate tephras and, in some cases,
magma batches within an eruptive sequence (Smith
et al., 2005; Preece et al., 2011b; Marcaida et al., 2014).
Trace-element and isotope analysis of glass
Although the microprobe is now the key analytical tool for
undertaking major-element analyses of glass, tephras or
cryptotephras cannot always be distinguishable uniquely
by major-element data alone. In these cases, other analyt-
ical methods are needed (Westgate et al., 2013c). Trace-
element analyses of glass separates from tephra or
cryptotephra deposits offer a greater range of elements
for use in correlation/provenance studies and may also
provide additional information on volcanic petrogenesis.
In the last decade, trace-element techniques have become
more common, the most widely available being laser abla-
tion inductively coupled plasma mass spectrometry
(LA-ICP-MS) (e.g., Albert et al., 2012; Ponomareva
et al., 2013a; Sulpizio et al., 2013) (Table 1). Pearce
(2014) provides protocols for the application of
LA-ICP-MS to glass shards. Around 150 individual glass
shards can be analyzed for about 30 trace elements in
1 day using LA-ICP-MS (Pearce et al., 2011). Compre-
hensive studies using this method were undertaken on
Pleistocene tephras in northern New Zealand (Pearce
et al., 2008b) and in marine cores from Ocean Drilling
Project (ODP) Site 1123 (Allan et al., 2008). The most
useful elements for correlating and distinguishing
between the ODP Site 1123 tephras were found to be the
abundances of Rb, Ba, Sr, Y, Zr, Hf, Mg, Mn, and Ti and
trace-element ratios including Rb/Sr, Ba/Sr, Zr/Y, Y/Th,
Ba/Th, and Rb/Sm. Using trace-element data for glass
shards, Allan et al. (2008) demonstrated that (i) tephras
with similar major-element compositions (e.g.,
Kawakawa tephra vs. Omataroa tephra) were easily distin-
guishable (Figure 7) and (ii) two previously unidentified
repeated sections of ODP cores 1123A (~4.5 m thick)
and 1123C (~7.9 m thick) were recognized (Allan
et al., 2008). As with the microprobe, there is a possibility
791
of microlites affecting trace-element characterizations of
individual shards via LA-ICP-MS, and so anomalous
assays need to be evaluated carefully (Abbott et al., 2013).
Analyses of isotopes of Sr, Nd, and Pb in glass
(or pumices) have been utilized in several studies to help
determine tephra source volcanoes (e.g., Roulleau
et al., 2009; Westgate et al., 2008, 2011; Giaccio
et al., 2013).
Comparing compositional data using graphical,
numerical, and statistical methods
Major- or trace-element datasets obtained from analyses
of glass or minerals may be displayed and compared,
and hence compositional variability evaluated, using vari-
ous methods: graphical (e.g., bivariate or trivariate plots),
numerical (e.g., similarity coefficients), or statistical (e.g.,
statistical distance measure, dendrograms, discriminant
function analysis, principal components analysis)
(Pearce et al., 2008a; Bourne et al., 2010; Lowe, 2011; Sell
and Samson, 2011). In many cases, the bivariate plots
alone can provide sufficient guidance to enable anomalous
data points to be identified (and possibly explained using
compositional databases for comparison) and potential
correlations or otherwise established with reasonable
likelihood, especially where multiple criteria provide
independent support, such as concordance of glass major-
and trace-element data together with Fe-Ti oxide data
(e.g., see Preece et al., 1999, 2011b). In other situations,
however, statistical methods such as principal components
analysis can allow units to be distinguished objectively on
the basis of multiple oxides or elements (not just two or
three) analyzed from glass shards or crystals (e.g., Tryon
et al., 2010; Chiasera and Cortés, 2011; Sell and Samson,
2011).
Dating tephras (tephrochronometry)
Tephras may be dated directly using primary minerals
(e.g., zircon, hornblende, K-feldspars, biotite, quartz) or
glass from within the tephra layer, or indirectly on either
enclosing or encapsulated material, using a range of
methods including radiometric, incremental (e.g., varves,
layering in ice), and age equivalence (Table 2) (Svensson
et al., 2008; Alloway et al., 2013; Zalasiewicz
et al., 2013). These ages can then be transferred from
one site to another where the tephra is recognized using
the lithostratigraphic and compositional-based methods
described above. Only the main radiometric methods,
and age modeling, are touched on here.
For tephras erupted within the past c. 60,000 calendar
years, the radiocarbon (14C) technique (e.g., Hogg
et al., 2007) remains the most important method for devel-
oping calibrated age models. In recent years, the advan-
tages of using pollen concentrates and terrestrial plant
macrofossils (e.g., leaves, twigs), rather than bulk organic
samples, as reliable dating materials via AMS have been
demonstrated (Newnham et al., 2007; Staff et al., 2011).
Together with dendrochronological wiggle-matching
792 TEPHROCHRONOLOGY

Tephrochronology, Figure 7 Bivariate plots for some major and trace elements derived from analyses of individual glass shards from
two New Zealand tephras, Omataroa (c. 31,600 cal. year BP, erupted from Okataina Volcanic Centre) and Kawakawa (c. 25,400 cal. year
BP, erupted from Taupo Volcanic Centre), identified in marine cores (a), (b), and (c) from ODP Site 1123 about 1,200 km east of New
Zealand (Modified after Allan et al., 2008, p. 2351, with permission of Elsevier). (a) CaO versus FeOt (total iron expressed as FeO)
derived by electron microprobe. Glass analyses from an onshore occurrence of Kawakawa tephra (at Irirangi) are also shown for
comparison. The plot shows that the analyses of marine and onshore samples of Kawakawa tephra are identical and that Kawakawa
and Omataroa tephras cannot be distinguished using these two oxides alone. In contrast, trace-element concentrations (in ppm),
derived by LA-ICP-MS, in (b) and (c), show that the tephras are distinctly different with respect to their elements/element ratios and
hence can be readily distinguished.

methods (Hogg et al., 2012; Yin et al., 2012), Bayesian
flexible depositional age modeling has added a revolution-
ary aspect to the construction of enhanced and more pre-
cise chronologies in tephrochronology involving 14C and
other dating methods (Blockley et al., 2008; Smith
et al., 2013). Lowe et al. (2013) dated late Quaternary
tephras in New Zealand by modeling 14C age data using
two Bayesian-based programs, Bacon (Blaauw and
Christen, 2011) and OxCal’s P_Sequence function
(Bronk Ramsey, 2009) and the IntCal09 dataset
(Figure 8). Lohne et al. (2013) also used P_Sequence
and IntCal09 to obtain high-precision ages of 12,066

42 and 10,210
35 cal. year BP (
1s) for the Vedde
and Saksunarvatn tephras, respectively, from sediments
in Kråkenes Lake, Norway. Similarly, the AT tephra of
Japan was dated to 30,009
189 cal. year BP (95 %
probability) using P_Sequence modeling of 14C data and
varves (Smith et al., 2013). Vandergoes et al. (2013) used
OxCal’s Tau_Boundary function, also Bayesian, to pre-
cisely date the Kawakawa/Oruanui tephra of New Zealand
to 25,358
162 cal. year BP (95 % probability).
Another key advance in tephrochronology has been the
development of the isothermal-plateau fission-track dat-
ing method (ITPFT) for glass (Westgate, 1989; Alloway
et al., 2004b, 2013; Westgate et al., 2013b). ITPFT and
the diameter-corrected fission-track method (Sandhu and
Westgate, 1995) have enabled ages to be obtained on
many distal vitric-rich tephras that previously were unable
to be dated because of low abundance of dateable mineral
constituents, fine grain size, and the presence of detrital
grains. Examples of such applications include dating Qua-
ternary glacioeustatic sedimentary cycles in the Wanganui
 TEPHROCHRONOLOGY
Tephrochronology, Table 2 Methods used for dating tephras
directly or indirectly (after Lowe, 2011)
Main method Applications
Radiometric Radiocarbon dating (radiometric/beta counting,
AMS)a
Fission-track dating of zircon or glass-ITPFT or
glass-DCFT dating
Argon isotopes (K/Ar, Ar/Ar including
SCLP/F, LIH)
Luminescence dating (TL, OSL, IRSL,
pIR-IRSL)
U-series including (U-Th)/He, U-Pb, and
238 230
U/ Th zircon dating (SIMS/TIMS,
SHRIMP, LA-ICP-MS)
Electron spin resonance
210
Pb, 137Cs, 3He, and 21Ne surface exposure
dating
Incremental Dendrochronology, varve chronology, layering
in ice cores (ice sheets/caps, glaciers)
Age equivalence Magnetopolarity, paleomagnetic secular
variation, astronomical (orbital) tuning,
correlation with marine oxygen isotope
stages, climatostratigraphy, biostratigraphy,
palynostratigraphy, paleopedology
Age modeling Various age-depth methods including Bayesian
flexible depositional modeling and wiggle
matching, spline-fit modeling
Relative Obsidian hydration dating, amino acid
racemization
Historical Eyewitness accounts or observations (e.g., via
remote sensing)
a
AMS accelerator mass spectrometry, ITPFT isothermal-plateau
fission track, DCFT diameter-corrected fission track, SCLP/F
single-crystal laser probe or fusion, LIH laser incremental heating,
TL thermoluminescence, OSL optically stimulated luminescence,
IRSL infrared stimulated luminescence, pIR-IRSL post infrared-
infrared-stimulated luminescence, SIMS secondary ionization mass
spectrometry, TIMS thermal ionization mass spectrometry, SHRIMP
sensitive high-resolution ion microprobe, LA-ICP-MS laser ablation
inductively coupled plasma mass spectrometry
Basin (Alloway et al., 1993; Pillans et al., 2005) and dat-
ing initial loess deposition in Alaska at ~3 million years
ago (Westgate et al., 1990). The glass-FT techniques have
been used, for example, also to test chronologies based on
alternative methods (such as magnetic polarity and astro-
nomical tuning) for marine tephra sequences (Alloway
et al., 2005; Allan et al., 2008) and for fossiliferous allu-
vial and lacustrine sediments in Beringia (Preece
et al., 2011b; Westgate et al., 2013a).
Where suitable minerals are available (e.g., sanidine,
biotite, leucite, anorthoclase), the 40Ar/39Ar method has
been useful, such as dating the Laacher See tephra in
Germany (Bogaard, 1995); the ultra-distal to distal
deposits of the Youngest Toba Tuff tephra of Indonesia
(Westgate et al., 1998; Storey et al., 2012); a widespread
Holocene tephra erupted from Ulleungdo stratovolcano
in South Korea (Smith et al., 2013); a 400,000-year-old
793
sequence of eruptives from Nemrut volcano preserved in
Lake Van, eastern Anatolia in Turkey (Sumita and
Schminke, 2013); and the distal tephras preserved in
lacustrine and fluvial sediments of intermontane basins
of central Italy (Giaccio et al., 2013). A relatively new
method for dating proximal pyroclastic deposits, previ-
ously applied to petrologic studies, is the use of U-Pb ana-
lyses to date zircons (Schmitt, 2006; Dickinson
et al., 2010; Wilson et al., 2010). Luminescence and
(U-Th)/He dating methods are also being applied system-
atically to tephra deposits (Danišík et al., 2012; Biswas
et al., 2013).
A trend in dating lake sediment sequences containing
tephras or cryptotephras is to apply multiple methods, as
exemplified by Staff et al. (2013) and Sirocko
et al. (2013).
Conclusions
Tephrochronology is the use of primary tephras or
cryptotephras as isochrons to link and synchronize geo-
logical, paleoenvironmental, or archaeological sequences
or events, or soils, and, uniquely, to transfer relative or
numerical ages to them using stratigraphic information
and mineralogical and geochemical compositional data,
especially from individual glass-shard analyses, obtained
for the tephras/cryptotephras. To function therefore as an
age-equivalent correlation and chronostratigraphic dating
tool, tephrochronology can be undertaken in three steps:
(i) mapping and describing tephras and determining
their stratigraphic relationships, (ii) characterizing tephras
or cryptotephras in the laboratory, and (iii) dating them
using a wide range of geochronological methods.
Tephrochronology is also an important tool in volcanol-
ogy, informing studies on volcanic petrology, volcano
eruption histories and hazards, and volcano-climate forc-
ing. Although limitations and challenges remain, multidis-
ciplinary applications of tephrochronology continue to
grow markedly (e.g., Alloway et al., 2013; Lane
et al., 2014).
Acknowledgments
We thank editors Jeroen Thompson and Jack Rink for
inviting us to write this entry and for their suggestions,
reviewer Vera Ponomareva for her helpful comments,
Megan Balks and Adam Brumm for providing photo-
graphs, and Chris Hayward and John Westgate for furnish-
ing preprints. Elsevier kindly allowed us to use previously
published material (Figures 7 and 8). The entry was
funded in part by the New Zealand Marsden Fund
(project 10-UOW-056 to DJL entitled “New views from
old soils”), administered by the Royal Society of New
Zealand, and it is an output also of the INTREPID
Tephra-II project (INQUA project 1307s) “Enhancing
tephrochronology as a global research tool through
improved fingerprinting and correlation techniques and
uncertainty modeling (phase II),” an initiative of the Inter-
national Focus Group on Tephrochronology and
794 TEPHROCHRONOLOGY

Rotoma

Poronui

Karapiti
Waiohau Okupata
Opepe
Konini

Rerewhakaaitu Rotorua

18000 16000 14000 12000 10000

cal. yr BP
Tephrochronology, Figure 8 Bayesian-derived age models (95 % probability) for nine late Quaternary tephras in New Zealand (From
Lowe et al., 2013, p. 179, with permission of Elsevier). Probability plots are colored according to tephra source volcanoes: red,
Okataina; orange, Taupo; green, Taranaki; and blue, Tongariro. Gray plots show the start and end ages of the late glacial cool episode,
designated climate event NZce-3 by Barrell et al. (2013), in part constrained chronologically by the ages on the adjacent tephras.

Volcanism (INTAV) supported by the Stratigraphy and
Chronology Commission of the International Union for
Quaternary Research (INQUA).
Bibliography
Abbott, P. M., and Davies, S. M., 2012. Volcanism and the Green-
land ice-cores: the tephra record. Earth-Science Reviews,
115, 173–191.
Abbott, P. M., Austin, W. E. N., Davies, S. M., Pearce, N. J. G., and
Hibbert, F. D., 2013. Cryptotephrochronology of the Eemian and
the last interglacial-glacial transition in the north-east
Atlantic. Journal of Quaternary Science, 28, 501–514.
Albert, P. G., Tomlinson, E. L., Smith, V. C., Di Roberto, A.,
Todman, A., Rosi, M., Marani, M., Muller, W., and Menzies,
M. A., 2012. Marine-continental tephra correlations: volcanic
glass geochemistry from the Marsili Basin and the Aeolian
Islands, southern Tyrrhenian Sea, Italy. Journal of Volcanology
and Geothermal Research, 229–230, 74–94.
Allan, A. S. R., Baker, J. A., Carter, L., and Wysoczanksi, R. J.,
2008. Reconstructing the Quaternary evolution of the world’s
most active silicic volcanic system: insights from a ~1.65 Ma
deep ocean tephra record sourced from the Taupo Volcanic Zone,
New Zealand. Quaternary Science Reviews, 27, 2341–2360.
Allan, A. S. R., Morgan, D. J., Wilson, C. J. N., and Millet, M.-A.,
2013. From mush to eruption in centuries: assembly of the super-
sized Oruanui magma body. Contributions to Mineralogy and
Petrology, 166, 143–164.
Alloway, B. V., Pillans, B. J., Sandhu, A. S., and Westgate, J. A.,
1993. Revision of the marine chronology in Wanganui Basin,
New Zealand, based on the isothermal plateau fission-track dat-
ing of tephra horizons. Sedimentary Geology, 82, 299–310.
Alloway, B. V., Westgate, J. A., Pillans, B. J., Pearce, N. J. G.,
Newnham, R. M., Bryami, M., and Aarburg, S., 2004a. Stratig-
raphy, age and correlation of middle Pleistocene silicic tephras
in the Auckland region, New Zealand: a prolific distal record
of Taupo Volcanic Zone volcanism. New Zealand Journal of
Geology and Geophysics, 47, 447–479.
Alloway, B. V., Pribadi, A., Westgate, J. A., Bird, M., Fifield, K. L.,
Hogg, A. G., and Smith, I. E. M., 2004b. Correspondence
between glass-FT and 14C ages of silicic pyroclastic flow
deposits sourced from Maninjau caldera, west-central Sumatra.
Earth and Planetary Science Letters, 227, 121–133.
Alloway, B. V., Pillans, B. J., Carter, L., Naish, T., and Westgate,
J. A., 2005. Onshore-offshore correlation of Pleistocene rhyolitic
eruptions from New Zealand: implications for TVZ eruptive his-
tory and paleoenvironmental construction. Quaternary Science
Reviews, 24, 1601–1622.
Alloway, B. V., Lowe, D. J., Barrell, D. J. A., Newnham, R. M.,
Almond, P. C., Augustinus, P. C., Bertler, N. A., Carter, L.,
Litchfield, N. J., McGlone, M. S., Shulmeister, J., Vandergoes,
M. J., Williams, P. W., and NZ-INTIMATE members, 2007.
Towards a climate event stratigraphy for New Zealand over the
past 30,000 years (NZ-INTIMATE project). Journal of Quater-
nary Science, 22, 9–35.
Alloway, B. V., Lowe, D. J., Larsen, G., Shane, P. A. R., and
Westgate, J. A., 2013. Tephrochronology. In Elias, S. A., and
Mock, C. J. (eds.), The Encyclopaedia of Quaternary Science,
2nd edn. Amsterdam: Elsevier, Vol. 4, pp. 277–304.
Austin, W. E. N., Abbott, P. M., Davies, S. M., Pearce, N. J. G., and
Wastegård, S., 2014. Marine tephrochronology: an introduction
to tracing time in the ocean. Geological Society, London, Special
Publications, 398, 1–5.
Balascio, N. L., Wickler, S., Narmo, L. E., and Bradley, R. S., 2011.
Distal cryptotephra found in a Viking boathouse: the potential
for tephrochronology in reconstructing the Iron Age in Norway.
Journal of Archaeological Science, 38, 934–941.
Barrell, D. J. A., Almond, P. C., Vandergoes, M. J., Lowe, D. J.,
Newnham, R. M., and NZ-INTIMATE Members, 2013.
A composite pollen-based stratotype for inter-regional evalua-
tion of climatic events in New Zealand over the past 30,000 years
 TEPHROCHRONOLOGY
(NZ-INTIMATE project). Quaternary Science Reviews,
74, 4–20.
Biswas, R. H., Williams, M. A. J., Raj, R., Juyal, N., and Singhvi,
A. K., 2013. Methodological studies on luminescence dating of
volcanic ashes. Quaternary Geochronology, 17, 14–25.
Blaauw, M., and Christen, J. A., 2011. Flexible paleoclimate
age-depth models using an autoregressive gamma process.
Bayesian Analysis, 6, 457–474.
Blockley, S. P. E., Bronk Ramsey, C., Lane, C. S., and Lotter, A. F.,
2008. Improved age modelling approaches as exemplified by the
revised chronology for the central European varved lake
Soppensee. Quaternary Science Reviews, 27, 61–71.
Bourne, A. J., Lowe, J. J., Trincardi, F., Asioli, A., Blockley, S. P. E.,
Wulf, S., Matthews, I. P., Piva, A., and Vigliotti, L., 2010. Distal
tephra record of the last c. 105,000 years from core PRAD 1–2 in
the central Adriatic Sea: implications for marine tephrostra-
tigraphy. Quaternary Science Reviews, 29, 3079–3094.
Brendryen, J., Haflidason, H., and Sejrup, H. P., 2010. Norwegian
Sea tephrostratigraphy of marine isotope stages 4 and 5: pros-
pects and problems for tephrochronology in the North Atlantic
region. Quaternary Science Reviews, 29, 847–864.
Bronk Ramsey, C., 2009. Bayesian analysis of radiocarbon dates.
Radiocarbon, 51, 337–360.
Brumm, A. G., Jensen, G. M., van den Bergh, G. D., Morwood,
M. J., Kurniawan, I., Aziz, F., and Storey, M., 2010. Hominins
on Flores, Indonesia, by one million years ago. Nature,
464, 748–752.
Chesner, C. A., and Luhr, J. F., 2010. A melt inclusion study of the
Toba Tuffs, Sumatra, Indonesia. Journal of Volcanology and
Geothermal Research, 197, 259–278.
Chiasera, B., and Cortés, J. A., 2011. Predictive regions for geo-
chemical compositional data of volcanic systems. Journal of Vol-
canology and Geothermal Research, 207, 83–92.
Churchman, G. J., and Lowe, D. J., 2012. Alteration, formation, and
occurrence of minerals in soils. In Huang, P. M., Li, Y., and Sum-
ner, M. E. (eds.), Handbook of Soil Sciences, 2nd edn. Boca
Raton: CRC Press. Properties and processes, Vol.
1, pp. 20.1–20.72.
Cioni, R., Pistolesi, M., Bertagnini, A., Bonadonna, C.,
Hoskuldsson, A., and Scateni, B., 2014. Insights into the dynam-
ics and evolution of the 2010 Eyjafjallajökull summit eruption
(Iceland) provided by volcanic ash textures. Earth and Planetary
Science Letters, 394, 111–123.
Coulter, S. E., Pilcher, J. R., Plunkett, G., Baillie, M. G. L., Hall,
V. A., Steffensen, J. P., Vinther, B. M., Clausen, H. B., and
Johnsen, S. J., 2012. Holocene tephras highlight complexity of
volcanic signals in Greenland ice cores. Journal of Geophysical
Research, [Atmospheres], 117(D21303), 1–11, doi:10.1029/
2012JD017698.
Cullen, V. L., Smith, V. C., and Arz, H. W., 2014. The detailed
tephrostratigraphy of a core from the south-east Black Sea span-
ning the last ~60 ka. Journal of Quaternary Science,
29, 675–690.
Danišík, M., Shane, P. A. R., Schmitt, A. K., Hogg, A. G., Santos,
G. M., Storm, S., Evans, N. J., Fifield, L. K., and Lindsay,
J. M., 2012. Re-anchoring the late Pleistocene tephrochronology
of New Zealand based on concordant radiocarbon ages and com-
bined 238U/230Th disequilibrium and (U-Th)/He zircon ages.
Earth and Planetary Science Letters, 349–350, 240–250.
Davies, S. M., Elmquist, M., Bergman, J., Wohlfarth, B., and
Hammarlund, D., 2007. Cryptotephra sedimentation processes
within two lacustrine sequences from west central Sweden. The
Holocene, 17, 319–330.
Davies, S. M., Wastegård, S., Abbott, P. M., Barbante, C., Bigler,
M., Johnsen, S. J., Rasmussen, T. L., Steffensen, J. P., and
Svensson, A., 2010. Tracing volcanic events in the NGRIP
ice-core and synchronising North Atlantic marine records during
795
the last glacial period. Earth and Planetary Science Letters,
294, 69–79.
Davies, S. M., Abbott, P. M., Pearce, N. J. G., Wastegård, S., and
Blockley, S. P. E., 2012. Integrating the INTIMATE records
using tephrochronology: rising to the challenge. Quaternary Sci-
ence Reviews, 36, 11–27.
de Fontaine, C. S., Kaufman, D. S., Anderson, R. S., Werner, A.,
Waythomas, C. F., and Brown, T. A., 2007. Late Quaternary dis-
tal tephra-fall deposits in lacustrine sediments, Kenai Peninsula,
Alaska. Quaternary Research, 68, 64–78.
Dickinson, W. R., Stair, K. N., Gehrels, G. E., Peters, L., Kowallis,
B. J., Blakely, R. C., Ammar, J. R., and Greenhalgh, B. W., 2010.
U-Pb and 40Ar/39Ar ages for a tephra lens in the Middle Jurassic
Page Sandstone: first direct isotope dating of a Mesozoic
eolianite on the Colorado Plateau. Journal of Geology,
118, 215–221.
Donoghue, S. L., Stewart, R. B., and Palmer, A. S., 1991. Morphol-
ogy and chemistry of olivine phenocrysts of Mangamate Tephra,
Tongariro Volcanic Centre, New Zealand. Journal of the Royal
Society of New Zealand, 21, 225–236.
Dugmore, A. J., and Newton, A. J., 2012. Isochrons and beyond:
maximising the use of tephrochronology in geomorphology.
Jökull, 62, 39–52.
Dunbar, N. W., and Kurbatov, A. V., 2011. Tephrochronology of the
Siple Dome ice core, West Antarctica: correlations and sources.
Quaternary Science Reviews, 30, 1602–1614.
Eden, D. N., Palmer, A. S., Cronin, S. J., Marden, M., and Berryman,
K. R., 2001. Dating the culmination of river aggradation at the end
of the last glaciation using distal tephra compositions, eastern
North Island, New Zealand. Geomorphology, 38, 133–151.
Feibel, C. S., 1999. Tephrostratigraphy and geological context in
paleoanthropology. Evolutionary Anthropology, 8, 87–100.
Gehrels, M. J., Lowe, D. J., Hazell, Z. J., and Newnham, R. M.,
2006. A continuous 5000-year Holocene cryptotephrostra-
tigraphic record from northern New Zealand: implications for
tephrochronology and volcanic hazard assessment. The Holo-
cene, 16, 173–187.
Gehrels, M. J., Newnham, R. M., Lowe, D. J., Wynne, S., Hazell,
Z. J., and Caseldine, C., 2008. Towards rapid assay of
cryptotephra in peat cores: review and evaluation of various
methods. Quaternary International, 178, 68–84.
Giaccio, B., Arienzo, I., Sottili, G., Castorina, F., Gaeta, M.,
Nomade, S., Galli, P., and Messina, P., 2013. Isotopic (Sr-Nd)
and major element fingerprinting of distal tephras: an application
to the Middle-Late Pleistocene markers from the Colli Albani
volcano, central Italy. Quaternary Science Reviews,
67, 190–206.
Guðmundsdóttir, E. R., Eiríksson, J., and Larsen, G., 2012. Holo-
cene marine tephrochronology on the Iceland shelf: an overview.
Jökull, 62, 53–72.
Hall, M., and Hayward, C., 2014. Preparation of micro- and crypto-
tephras for quantitative microbeam analysis. Geological Society,
London, Special Publications, 398, 21–28.
Hall, V. A., and Pilcher, J. R., 2002. Late-Quaternary Icelandic
tephras in Ireland and Great Britain: detection, characterization
and usefulness. The Holocene, 12, 223–230.
Hayward, C., 2012. High spatial resolution electron probe micro-
analysis of tephras and melt inclusions without beam-induced
chemical modification. The Holocene, 22, 119–125.
Hodder, A. P. W., de Lange, P. J., and Lowe, D. J., 1991. Dissolution
and depletion of ferromagnesian minerals from Holocene
tephras in an acid bog, New Zealand, and implications for tephra
correlation. Journal of Quaternary Science, 6, 195–208.
Hogg, A. G., Fifield, L. K., Palmer, J. G., Turney, C. S. M., and Gal-
braith, R., 2007. Robust radiocarbon dating of wood samples by
high-sensitivity liquid scintillation spectroscopy in the 50–70
kyr age range. Radiocarbon, 49, 379–391.
796 TEPHROCHRONOLOGY
Hogg, A. G., Lowe, D. J., Palmer, J. G., Boswijk, G., and Bronk
Ramsey, C. J., 2012. Revised calendar date for the Taupo erup-
tion derived by 14C wiggle-matching using a New Zealand kauri
14
C calibration data set. The Holocene, 22, 439–449.
Housley, R. A., Lane, C. S., Cullen, V. L., Weber, M.-J., Riede, F.,
Gamble, C. S., and Brock, F., 2012. Icelandic volcanic ash from
the Late-glacial open-air archaeological site of Ahrenshöft LA
58 D, north Germany. Journal of Archaeological Science,
39, 708–716.
Housley, R. A., MacLeod, A., Nalepka, D., Jurochnik, A., Masojć,
M., Davies, L., Lincoln, P. C., Bronk Ramsey, C., Gamble, C. S.,
and Lowe, J. J., 2013. Tephrostratigraphy of a Lateglacial lake
sediment sequence at Węgliny, southwest Poland. Quaternary
Science Reviews, 77, 4–18.
Jensen, B. J. L., Pyne-O’Donnell, S., Plunkett, G., Froese, D. G.,
Hughes, P. D. M., Sigl, M., McConnell, J. R., Amesbury,
M. J., Blackwell, P. G., van den Bogaard, C., Buck, C. E.,
Charman, D. J., Clague, J. J., Hall, V. A., Koch, J., Mackay,
H., Mallon, G., McColl, L., and Plicher, J. R., 2014. Transatlan-
tic distribution of the Alaskan White River Ash. Geology,
42, 875–878.
Juvigné, E. T., and Porter, S. C., 1985. Mineralogical variations
within two widespread Holocene tephra layers from Cascade
Range volcanoes, U.S.A. Géographie Physique et Quaternaire,
39, 7–12.
Kilgour, G., Blundy, J., Cashman, K., and Mader, H. M., 2013.
Small volume andesite magmas and melt-mush interactions at
Ruapehu, New Zealand: evidence from melt inclusions. Contri-
butions to Mineralogy and Petrology, 166, 371–392.
Kuehn, S. C., and Negrini, R. M., 2010. A 250 k.y. record of
Cascade arc pyroclastic volcanism from late Pleistocene lacus-
trine sediments near Summer Lake, Oregon, USA. Geosphere,
6, 397–429.
Kuehn, S. C., Froese, D. G., Shane, P. A. R., and Intercomparison
Participants, I. N. T. A. V., 2011. The INTAV intercomparison
of electron-beam microanalysis of glass by tephrochronology
laboratories: results and recommendations. Quaternary Interna-
tional, 246, 19–47.
Lane, C. S., Chorn, B. T., and Johnson, T. C., 2013. Ash from the
Toba supereruption in Lake Malawi shows no volcanic winter
in East Africa at 75 ka. Proceedings of the National Academy
of Sciences of the United States of America, 110, 8025–8029.
Lane, C. S., Cullen, V. L., White, D., Bramham-Law, C. W. F., and
Smith, V. C., 2014. Cryptotephra as a dating and correlation tool
in archaeology. Journal of Archaeological Science, 42, 42–50.
Larsen, G., Eiríksson, J., and Gudmundsdóttir, E. R., 2014. Last
millennium dispersal of air-fall tephra and ocean-rafted pumice
towards the north Icelandic shelf and the Nordic seas. Geological
Society, London, Special Publications, 398, 113–140.
Lawson, I. T., Swindles, G. T., Plunkett, G., and Greenberg, D.,
2012. The spatial distribution of Holocene cryptotephras in
north-west Europe since 7 ka: implications for understanding
ash fall events from Icelandic eruptions. Quaternary Science
Reviews, 41, 57–66.
Liu, E. J., Cashman, K. V., Beckett, F. M., Witham, C. S.,
Leadbetter, S. J., Hort, M. C., and Guðmundsson, S., 2014.
Ash mists and brown snow: remobilization of volcanic ash from
recent Icelandic eruptions. Journal of Geophysical Research,
[Atmospheres], 119, 9463–9480.
Lohne, Ø. S., Mangerud, J., and Birks, H. H., 2013. Precise 14C ages
of the Vedde and Saksunarvatn ashes and the Younger Dryas
boundaries from western Norway and their comparison with
the Greenland Ice Core (GICC05) chronology. Journal of
Quaternary Science, 28, 490–500.
Lowe, D. J., 1986. Controls on the rates of weathering and clay min-
eral genesis in airfall tephras: a review and New Zealand case
study. In Colman, S. M., and Dethier, D. P. (eds.), Rates of
Chemical Weathering of Rocks and Minerals. Orlando: Aca-
demic, pp. 265–330.
Lowe, D. J., 2011. Tephrochronology and its application: a review.
Quaternary Geochronology, 6, 107–153.
Lowe, D. J., and Tonkin, P. J., 2010. Unravelling upbuilding pedo-
genesis in tephra and loess sequences in New Zealand using
tephrochronology. In Proceedings 19th World Congress of Soil
Science. Available at http://www.iuss.org/19th%20WCSS/
WCSS_Main_Page.html, Geochronological techniques and soil
formation symposium 1.3.2, pp. 34–37.
Lowe, D. J., Shane, P. A. R., Alloway, B. V., and Newnham, R. M.,
2008. Fingerprints and age models for widespread New Zealand
tephra marker beds erupted since 30,000 years ago: a framework
for NZ-INTIMATE. Quaternary Science Reviews, 27, 95–126.
Lowe, J. J., Barton, N., Blockley, S., Bronk Ramsey, C., Cullen,
V. L., et al., 2012. Volcanic ash layers illuminate the resilience
of Neanderthals and early modern humans to natural hazards.
Proceedings of the National Academy of Sciences of the United
States of America, 109, 13532–13537.
Lowe, D. J., Blaauw, M., Hogg, A. G., and Newnham, R. M., 2013.
Ages of 24 widespread tephras erupted since 30,000 years ago in
New Zealand, with re-evaluation of the timing and
palaeoclimatic implications of the Lateglacial cool episode
recorded at Kaipo bog. Quaternary Science Reviews,
74, 170–194.
Manville, V., and Wilson, C. J. N., 2004. The 26.5 ka Oruanui erup-
tion, New Zealand: a review of the roles of volcanism and cli-
mate in the post-eruptive sedimentary response. New Zealand
Journal of Geology and Geophysics, 47, 525–547.
Marcaida, M., Mangan, M. T., Vazquez, J. A., Bursik, M., and
Lidzbarski, M. I., 2014. Geochemical fingerprinting of Wilson
Creek formation tephra layers (Mono Basin, California) using
titanomagnetite compositions. Journal of Volcanology and Geo-
thermal Research, 273, 1–14.
Matsu’ura, T., Ueno, T., and Furusawa, A., 2011. Characterization
and correlation of cryptotephras using major-element analyses
of melt inclusions preserved in quartz in last interglacial marine
sediments, southeastern Shikoku, Japan. Quaternary Interna-
tional, 246, 48–56.
Matsu’ura, T., Furusawa, A., and Yanagida, M., 2012. Detection
and correlation of widespread cryptotephras in middle Pleisto-
cene loess in NE Japan using cummingtonite geochemistry.
Journal of Asian Earth Sciences, 60, 49–67.
Moebis, A., Cronin, S. J., Neall, V. E., and Smith, I. E., 2011.
Unravelling a complex volcanic history from fine-grained, intri-
cate Holocene ash sequences at the Tongariro Volcanic Centre,
New Zealand. Quaternary International, 246, 352–363.
Mullen, P. O., 2012. An archaeological test of the effects of the
White River Ash eruptions. Arctic Anthropology, 49, 35–44.
Naranjo, J. A., and Stern, C. R., 2004. Holocene tephrochonology of
the southernmost part (42 300 45 S) of the Andean Southern
Volcanic Zone. Revista Geologica de Chile, 31, 225–240.
Newnham, R. M., and Lowe, D. J., 1999. Testing the synchroneity
of pollen signals using tephrostratigraphy. Global and Planetary
Change, 21, 113–128.
Newnham, R. M., Eden, D. N., Lowe, D. J., and Hendy, C. H., 2003.
Rerewhakaaitu Tephra, a land-sea marker for the last termination
in New Zealand, with implications for global climate change.
Quaternary Science Reviews, 22, 289–308.
Newnham, R. M., Vandergoes, M. J., Garnett, M. H., Lowe, D. J.,
Prior, C., and Almond, P. C., 2007. Test of AMS 14C dating of
pollen concentrates using tephrochronology. Journal of Quater-
nary Science, 22, 37–51.
O’Sullivan, P. B., Morwood, M. J., Hobbs, D., Aziz, F., Suminto,
A. F., Situmorang, M., Raza, A., and Maas, R., 2001. Archaeo-
logical implications of the geology and chronology of the Soa
basin, Flores, Indonesia. Geology, 29, 607–610.
 TEPHROCHRONOLOGY
Óladóttir, B. A., Sigmarsson, O., Larsen, G., and Thordarson, T.,
2008. Katla volcano, Iceland: magma composition, dynamics
and eruption frequency as recorded by Holocene tephra layers.
Bulletin of Volcanology, 70, 475–493.
Óladóttir, B. A., Larsen, G., and Sigmarsson, O., 2012. Deciphering
eruption history and magmatic processes from tephra in Iceland.
Jökull, 62, 21–38.
Payne, R., and Gehrels, M. J., 2010. The formation of tephra layers
in peatlands: an experimental approach. Catena, 81, 12–23.
Payne, R., Blackford, J., and van der Plicht, J., 2008. Using
cryptotephras to extend regional tephrochronologies: an exam-
ple from southeast Alaska and implications for hazard assess-
ment. Quaternary Research, 69, 24–55.
Pearce, N. J. G., 2014. Towards a protocol for the trace element
analysis of glass from rhyolitic shards in tephra deposits by laser
ablation ICP-MS. Journal of Quaternary Science, 29, 627–640.
Pearce, N. J. G., Bendall, C. A., and Westgate, J. A., 2008a. Com-
ment on “Some numerical considerations in the geochemical
analysis of distal microtephra” by A.M. Pollard,
S.P.E. Blockley, C.S. Lane, Applied Geochemistry
21, 1692–1714. Applied Geochemistry, 23, 1353–1364.
Pearce, N. J. G., Alloway, B. V., and Westgate, J. A., 2008b.
Mid-Pleistocene silicic tephra beds in the Auckland region,
New Zealand: their correlation and origins based on the trace ele-
ment analyses of single glass shards. Quaternary International,
178, 16–43.
Pearce, N. J. G., Westgate, J. A., Perkins, W. T., and Wade, S. C.,
2011. Trace-element microanalysis by LA-ICP-MS: the quest
for comprehensive chemical characterisation of single,
sub-10 mm volcanic glass shards. Quaternary International,
246, 57–81.
Pearce, N. J. G., Abbott, P. M., and Martin-Jones, C., 2014.
Microbeam methods for the analysis of glass in fine-grained
tephra deposits: a SMART perspective on current and future
trends. Geological Society, London, Special Publications, 398,
29–46.
Pillans, B., Alloway, B. V., Naish, T., Westgate, J. A., Abbot, S., and
Palmer, A. S., 2005. Silicic tephras in Pleistocene shallow
marine sediments of Wanganui Basin, New Zealand. Journal
of the Royal Society of New Zealand, 35, 43–90.
Platz, T., Cronin, S. J., Smith, I. E. M., Turner, M. B., and
Stewart, R. B., 2007. Improving the reliability of microprobe-
based analyses of andesitic glasses for tephra correlation. The
Holocene, 17, 573–583.
Ponomareva, V., Kyle, P. R., Melekestsev, I. V., Rinkleff, P. G.,
Dirksen, O. V., Sulerzhitsky, L. D., Zaretskaia, N. E., and
Rourke, R., 2004. The 7600 (14C) year BP Kurile Lake
caldera-forming eruption, Kamchatka, Russia: stratigraphy and
field relationships. Journal of Volcanology and Geothermal
Research, 136, 199–222.
Ponomareva, V., Portnyagin, M., Derkachev, A., Pendea, L. F.,
Bourgeois, J., Reimer, P. J., Garbe-Schonberg, D.,
Krasheninnikov, S., and Nurnberg, D., 2013a. Early Holocene
M ~6 explosive eruption from Plosky volcanic massif
(Kamchatka) and its tephra as a link between terrestrial and
marine paleoenvironmental records. International Journal of
Earth Sciences, 102, 1673–1699.
Ponomareva, V., Portnyagin, M., Derkachev, A., Juschus, O.,
Garbe-Schonberg, D., and Nurnberg, D., 2013b. Identification
of a widespread Kamchatkan tephra: a middle Pleistocene
tie-point between Arctic and Pacific paleoclimatic records. Geo-
physical Research Letters, 40, 3538–3543.
Preece, S. J., Westgate, J. A., Stemper, B. A., and Péwé, T. L., 1999.
Tephrochronology of late Cenozoic loess at Fairbanks, central
Alaska. Geological Society of America Bulletin, 111, 71–90.
Preece, S. J., Westgate, J. A., Alloway, B. V., and Milner, M. W.,
2000. Characterization, identity, distribution, and source of late
797
Cenozoic tephra beds in the Klondike district of the Yukon,
Canada. Canadian Journal of Earth Sciences, 37, 983–996.
Preece, S. J., Westgate, J. A., Froese, D. G., Pearce, N. J. G., and
Perkins, W. T., 2011a. A catalogue of late Cenozoic tephra beds
in the Klondike goldfields, Yukon. Canadian Journal of Earth
Sciences, 48, 1386–1418.
Preece, S. J., Pearce, N. J. G., Westgate, J. A., Froese, D. G., Jensen,
B. J. L., and Perkins, W. T., 2011b. Old Crow tephra across east-
ern Beringia: a single cataclysmic eruption at the close of Marine
Isotope Stage 6. Quaternary Science Reviews, 30, 2069–2090.
Pyne-O’Donnell, S. D. F., 2011. The taphonomy of Last Glacial-
Interglacial Transition (LGIT) distal volcanic ash in small Scot-
tish lakes. Boreas, 40, 131–145.
Pyne-O’Donnell, S. D. F., Hughes, P. D. M., Froese, D. G., Jensen,
B. J. L., Kuehn, S. C., Mallon, G., Amesbury, M. J., Charman,
D. J., Daley, T. J., Loader, N. J., Mauquoy, D., Street-Perrott,
F. A., and Woodman-Ralph, J., 2012. High-precision ultra-distal
Holocene tephrochronology in North America. Quaternary Sci-
ence Reviews, 52, 6–11.
Riede, F., and Thastrup, M. D., 2013. Tephra, tephrochronology and
archaeology – a (re-) view from northern Europe. Heritage Sci-
ence, 1(15), 1–17.
Roulleau, E., Pinti, D. L., Rouchon, V., Quidelleur, X., and Gillot,
P.-Y., 2009. Tephro-chronostratigraphy of the lacustrine intergla-
cial record of Piànico, Italian southern Alps: identifying the vol-
canic sources using radiogenic isotopes and trace elements.
Quaternary International, 204, 31–43.
Sanborne, P. T., Smith, C. A. S., Froese, D. G., Zazula, G., and
Westgate, J. A., 2006. Full-glacial paleosols in perenially frozen
loess sequences, Klondike goldfields, Yukon Territory, Canada.
Quaternary Research, 66, 147–157.
Sandhu, A. S., and Westgate, J. A., 1995. The correlation between
reduction in fission-track diameter and areal track density in vol-
canic glass shards and its application in dating tephra beds. Earth
and Planetary Science Letters, 131, 289–299.
Scaillet, S., Vita-Scaillet, G., and Rotolo, S. G., 2013. Millennial-
scale phase relationships between ice-core and Mediterranean
marine records: insights from high-precision 40Ar/39Ar dating
of the Green Tuff of Pantelleria, Sicily Strait. Quaternary Sci-
ence Reviews, 78, 141–154.
Schmitt, A. K., 2006. Laacher See revisited: high-spatial-resolution
zircon dating indicates rapid formation of a zoned magma cham-
ber. Geology, 34, 597–600.
Sell, B. K., and Samson, S. D., 2011. A tephrochronologic method
based on apatite trace-element chemistry. Quaternary Research,
76, 157–166.
Shane, P. A. R., 1998. Correlation of rhyolitic pyroclastic eruptive
units from the Taupo volcanic zone by Fe-Ti oxide composi-
tional data. Bulletin of Volcanology, 60, 224–238.
Shane, P. A. R., 2000. Tephrochronology: a New Zealand case
study. Earth-Science Reviews, 49, 223–259.
Shane, P. A. R., and Hoverd, J., 2002. Distal record of multi-sourced
tephra in Onepoto Basin, Auckland, New Zealand: implications
for volcanic chronology, frequency and hazards. Bulletin of Vol-
canology, 64, 441–454.
Shane, P. A. R., and Zawalna-Geer, A., 2011. Correlation of basaltic
tephra from Mt Wellington volcano: implications for the penulti-
mate eruption from the Auckland Volcanic Field. Quaternary
International, 246, 374–381.
Shane, P. A. R., Smith, V. C., and Nairn, I. A., 2003. Biotite compo-
sition as a tool for the identification of Quaternary tephra beds.
Quaternary Research, 59, 262–270.
Shane, P. A. R., Sikes, E. L., and Guilderson, T. P., 2006. Tephra
beds in deep-sea cores off northern New Zealand: implications
for the history of Taupo Volcanic Zone, Mayor Island and White
Island volcanoes. Journal of Volcanology and Geothermal
Research, 154, 276–290.
798 TEPHROCHRONOLOGY
Shane, P. A. R., Nairn, I. A., Martin, S. B., and Smith, V. C., 2008a.
Compositional heterogeneity in tephra deposits resulting from
the eruption of multiple magma bodies: implications for
tephrochronology. Quaternary International, 178, 44–53.
Shane, P. A. R., Smith, V. C., and Nairn, I. A., 2008b. Millennial
timescale resolution of rhyolite magma recharge at Tarawera vol-
cano: insights from quartz chemistry and melt inclusions. Con-
tributions to Mineralogy and Petrology, 156, 397–411.
Shane, P., Gehrels, M. J., Zawalna-Geer, A., Lindsay, J., and
Chaillou, I., 2013. Longevity of a small shield volcano revealed
by crypto-tephra studies (Rangitoto volcano, New Zealand):
change in eruptive behavior of a basaltic field. Journal of Volca-
nology and Geothermal Research, 257, 174–183.
Sigl, M., McConnell, J. R., Layman, L., Maselli, O., McGwire, K.,
Pasteris, D., Dahl-Jensen, D., Steffensen, J. P., Vinther, B.,
Edwards, R., Mulvaney, R., and Kipfstuhl, S., 2013. A new bipo-
lar ice core record of volcanism from WAIS Divide and NEEM
and implications for climate forcing of the last 2000 years. Jour-
nal of Geophysical Research, [Atmospheres], 118, 1151–1169.
Sirocko, F., Dietrich, S., Veres, D., Grootes, P. M., Schaber-Mohr,
K., Seelos, K., Nadeau, M.-J., Kromer, B., Rothacker, L.,
Röhner, M., Krbetschek, M., Appleby, P., Hambach, U., Rolf,
C., Sudo, M., and Grim, S., 2013. Multi-proxy dating of Holo-
cene maar lakes and Pleistocene dry maar sediments in the Eifel,
Germany. Quaternary Science Reviews, 62, 56–76.
Smith, V. C., Shane, P., and Nairn, I. A., 2005. Trends in rhyolite
geochemistry, mineralogy, and magma storage during the last
50 kyr at Okataina and Taupo volcanic centres, Taupo Volcanic
Zone, New Zealand. Journal of Volcanology and Geothermal
Research, 148, 372–406.
Smith, V. C., Pearce, N. J. G., Matthews, N. E., Westgate, J. A.,
Durant, A. J., Lane, C., Petraglia, M. D., Haslam, M., Korisettar,
R., and Pal, J. N., 2011. Chemically fingerprinting volcanic ash
from Toba using biotite compositions. Quaternary Interna-
tional, 246, 97–104.
Smith, V. C., Staff, R. A., Blockley, S. P. E., Bronk Ramsey, C.,
Nakagawa, T., Mark, D. F., Takemura, K., Danhara, T., and
Suigetsu 2006 Project Members, 2013. Identification and corre-
lation of visible tephras in the Lake Suigetsu SG06 sedimentary
archive, Japan: chronostratigraphic markers for synchronising of
east Asian/west Pacific palaeoclimatic records across the last
150 ka. Quaternary Science Reviews, 67, 121–137.
Staff, R. A., Bronk Ramsey, C., Bryant, C. L., Brock, F., Payne,
R. L., Schlolaut, G., Marshall, M. H., Brauer, A., Lamb, H. L.,
Tarasov, P., Rokoyama, Y., Haraguchi, T., Gotanda, K.,
Yonenobu, H., Nakagawa, T., and Suigetsu 2006 Project Mem-
bers, 2011. New 14C determinations from Lake Suigetsu, Japan:
12,000 to 0 cal BP. Radiocarbon, 53, 511–528.
Staff, R. A., Nakagawa, T., Schlolaut, G., Marshall, M. H., Brauer,
A., et al., 2013. The multiple chronological techniques applied to
the Lake Suigetsu SG06 sediment core, central Japan. Boreas,
42, 259–266.
Stevenson, J. A., Loughlin, S., Rae, C., Thordarson, T.,
Milodowski, A. E., Gilbert, J. S., Harangi, S., Lukács, R.,
Højgaard, B., Arting, U., Pyne-O’Donnell, S., MacLeod, A.,
Whitney, B., and Cassidy, M., 2012. Distal deposition of tephra
from the Eyjafjallajökull 2010 summit eruption. Journal of Geo-
physical Research, 117(B00C10), 1–10, doi:10.1029/
2011JB008904.
Storey, M., Roberts, R. G., and Saidin, M., 2012. Astronomically
calibrated 40Ar/39Ar age for the Toba supereruption and global
synchronization of late Quaternary records. Proceedings of the
National Academy of Sciences of the United States of America,
109, 18684–18688.
Streeter, R. T., and Dugmore, A. J., 2013. Reconstructing late-
Holocene environmental change in Iceland using high-
resolution tephrochronology. The Holocene, 23, 197–207.
Sulpizio, R., Alçiçek, M. C., Zanchetta, G., and Solari, L., 2013.
Recognition of the Minoan tephra in the Acigöl Basin, western
Turkey: implications for inter-archive correlations and fine ash
dispersal. Journal of Quaternary Science, 28, 329–335.
Sumita, M., and Schminke, H.-U., 2013. Impact of volcanism on the
evolution of Lake Van II: temporal evolution of explosive volca-
nism of Nemrut Volcano (eastern Anatolia) during the past
ca. 0.4 Ma. Journal of Volcanology and Geothermal Research,
253, 15–34.
Suzuki, T., Kasahara, A., Nishizawa, F., and Saito, H., 2014. Chem-
ical characterization of volcanic glass shards by energy disper-
sive X-ray spectrometry with EDAX Genesis APEX2 and
JEOL JSM-6390. Geographical Reports of Tokyo Metropolitan
University, 49, 1–12.
Svensson, A., Andersen, K. K., Bigler, M., Clausen, H. B., Dahl-
Jensen, D., Davies, S. M., Johnsen, S. J., Muscheler, R.,
Parrenin, F., Rasmussen, S. O., Röthlisberger, R., Seierstad, I.,
Steffensen, J. P., and Vinther, B. M., 2008. A 60,000 year Green-
land stratigraphic ice core chronology. Climate of the Past, 4,
47–57.
Swindles, G. T., De Vleeschouwer, F., and Plunkett, G., 2010.
Dating peat profiles using tephra: stratigraphy, geochemistry
and chronology. Mires and Peat, 7, 1–9.
Swindles, G. T., Lawson, I. T., Savov, I. P., Connor, C. B., and
Plunkett, G., 2011. A 7000 yr perspective on volcanic ash clouds
affecting northern Europe. Geology, 39, 887–890.
Swindles, G. T., Galloway, J., Outram, Z., Turner, K., Schofield,
J. E., Newton, A. J., Dugmore, A. J., Church, M. J., Watson,
E. J., Batt, C., Bond, J., Edwards, K. J., Turner, V., and Bashford,
D., 2013. Re-deposited cryptotephra layers in Holocene peats
linked to anthropogenic activity. The Holocene, 23, 1493–1501.
Todd, J. A., Austin, W. E., and Abbott, P. E., 2014. Quantifying bio-
turbation of a simulated ash fall event. Geological Society, Lon-
don, Special Publications, 398, 195–207.
Torres, R., Mouginis-Mark, P., Self, S., Garbeil, H., Kallianpur, K.,
and Quiambao, R., 2004. Monitoring the evolution of the Pasig-
Porero alluvial fan, Pinatubo Volcano, using a decade of remote
sensing data. Journal of Volcanology and Geothermal Research,
138, 371–392.
Tryon, C. A., Faith, J. T., Peppe, D. J., Fox, D. L., Holt, K.,
Dunsworth, H., and Harcourt-Smith, W., 2010. The Pleistocene
archaeology and environments of the Wasiriya Beds, Rusinga
Island, Kenya. Journal of Human Evolution, 59, 657–671.
Turner, M. B., Cronin, S. J., Bebbington, M. S., Smith, I. E., and
Stewart, R. B., 2011a. Integrating records of explosive and effu-
sive activity from proximal and distal sequences: Mt. Taranaki,
New Zealand. Quaternary International, 246, 364–373.
Turner, M. B., Cronin, S. J., Bebbington, M. S., Smith, I. E. M., and
Stewart, R. B., 2011b. Relating magma composition with erup-
tion variability at andesitic volcanoes. A case study from
Mt. Taranaki, New Zealand. Geological Society of America Bul-
letin, 123, 2005–2015.
Turney, C. S. M., and Lowe, J. J., 2001. Tephrochronology. In Last,
W. M., and Smol, J. P. (eds.), Tracking Environmental Changes
in Lake Sediments: Physical and Chemical Techniques. Dor-
drecht: Kluwer, pp. 451–471.
Ukstins Peate, I., Kent, A. J. R., Baker, J. A., and Menzies, M. A.,
2008. Extreme geochemical heterogeneity in Afro-Arabian
Oligocene tephras: preserving fractional crystallization and
mafic recharge processes in silicic magma chambers. Lithos,
102, 260–278.
van den Bogaard, P., 1995. Ar-40/Ar-39 ages of sanidine pheno-
crysts from Laacher-See tephra (12,900 yr BP) – chronostra-
tigraphic and petrological significance. Earth and Planetary
Science Letters, 133, 163–174.
Van den Bogaard, P., and Schminke, H.-U., 1985. Laacher See
tephra: a widespread isochronous late Quaternary tephra layer
 TERRESTRIAL COSMOGENIC NUCLIDE DATING 799

in central and northern Europe. Bulletin of the Geological Soci- Changbaishan, northeastern China. Quaternary Science
ety of America, 96, 1554–1571. Reviews, 47, 150–159.
Vandergoes, M. J., Hogg, A. G., Lowe, D. J., Newnham, R. M., Zalasiewicz, J., Cita, M. B., Hilgen, F., Pratt, B. R., Strasser, A.,
Denton, G. H., Southon, J., Barrell, D. J. A., Blaauw, M., Wilson, Thierry, J., and Weissert, H., 2013. Chronostratigraphy and geo-
C. J. N., McGlone, M. S., Allan, A. S. R., Almond, P. C., chronology: a proposed realignment. GSA Today, 23, 4–8.
Petchey, F., Dalbell, K., and Dieffenbacher-Krall, A. C., 2013. Zdanowicz, C. M., Zielinski, G. A., and Germani, M. S., 1999.
A revised age for the Kawakawa/Oruanui tephra, a key marker Mount Mazama eruption: calendrical age verified and atmo-
for the Last Glacial Maximum in New Zealand. Quaternary Sci- spheric impact assessed. Geology, 27, 621–624.
ence Reviews, 74, 195–200.
Westgate, J. A., 1989. Isothermal plateau fission track ages of
hydrated glass shards from silicic tephra beds. Earth and Plane- Cross-references
tary Science Letters, 95, 226–234. Accelerator Mass Spectrometry
Westgate, J. A., Smith, D. G. W., and Nichols, H., 1969. Late Qua- Ar–Ar and K–Ar Dating
ternary pyroclastic layers in the Edmonton area, Alberta. In Biostratigraphy
Pawluk, S. (ed.), Pedology and Quaternary Research. Edmon- 14
C in Plant Macrofossils
ton: University of Alberta, pp. 179–186. Dendrochronology, Volcanic Eruptions
Westgate, J. A., Stemper, B. A., and Péwé, T. L., 1990. A 3 m.y. Fission Track Dating and Thermochronology
record of Pliocene-Pleistocene loess in interior Alaska. Geology, Ice Cores
18, 858–861. Laser Ablation Inductively Coupled Mass Spectrometer (LA ICP-
Westgate, J., Shane, P., Pearce, N., Perkins, W., Korisettar, R., MS)
Chesner, C. A., Williams, M., and Acharyya, S. K., 1998. All Lacustrine Environments (14C)
Toba tephra occurrences across Peninsular India belong to the Luminescence Dating
75,000 yr B.P. eruption. Quaternary Research, 50, 107–112. Magnetostratigraphic Dating
Westgate, J. A., Preece, S. J., Froese, D. G., Pearce, N. J. G., Marine Isotope Stratigraphy
Roberts, R. G., Demuro, M., Hart, W. K., and Perkins, W., Paleosol
2008. Changing ideas on the identity and stratigraphic signifi- 210
Pb Dating
cance of the Sheep Creek tephra beds in Alaska and the Yukon Peat (14C)
Territory, northwestern North America. Quaternary Interna- Radiocarbon Dating
tional, 178, 183–209. Sedimentary Rocks (Rb-Sr Geochronology)
Westgate, J. A., Pearce, N. J. G., Perkins, W. T., Shane, P. A. R., and Single-Crystal Laser Fusion
Preece, S. J., 2011. Lead isotope ratios of volcanic glass by laser Uranium–Lead, Zircon
ablation inductively-coupled plasma mass spectrometry: appli- U–Th/He Dating
cation to Miocene tephra beds in Montana, USA and adjacent Volcanic Glass (Fission Track)
areas. Quaternary International, 246, 89–96. Zircon
Westgate, J. A., Pearce, G. W., Preece, S. J., Schweger, C. E.,
Morlan, R. E., Pearce, N. J. G., and Perkins, W. T., 2013a.
Tephrochronology, magnetostratigraphy and mammalian faunas
of Middle and Early Pleistocene sediments at two sites on the
Old Crow River, northern Yukon Territory, Canada. Quaternary TERRESTRIAL COSMOGENIC NUCLIDE DATING
Research, 79, 75–85.
Westgate, J. A., Naeser, N. D., and Alloway, B. V., 2013b. Fission-
track dating. In Elias, S. A., and Mock, C. J. (eds.), The Encyclo- John Gosse1 and Jeff Klein2
1
paedia of Quaternary Science, 2nd edn. Amsterdam: Elsevier, Department of Earth Sciences, Dalhousie University,
Vol. 1, pp. 643–662. Halifax, NS, Canada
Westgate, J. A., Pearce, N. J. G., Perkins, W. T., Preece, S. J., 2
Department of Physics and Astronomy, University of
Chesner, C. A., and Muhammad, R. F.,
2013c. Tephrochronology of the Toba tuffs: four primary glass Pennsylvania, Philadelphia, PA, USA
populations define the 75 ka Youngest Toba Tuff, northern
Sumatra, Indonesia. Journal of Quaternary Science, 28, Synonyms
772–776. Burial dating; Cosmic ray exposure (CRE) dating;
White, J. F. L., and Houghton, B. F., 2006. Primary volcaniclastic
rocks. Geology, 34, 677–680. Cosmogenic nuclide isochron dating; Cosmogenic radio-
Wilson, C. J. N., Gravley, D. M., Leonard, G. S., and Rowland, J. V., nuclide (CRN) dating; Depth profile dating; Surface expo-
2009. Volcanism in the central Taupo Volcanic Zone, New sure dating (SED); Terrestrial in situ cosmogenic nuclide
Zealand: tempo, styles and controls. In Thordarson, T., Self, S., (TCN) exposure histories
Larsen, G., Rowland, S. K., and Hoskuldsson, A. (eds.), Studies
in Volcanology: The Legacy of George Walker. London: Geolog-
ical Society. Special publications of IAVCEI, Vol. Definitions
2, pp. 225–247. Cosmic rays are high-energy (104 to 1020 eV) particles
Wilson, C. J. N., Charlier, B. L. A., Rowland, J. V., and Browne, composed mostly (90 %) of atomic nuclei but with some
P. R. L., 2010. U-Pb dating of zircon in subsurface, hydrother- photons (gamma rays), electrons, and positrons. The
mally altered pyroclastic deposits and implications for subsi-
dence in a magmatically active rift: Taupo Volcanic Zone, New
majority (90 %) of the atomic nuclei are protons
Zealand. Journal of Volcanology and Geothermal Research, (H nuclei), the rest are alpha particles (He nuclei, 9 %)
191, 69–78. and nuclei heavier atoms (1 %). Two major sources of this
Yin, J., Jull, A. J. T., Burr, G. S., and Zheng, Y., 2012. A wiggle- primary cosmic radiation are the Sun whose flux peaks
match age for the Millennium eruption of Tianchi Volcano at at ~100 MeV (solar cosmic rays, SCR) and our galaxy,
800 TERRESTRIAL COSMOGENIC NUCLIDE DATING
the Milky Way, with the flux peak between 300 MeV and
1 Gev (galactic cosmic rays, GCR). We believe GCR is
produced primarily in supernova events (an inference
suggested by the excess over cosmic abundance of heavy
nuclei).
Secondary cosmic radiation (or “secondaries”) refers to
the cascade of particles produced by the interaction of pri-
mary cosmic rays with the atmosphere and the surface of
the Earth. Because of the high energies of the incoming
particles, a wide spectrum of hadrons, mesons, and leptons
are produced. The lifetimes of many of these secondary
particles are so short that they decay in a shower of lighter
particles before they can interact with another atom of the
material through which they are passing, but others do
interact, producing yet another cascade. Ultimately, the
flux is dominated by the longer-lived particles and by
the particles with small interaction cross sections. By the
time the primaries have passed through 10 % of the atmo-
sphere, the secondary flux is predominately pions, kaons,
protons, neutrons, muons, electrons, positrons, and
gamma rays (photons). These secondaries will produce
larger nuclei (cosmogenic nuclides) in the atmosphere
and minerals.
Cosmogenic nuclides are produced through any inter-
action of primary or secondary cosmic radiation with mat-
ter. While some notable cosmogenic nuclides (isotopes)
are produced from SCR (e.g., atmospheric radiocarbon is
produced from low-energy (thermal) neutron capture by
nitrogen: 14N(n, p)14C), the majority of TCN production
is through spallation of a target nucleus which requires
GCR primaries or secondaries with sufficient energy to
exceed the binding energy of the target nucleus
(a minimum of >3 MeV per target nucleon). Cosmogenic
nuclides need to be distinguished from radiogenic and
nucleogenic nuclides. A radiogenic nuclide is a daughter
product of a radioactive decay (fission, beta, alpha).
For U-dating methods or (U–Th)/He thermochronology,
234
Th + 4He are radiogenic products of the a-decay of
238
U. A nucleogenic nuclide is produced when an ener-
getic radiogenic particle interacts with another nucleus
just after decay of its parent. For example, 10Be can be pro-
duced in a lithium-rich mineral via 7Li(a, p)10Be.
Cosmogenic nuclide dating uses the accumulation, pro-
duction, or decay of cosmogenic nuclides to determine the
exposure history of near-surface samples (top tens of
meters on Earth or top hundreds of meters in space). This
can be accomplished by measuring a cosmogenic isotope
that has been produced in situ in a mineral on a rock sur-
face or that has been produced in the atmosphere and sub-
sequently accumulated in a soil. For instance, the
concentration of 10Be accumulated in a saprolite will
depend on the flux of 10Be in precipitation and dust deliv-
ered to the soil, the abundance and type of clay that holds
the 10Be, and the duration that the soil is at the surface. In
addition to exposure ages, the method can determine
whether a sample was at the surface or beneath it, whether
the sample’s position was stable or eroding, and whether
the sample’s surface history was interrupted by periods
of burial. Examples of applications of cosmogenic
nuclides that are produced in situ, but not in Earth’s litho-
sphere, include surface exposure dating on Martian
(Farley et al., 2014) and lunar surfaces (Drozd
et al., 1974) and the size and exposure history of meteor-
ites (Eugster et al., 2006). Cosmogenic nuclides produced
in the atmosphere include 14C (thermal-neutron capture by
N). Radiocarbon dating has been widely applied for dating
objects of importance in defining geologic and
archaeologic chronologies; in testing the authenticity of
objects of art, or worship, or historic significance; and,
on even shorter time scales, in studying the inception
and progress of disease in a living organism. 14C has been
used to study atmospheric mixing, verify the global circu-
lation model of the oceans, and follow the fate of anthro-
pogenic carbon in the biosphere and oceans. 36Cl
produced in the atmosphere by cosmic ray-induced spall-
ation of argon has been used primarily by hydrologists
to study the recharge histories of hydrologic systems.
10
Be and 7Be produced in the atmosphere have been used
to estimate ages of soils and ground ice (e.g., Gilichinsky
et al., 2007) or rates of erosion of catchments (Brown
et al., 1988) and study the behavior of aerosol deposition
and air mass mixing.
Broadly, there are three types of systems that are dated
using cosmogenic nuclides: (1) those in which the cosmo-
genic nuclide is produced in situ in a solid, e.g., 10Be in
rock; (2) closed systems in which the in situ production
has stopped because of the complete shielding of a sample
from cosmic rays, e.g., 26Al/10Be in deeply buried quartz;
and (3) a now-closed system that formerly was in equilib-
rium with a reservoir that had a constant or known concen-
tration of a cosmogenic nuclide, e.g., 14C in any biogenic
material; the reservoir is either the atmosphere or ocean.
This entry treats the open system and closed system
(1 and 2) applications of nuclides specifically produced
in exposed minerals near Earth’s surface, to determine
an exposure age of a surface, burial duration of minerals,
or erosion rate of surfaces and catchments (Lal, 1991).
The term terrestrial in situ cosmogenic nuclide (TCN )
exposure history describes this method, which utilizes a
growing knowledge of how production rates spatially
and temporally vary on Earth owing to variations in the
geomagnetic field and atmospheric shielding. TCN
methods are distinguished from methods using meteoric
cosmogenic nuclides, even when the latter are absorbed
by clays in soil and marine sediments or accumulated in
various water reservoirs (e.g., ice sheets, groundwater).
The term CRN excludes the stable cosmogenic nuclides
(e.g., 3He, 21Ne, 38Ar), SED can be achieved with other
dating methods (e.g., luminescence), and CRE dating
can be conducted on lunar and other planetary surfaces.
The TCN dating method is a numerical dating method.
Currently the TCN production rates have been calibrated
against other dating methods owing to the difficulty of
obtaining precise cross sections for the production of
each isotope through different interaction pathways.
This would suggest that it is calibrated, such as
 TERRESTRIAL COSMOGENIC NUCLIDE DATING
lichenometry or the geomagnetic secular variation dating
method. However, with improvements in our understand-
ing of the temporal and spatial variations in the geomag-
netic field, cosmic ray interactions, nuclear cross
sections, production rates, and depth profile shapes are
becoming reliably predicted by theory (Masarik and
Reedy, 1995; Argento et al., 2013). In this sense, the
TCN dating method is a numerical, not calibrated,
method. The TCN method is distinct from cosmogenic
dating methods based on correlation. For instance,
meteoric cosmogenic nuclides measured in time
series media – such as 10Be in glacier ice (Beer
et al., 1987) or marine sediment (Morris et al., 2002) –
can provide a reliable method for correlating undated
records to meteoric cosmogenic nuclide records in inde-
pendently dated cores.
Introduction
TCN exposure methods are used to determine the duration
of exposure to cosmic rays, duration and degree of
shielding from cosmic rays, and rate and style of erosion,
on or near Earth’s surface. The premise is that the mea-
sured concentration of TCN is proportional to the duration
that the mineral has been exposed to cosmic rays
(Figure 1) and, in the case of a radioactive TCN in a
shielded sample, how much time the concentration has
decayed.
Depending on the energy and type of incident cosmic
ray particle and target nucleus, cosmic ray interactions
produce a variety of secondary particles over a wide
energy spectrum, including gammas, electrons, kaons,
pions, muons, neutrons, protons, deuterons, tritons, 3He
x 105
2 4 atoms g1 year1 at sea level, high latitude, under cur-
1.8
Concentration in quartz, SLHL
1.6
Concentration (atoms/g)
1.4
1.2
1 Thus, every TCN-mineral system is a unique clock.
0.8
0.6
0.4
0.2
0
0 5 10 15 20
Exposure age (ka)
Terrestrial Cosmogenic Nuclide Dating, Figure 1 Gradual
buildup of four TCNs in quartz at sea level high latitude. Black,
14
C; green, 21Ne; blue, 10Be; and red, 26Al. Continuous production
by spallation only, with rates according to Borchers
et al. (submitted) and Kober et al. (2011). Solid lines represent
surfaces with zero erosion. Constant continuous erosion of 5 and
10 mm ka1 shown with dashed and dotted curve for 14C.
801
nuclei, alphas, and heavier nuclides. It is the heavier
nuclides that are of interest. However, to be useful for
exposure dating, the TCN should:
• be stable or have suitably slow decay rates
• be more abundant than the radiogenic or nucleogenic
species in the target mineral
• have in situ production rates that yield extractable con-
centrations that can be resolved against its elemental
isotopes
• be precisely measurable considering blanks, detection
limits, and isobaric interferences
• be produced in stable, chemically resistant minerals that
are common and have low thermal diffusivities for
noble gases at surface temperatures.
This greatly reduces the number of useful particles, and
currently only 6 TCNs are widely used: 3He, 10Be, 14C,
21
Ne, 26Al, and 36Cl (Table 1). The probability for cosmic
ray interaction – cross section, measured in barns – takes
into account the type and energy of the incident secondary
and target nuclei. Thus, each TCN is produced at a unique
rate, depending on the energy of the cosmic radiation, the
type of interactions of which there are five (spallation,
thermal and epithermal neutrons, and interactions of neg-
ative and fast muons), and the chemistry of the mineral.
Although quartz is the most commonly used mineral,
other minerals include calcite, feldspar, garnet, horn-
blende, magnetite, olivine, pyroxene, and sylvite. With
the general exception of quartz, the chemical composition
of each mineral should be measured to determine the
abundance of suitable target nuclei for each of the five
interactions. Quartz is unique in that its composition is
stoichiometric, eliminating the need to analyze its chemi-
cal composition to determine the different production
rates. For instance, 10Be is produced in quartz from spall-
ation and muonic interaction on Si and O at a rate of about
rent atmospheric and magnetic field conditions. Until
recently, even whole rock samples have been analyzed
for 36Cl, 10Be, and 14C, giving the possibility that any
lithology can be used if the TCN production rates for
the specific rock composition can be determined.
Ratios of multiple TCN can be measured in a single min-
eral (e.g., 10Be, 14C, 21Ne, and 26Al in quartz, Figure 1)
or in the same rock (36Cl in feldspar and 3He in olivine
in basalt).
The TCN exposure methods differ from other dating
methods. Both radioactive and stable (noble gas) isotopes
25 30 can be used. The concentration of the isotopes is generally
depth dependent, unlike other radiometric chronometers
which are dependent on the parent isotope abundance.
The technique is not temperature dependent as is the case
for thermochronometry and amino racemization. Unlike
most geochronometers, the TCN method is not necessarily
restricted to specific minerals with a certain isotopic abun-
dance, although production rates are currently well known
for relative few minerals.
802 TERRESTRIAL COSMOGENIC NUCLIDE DATING

Terrestrial Cosmogenic Nuclide Dating, Table 1 Commonly used terrestrial cosmogenic nuclides

Nuclide ET ATM Half-life (ka) Isotope Isobar TCN analysis TCN production mechanisms Target minerals
3
He • Stable 4
He 3
H NGMS nf Ol
10
Be • • 1,390 9
Be 10
B AMS nf, m Qtz
14
C • 5.73 12/13
C 14
C AMS nf, m, nt Qtz
21
Ne • Stable 20/22
Ne NGMS nf, m Qtz
26
Al • • 720 27
Al 26
Mg AMS nf, m Qtz
36
Cl • • 300 35/37
Cl 36
S AMS nf, nt, m K, Ca-rich

Notes
Cosmogenic 37–38Ar is not listed as it is not widely used as a TCN owing to difficulties in isolating it from the radiogenic components
ET, ATM: these nuclides are also measured in extraterrestrial and atmospheric/meteoric targets
Isobar: listed are the most important isobaric interferences; recent isobaric separation innovations have greatly reduced the impact of isobars
on low-level TCN measurements
Production pathways: nf, spallation by fast neutrons; nt, thermal (and epithermal) neutron capture; m, fast and negative muonic interactions
Target minerals: only the most commonly utilized minerals are listed. For 36Cl and 37–38Ar feldspar separates are now used more than
whole rock

Examples of TCN exposure and burial dating
Details of the measurement and interpretation of TCN are
provided below. Table 2 provides an indication of the wide
range of applications of the method. Currently the most
widely used application of TCN methods is exposure dat-
ing of glacial boulders for moraine chronologies, followed
by the application of TCN to provide age control on allu-
vial strain markers for fault kinematics, and catchment
average erosion rates.
Range of applicability
The applicable range for TCN dating crosses seven orders
of magnitude – more than any other dating method. The-
oretically, the lower limit of the surface exposure dating
method is ~100 years. While exposure ages of years have
been determined, the technique is practically limited by
the difficulty of measuring low concentrations (i.e., few
hundreds of atoms per gram of mineral, requiring a large
sample mass) and our knowledge of production rate con-
trolling factors such as atmospheric dynamics and the geo-
magnetic field variations over such a short integration
time. Although the stable noble gas cosmogenic isotopes
do not decay and have low thermal diffusivities in selected
minerals at the Earth’s surface, an effective upper dating
limit on Earth is 107 years (e.g., alluvial surfaces in the
Atacama Desert, Dunai et al., 2005). This is because the
concentration of cosmogenic nuclides typically decreases
with depth and because all surfaces on Earth over such
long exposure durations will experience erosion. The
applicable range for burial dating with ratios of TCN is
narrower because short periods of shielding cannot be dis-
tinguished from the effects of erosion on the ratio.
The range of erosion rates over which the technique is
applicable is narrow and depends on the approach,
isotope-mineral system, and measurement precision. Sur-
face erosion rates of continually exposed bedrock
approaching 0.01 mm ka1 can be determined with ana-
lytical precisions (~3 %) and are usually averaged over a
wide exposure time. Faster erosion rates, up to a maxi-
mum of 103 mm ka1 for rock, may be resolved in the rare
instance that the erosion rate is constant over much shorter
exposure durations. Only recently has the TCN method
been able to establish rates of the complex situation of
nonconstant episodic erosion.
The TCN dating method extends beyond the normal
limits of common Quaternary dating methods such as
radiocarbon (5  104 years) and luminescence (105 years).
Therefore, the technique provides a useful bridge between
Quaternary and older dating methods. Furthermore, its
ability to determine erosion and burial histories over the
past 107 years can support exhumation models based on
thermochronology.
Historical developments
While the TCN dating method became established in the
mid-1980s, the theory of the geochronometer was devel-
oped much earlier. Following the discovery of X-rays
and radioactivity at the turn of the nineteenth century,
experiments were undertaken to establish how high above
Earth’s surface atmospheric ionization would continue.
As observation heights increased from towers to balloon
experiments more than 5,000 m high in the atmosphere,
it became apparent that above a few hundred meters the
ionization rate actually increased with height and that the
radiation did not diminish at night or during an eclipse.
The demonstration of a non-terrestrial and nonsolar radia-
tion (ionization with constant day and night and during an
eclipse, Hess, 1912) initiated a new field of physics. Two
decades later, the first cosmic “radioelement” measured
in rock was protactinium in eucolite and eudialyte, col-
lected from surface outcrops in Greenland. Grosse
(1934) carefully demonstrated that the amount of protac-
tinium from the already well-established U-series decay
rates was more than 3 % higher than the parent activity
could explain. He indicated that a better understanding
of the interaction of cosmic rays with matter and cosmic
 TERRESTRIAL COSMOGENIC NUCLIDE DATING
Terrestrial Cosmogenic Nuclide Dating, Table 2 Examples of applications of TCN methods
Method Landform/material
Exposure dating Lava surface
Bedrock fault scarp
Landslide scarp face
Erratics on moraine, esker, or plain
Boulders on alluvial fan
Boulders on a beach
Impact ejecta
Bedrock mountain peaks, tors
Precariously perched boulders
Overturned boulders
Colluvial blocks
Submarine cobble on continental shelf
Debitage, quarried tools
Depth profile dating Multiple sand samples in a terrace
Multiple samples in raised delta foresets
Burial dating/isochron Sediment in caves
dating Lava stack
Sediment buried under sediment
Cobbles buried by water
Bedrock or sediment in glaciated area
Erosion rate Single rock surface
Single outcrop with TCN at saturation
Modern stream sediment
Older fluvial sediment
Sediment sample
Multiple samples in a depth profile
Two adjacent samples at different depth
Paleoaltimetry Single lava surface of known age with large vertical offset Rate of offset
Rapidly uplifted surface
radiation variation with time was necessary to fully under-
stand the production of cosmic radioelements. Knowledge
of cosmic ray flux and interactions continued only through
measurements of cosmogenic nuclides at higher concen-
trations in meteorites, lunar samples, and meteoric sam-
ples (tree rings, manganese nodules, ice sheets) and
sophisticated transport codes capable of simulating parti-
cle interactions at various energies and magnetic field
(Reedy et al., 1983). A current and very comprehensive
review by Olive et al. (2014) provides the state of knowl-
edge of the particle physics and cosmology underlying the
theory.
Measurement of TCN lagged the theory. The low pro-
duction rate of cosmogenic nuclides on Earth’s surface
even at mountain elevation (hundreds of atoms per gram
of mineral per year) resulted in few successful measure-
ments of TCN (Davis and Schaeffer, 1955; Hampel
et al., 1975) until significant improvement in isotope mass
spectrometry. Unlike radiocarbon dating (based on the
decay of 14C produced in atmosphere by cosmic rays
and invented by Willard Libby in 1948), application of
TCN methods has only been practical since the advent of
improved spectrometry of noble gas isotopes and
803
Purpose/test
Eruption age or recurrence rate
Variation in slip rate with time
Acceleration of mass movement
Timing of deglaciation, retreat rate
Age of strain markers for fault kinematics
Isostatic uplift rate or tilt rate of raised shorelines
Age of impact
Age of formation, timing of last glaciation
Timing of seismicity for a given acceleration
Timing of last major flood event
Timing of last seismic shaking
Duration of shelf exposure
Timing of human occupation
Stream incision rate, age of strain marker
Relative sea-level curve, timing of marine limit
Burial age of associated fossils, stream piracy
Eruption recurrence rate
Age of paleoecological, paleontological, or
anthropology records
Duration of transgression
Duration of cold-based ice cover
Outcrop-scale maximum constant erosion rate
Outcrop-scale average constant erosion rate
Average erosion rate for catchment above sample
Paleo-erosion rates of catchment above sample
Erosion rate of surface with short-lived TCN
Erosion rate of sediment surface
Episodic erosion rate on glaciated summits
Rate of uplift using atmospheric depth-dependent ratio
of TCNs
ultimately accelerator mass spectrometry (AMS) in
1977. The technical difficulty lies in the accurate measure-
ment of an extremely small number of atoms, typically 106
to 108 in a sample that might range from a few milligrams
(~1019 atoms) to a kilogram of material (~1025 atoms).
Even after extracting and concentrating the chemical moi-
ety of interest, say aluminum for 26Al measurements, the
isotopic ratios are still daunting, <1014.
During an experiment in which atmospheric 10Be (half-
life 1.39 Ma) was used as a radioactive tracer to prove that
pelagic sediments from the ocean floor were carried
through the subduction zone, and ultimately incorporated
in the lavas erupted by arc volcanoes, Brown et al. (1982)
obtained high concentrations (1  106 atoms g1) of 10Be
in the Columbia Plateau basalt. The Columbia Plateau
basalt had been chosen as a control sample: since it was
produced by hot-spot volcanism which has no connection
with subduction, sediments, or any other “surface” mate-
rial, it was expected to be a blank (<102 atoms g1). All
other samples from hot-spot volcanism in Hawaii and Ice-
land had been blank. The significant difference between
the Columbia Plateau sample and these other samples
was that the other samples were collected within a year
804 TERRESTRIAL COSMOGENIC NUCLIDE DATING
of their eruption, sometimes days, while the Columbia
Plateau sample had lain subaerially exposed for 14 Ma.
It was first thought that the 10Be contamination might be
meteoric 10Be, brought to the basalt by rain. But sequential
etching of the surface proved that the “contamination” was
distributed uniformly throughout the sample. The only
remaining possibility was that the 10Be was produced in
situ as the sample lay on the surface. A quick calculation
showed it was possible. This was the first measurement
of an in situ-produced terrestrial cosmogenic nuclide with
modern methods. Shortly after, five papers in 1986
announced the birth of TCN exposure history and erosion
methods (Craig and Poreda, 1986; Klein et al., 1986;
Kurz, 1986; Nishiizumi et al., 1986; Phillips et al.,
1986). The number of different applications of the TCN
method has continued to grow (Table 2; for reviews, see
Lal, 1991; Gosse and Phillips, 2001; Dunai, 2010;
Granger et al., 2013).
Since Lal’s seminal 1991 paper, the number of parame-
ters considered when interpreting the TCN data has
increased, and uncertainty in their values has improved.
During the last three decades, the community has benefit-
ted from over a hundred studies from both individual
and group efforts to improve sampling strategies, isotope
geochemistry procedures, our knowledge of production
rate through the atmosphere and lithosphere, and our
capacity to process the data. The community shared a
number of different calculators – e.g., CHLOE for 36Cl
ages, erosion rates, and depth profiles (Phillips and
Plummer, 1996), CosmoCALC for ages with any of the
TCN (Vermeesch, 2007), and a Bayesian-like Monte
Carlo-based calculator for depth profile dating (Hidy
et al., 2010). The most significant development began in
the mid-2000s, when two multinational programs
were funded for the primary purpose of improving our
knowledge of TCN systematics. CRONUS-Earth was
funded by the US National Science Foundation and
CRONUS-EU was funded by the European Union. Early
in the process, an online calculator was provided by
members of the CRONUS-Earth group (Balco
et al., 2008). Over the last decade, important data gaps
were filled and improvements in our understanding were
made in nuclear cross sections, radioactive decay rates,
muonic interactions, spatial scaling of production rates,
temporal variations in production rates, and a more robust
assessment of error. The results of these programs have
been recently published or submitted (Phillips et al.,
2014 submitted). One major improvement is the manner
in which the secondary flux is scaled through the atmo-
sphere and the changing geomagnetic field (Lifton
et al., 2014). An updated CRONUS online calculator is
available at http://web1.ittc.ku.edu:8888/. Currently there
remains a lack of consensus regarding some aspects of the
computation used in this and other calculators and
the values of some of the parameters chosen, particularly
the production rates. Although the CRONUS programs
have ended, efforts to improve the TCN methods
continue.
TCN dating systematics
TCN production pathways
At the base of the atmosphere, the most abundant second-
aries capable of producing TCN are neutrons and muons.
Other secondaries either lack energy, have rapid decay
rates, are uncommon, or are absorbed so quickly in the
atmosphere that they generate TCN at a much lower rate.
TCN are produced in exposed minerals by three main
interactions: spallation by high-energy neutrons, capture
of thermal or epithermal neutrons, and muonic
interactions.
At the base of the atmosphere, at sea level, the most
abundant cosmic ray is muons with an average energy of
4.2 GeV. They have a vertical flux of ~70 m2 s1 sr1
(for momentum > 1 GeV c1); incident on a horizontal
surface, the equivalent flux is ~1 cm2 min1. Nucleons
are the next most abundant: the integral vertical intensity
of protons above 1 GeV c1 is ~0.9 m2 s1 sr1 and
for neutrons ~0.45 m2 s1 sr1. Other components of
the flux including the electromagnetic pions and kaons
play a little role in the production of TCNs.
Spallation
A spallation reaction occurs when a low-mass (e.g., neu-
tron, proton) energetic (>30 MeV) particle interacts with
a nucleus. The interaction causes an intranuclear or
hadronic cascade that at first evaporates neutrons and pro-
tons and other light nuclei and then may eventually result
in the breaking apart (fission) of the target nucleus. In gen-
eral, the most probable products have masses close to the
target nucleus or close to the projectile. All TCNs are pro-
duced by spallation, and it is the dominant production
pathway for TCN within the upper 1,000 g cm2 (~4 m)
of the lithosphere.
Thermal and epithermal neutron capture
Neutron capture occurs when a slow neutron is absorbed
by a nucleus to create a nucleus with a mass of one amu
larger. Fast neutrons lose their energy through momentum
transfer during elastic collisions with nuclei. Epithermal
neutrons (0.1 MeV to 0.5 eV) and thermal neutrons
(<0.025 eV) can be captured by a nucleus. The resulting
nucleus is generally unstable and it decays by the immedi-
ate emission of a proton, or beta decay. Excited nuclei
relax by emission of a gamma. While thermal-neutron
capture is possible by many nuclides, the only cosmogenic
nuclides that are produced significantly by thermal-
neutron capture are 14N(n, p)14C, 35Cl(n, g)36Cl,
and 40Ca(n, g)41Ca.
Muonic interactions
Muons (leptons that are 207 more massive than elec-
trons) are produced in cosmic ray showers from the decay
of mesons and comprise more than eighty percent of the
non-electromagnetic cosmic ray flux at Earth’s surface.
However, like other leptons, they do not interact through
the strong force and thus their penetration is much greater
 TERRESTRIAL COSMOGENIC NUCLIDE DATING
than neutrons. Despite their short half-life (106 s) and
their relatively small contribution (<2 %) to total TCN
production at the lithosphere surface, slow negative
muons (e-folding length L ~1,300 g cm2, Heisinger
et al., 2002a) and fast muon (L ~4,320 g cm2, Heisinger
et al., 2002b) are the dominant production pathways at
depths greater than 4 m in rock (>1,000 g cm2). For
comparison, the L for fast neutrons is ~145 g cm2. Muon
contributions to TCN production are therefore important
when applying TCN burial dating methods or when sam-
ples are collected from rock or sediment depths greater
than 1 m.
TCN production
Exposure duration (t) of a basalt boulder on a moraine is
determined by the measurement of the TCN concentration
in a target mineral, such as 3He in pyroxene:
C meas  C i
t¼ ð1Þ
PT , t, 0
where Cmeas is the measured abundance of 3He per gram of
pyroxene, Ci is any concentration of 3He that is either
non-cosmogenic or in situ cosmogenic 3He produced prior
to the deposition of the boulder on the moraine
(i.e., “inherited concentration”), and PT,t,0, the total pro-
duction rate at the surface, is the time-integrated sum of
all cosmogenic nuclide production pathways for 3He in
pyroxene at the surface for the particular time, in
atoms g1a1.
The state of knowledge of production rates and how
they scale spatially and temporally is rapidly improving
(Lifton et al., 2014; Phillips et al., 2014 submitted). Site-
specific time-integrated production rates have been deter-
mined (calibrated) by measuring TCN concentrations in
minerals with simple exposure histories over an indepen-
dently determined duration (Ci is considered zero or is
measured). The sum of production rates from the different
cosmogenic nuclide interactions (fast neutrons, Pnf; ther-
mal neutrons, Pnt; epithermal neutrons, Pne; negative
muons, Pmn; and fast muons, Pmf) depends on the
nuclide-target system, location, environmental factors,
and the particular part of Earth’s geomagnetic field and
atmosphere history corresponding to the exposure period.
For reference, the calibrated TCN production rates are
scaled to sea level, high latitude, under today’s geomag-
netic and atmospheric conditions (e.g., 10Be is produced
in quartz at ~4 atoms g1 year1). Thus, first-order
(up to two orders of magnitude) adjustments to the refer-
ence production rate may be required to account for spatial
variability in atmospheric and geomagnetic controls on
the cosmic ray flux to any location on Earth.
Earlier simplifications with a symmetric standard atmo-
sphere and dipolar magnetic field have now been replaced
by more sophisticated models for both. In general, lower
magnetic field strength will result in a higher production
rate and a softer (lower average) secondary energy spec-
trum that will not penetrate as far. Sites at higher
805
geomagnetic latitudes have higher surface production
rates. The flux of each type of secondary decreases with
penetration through the atmosphere, so production rates
decrease with atmospheric depth. Atmospheric shielding
is a greater control than geomagnetic field effects
(except during reversals) so the highest production rates
are at mountain tops.
The intensity and geometry of the geomagnetic field
and the distribution of atmospheric mass has changed with
time. The effects can cause changes of >25 % over several
millennia for a given site. Periods with higher
paleointensities had lower instantaneous TCN production
rates. During colder periods when the atmosphere
contracted, production rates at high altitudes were less
shielded. Such temporal variation in production is clearly
evident from the variation in abundances of atmospheric
14
C or 10Be abundances recorded in ice sheets or marine
sediments. Fortunately, unlike the instantaneous effect of
these changes on atmospheric cosmogenic nuclide pro-
duction used for radiocarbon dating, terrestrial in situ pro-
duction is integrated over the entire exposure duration, so
TCN dating is not as sensitive to short-term high-
amplitude geomagnetic or atmospheric variations.
Other first-order adjustments are needed to account for
the decrease in production as the average energy of each
type of secondary flux decreases with mass depth
z (g cm2) through rock:
Pz, nf ¼ P0, nf ez=Lnf ð2Þ
where P0 is the surface production rate and Lnf is the effec-
tive e-folding attenuation length for fast neutrons at that
depth. While the concentration of TCN produced from
spallation by fast neutrons generally follows this simple
exponential, thermal-neutron capture interactions do not.
Owing to the moderating effect of water (water vapor in
air, pore water in rock or sediment, snow cover) on ther-
mal neutrons, initially there is a reduced thermal and
epithermal neutron flux in the upper 60 g cm2 of rock
or sediment until the flux re-equilibrates with the expo-
nentially attenuating fast neutron flux. Unlike spallogenic
TCN, the highest concentration of thermal-neutron cap-
ture TCN occurs a few dm below the surface.
Second-order adjustments to the production rate may
be negligible or significant. These include corrections for
erosion or exhumation of the samples, cover by snow,
water, ice, loess, or colluvium (shielding the surface from
fast neutrons); geometry of the sampled surface; partial
shielding by vegetation; elevation changes due to iso-
static, tectonic, or surface processes; and partial shielding
of cosmic ray secondaries due to topography.
Laboratory and field considerations
The field and laboratory procedures will depend on the
objective of the study. Choices of TCN will mainly
depend on the abundance of target minerals with known
production rates, the radionuclide half-life, and whether
806 TERRESTRIAL COSMOGENIC NUCLIDE DATING

or not multiple nuclides or a particular production path-
way is sought.
There is no standardized list of field observations for
TCN methods, partly because the types of TCN applica-
tions vary widely. Table 3 includes most of the important
observations that will be needed to calculate an age or ero-
sion rate, adjust for complexities in exposure history, inter-
pret the data, and evaluate uncertainty. The number of
samples necessary will depend on many factors. If inheri-
tance is possible (e.g., due to exposure prior to the

Terrestrial Cosmogenic Nuclide Dating, Table 3 Field observations

Observation Example Comment

Latitude 72.4472 Convention: south is negative; high precision is useful for site recovery
Longitude 130.3362 Convention: west is negative; high precision is useful for site recovery
Elevation 3,232 m Record method; evidence for surface elevation changes during exposure should be
considered
Pressure Varies Used if atmospheric model does not adequately describe the time-averaged barometric
pressure for the average temperature history
Bulk density 2.7 g cm3 Bulk density of the material sampled; if subsurface sampling, bulk density of the material
above each sample must also be determined
Thickness 1.0 cm Average thickness of the sample
Depth below top 0 cm or 10 g cm2 For surface samples, this is taken as 0 cm. For subsurface samples, the vertical distance
between the landform surface and the top of the sample is recorded. Units vary with
calculator
Surface slope 19 ! 235 This is the dip and dip direction of the slope of the sampled surface. Generally dips less
than 10 will have negligible foreshortening effects, unless there is considerable
topographic shielding
Topographic 235, 310, 40, 130, 170 Trend and plunge ( ) measurements taken at major topographic inflections in the skyline.
shielding 22, 13, 9, 17, 25 This shielding becomes more critical with greater shielding (e.g., in a canyon) and with
greater sampled surface slope
Sample height 130 cm; geometry If the surface of the sampled landform is above the local ground level, such as a boulder
top, it may be useful to provide the entire geometry (L, W, H) when considering neutron
loss
Snow shielding 100 cm for 3 months, An estimate of the thickness and density of snow that may have covered the sampled
50 cm for 2 months, surface. If the surface is above the local ground level, consideration for the sample
0.25 g cm3 height may be required
Geometry effects 30 cm from edge; 25 cm This is needed to compute effective attenuation lengths, compute thermal-neutron
under an overhang that production rates, and evaluate neutron loss
protrudes 80 cm from
cliff
Forest shielding Temperate rainforest for Record the type of forest and other vegetation cover. For boulders, this could account for
majority of time 2 % to >9 % shielding of secondary cosmic ray flux (e.g., boreal forest or temperate
rainforest). For subsurface samples, vegetation shielding could be even greater
Erosion 1 mm ka1; 30 cm max If known, indicate method or evidence (height of protruding quartz vein; striation; soil
development; lichen cover; weathering features; measured). Units vary with calculator
Aggradation 50 cm Gradual or episodic cover by loess, ash, colluvium, ice, water, and other material currently
or previously existing that was added to the surface of the landform after exposure began
Sample type Boulder; braided stream Describe lithology, sediment structures
deposits
Landform type Broad moraine ridge; Provide stratigraphy information, e.g., Qf3. Describe where on the landform the sample
fluvial terrace was collected (subsurface samples under a longitudinal bar in the alluvial fan)
Sample method Cutoff saw; chisel; hand Indicate any difficulties, particularly if thickness is not constant
dug pit
Inheritance High catchment erosion Note evidence of catchment erosion for sediments (e.g., alluvial fans); low rates may
indicate a higher probability of inheritance, yielding older apparent ages or causing
scatter among individual ages on a given landform
Volumetric water 5 % water content in soil The volumetric water content, Qv, is needed to account for moderation of the thermal and
epithermal neutron flux through rock and sediment. Each sedimentary layer and soil
horizon may need its own measurement
Soil properties 15 cm below the 30 m Information about the soil that can help constrain erosion rate (e.g., a local catena, or
mixed zone evidence that the soil has been truncated); evidence that the samples were collected
below any zone of cryoturbation or bioturbation; consideration that a soil horizon may
have gradually altered the bulk density and moisture properties over time
Exhumation of the Cobble may have been Boulders and small clasts may have moved vertically in the sediment and been completely
surface frost heaved, or till may exposed for a shorter time than the actual landform. Exhumation by frost heave or
have been eroded denudation of the landscape should be noted
 TERRESTRIAL COSMOGENIC NUCLIDE DATING
exposure period of interest), concordance of three or more
surface exposure ages may be required to establish if
inheritance is negligible, or subsurface samples in sedi-
ment collected in a depth profile may be needed to adjust
for the average inheritance of a sediment.
While the physical and chemical processing required
will vary with the isotope-mineral system, the objective
is the same: provide a pure and representative target with
sufficient TCN concentration to obtain the desired preci-
sion. Mineral isolation processes after cleaning, crushing,
grinding, and sieving include froth flotation; magnetic,
electrostatic, heavy liquid, air abrasion; and differential
leaching with hydrofluoric, hexafluorosilicic, or hot phos-
phoric acid. For the noble gas measurement, no future
chemistry may be required on the <100 mg of target min-
eral. For AMS, extracting and concentrating the radionu-
clides after dissolution is achieved with isotope dilution
(spiking) followed by dissolution, ion chromatography,
and pH-controlled precipitation, yielding BeO, Al2O3,
AgCl, or other targets which are mixed with a pure metal
to enhance sputtering in the AMS source. Masses between
2 and 200 g will be needed, depending on the concentra-
tion. Extraction of 14C is by thermal extraction of 14C in
the presence of ultrapure O and can be in graphite or
CO2 gas. Most extraction methods require less than 10 g
of quartz. Some elemental analyses of the naturally abun-
dant isotope or a measurement of target nuclei abun-
dances, radiogenic and nucleogenic sources, or neutron
moderators is usually required.
TCN applications: dating of surfaces and
sediments with simple exposure histories
Exposure dating has been conducted in just about every
landscape element to address questions relevant to classi-
cal geomorphology, landscape responses to climatogenic
and tectonic change, kinematics of brittle and ductile
deformation, and applied science regarding geohazard
risks associated with seismicity, volcanism, mass wasting,
aridity, and sea-level change. Various combinations of iso-
topes and sampling strategies can reduce the degrees for
freedom in constraining the factors that control the calcu-
lation of an exposure or burial age.
Single-nuclide surface exposure dating
A simple exposure age on a surface that is not eroding can
be determined with a stable or radionuclide with decay
constant l:
 
1 Cl
t ¼  ln 1  ð3Þ
l P
However, if the surface experienced a constant erosion or
aggradation during the equation, it is more complicated.
Constant erosion and aggradation can be treated explicitly.
As the absorption of the cosmic ray secondaries with mass
is well known, we are not required to sample at the sur-
face. An approximation for the concentration of a nuclide
807
in a rock at depth z (cm) with an erosion rate e (cm a1)
was provided by Lal (1991)
Pðt, 0Þ re  
ðlþre
L Þt
C ðz, tÞ ¼ re e L 1  e þ C ðz, inhÞ elt
lþ
L
ð4Þ
The latest CRONUS calculator (Borchers
et al., submitted) iteratively seeks an exposure time with
a set of equations that allow production rates through the
five interaction pathways to vary over the exposure
period, simultaneously considering changes in effective
attenuation lengths during erosion. If erosion rate is
known and constant, exposure age can be solved. How-
ever, if the erosion rate is uncertain or appears to be epi-
sodic, the normal protocol with a dominantly spallogenic
TCN is to assume e ¼ 0 to obtain a minimum exposure
age for the surface. If the inherited concentration
(C(z,inh)) is also not negligible, but erosion is zero, the
exposure age would be a maximum age. The exposure
age cannot be constrained with one isotope (maximum
nor minimum limit) if both inheritance and erosion are
nonzero. However, if single-nuclide exposure ages from
multiple, spatially distinct samples from the same land-
form are identical, then it may be possible to suggest a
negligible inheritance in all samples.
Multiple nuclide surface exposure dating
To overcome the difficulty related to erosion, two nuclides
can be measured in the same rock surface. If the surface
has been exposed and eroding at a steady state for a long
time, the concentration of radionuclides will reach a
dynamic equilibrium where loss due to decay is matched
by the production according to the above equation. A
short-lived radionuclide that has reached its saturation
concentration C* for a given erosion rate over sufficient
time can be used to determine the erosion rate (Lal, 1991):
 
L P
e¼ l ð5Þ
r C
The surface exposure time can be determined by adjusting
the unsaturated concentration of a second nuclide for this
erosion rate. Alternatives to the saturation approach
include a combination of a TCN with a significant
thermal-neutron capture production rate (e.g., 36Cl) and a
spallogenic isotope (e.g., 10Be). The thermal-neutron loss
near the surface will cause the concentration to increase
with erosion, while the spallogenic nuclide will decrease
for the same erosion rate. Their erosion-adjusted ages will
converge provided that erosion rate is not too fast and that
sufficient 35Cl is in the mineral.
The two-isotope approach can also solve exposure age
and erosion rate simultaneously for two radionuclides
without the need for saturation or thermal-neutron produc-
tion (Figure 2). However, the 26Al/10Be ratio plot
(Nishiizumi et al., 1986) reveals a narrow banana-shaped
808 TERRESTRIAL COSMOGENIC NUCLIDE DATING
Terrestrial Cosmogenic Nuclide Dating, Figure 2 26Al/10Be
versus 10Be plot. Production systematics as per Balco et al. (2008)
for the Lal (1991) and Stone (2000) time invariant scaling model,
initial 10Be production rate of 4.9 atoms g1 and surface
26
Al/10Be of 6.75, and muogenic production from negative and
fast muons (Heisinger et al., 2002a; Heisinger et al., 2002b). Plot
and sample represent a surface of 30 m depth under gravel with
1.9 g cm3 density. The thick red upper curve represents the
history of a sample that experienced continuous exposure with
no erosion (exposure duration in Ma increases to the right).
Ratios plotting within the shaded field (“steady-state erosion
field” (Lal, 1991)) have experienced erosion or minor burial
(curves indicated for 0.1 and 1.0 mm ka1 constant and
continuous erosion). Samples plotting below the shaded field
have experienced a complicated exposure history in which an
exposure was by at least interrupted one burial event. Even at
30 m, the shielding is not complete because of muogenic
production. The duration of burial indicated by the thin blue
curves ranging from 0.25 to 8 Ma represents the burial age for a
single even or a minimum estimate of burial duration if multiple
burial events have occurred, as in cyclic glacial cover. The
hypothetical sample shown with 1s error ellipse could be
interpreted to represent a minimum exposure duration of
0.2 Ma and minimum burial duration of 2 Ma. However, it is
possible that the sediment was buried and reexposed many
times, so the total time represented would be greater than
2.2 Ma. If an amalgamated sediment had a depressed initial ratio
(owing to complex exposure prior to final deposition), then the
burial age would overestimate the timing of the last burial
event.
steady-state island and therefore requires very high preci-
sion to establish erosion and exposure age. Phillips
et al. (1997) have suggested that the 10Be/36Cl may have
the advantage of being more sensitive for experiments
requiring simultaneous solutions of exposure age and ero-
sion rate. Ratios of a noble gas and radionuclide have also
been used, and a three-isotope approach may have the
advantage of a reduction of a degree of freedom. The
multi-isotope approach may also simplify the requirement
to know how production rates have changed with time if,
in some circumstances, the amount of variation is the same
for all measured isotopes in the same sample.
Exposure dating of sediments
The difference between dating the exposure age of a rock
surface and dating the depositional age of a sediment is
that each grain of the allochthonous sediment will have a
different Cinh which must be considered in the calcula-
tions. There are many approaches to dating a sediment
(see Gosse, 2012 for a review). If the objective is to deter-
mine the age of a cut or fill alluvial terrace surface, a fan or
delta surface, moraine, or mass wasting deposit, it is pos-
sible to date large boulders (reducing the effect of snow
or loess cover and exhumation) or cobbles and pebbles
on the surface. As in dating a bedrock surface, a single-
nuclide age on a clast (or amalgamation of clasts) on the
surface of a deposit requires knowledge of erosion
(or aggradation) history and inheritance.
Erosion of the surface can be determined by different
TCN approaches as discussed above. An alternative
approach is to use subsurface samples. The concentration
versus depth profile through a rock or sediment is con-
trolled by erosion rate. The faster the erosion rate, the
steeper the profile until ultimately only a fast muonic
attenuation would be measured. Unfortunately, constant
erosion rates on Earth are generally not that fast, so most
measured depth profiles will be dominantly controlled
by the simple e-folding length for a fast neutron. There-
fore, a single spallogenic nuclide cannot establish a unique
erosion rate or exposure age. However, a depth profile for
a single TCN with sufficient thermal-neutron capture pro-
duction (e.g., 36Cl) can provide this uniqueness. Alterna-
tively, two TCNs can be measured in the same profile to
solve for erosion rate and age (Mercader et al., 2012).
The Cinh in sediment clasts is equally problematic. If
not treated, the apparent exposure age will overestimate
the actual timing of deposition. There are many
approaches, each with assumptions that may be validated
in certain scenarios. One option is to measure the exposure
age for many samples on the surface of the deposit and
assume that all clasts have been eroded similarly. If there
is no variability among ages, then it may be possible to
interpret that inheritance is negligible, as it is very unlikely
that all clasts would have the same inheritance if the Cinh
was a significant component of the total concentration.
This is often the case where sediment is eroded from gla-
ciated catchments or regions with otherwise very high ero-
sion rates. In regions where the sediment is sourced from a
catchment that has not been rapidly eroded, determining
the Cinh is necessary. One option is to measure a sample
that is essentially shielded from cosmic rays (e.g.,
>4,000 g cm2 deep) for long exposure to muons
(Brown et al., 2002) and assume that the sediment at
shallower depths have the same Cinh. This assumption
would only be true in special cases where, for instance,
the sediment flux from the catchment remained constant
 TERRESTRIAL COSMOGENIC NUCLIDE DATING
over that entire thickness. A similar approach that uses the
TCN concentration in modern stream sediment as an indi-
cation of Cinh is similarly invalid for many catchments in
which millennial-scale climate variations have changed
catchment erosion rates. A better and more widely used
treatment of inheritance in alluvium (Anderson
et al., 1996) uses the TCN concentration in at least four sub-
surface samples of amalgamated clasts along a depth pro-
file. The profile will reveal the time-integrated
concentrations from the different interaction pathways as
the rock is advected toward the surface due to erosion.
The exponential decrease in concentration will trend toward
zero with depth if there is no inheritance. However, if Cinh is
nonzero, the concentration versus depth curve will be
asymptotic along the depth axis. The offset corresponds to
the Cinh and can be subtracted from the measured concen-
tration to solve for erosion or exposure age. Hidy
et al. (2010) showed that amalgamations of hundreds of
thousands of sand clasts provide a more precise solution
than dozens of pebbles (pebbles seem to work in rapidly
eroding catchments, whereas many more sand grains are
necessary to characterize the average Cinh where catchment
erosion rate is low, and Cinh is high and more variable).
Burial dating
When rock or sediment is buried and effectively shielded
from any secondary cosmic ray flux, the concentration of
TCNs in the buried minerals will change according to their
decay rate. The noble gases will remain constant
(assuming no unaccounted diffusion) and radionuclides
with faster decay rates will decrease faster. The change
in the ratio of the concentration of two (or more) different
TCNs in the same sample, as long as one is a radionuclide,
depends on burial duration, erosion rate, and depth. Burial
dating has been contributed to many different disciplines:
• Glaciology and paleoclimatology: burial of bedrock by
one or more episodes of glacier cover (e.g., Nishiizumi
et al., 1991)
• Anthropology and archaeology: sediment and associ-
ated human fossils that were shielded in caves (e.g.,
Shen et al., 2009)
• Geomorphology and paleontology: fluvial terraces and
alluvial back to the Pliocene (e.g., Rybczynski
et al., 2013)
Simple burial dating with two isotopes
In the simplest case of a surface that had build up over
exposure time (t) interrupted by a single period of com-
plete shielding from cosmic rays (tb), the burial duration
can be determined using two TCNs with different mean
lives (t) in the same sample:
 
tshort tlong Plongðt, z, eÞ C short
tb ¼  ln
tlong  tshort Pshortðt, z, eÞ C long
where C is the measured conc
809
of exposure for a given depth and erosion rate for the
short- or long-lived isotope (Granger and Muzikar,
2001). In this simple scenario, such as burial of a lava by
another thick lava, there Cinh ¼ 0 atoms g1 and
e ¼ 0 cm a1. Even if the initial exposure duration is not
known, the pre-burial and burial durations may be solved
if the ratio of the two isotopes fall below the steady-state
island in a ratio plot (e.g., 26Al/10Be vs. 10Be, Figure 2),
erosion is assumed negligible, and the surface or sediment
had never experienced a previous burial (or storage) event.
This method requires two isotope measurements in only
one sample. It assumes that burial was complete and
instantaneous; otherwise, production during partial expo-
sure at varying depths with time will need to be
considered.
Isochron burial dating
Balco and Rovey (2008) provided an isochron solution for
the common instance where a surface or sediment has
undergone previous burial episodes. In these cases, the ini-
tial ratio of the isotopes measured in a sample does not cor-
respond
 to the concentration for the production ratio
Pshortðt, z, eÞ
Plongðt, z, eÞ . Instead, prior burial durations will cause the
concentration ratio to be lower at the beginning of the last
burial episode. If ignored, the calculated burial duration
will overestimate the burial age. By using an isochron
approach (Figure 3), the concentration ratio prior to the
last burial episode does not need to be known. The iso-
chron approach requires that multiple samples are col-
lected in the buried rock or sediment and that the
concentrations of the TCN in each sample are different
but that the ratio of the two TCN samples is the same.
One option is to collect samples within a short (<1 m)
depth profile interval below the burial contact. In this case
the concentration for each TCN will diminish with depth if
the buried minerals were exposed prior to burial. The
greater the pre-burial exposure time, the greater the differ-
ences in concentration in the samples. However, the ratio
of two spallogenic nuclides will be approximately con-
stant over this short and shallow interval. For instance,
on a 26Al/10Be isochron plot, the initial slope will be close
to the production ratio (~6.75). During burial, the isotopic
ratios will decrease at a rate proportional to their concen-
tration, so the slope of their concentrations will decrease.
The slope is proportional to the burial duration. In
instances where exposure prior to burial was short (e.g.,
if eruption recurrence intervals were short) or there is no
evidence of a paleosol because of significant erosion prior
to burial, there are other options for obtaining multiple
samples with different concentrations. For instance, there
is a well-documented grain size dependency on concentra-
tion in sediments where deep-seated landsliding has
occurred (Brown et al., 1995), so 26Al and 10Be concentra-
tions in different grain size fractions from a single