Farias Etal Tectonics 2010

Click
Here
TECTONICS, VOL. 29, TC3006, doi:10.1029/2009TC002480, 2010
for
Full
Article
Crustalscale structural architecture in central Chile based

on seismicity and surface geology: Implications for Andean
mountain building
Marcelo Faras,1,2,3 Diana Comte,2 Reynaldo Charrier,1 Joseph Martinod,3
Claire David,1,2,3 Andrs Tassara,2,4 Felipe Tapia,1 and Andrs Fock1,5
Received 26 February 2009; revised 21 October 2009; accepted 23 December 2009; published 29 May 2010.
[1] We document a crustalscale structural model for
the central Chile Andes based on seismicity and surface geology, which consists in a major east verging
rampdetachment structure connecting the subduction
zone with the cordillera. The ramp rises from the subducting slab at 60 km depth to 1520 km below the
western edge of the cordillera, extending eastward as a
10 km depth flat detachment. This structure plays a
fundamental role in the Andean orogenesis because
most of the shortening has been accommodated by
structures rooted in it and allows the distribution of
crustal thickening in a simple shear deformation
mode. Indeed, despite shortening distribution being
very asymmetric (16 km versus 70 km in the western
and eastern side, respectively), the western side is
higher and thicker than what is expected. Yield strength
envelopes show strong rheological control on this
structure. Vp and Vp/Vs variations in the upper mantle
and in the deepest limit of the seismogenic interplate
contact mark the intersection of the ramp with the slab,
which coincides with the blueschisteclogite transition.
Therefore, subduction processes would control the
depth where the major east verging structure may
merge with the slab. Such a rampflat structure is
observed in other parts of the Chilean margin; hence,
it seems to be a firstorder feature in the Andean subduction zone. This structure delimitates upward the
rocks, transmitting part of the plate convergence stress
from the plate interface, and controls mountainbuilding
tectonics, thus playing a key role in the Andean orogeny.
Citation: Faras, M., D. Comte, R. Charrier, J. Martinod, C. David,
A. Tassara, F. Tapia, and A. Fock (2010), Crustalscale structural
architecture in central Chile based on seismicity and surface
1
Departamento de Geologa, FCFM, Universidad de Chile, Santiago,
Chile.
2
Departamento de Geofsica, FCFM, Universidad de Chile, Santiago,
Chile.
3
LMTG, CNRSIRDUniversit de Toulouse, Toulouse, France.
4
Now at Departamento de Ciencias de la Tierra, Universidad de
Concepcin, Concepcin, Chile.
5
Now at Sociedad Qumica y Minera S.A., Antofagasta, Chile.
Copyright 2010 by the American Geophysical Union.

02787407/10/2009TC002480
geology: Implications for Andean mountain building, Tectonics,

29, TC3006, doi:10.1029/2009TC002480.
1. Introduction
[2] The Andes are the Earths longest and highest active
mountain chain formed in an oceancontinent subduction
margin. It is widely recognized that the elevation and crustal
thickness of this mountain range have been mainly produced
by crustal shortening because of the interaction between the
subducting oceanic Nazca plate and the overriding continental South American plate. In the shallow levels of the
continental crust, most of the shortening has been accommodated in the eastern flank of the range in several fold
andthrust belts (Figure 1). Therefore, deformation occurs in
the hanging walls of large east verging detachments in
which ramps are predominantly east verging, showing that
tectonic transport in the Andes preferentially goes to the east
[e.g., Isacks, 1988; Allmendinger et al., 1990; Allmendinger
and Gubbels, 1996; Ramos et al., 1996; Allmendinger and
Zapata, 2000; Cristallini and Ramos, 2000; Giambiagi
and Ramos, 2002; McQuarrie, 2002; Giambiagi et al.,
2003a; Ramos et al., 2004; Arriagada et al., 2006; Vergs
et al., 2007; McQuarrie et al., 2008]. Therefore, detachments related to eastern peripheral deformation belts appear
to be the most relevant structures controlling the Andean
orogeny, as occurs in many modern and old mountain
chains [e.g., Cook and Varsek, 1994].
[3] Shortening accommodated in the eastern flank of the
Andes has been widely studied, usually favored by the
sedimentary constitution of the rocks involved in the fold
and thrust belts, as well as by the great database of geophysical data and core drilling performed by oil exploration
companies. These factors have allowed determining the
depth and geometry of detachment levels, as well as
performing accurate temporal and geometrical reconstruction of deformation. The relevance of determining the depth
and geometry of detachments resides in the fact that they
define the boundary conditions for structural restoration.
The lack of a wellconstrained determination of the detachment (in which superficial structures are rooted) may lead to
overestimate or underestimate the amount of shortening;
therefore, the location of this rooting structure is fundamental.
[4] The westward prolongation of detachments into the
arc and forearc region, in which plate interaction occurs, is
poorly understood. This situation is mainly produced by the
fact that shortening is very much smaller here, thrusts verge
mainly to the west (in northern Chile, from Muoz and
TC3006
1 of 22
TC3006
FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE
Figure 1. (a) Tectonic framework of the Andean margin. (b) Maximum elevation and Moho depth from
32C to 37S. Elevations calculated from SRTM90m DEM, and crustal thickness is after Tassara et al.
[2006]. (c) Main tectonic and morphological features of the Andes of central Chile and western Argentina.
Seismologic stations of the permanent network of the University of Chile (white inverted triangles) and
temporary network deployed during JanuaryApril 2004 (dark inverted triangles) are shown. Grid in
Figure 1c corresponds to the region and cells in which tomography was performed. Absolute plate motion
velocity is after Gripp and Gordon [2002]. Focal mechanisms of the two greatest shallow crust earthquakes of the last years are those calculated by Harvard CMT.
2 of 22
TC3006
TC3006
Charrier [1996], Victor et al. [2004], Faras et al. [2005],

and Garca and Hrail [2005] and in central Chile, from
Ramos et al. [1996], Charrier et al. [2005], and Fock et al.
[2006]), numerous intrusive bodies perturb the structure and
the lithological continuity, and there is no evidence for the
downdip prolongation of these structures nor their rooting
detachments. These circumstances have led to a poor knowledge on the structural connection between the mountain belt
and the subduction zone, as well as on the mechanisms of
stressstrain transfer from the interplate contact toward the
continent. Although some studies have attempted to balance
the forces released in the subduction zone and those transferred to the orogen [e.g., Isacks, 1988; Kono et al., 1989;
Lamb and Davis, 2003; Yez and Cembrano, 2004; Lamb,
2006], these models did not considered the geometry of the
structures involved in the resulting strain transfer.
[5] In the Andes of central Chile (3335S), the construction of wellconstrained crustalscale cross sections has
presented several obstacles, such as the difficulties to
identify stratigraphic markers and to determine ages of
deposition and deformation. Among other reasons, this situation resides in the fact that outcropping sequences in this
region are predominantly of an igneous composition, they
exhibit large lateral variations in facies and thickness, and
there is a pervasive lowgrade metamorphism affecting them
[e.g., Levi et al., 1989; Vergara et al., 1993; Muoz et al.,
2006].
[6] However, the knowledge on the tectonic evolution of
the western side of the range in this region has increased
during the last years due to a more accurate mapping and
dating of tectonic and depositional events. This has been
possible due to a more systematic search for unaltered levels
[Fuentes, 2004; Muoz et al., 2006], the use of more penetrative radioisotopic dating [Charrier et al., 1996, 2002,
2005; Fock et al., 2006; Montecinos et al., 2008; Oliveros et
al., 2008], the discovery of abundant Cenozoic mammal
fauna [Wyss et al., 1990; Flynn et al., 1995; Croft et al.,
2003; Flynn et al., 2003], and the use of quantitative geomorphology [Faras et al., 2008a] and lowtemperature
thermochronology [Faras et al., 2008a; Maksaev et al.,
2009]. These new studies have allowed to establish that
the contractive development of the Andes in central Chile
occurred fundamentally since the latest Oligoceneearliest
Miocene [e.g., Charrier et al., 2002, 2005] (similar to the
central Argentinean Andes [Ramos et al., 1996; Giambiagi
and Ramos, 2002; Giambiagi et al., 2003a; Ramos et al.,
2004]), and that the main stage of surface uplift occurred
during the late Mioceneearly Pliocene [Faras et al.,
2008a].
[7] Another source of data that has been used to understand the forearc evolution comes from the seismology.
Although most of the studies in this discipline performed in
the Andean region have paid attention to the subduction
zone (nucleation and propagation of megathrust earthquakes
[M 8]), last years improvements of permanent seismological networks and the deployment of temporary networks
have permitted detecting abundant shallow (<20 km depth)
crustal seismicity beneath the Chilean forearcarc region.
On the recorded data, three events stand out in the Chilean
TC3006
Andes (Figure 1): the Aroma earthquake (24 July 2001, Mw =

6.3) at 1930S [Faras et al., 2005; Legrand et al., 2007],
the Altos del Teno event (28 August 2004, Mw = 6.5) at
3510S [Faras et al., 2006; Comte et al., 2008], and the
more recent Aysn event (21 April 2007, ML = 6.2) at
4515S. These earthquakes, as well as most of the detected
shallow seismicity, occurred in association with structural
systems very relevant during the Cenozoic mountain
building [Barrientos et al., 2004; Faras, 2007; Faras et
al., 2005, 2006; Comte et al., 2008; Lange et al., 2008].
Shallow crustal seismicity occurring in relationship to great
fault systems emphasizes the relevance of this kind of
earthquakes in order to explore the relevance of these
structures during mountain building, their subsurface prolongation into the continental lithosphere, and their association with main detachments and links to the subduction
interface.
[8] In this study, we integrate seismological and geological data obtained in the Andes of central Chile (3335S) in
order to address two main questions: (1) which is the fore
arcarc structural array that relates tectonics in the subduction zone to the mountain belt, and (2) which has been the
contribution of this structural array for the construction of
the mountain belt. This study is based on the analysis of
seismicity recorded by the permanent network of the Seismologic Survey at the Universidad de Chile and a temporary
network deployed along one of the largest and longest
structural systems in the high Andes of central Chile. We
correlate these data with surface geology in order to infer the
crustalscale structural architecture across the whole
mountain range. Based on our findings, we show a major
detachment running below the entire mountain belt, which
would be connected to the subduction zone through a ramp
that cuts across the forearc lithosphere wedge. We finally
propose a qualitative approach to the mechanisms controlling the transference of deformation from the interplate
interface toward the mountain belt, which could be very
significant for understanding orogenic mechanisms not only
in the study region, but also along the Chilean subduction
margin and even in other subduction orogens.
2. Tectonic and Geological Settings

2.1. Generalities
[9] The subduction of the oceanic Nazca (former Farallon)
plate beneath the South American continent is the main
tectonic process along the Andean margin since Jurassic
times [e.g., see Mpodozis and Ramos, 1989; Charrier et al.,
2007, and references therein]. Late Cenozoic evolution of
convergence rates shows a strong diminution to its current
rate (N78E, 8 cm yr1 relative plate motion; Figure 1a).
Current absolute plate motion relative to hot spots frame for
the South American and Nazca plates are 4.8 and 3.2 cm yr1,
respectively [Gripp and Gordon, 2002].
[10] The ChileanPampean flatslab subduction region
(27S33S), where slab dip is <10 between 100 and 150 km
depth [e.g., Jordan et al., 1983; Pardo et al., 2002],
represents a major alongstrike change on the Andean
subduction system. North and south of this segment, the dip
3 of 22
TC3006
of the slab is nearly 30E, being the typical example of the

Chileantype subduction (in the sense of Uyeda and
Kanamori [1979]). This segmentation has been interpreted
as the result of buoyancy forces exerted by the subduction of
the Juan Fernndez ridge at 32.5S (Figure 1) [Pilger, 1981;
Nur and BenAvraham, 1981; Gutscher et al., 2000; Yez
et al., 2001]. In addition, some distinctive features appear at
approximately 33S (Figure 1): (1) strike change (Maipo
Orocline in this work) of the trench, coast, and mountain belt
from NS, northward, to NNESSW, southward; (2) decrease
of magmatic activity in the late Miocene until its extinction
at 1.9 Ma in the flatslab region [Kay et al., 1991; Ramos et
al., 2002]; (3) decrease in mountain elevation and crustal
thickness south of 33S; (4) increase in flexural rigidity
south of 34S along the mountain belt [e.g., PrezGussiny
et al., 2007; Tassara et al., 2007]; (5) southward segmentation of the mountain range by a longitudinal valley
(Central Depression); and (6) absence of great basement
involved ranges as the Frontal Cordillera, Pampean Ranges,
and Precordillera south of 3415S.
2.2. Geologic Evolution
[11] The Andean region between 33S and 35S comprises
five major continental morphostructural units (Figures 1
and 2), from west to east: Coastal Cordillera, Central
Depression, Principal Cordillera (in this work, subdivided
into a western, central and eastern Principal Cordillera, see
Figure 2), Frontal Cordillera (not present south of 3415S),
and active the foreland.
[12] The Coastal Cordillera consists of a late Paleozoic
Triassic basement in its western flank and east dipping
Jurassic to Cretaceous intraarc sequences in its eastern
flank, whose series extends to the middle of the Central
Depression Thomas [1958] (see Figure 2).
[13] From the eastern half of the Central Depression to the
central Principal Cordillera, an extensional basin developed
during late Eocene to late Oligocene times (Abanico basin),
which began to be inverted in the latest Oligoceneearliest
Miocene [Godoy et al., 1999; Charrier et al., 2002]. This basin
was filled by the predominantly volcanicvolcanoclastic
Abanico Formation. In early stages of inversion, folding and
highangle reverse faulting mainly affected both basin edges
[Fock et al., 2006]. In the middle zone of the former basin,
the Farellones Formation, predominantly volcanic, was
deposited during early to middle Miocene times. This unit is
generally mildly folded, excepting near the edges of the
Abanico basin, where overlays the Abanico Formation either
unconformably or developing growth strata in its lower
layers (older than 16 Ma, see Figure 3) [Charrier et al., 2002,
2005; Fock et al., 2006]. After 16 Ma, contractive deformation migrated toward the eastern Principal Cordillera.
TC3006
[14] The eastern Principal Cordillera domain is characterized by the presence of Mesozoic sequences of predominant sedimentary composition. These units were deposited
into the northern sector of the Neuqun backarc basin,
which developed in this zone during most of the Mesozoic
[Uliana et al., 1989; Giambiagi et al., 2003b, and references
therein]. This part of the Principal Cordillera has accommodated most of the shortening in this region since 16 Ma
[Giambiagi and Ramos, 2002; Giambiagi et al., 2003a;
Ramos et al., 2004] when shortening almost ended in western
sectors.
[15] After 8.5 Ma, just east from the eastern Principal
Cordillera, the uplift of Proterozoiclower Triassic basement
by the means of the inversion of riftrelated highangle faults
concluded in the raise of the Frontal Cordillera [Giambiagi
2004]. Simultaneously or shortly after, highangle outof
sequence reverse faults deformed the eastern Abanico basin
and the eastern Principal Cordillera [Giambiagi and Ramos,
2002; Giambiagi et al., 2003a; Fock et al., 2006]. At circa
4 Ma, shortening migrated farther east to the foreland
[Giambiagi et al., 2003a], and the Chilean belt reached most
of its presentday elevation, diminishing drastically the
uplift rates in the Chilean belt since then in about 1 order of
magnitude from 1 to 2 mm yr1 during the late Miocene to
0.1 mm yr1 [Faras et al., 2008a].
[16] Magmatic history in the central Chile Andes is
mainly related to an almost continuous eastward migration
of the arc since the Jurassic [e.g., Kay et al., 2005; Charrier
et al., 2007]. Coeval to the beginning of contractive tectonics in the early Miocene, many granitic intrusions
emplaced in the westernmost Principal Cordillera (lower
Miocene intrusive belt, Figure 2) [e.g., Kay et al., 2005;
Charrier et al., 2007]. Shortly after, the arc migrated
slightly to the east, as evidenced by the volcanic rocks of the
Farellones Formation, which represent the locus of the arc
until the Langhian in a zone that did not accommodate
significant deformation. At the end of the volcanic pulses
related to this formation, the magmatic arc migrated again to
the east, intruding the eastern flank of the Farellones and
Abanico formations. This magmatic activity formed a long
intrusive chain that was active in the late Miocene between
13 and 7 Ma [e.g., Kurtz et al., 1997; Kay et al., 2005;
Charrier et al., 2007] (hereafter we will refer to this chain as
late Miocene intrusive belt). After this event, volcanic/
magmatic activity declined, but some pulses shifted to the
west, forming the porphyry copper deposits of the El Teniente
and Ro BlancoLos Bronces. This magmatic arc was active
between 9 and 4 Ma [e.g., Maksaev et al., 2004; Deckart
et al., 2005] being coeval with the outofsequence event
and the uplift of the cordillera [Faras et al., 2008a]. After
this time, magmatism migrated again eastward to the eastern
Figure 2. (a) Simplified geological map and (b) cross sections of the Andes of central Chile and western Argentina. Only
main inverse faults active during the Neogene are plotted. El Diablo and Las Leas faults belong to the El Fierro fault system. Modified after Servicio Geolgico Minero Argentino [1997], Godoy et al. [1999], Charrier et al. [2002], Servicio
Nacional de Geologa y Minera [2002], Giambiagi et al. [2003a], and Fock et al. [2006]. Cross sections only show
what can be observed in the field. AFTB, Aconcagua foldandthrust belt; MFTB, Malarge foldandthrust belt.
4 of 22
TC3006
Figure 2
5 of 22
TC3006
TC3006
Figure 3
6 of 22
TC3006
TC3006
Principal Cordillera, place in which the presentday volcanic

arc is located.
TC3006
[17] We present two cross section (Figure 2b) in which we

integrate previous works (Thiele [1980], Giambiagi et al.
[2003a], and Fock et al. [2006] for the Santiago section,
P1 in Figure 2b; Charrier [1981], Godoy et al. [1999], and
Charrier et al. [2002, 2005] for the Cachapoal section, P2 in
Figure 2b) with new data obtained in this study. In the
following paragraph, we describe the major structural and
geological features along these sections that can be considered representatives of the study region because of their
location.
that exhumation has been very low and not capable of raise
the 100120C isotherm since the Abanico Formation
deposition (that is, no more than 34 km considering a
thermal gradient of 2540C km1) (see dating in the work
by Faras et al. [2008a]).
[20] Just east from the mountain front, a series of synclines and anticlines were developed between circa 22 and
16 Ma, which is evidenced by growth strata developed in
the lower layers of the Farellones Formation [Fock et al.,
2006] (Figures 2b, 3a, and 3b). To the east, some folds
and faults deform subtly the Cenozoic sequence, showing a
predominant east vergence in both sections (Figure 2b).
Particularly, the San Jos fold (Figures 2b and 3b) stands out
because in its core it is developed an east vergent duplex
involving sedimentary layers of the Abanico Formation.
Above this zone, the Farellones Formation presents growth
strata developed prior 16 Ma [Fock et al., 2006].
3.1. Eastern Central Depression and Western Principal

Cordillera
[18] Only Cenozoic deposits crop out in this sector
(Abanico and Farellones formations, and intrusive bodies).
The western edge of this region is characterized by east
dipping partially inverted normal faults (Los Angeles fault
system according to Carter and Aguirre [1965], Infiernillo
fault according to Fock et al. [2006]) (Figure 2) in which
Cenozoic units override Mesozoic sequences. This fault
system would be the western edge of the Abanico Basin
[Fock et al., 2006].
[19] The western edge of the Principal Cordillera is defined
by a west vergent reverse fault system (San RamnPocuro
fault and its southward prolongation [Thiele, 1980; Charrier
et al., 2005; Fock et al., 2006; Rauld et al., 2006; Armijo et
al., 2010] (Figure 2). On the basis of geomorphologic
markers, the cordilleran front structural system would have
accommodated a maximum vertical throw of 0.71.1 km
since the late Miocene at the latitude of Santiago, and 600
800 m at 35S [Faras et al., 2008a]). This structural system
has not accommodated much more throw since basin
inversion than the presentday elevation of this zone. This
because apatite fission tracks dating coincides with stratigraphic ages in the rocks of the Abanico Formation and
subsidence related to the orogenic load in the Central
Depression has been almost negligible (no more than 500 m
in Santiago [Faras et al., 2008b]. The former reason means
3.2. Central Principal Cordillera

[21] As in the previous sector, only Cenozoic deposits
crop out here. These units are bounded to the east by east
verging faults that produce the overriding of the Abanico
Formation on the Mesozoic sequences of the eastern Principal Cordillera. These faults would constitute the eastern
edge of the Abanico basin, and they can be traced over more
than 300 km along strike (Figures 2a and 3c). The main
faults of this system received different names according to
their latitude: El Diablo fault at 3345S [Fock et al., 2006],
Las LeasEspinoza fault at 3430S [Charrier et al., 2002],
and El Fierro fault at approximately 35S [Davidson and
Vicente, 1973]. Hereafter, we will name these faults as El
Fierro fault system.
[22] In the central Principal Cordillera, very significant
folds and thrusts affect the Abanico Formation. Folds
exhibit maximum amplitudes over 23 km. Folding in this
sector can be mostly related to the development of east
verging faultbend and faultpropagation folds (Figures 2b,
3c, and 3d). In this domain, the geometry of folds and
related faults suggest tectonic inversion of both Cenozoic
and Mesozoic extensional basins (Figures 3c and 3e). Folds
are usually cut by breakthrough reverse faults, as well as by
outofsequence faults and back thrusts likely related to the
late Mioceneearly Pliocene event [Fock et al., 2006] or at
least posterior to the basin inversion. Among them, the
3. Structural Features of the Mountain Belt in

Central Chile and Western Central Argentina
Figure 3. (a) Growth strata developed in the eastern limb of the San Ramn anticline. (b) East vergent fold and thrusts
related to the San Jos anticline, immediately south of Maipo river. Deformation in less competent layers of the Abanico
Formation (sedimentary layers) is produced by progressive development of east vergent thrusts, which also produced eastward growth strata in the Farellones Formation during the lower Miocene and the progressive increase of slope in the western limb of the fold. Asterisk indicates zircon SHRIMP UPb age [Fock et al., 2006]. (c) Structure in the eastern flank of the
Abanico basin in the Volcn valley (upper course of the Maipo drainage basin). Lo Valds Formation is a Tithonean
Neocomian marine unit and the Colimapu Formation is an early Cretaceous clasic unit. Double asterisk indicates zircon
SHRIMP UPb unpublished age obtained in this work. (d) View to the structure along the Las Leas valley (central
Principal Cordillera in profile P2, Figure 2b). (e) View to the southern slope of the Las Leas valley immediately east of the
El Fierro thrust (eastern Principal Cordillera in profile P2, Figure 2b). The Ro Damas Formation is a Kimmeridgian unit,
the LeasEspinoza Formation is a Callovian volcanicmarine unit, and the Termas del Flaco Formation is a Tithonian unit
(equivalent to the Lo Valds Formation).
7 of 22
TC3006
TC3006
structures would be rooted in a detachment fault located at

midcrustal levels.
4. Seismology
4.1. Data Acquisition
[25] We used the seismologic data recorded by the permanent network of the Seismologic Survey at the Universidad de Chile between 1980 and 2004 complemented
with a temporary network deployed from January to April
2004 (Figures 1 and 2). The permanent network has
24 seismologic stations in the study region and the temporary
network consisted in seven shortperiod threecomponent
stations. The final database includes 23,449 events, from
which 217 earthquakes were detected by the temporary
network.
Figure 4. Initial 1D P wave velocity models.
highangle west verging ChacayesYesillo fault (see profile

P1 in Figures 2b and 3c) stands out because of its >2 km of
vertical throw [Charrier et al., 2005] cutting across the
previously deformed series.
3.3. Eastern Principal Cordillera and Frontal
Cordillera
[23] The eastern Principal Cordillera extends east of the El
Fierro fault system and consists of Mesozoic marine and
continental sedimentary deposits including an Oxfordian
gypsum level (Figures 2b, 3c, and 3d) and some volcanic
layers. Neogene syntectonic foreland basin deposits (Altos
de Tunuyn Conglomerates) overlying the Mesozoic
sequence record the beginning of shortening and the eastward migration of deformation to the eastern Principal
Cordillera shortly after 16 Ma [Giambiagi et al., 2003a].
Shortening in this region (47 km) has been mainly
accommodated by east verging thrusts and related folds
[Giambiagi and Ramos, 2002]. Structural restoration based
on balanced cross sections [Giambiagi and Ramos, 2002]
suggests that these thrusts would be rooted in a 10 km
depth east verging detachment fault along the latitude of the
profile P1 (Figure 2b).
[24] North of 3415S, the crystalline basement of the
Frontal Cordillera crops out east of the Principal Cordillera.
There, several basement blocks were uplifted by highangle
east verging reverse faults accommodating at least 15 km of
shortening between 9 and 6 Ma [Giambiagi and Ramos,
2002]. Giambiagi et al. [2003a] and Ramos et al. [2004]
interpreted this thickskinned contractive style as a result
of the inversion of faults related to the Mesozoic extensional
basins. According to these authors, the reactivation of these
4.2. Tomographic Inversion and Hypocentral

Relocalization Procedure
[26] Once arrival times were picked, hypocenters were
first determined using the HYPOINVERSE program [Klein,
1978] with a 1D P wave velocity model based on that of
Thierer et al. [2005] (Figure 4). Onedimensional S wave
velocity model was determined using 1.75 as Vp/Vs. Each
event was located with different trial depths in order to
minimize the effect of the initial conditions on the final
hypocentral determination. Trial depths were varied between
0 and 250 km with an increment of 5 km. The location with
the lowest root mean square misfit and with the maximum
number of body waves first arrivals was selected for each
event. This procedure ended with 140,000 and 95,000 P
and S arrival times, respectively. Because this procedure
determines preliminary hypocentral locations, these results
are considered preliminary travel times.
[27] In order to obtain better hypocentral locations, we
constructed a simple 3D velocity structure that was used to
relocate the earthquakes. This procedure consists in using
the preliminary hypocenters and seismic wave travel times
to construct a 3D velocity structure using the SPHYPIT90/
SPHREL3D90 program (see details in the work by Roecker et
al. [1993] and at http://gretchen.geo.rpi.edu/roecker/manuals/
sphypit90/Sphypit90.html). Inversion was performed on a
region divided into 6 7 blocks with a grid spacing of 30
30 km2 (see Figures 1 and 6) and 12 layers of 10 km thick,
except the shallowest one, which is 13 km thick. P wave and
S wave velocities were independently inverted.
[28] The resulting Vp and Vs models were used to relocate
the hypocenters and adjust travel times. Hypocenters were
classified and filtered according to the following criteria
(used by Abers and Roecker [1991]): (1) standard deviation
of the residual travel times <0.7 s, (2), maximum spatial
standard error < 0.2 s, and (3) maximum spatial correction
the hypocenter <0.2 s, and total change in location from start
to finish < 40 km. These filtered events were used to generate a new inversion. Iterations continued until resulting
models converged to a solution: three iterations were needed
to achieve convergence. Inversion resulted in a final model
consisting in 1008 final blocks with 877 blocks considered
8 of 22
TC3006
TC3006
Figure 5. Standard deviation calculated for each event using the nine velocity models. Data filtering
was made considering the most reliable earthquakes.
as reliable (those having >20 rays hits, however most of
blocks are hit by >1000 rays). The resulting velocity model
was used to relocate the hypocenters, which were classified
and filtered again.
4.3. Model Validation and Filtering
[29] It is necessary to point out that the tomography was
performed with the aim of determining better hypocentral
locations rather than obtaining an accurate velocity field
because of its rough resolution (30 30 10 km each cell).
However, the following tests show that this final model has
a relevant regional validity that can give insight about the
real velocity field in the study region.
[30] Tomography testing consisted in the development of
eight additional models using 1D velocity structures randomly perturbed from the initial model of Thierer et al.
[2005] (Figure 4). Comparison among the nine resulting
velocity models shows a minimal deviation at each cell
(<3%). Because of this reason, we took the mean velocity of
the nine models as the final Vp and Vs models that will be
used in the rest of this work (considering the deviation as the
involved error at each cell).
[31] In order to test the sensibility on hypocenter relocalization, we compare the hypocenters determined by the
nine models (Figure 5). This comparison consisted in the
determination of the standard deviation in latitude, longitude
and depth for each event. Figure 5 illustrates that more than
the half of the events has minimal standard deviation,
showing that the different models tend to a similar hypocentral relocalization neglecting the deviation of the initial
velocity models. Actually, the events that exhibited minimal
standard deviation coincide with those events having minimal RMS derived from the procedure (<0.3 s).
[32] In order to use the most reliable hypocenters, we only

considered those having standard deviations smaller than
1 km in latitude and longitude and depth deviations minor
than 2.5 km. In addition, we used those events having
procedural RMS < 0.3 s. This course of action resulted in
7077 events that will be used in the rest of this work. The
filtering treatment used in the earthquakes detected by the
temporary network (217 events) was different: we consider
that they are a priori better constrained because of the
proximity of stations to the quakes (less than 30 km); this
filtering consisted in selecting those event having RMS <
0.35 s, resulting in 149 events.
4.4. Results and Analysis of Seismologic Data
4.4.1. General Results
[33] The final distribution of hypocenters shows that most
of the crustal seismicity is located beneath the Principal
Cordillera and eastern Central Depression at depths shallower than 20 km (Figure 6). Superficial seismicity in the
offshore and coastal fore arc mainly corresponds to events
occurring in the interplate contact, except some earthquakes
located in the overriding plate. As previously suggested by
Barrientos et al. [2004] and Charrier et al. [2005], the most
relevant cluster of shallow crustal seismicity is located close
to the ChileArgentina boundary aligned with the El Fierro
fault system (Figure 6).
4.4.2. Ramp Flat Seismologic Structure
[34] In order to analyze the relationship between seismicity, seismic velocity fields and lithospheric structure,
Figure 7 shows three profiles crossing perpendicularly the
orogen strike and covering most of the study region (note
that these profiles have different azimuths because the
orogen strike changes in the Maipo orocline, see Figure 2).
9 of 22
TC3006
TC3006
Figure 6. Regional distribution of final hypocenters. Solid lines correspond to the locations of cross
sections in Figure 7. Grid corresponds to that used as the inversion region (each cell has a 30 30 km
planar size).
At first sight, the geometry displayed by seismicity can be
interpreted as a rampflat crustalscale structure.
[35] Seismicity associated with the flat geometry is
located beneath the Chilean Principal Cordillera. This
structure dips 10W in the western Principal Cordillera,
where it is located at 1510 km depth. In the central eastern
Principal Cordillera, the structure is located at 105 km
depth and dips 5W. In this sector, the flat structure
coincides fairly well with the geometry and depth proposed
by Giambiagi et al. [2003a] for the detachment that has
controlled shortening in the eastern Principal Cordillera.
[36] The ramp segment dips 40W and extends downward from the western edge of the Principal Cordillera to the
Moho below the Central Depression. Although seismicity in
the ramp segment is not present in the lithosphere mantle in
the southern sections (possibly reflecting the aseismic
behavior of this part of the lithosphere), it is well detected
in the section across the Central Depression at 33.2S
(Figure 7a). In this profile, the seismic ramp intersects the
WadattiBenioff zone at 60 km depth.
[37] The seismic ramp, and its likely location in zones in
which it is not seismically present, can be correlated with
slight discontinuities on Vp and Vp/Vs within the lithosphere
mantle wedge (Figure 7). Discontinuity on the P wavefield
consists in an eastward velocity increase from 7.37.7 to
7.98.2 km s1. Discontinuities on Vp/Vs are not much
obvious; however, the ramp (or what could be its southward
prolongation in Figures 7b and 7c) is surrounded by high
Vp/Vs regions (>1.80) (Figure 7).
[38] In spite of the former features related to the ramp or

its inferred southward prolongation, it is also likely that
seismicity defining the ramp in the mantle could be related
to dehydrating associated with fluid releasing from the slab.
In this case, faulting into the lithosphere mantle associated
with the ramp structure would not occur, and thus this zone
of the fore arc would only transmit the stress that is released
in deformation along and above the crustal rampflat
structure that is very well observed seismically in the study
region (see section 8.2).
5. Rheological Analysis
[39] The strength of the continental lithosphere is controlled by its depthdependent rheological structure. This is
mainly depending on the thickness and composition of
crustal layers, the thickness of the lithosphere mantle, the
temperature structure, the strain rate, and the presence or
absence of fluids [e.g., Carter and Tsenn, 1987; Kirby and
Kronenberg, 1987; Burov and Diament, 1995, 1996;
Cloetingh et al., 2005].
[40] In order to analyze the probable rheological control on
the rampflat structure, we constructed four onedimensional
columns of compressive yield strength envelopes (Figure 8)
across the profile BB (Figure 7b). They are based on the
3D lithospheric compositional and geometrical model of
Tassara et al. [2006] (which consists in two crustal layers
and a lithosphere mantle; Figure 8a), a 2D geothermal
gradient (based on the work by Oleskevich et al. [1999] and
10 of 22
TC3006
Figure 7
11 of 22
TC3006
TC3006
Yez and Cembrano [2004]) (Figure 8b), and experimental

rheological parameters.
[41] The 2D geothermal gradient used in this analysis
was mainly constructed using a heat diffusion model in
which the downgoing slab produces the downward isotherm
advection [Oleskevich et al., 1999; Yez and Cembrano,
2004]. The effects of the volcanic arc on the thermal gradient were minimized: we consider that the longterm
effects of the slab are more relevant than the continuously
migrating magmatic/volcanic arc.
[42] The rheological parameters used in this work are
exactly those used by Tassara et al. [2006]. They assumed a
quartzite rheology for a granitic upper crust, a dry diabase
rheology for a quartzdiorite lower crust, and a wet dunite
rheology for a continental harzburgite lithosphere mantle,
whose values are those determined by Carter and Tsenn
[1987] and Burov and Diament [1995] (see Figure 8 caption
for details).
[43] Resulting yield strength envelopes (Figure 8c) illustrate that in the western flank of the eastern Principal Cordillera (column IV), lowviscosity ductile rocks should
prevail between 8 and 20 km depth (upper crust) and
between 35 and 60 km depth (lower crust). These low
viscosity ductile zones are confined by three rigid (high
strength brittle and/or highviscosity ductile) layers located
between 12 and 3 km depth, between 20 and 35 km depth,
and immediately below the Moho.
[44] The geothermal gradient diminishes to the west. Its
effect is clearly observed in column III (western edge of the
central Principal Cordillera), showing that the upper low
viscosity ductile zone wedges out toward the west and disappears beneath the western Principal Cordillera (column II).
This analysis predicts mechanic coupling between the upper
and lower crusts beneath the westernmost Principal Cordillera and Central Depression (column I). Yield strength
envelopes also suggest that the viscosity in the base of the
crust would remain low beneath the western edge of the
Principal Cordillera. However, farther west, strong coupling
between the crust and the upper mantle is observed (column I).
This is produced by the cooling exerted by the subduction of
a cold oceanic lithosphere that deflects downward the isotherms by advection. Such a kind of strong and cold continental lithosphere has been proposed for the northern Chile
fore arc, which would furthermore behave as a strong
eastward intender [Tassara, 2005]. In central Chile, this
characteristic also appears in this analyze; that is, the lithosphere fore arc west of the Central Depression is expected
to be highly rigid and cold.
[45] Beneath the Principal Cordillera (columns III and IV),
the top of the upper crustal lowviscosity ductile zone is
fairly well correlated with the flat segment of the seismic
structure (Figure 8c). The western edge of the flat segment
TC3006
correlates well with the western edge of this lowviscosity

ductile zone (columns II and III). The ramp segment extends
from there downward to the top of the lower crustal low
viscosity ductile zone near the Moho (column II). This
shift from an upper to a lower detachment level is usual in
ramp structures, which generally connect weak zones across
a rigid layer [see Cook and Varsek, 1994, and references
therein]. In spite of the lack of direct evidence for the lower
ductile zone, the activity of its top as a deep east verging
detachment could explain the reactivation of the deeply
rooted basement structures that controlled the uplift of the
Frontal Cordillera in the late Miocene [Giambiagi and
Ramos, 2002; Giambiagi et al., 2003a; Ramos et al., 2004],
as well as those that are now controlling the uplift of the San
Rafael block farther east [Ramos et al., 2004].
[46] Yield strength envelopes predict that the rocks
located above the detachment beneath the Principal Cordillera
should prevail very rigid. However, this zone concentrates
abundant seismicity normally aligned with some structures
observed at the surface. Because this zone has been widely
deformed during the Neogene (Figure 2b), it is likely that
seismic activity in this region is related to the reactivation of
ancient discontinuities and fractures. Indeed, Charrier et al.
[2002, 2005] proposed that the deformation within the
Abanico Formation is mostly related to the inversion of the
normal faults that controlled the development of the Eocene
Oligocene extensional Abanico basin. Likewise, Neogene
deformation in the eastern Principal Cordillera and Frontal
Cordillera is related to the reactivation of older normal faults
as well as fracturing along less competent levels such as
gypsum layers, old Paleozoic sutures and basement fabrics
[Ramos et al., 1996; Giambiagi and Ramos, 2002; Giambiagi
et al., 2003a]. Therefore, seismicity above the detachment
would be related to the reactivation of older structures and
discontinuities rather than new faults.
[47] Yield strength envelopes analysis shows that the fore
arc beneath the Coastal Cordillera should be very rigid, even
though some earthquakes related to the ramp segment are
located in this zone (Figure 7a). This fact suggests that the
rheology of this zone may differ from what is expected,
which could be explained by the presence of fluid released
from the subducting slab or lithosphere discontinuities
derived from the forearc evolution. We will discuss this
subject in section 8.
6. Integrating Seismologic Data and Surface

Geology
6.1. Construction of Balanced Cross Sections
[48] In spite of the good correlation existing between yield
strength envelopes and the locus of seismicity, both are
Figure 7. Crustalscale cross sections perpendicular to the orogen strike showing velocity structures and relocated hypocenters. Location and orientation of sections are indicated in Figure 6. Moho depth is after Tassara et al. [2006]. In all
sections, earthquakes were projected using a 40 km width box (+20/20 km from the profile). White circles are the events
recorded by the permanent network, and red circles are those obtained from the temporary network. The rampflat structure
has been drawn in Vp/Vs sections. All sections are at the same scale, using the same color palette for velocities, and there is
no vertical exaggeration.
12 of 22
TC3006
Figure 8. Compressional yield strength envelopes analysis. (a) Figure 6b used as reference to indicate
the location of the columns where yield strength envelopes were calculated. Moho and intracrustal density
discontinuity (ICD) depth are after Tassara et al. [2006]. The 400C isotherm is reported for reference.
(b) Geothermal gradient for the four columns used for yield strength envelop calculation. (c) Resulting
yield strength envelops. Geothermal gradient is approximated from Oleskevich et al. [1999] and Yez
and Cembrano [2004]. The ICD delimitates an upper crust with quartzite composition (H = 1.9 105
[J mol1], A = 5 1012 [N3 m6 s1] [Burov and Diament, 1995]) from a lower crust with quartzdiorite
composition (H = 2.12 105 [J mol1], A = 5.1 1015[N2.4 m5.76 s1] [Burov and Diament, 1995]).
Mantle has been considered with a wet dunite composition (H = 4.44 105 [J mol1], A = 7.94 1017
[N3.35 m11.22 s1] [Carter and Tsenn, 1987]). Maximum deviatoric compressive stress in the continental
lithosphere (100 MPa) is according to England and Molnar [1991]. Locations of columns are as follows:
column I is below the Central Depression, column II is below the western edge of the Principal Cordillera,
column III is below the central Principal Cordillera, and column IV is beneath the Chilean side of the
eastern Principal Cordillera.
13 of 22
TC3006
TC3006
TC3006
Figure 9. Structural cross sections and shallow seismicity. (a) Maipo profile (P1 in Figure 2b)
(b) Cachapoal profile (P2 in Figure 2b). Structural restoration shows that less than 20 km (16 km of
shortening minimum) has been accommodated within the Cenozoic sequences (Abanico and Farellones
formations). This shortening is distributed almost equitably on both flanks of the former extensional
basin. Note the features of the bordering fault systems, which present evidence for tectonic inversion.
presentday features, and hence, they cannot be directly
associated with a longterm situation. However, as we will
expose in the following paragraphs, they are also well correlated with surface geology and structural reconstruction
made on the eastern flank of the Andes [Giambiagi and
Ramos, 2002; Giambiagi et al., 2003a].
[49] At depth, the seismic flat structure coincides fairly
well with the detachment proposed by Giambiagi and
Ramos [2002], Giambiagi et al. [2003a], and Ramos et al.
[2004] based on balanced cross sections. In order to analyze the role of this structure on mountain building, we
constructed two upper crustal cross sections integrating

surface geology and seismicity (Figure 9). These sections
display the structure only in the Chilean side of the cordillera: the Argentinean side of the belt has been already
studied in detail by Giambiagi and Ramos [2002] and
Giambiagi et al. [2003a] and the resolution of recorded
seismicity does not extent farther east to Argentina.
[50] Cross sections are constrained by downplug projection of surface structure and its correlation with seismicity, geometrical constraints, and the age of deformation.
Despite the precise downward prolongation of some par-
14 of 22
TC3006
TC3006

Table 1. Shortening Across the Andes at 33.8Sa
First Stage
2215 Ma
Abanico basin
Eastern Principal Cordillera
Frontal Cordillera
Foreland
Total
16
6
Second Stage
158.5 Ma
Third Stage
8.54 Ma
24
17
15
24
32
Fourth Stage
40 Ma
Total
6
6
16
47
15
6
84
22
Approximated shortening in km. The values are according to Giambiagi and Ramos [2002].
Estimated in this work.
ticular faults may be debated, these cross sections show the

geometry of the major orogenscale structures. It is necessary to point out that this restoration could underestimate the
shortening because of minimization of fault slip when they
are not related to folding or where the cut across the folds. It
is likely that this minimization will not modify significantly
the actual shortening because erosionexhumation related to
vertical movements of main faults are not very relevant and
even fission track dating did not record exhumation in rocks
of the hanging walls [Faras et al., 2008a], at least in the
western Principal Cordillera.
[51] We also expect that the use of seismicity to determine
the depth of the detachment fault controlling the deformation observed at the surface would better constrain the
structural construction. In fact, it is widely recognized that
the depth of the detachment respect to the surface structures
does determine the amount of shortening and that assuming
a certain depth without more constraints than those observed
at the surface may lead to underestimating or overestimating
the total deformation along a cross section. Moreover,
performing of balanced cross sections usually consider that
thickness of sequences involved in deformation is constant.
However, the Abanico Formation reports important changes
in its thickness [e.g., Charrier et al., 2002]. Moreover, the
Mesozoic sequences also present great difference in thickness, from about 30 km between the Coastal Cordillera and
Central Depression (intraarc domain) to about 10 km in the
eastern Principal Cordillera (Neuqun basin).
6.2. Shortening Magnitude and Timing
[52] In the zone where the Abanico basin developed
(eastern Central Depression, western and central Principal
Cordillera), the orogen approximately displays a symmetric
doublevergency system of faults, preserving a central
portion that remains scarcely deformed (Figure 9). A minimum estimation of shortening in this part of the chain is
16 km (20% of the total shortening across the mountain
belt, see Table 1). At this place, most of the contractive
deformation occurred in the lower Miocene, resulting in the
partial inversion of the Abanico basin. Deformation is distributed almost equitably on both flanks. Thus, the resulting
geometry is consistent with the inversion of an extensional
rampflat listric fault system [McClay, 1995] (Figure 9), as
well as local structural geometry evidences basin inversion
(see also Figure 3). During the basin inversion, Mesozoic
sequence in the eastern Principal Cordillera accommodated
6 km of shortening [Giambiagi and Ramos, 2002].
[53] After circa 16 Ma, shortening migrated from the

Abanico basin to the east. Initially, shortening affected the
eastern Principal Cordillera, propagating to the Frontal
Cordillera at circa 9 Ma, and returning into the eastern
Principal Cordillera during an outofsequence event (84 Ma)
in the study region. It is necessary to point out that timing
and shortening magnitude present wide variation along
strike: shortening has clear decreasing trending to the south
and being slightly younger to the south [Ramos et al., 1996;
Giambiagi et al., 2008, 2009]. Finally, shortening migrated
to the foreland at circa 4 Ma [Giambiagi et al., 2003a].
Deformation in the eastern side of the mountain range has
been predominantly accommodated by east verging thrusts
and some backthrusts. Shortening in the eastern Principal
Cordillera, Frontal Cordillera, and foreland is at least 62 km
at 33.8S [Giambiagi and Ramos, 2002], which represents
about 80% of the total shortening across the chain at this
latitude (see Table 1).
6.3. Synthesis of Deformation Events
[54] Considering the timing of deformation, most of the
shortening occurred during three major events at 33.8S
(Table 1): (1) Abanico basin inversion (16 km of shortening between 22 and 16 Ma, and 6 km of shortening in the
eastern Principal Cordillera before 15 Ma), (2) thinskinned
foldandthrust belt development in the eastern Principal
Cordillera (24 km of shortening between 16 and 8.5 Ma
[Giambiagi and Ramos, 2002]), and (3) uplift of the Frontal
Cordillera (15 km of shortening between 8.5 and 6 Ma) and
outofsequence thrusting in the central eastern Principal
Cordillera (17 km of shortening between 8.5 and 4 Ma
[Giambiagi and Ramos, 2002]).
[55] The outofsequence thrusting event in the eastern
Principal Cordillera represents a disruption of the eastward
migration of shortening. Shortening migrated eastward
along the detachment until the highangle basement faults
rooted in the Frontal Cordillera were reactivated. This
reactivation caused the return of the deformation to the axis of
the mountain belt as outofsequence thrusting [Cristallini
2004]. Considering that this event was coeval with the
migration of the magmatic activity to west [e.g., Kay et al.,
2005] and that in this new arc developed the porphyry
copper deposits, it is very likely a relationship. Preliminarily, we think that the westward migration of shortening and
the high crustal thickening reached at this time would alter
the magmatic ascent and differentiation. After 4 Ma, the
15 of 22
TC3006
magmatic activity ended in the porphyry copper deposits

[Maksaev et al., 2004; Deckart et al., 2005], coeval with the
migration of shortening to the foreland. During this new
stage, about 6 km of shortening has been accommodated
[Giambiagi and Ramos, 2002]. At this time, the high cordillera did not accommodate significant shortening in the
study region. In turn, strikeslip deformation is presently
reported by seismic activity in earthquakes M > 5 [Faras et
al., 2006; Faras, 2007]. According to Faras et al. [2008a],
the migration of contractive tectonics to the foreland and the
beginning of a strikeslip regime in the mountain belt would
be related to the opposition of shortening exerted by the
high elevation reached by the mountain belt in the Pliocene.
7. Discussion: Implications for Mountain

Building in Subduction Zones
7.1. Simple Shear and Pure Shear Mountain Building
[56] Before 22 Ma, crustal thickness would have been
moderate (<35 km thick) and approximately uniform across
the Abanico basin according to studies on the geochemistry
of the basinrelated volcanic deposits [e.g., Fuentes, 2004;
Kay et al., 2005; Muoz et al., 2006; Montecinos et al.,
2008]. In fact, there is no structural evidence of relevant
shortening after the Permian and before the late Oligocene
in this Andean region [Charrier et al., 2002; Giambiagi et al.,
2003a; Ramos et al., 2004; Fock et al., 2006; Giambiagi et al.,
2009].
[57] Although in both flanks of the Abanico basin crustal
thickness was approximately similar before inversion and
surface shortening accommodated in both sides has been
approximately the same amount, the presentday crust is
20 km thicker below the eastern boundary of the former
basin (central Principal Cordillera) than beneath the western
Principal Cordillera (Figure 10). This suggests that in situ
surface deformation cannot explain the thickening of the
crust in a pure shear mode (in the sense of Allmendinger
and Gubbels [1996]). Therefore, other thickening mechanisms would have driven the construction of the Chilean
flank of the belt.
[58] We propose that some of the 70 km of surface
shortening accommodated east of the Abanico basin has
been transferred to the west beneath the detachment in a
simple shear mode rather than in a pure shear mode
(both in the sense Allmendinger and Gubbels [1996] and see
Figure 10), similarly to the proposition of Ramos et al.
[2004]. It is necessary to point out that this simple shear
mode of deformation does not mean that all shortening is
transferred to the west, but only partially. The importance of
deep shortening in the western part of the chain is also
supported by the fact that most of the surface and rock uplift
of the western and central Principal Cordillera occurred
between 10 and 4 Ma, even though most of the surface
shortening was accommodated before 16 Ma [Faras et al.,
2008a]. Furthermore, the fact that subsidence is almost
negligible in the peripheral Central Depression [Faras et
al., 2008a, 2008b], despite the lack of enough shortening
in the westernmost Principal Cordillera supporting this part
of the chain [Faras et al., 2008a], implies that deep
TC3006
thickening should have occurred in order to compensate the

elevation of the westernmost Andes. This implies that the
amount of crustal shortening inferred from geophysical data
(supported by determining of crustal thickness and its history of thickening [e.g., Introcaso et al., 1992; Pose et al.,
2005]) matches the amount of shortening derived from
structural cross sections, as well as the uplift timing and
magnitude [Faras et al., 2008a].
[59] The advance to west of the crust beneath the detachment would be opposed by the westward increasing of
the lithosphere rigidity in the fore arc (Figures 8 and 10).
This opposition seems to be evidenced by the seismic cluster
located in the ramp immediately above the Moho (Figure 9a),
which presents compressive focal mechanisms [Pardo et al.,
2008]. Decreasing westward movement of the crust beneath
the rampflat structure means that thickening occurred in a
mixed pure and simple shear mode. Thus, east vergent
crustal wedging would have controlled the accommodation
of the thick mountain root beneath the High Andes. This
situation can finally explain the decreasing crustal thickness
and mountain elevation to the west (Figure 10), as well as
the minimal subsidence occurring in the Central Depression.
Moreover, decreasing westward advance of the crust below
the detachment could explain the fact that seismicity related
to the ramp contacting the slab or occurring in the upper
mantle is not be observed everywhere. If the Moho has been
displaced along the ramp fault will remain unsolved until
better accurate geophysical images are obtained, even
though receiver function derived images along transects at
30S and 36S do not show this situation [Gilbert et al.,
2006]. It is likely that if displacements occur in the Moho,
ductile behavior in the base of the crust would allow the
redistribution of the displaced material.
[60] A major east verging rampflat structure with a
geometry also controlled by the rheology of the continental
plate has been already proposed for the northern Chile
margin (at 1930S) by Faras et al. [2005] and Tassara
[2005] based on the works of Isacks [1988] and Lamb et
al. [1997], among others. These authors proposed that this
structure would be connected to the detachment fault that
extends through the Altiplano to the Eastern Cordillera and
Subandean zone where most of the shortening has been
accommodated [e.g., McQuarrie, 2002; see also Ramos et
al., 2004]. According to this model, the ramp encloses
west and upward a rigid fore arc acting as a pseudoindenter
that resists the westward advance of the crustal mass located
beneath the detachment (see Figure 10). It results in crustal
thickening in the Precordillera and Western Cordillera, in a
zone where the surface shortening has been moderate (less
than 20 km) and essentially older than uplift [e.g., Garca,
2002; Victor et al., 2004; Faras et al., 2005; Hoke et al.,
2007; Riquelme et al., 2007], and where subsidence in the
Central Depression can be also considered negligible
[Faras et al., 2005, 2008b].
[61] Our model for the structural architecture in which the
Andean Cordillera developed is consistent not only with
evidence from the central Chilecentral Argentina Andes,
but also from the AltiplanoPuna central Andes. This model
is based on the fact that reported shortening along this
Andean region has been mostly accommodated by east
16 of 22
TC3006
vergent thrusts, in which west vergent thrusts correspond to

back thrust. In spite of that, relevant west vergent thrusts
occur along the western cordillera front (west vergent thrust
system in northernmost Chile [e.g., Muoz and Charrier,
1996; Garca, 2002; Victor et al., 2004; Faras et al., 2005;
Garca and Hrail, 2005] and the San RamnPocuro fault
in central Chile [Charrier et al., 2002, 2005; Fock et al.,
TC3006
2006; Rauld et al., 2006; Armijo et al., 2010]). Nevertheless, these structures are not much relevant in the context of
the Andean shortening. In fact, west vergent structures do
not exhibit more than 23 km of vertical throw along the
Chilean flank of the belt (which is evidenced by thermochronological data in central Chile [Faras et al., 2008a] and
by direct structural observations in northern Chile [Victor et
Figure 10
17 of 22
TC3006
al., 2004; Faras et al., 2005]) and its accommodated

shortening has been only a minimal fraction of the total
shortening across the belt [cf. Garca, 2002; Victor et al.,
2004; Faras et al., 2005; this work]. Furthermore, considering that subsidence in the peripheral Central Depression
has been almost negligible, the relevance of west vergent
thrusts along the western cordillera flank on crustal thickening (and therefore on Andean building) is therefore minimal.
Consequently, evidence supports our proposition of crustal
deep thickening related to the east vergent rampdetachment
structure; on the contrary, evidence does not support west
vergent model for the Andean development proposed by
Armijo et al. [2010].
7.2. StressStrain Transfer From the Interplate
Contact
[62] The east vergent ramp emerging from the interplate
contact area or the downward prolongation of its trace from
the crust to the mantle has been visualized at different latitudes along the margin, intersecting the slab approximately
at the same depth (60 km). Hypocentral location of
earthquakes within the overriding plate show that such a
structure is active at 19S [Comte et al., 1999; David et al.,
2002] and 27S [Comte et al., 2002; Pardo et al., 2002], and
seismic images visualize a strong west dipping reflector
immediately above the slab at 38S [Gross et al., 2007].
Furthermore, comparable features at a similar depth have
been reported in other subduction zones (Central America
[e.g., DincAkdogan et al., 2007; Syracuse et al., 2008],
northwestern Pacific [e.g., Zhao et al., 2000; Mishra et al.,
2003], and Cascadia [e.g., Zhao et al., 2001]). Furthermore,
in northern Chile, David et al. [2002] determined that
earthquakes along the ramp immediately above the slab
have compressive mechanisms. Nevertheless, we agree that
there is no enough evidence supporting the real activity of
the ramp in the lithosphere mantle because of the absence of
seismicity in all the sections. Therefore, it is likely that
faulting extents into this deep region only in some places,
probably in relationship with ancient discontinuities and/or
with mineral reactions associated with slab dehydration and
uppermost continental mantle hydration (serpentinization),
which downdip limit should be related to the transformation
from blueschist to eclogite facies (see below).
[63] If the intersection of the ramp or its deep prolongation
into the mantle with the slab is located everywhere at similar
depths in the Chilean fore arc, it is likely that subduction
processes control this feature. In fact, this intersection
TC3006
coincides with the deepest limit of the seismogenic contact

along the Chilean subduction zone (i.e., along the place in
which thrust interplate earthquakes propagate [Suarez and
Comte [1993]) (Figure 10). In addition, the continental
mantle in this zone presents relevant variations on Vp and
Vp/Vs respect to the initial onedimensional model (Figure 7).
In spite of the uncertainties related to the velocity fields,
these variations could be related to mantle serpentinization
because the referred intersection would also coincide with
the 400500C isotherm (Figures 7 and 9), which is the
upper limit of serpentinite stability [Carlson and Miller,
2003]. Empirical relationships between the degree of upper
continental mantle serpentinization and Vp [Carlson and
Miller, 2003] predict that the observed discontinuity would
correspond to a change from 0% to 20% of serpentinization
at the conditions of pressuretemperature expected for this
zone. In addition, high Vp/Vs ratios at depths minor than
60 km in the mantle wedge (Figure 7) are also consistent with
serpentinization [e.g., Kamiya and Kobayashi, 2000].
[64] Nonetheless, the fact that in northern and central
Chile the depth at which the ramp would intersect the slab is
approximately the same despite the different thermal gradient
expected in both regions suggests that not only the mainly
temperaturecontrolled serpentinization would determine
this feature. Therefore, it is likely that pressure also exerts a
very significant control. In fact, approximately at this depth a
strong slab dewatering occurs due to the metamorphic
reactions related to the blueschisteclogite transition. In order
to test whether metamorphic reactions, serpentinization, or
the interplay between both control this feature, thermomechanical modeling is required.
[65] It is clear that an important shift on the mechanic
behavior along the interplate contact occurs in the intersection of the ramp with the slab at nearly 60 km depth.
Following Lamb and Davis [2003] and Lamb [2006], the
stress produced by the plate convergence is mostly transferred to the overriding plate along the seismogenic interplate contact, which also occurs at this depth (Figure 10).
Moreover, Tassara [2005] proposed that the rigid behavior
of the fore arc (which is expected to be strongly coupled
with the slab) would promote more effectively the stress
transfer to the continent. In this context, the ramp structure
in the lithosphere mantle or its geometrical prolongation into
this zone control the strain transfer and delimitate upward
the rocks that transmit part of the plate convergence forces
toward the continental crust. In this way, this zone is crucial
in controlling mountain building.
Figure 10. Model for crustal growth and the relevance of the rampdetachment structure at approximately 3350S.
(a) Two different models of mountain building (models modified after Allmendinger and Gubbels [1996]). In pure
shear mode of deformation, the crust thickens in the same place where surface shortening occurs. In turn, in simple shear
mode, thickening occurs far from the place where surface shortening occurs due to the presence of a detachment that
transports the deep crust to the left. (b) Initial setting before shortening at 22 Ma. (c) Presentday (shortened) crustal
configuration of the continental plate. Shortening in the eastern Principal Cordillera, Frontal Cordillera, and foreland
according to Giambiagi and Ramos [2002]. Moho depth is after Tassara et al. [2006]. The 400C isotherm is based on the
work by Oleskevich et al. [1999] and Yez and Cembrano [2004]. Note that the deepest limit of the seismogenic contact
also coincides with the blueschisteclogite transition. It can be observed that despite the differences of shortening at the
surface, crustal thickening has been distributed to the west in a partial simple shear mode of deformation. LVZ is a low
velocity zone expected in this area because of magmatic plume. Cuyania Terrane is according to Ramos et al. [2004].
18 of 22
TC3006
[66] On the basis of the evidence presented in this work,

the rampdetachment structure seems to be the firstorder
feature controlling the transference of strain and stress from
the subduction zone to the mountain belt not only in the
Andes of central Chile, but also probably along the entire
Andean margin. Because this structure should be strongly
controlled by subduction processes and similar features are
observed in other subduction regions, it is likely that the
model presented in this work holds for other mountain belts
formed in a subduction regime.
8. Conclusion
[67] Using the seismicity recorded in central Chile by
permanent and temporary networks, we performed a 3D
tomography inversion that led to the relocalization of the
most reliable hypocenters. We showed the presence of a
crustalscale rampflat structure that connects the subduction zone at 60 km depth with the mountain belt at 10 km.
The flat segment crosses the entire mountain belt and correlates with the east verging detachment that accommodated
most of the upper crust shortening during the Neogene
mountain building.
[68] Geological cross sections show that in the Chilean
side of the belt, upper crustal shortening was much smaller
than in the Argentinean foldandthrust belts (1/5 versus 4/5
of the total shortening). In the western part of the central
Chile Andes, neither the presentday crustal thickness nor
the uplift of this side of the mountain belt can be explained
by surface shortening. In fact, most of the Neogene crustal
thickening and uplift of the western part of the central Chile
Andes would result from the shortening accommodated
beneath the detachment.
[69] Despite huge latitudinal contrasts in the morphological and tectonic evolution of the Chilean Andes, a similar
general lithosphere structural scheme in which a major east
verging fault system emerges from the interplate contact
TC3006
area at 60 km depth and controls the structuring of the

Andes has been proposed for northern Chile. Inferences
made on the south central Chile region also suggest such a
structure there, as well as observations made in different
subduction zones around the Pacific basin. The intersection
of the ramp with the slab coincides with the deepest limit of
the seismogenic interplate contact. It is also marked by sharp
variations in the overriding mantle seismic velocities that
can be interpreted as a result of serpentinization of the
lithospheric mantle wedge probably in relationship with
blueschisteclogite reactions in the slab.
[70] Therefore, we propose that subduction controls the
tectonic behavior of the fore arc, weakening the zones by
which strain stress is transferred to the mountain range from
the plate interface. Likewise, the major east verging ramp
delimitates upward rigid rocks that transmit part of the plate
convergence forces toward the continental lithosphere.
[71] This model suggests that the east verging rampflat
structure is the firstorder structure in the Andean mountain
belt orogeny. Because this structural architecture is strongly
controlled by the subduction factory and similar features can
be observed in other regions around the Pacific basin, this
model could be potentially applicable to other subduction
margins.
[72] Acknowledgments. This work was supported by FONDECYT
grants 1030965 and 11085022, PBCT ANILLO ACT 18 and PDA07
Project, INSU grant Relief de la Terre. Impact du climat sur la dynamique
du relief des Andes: quantification et modlisation, and an IRD Phd scholarship to M. Faras. The authors particularly recognize the labor made by
the Seismologic Survey at the University of Chile. We acknowledge Steven
Roecker for providing the SPHREL90/SPHYPIT programs. Useful discussion with C. Arriagada, M. Pardo, G. Hrail, A. Reynaldo, and G. Yez
helped develop and clarify our ideas. Some figures were made using
GMT 4 [Wessel and Smith, 1998] and GRASS 6.3 programs; GTOPO30,
SRTM90 topographic data, and the 2 min gridded ocean bathymetry of
Smith and Sandwell [1997] were used in some figures.
References
Abers, G. A., and S. Roecker (1991), Deep structure of
an arccontinent collision: Earthquake relocation
and inversion for upper mantle P and S wave velocities beneath Papua New Guinea, J. Geophys. Res.,
96, 63796401, doi:10.1029/91JB00145.
Allmendinger, R. W., and T. Gubbels (1996), Pure
and simple shear plateau uplift, AltiplanoPuna,
Argentina and Bolivia, Tectonophysics, 259, 114,
doi:10.1016/0040-1951(96)00024-8.
Allmendinger, R. W., and T. R. Zapata (2000), The
footwall ramp of the Subandean decollement, northernmost Argentina, from extended correlation of
seismic reflection data, Tectonophysics, 321, 37
55, doi:10.1016/S0040-1951(00)00077-9.
Allmendinger, R. W., D. Figueroa, D. Snyder, J. Beer,
C. Mpodozis, and B. L. Isacks (1990), Foreland
shortening and crustal balancing in the Andes at
30S latitude, Tectonics, 9(4), 789809,
doi:10.1029/TC009i004p00789.
Armijo, R., R. Rauld, R. Thiele, G. Vargas, J. Campos,
R. Lacassin, and E. Kausel (2010), The West
Andean Thrust, the San Ramn Fault, and the seismic hazard for Santiago, Chile, Tectonics, 29,
TC2007, doi:10.1029/2008TC002427.
Arriagada, C., P. R. Cobbold, and P. Roperch (2006),
Salar de Atacama basin: A record of compressional tectonics in the central Andes since the mid
Cretaceous, Tectonics, 25, TC1008, doi:10.1029/

2004TC001770.
Barrientos, S., E. Vera, P. Alvarado, and T. Monfret
(2004), Crustal seismicity in central Chile, J. South
Am. Earth Sci., 16, 759768, doi:10.1016/j.
jsames.2003.12.001.
Burov, E. B., and M. Diament (1995), The effective
elastic thickness (Te) of continental lithosphere:
What does it really mean?, J. Geophys. Res., 100,
39053927, doi:10.1029/94JB02770.
Burov, E., and M. Diament (1996), Isostasy, equivalent
elastic thickness, and inelastic rheology of continents
and oceans, Geology, 24, 419422, doi:10.1130/
0091-7613(1996)024<0419:IEETAI>2.3.CO;2.
Carlson, R. L., and D. J. Miller (2003), Mantle wedge
water contents estimated from seismic velocities in
partially serpentinized peridotites, Geophys. Res.
Lett., 30(5), 1250, doi:10.1029/2002GL016600.
Carter, W. D., and L. Aguirre (1965), Structural geology of the Aconcagua province and its relationship to
the Central Valley graben, Chile, Geol. Soc. Am.
Bull., 76, 651664, doi:10.1130/0016-7606(1965)
76[651:SGOAPA]2.0.CO;2.
Carter, N. L., and M. C. Tsenn (1987), Flow properties
of continental lithosphere, Tectonophysics, 136, 27
63, doi:10.1016/0040-1951(87)90333-7.
Charrier, R. (1981), Geologie der chilenischen Hauptkordillere zwischen 34 und 3430 sdlicher Breite
19 of 22
und ihre tektonische, magmatische und palogeographische Entwicklung, Ph.D. thesis, 270 pp.,
Freie Univ. Berlin, Berlin, Germany.
Charrier, R., A. R. Wyss, J. J. Flynn, C. C. Swisher,
M. A. Norell, F. Zapatta, C. McKenna, and M. J.
Novacek (1996), New evidence for late Mesozoic
early Cenozoic evolution of the Chilean Andes in
the upper Tinguiririca valley (35S), central Chile,
J. South Am. Earth Sci., 9, 393422, doi:10.1016/
S0895-9811(96)00035-1.
Charrier, R., O. Baeza, S. Elgueta, J. J. Flynn, P. Gans,
S. M. Kay, N. Muoz, A. R. Wyss, and E. Zurita
(2002), Evidence for Cenozoic extensional basin
development and tectonic inversion south of the
flatslab segment, southern central Andes, Chile
(3336S.L.), J. South Am. Earth Sci., 15, 117
139, doi:10.1016/S0895-9811(02)00009-3.
Charrier, R., M. Bustamante, D. Comte, S. Elgueta, J. J.
Flynn, N. Iturra, N. Muoz, M. Pardo, R. Thiele,
and A. R. Wyss (2005), The Abanico Extensional
Basin: Regional extension, chronology of tectonic
inversion, and relation to shallow seismic activity
and Andean uplift, Neues Jahrb. Geol. Palaeontol.
Abh., 236, 4347.
Charrier, R., L. Pinto, and M. P. Rodriguez (2007), Tectonostratigraphic evolution of the Andean Orogen in
Chile, in The Geology of Chile, edited by T. Moreno
and W. Gibbons, pp. 21114, Geol. Soc., London.
TC3006
Cloetingh, S., P. A. Ziegler, F. Beekman, P. A. M.

Andriessen, N. Hardebol, and P. Dzes (2005),
Intraplate deformation and 3D rheological structure
of the Rhine Rift System and adjacent areas of the
northern Alpine foreland, Int. J. Earth Sci., 94,
758778, doi:10.1007/s00531-005-0502-3.
Comte, D., L. Dorbath, M. Pardo, T. Monfret, H. Haessler,
L. Rivera, M. Frogneux, B. Glass, and C. Meneses
(1999), A doublelayered seismic zone in Arica,
northern Chile, Geophys. Res. Lett., 26, 1965
1968, doi:10.1029/1999GL900447.
Comte, D., H. Haessler, L. Dorbath, M. Pardo, T. Monfret,
A. Lavenu, B. Pontoise, and Y. Hello (2002), Seismicity and stress distribution in the Copiapo, northern Chile subduction zone using combined on and
offshore seismic observations, Phys. Earth Planet.
Inter., 132, 197217, doi:10.1016/S0031-9201(02)
00052-3.
Comte, D., M. Faras, R. Charrier, and A. Gonzlez
(2008), Active tectonics in the central Chilean
Andes: 3D tomography based on the aftershock
sequence of the 28 August 2004 shallow crustal
earthquake, Eos Trans, AGU, 89(53), Fall Meet.
Suppl., Abstract T41A1932.
Cook, F. A., and J. L. Varsek (1994), Orogenscale decollements, Rev. Geophys., 32, 3760, doi:10.1029/
93RG02515.
Cristallini, E. O., and V. A. Ramos (2000), Thick
skinned and thinskinned thrusting in La Ramada
fold and thrust belt: Crustal evolution of the High
Andes of San Juan, Argentina (32 SL), Tectonophysics, 317, 205235, doi:10.1016/S0040-1951
(99)00276-0.
Croft, D. A., J. J. Flynn, and A. R. Wyss (2003), Diversification of mesotheriids (Mammalia:Notoungulata:Typotheria) in the middle latitudes of South
America, J. Vertebr. Paleontol., 23(3, Suppl), 43A.
David, C., J. Martinod, D. Comte, G. Hrail, and
H. Haessler (2002), Intracontinental seismicity and
Neogene deformation of the Andean forearc in the
region of Arica (18.5S19.5S), paper presented
at 5th International Symposium on Andean Geodynamics, Inst. de Rech. pour le Dv., Toulouse,
France.
Davidson, J., and J.C. Vicente (1973), Caractersticas
paleogeogrficas y estructurales del rea fronteriza
de las Nacientes del Teno (Chile) y Santa Elena
(Argentina) (Cordillera Principal, 35 a 3515 de
latitud sur), Actas Congr. Geol. Argent., V, 1155.
Deckart, K., A. H. Clark, C. Aguilar, R. Vargas, A. Bertens,
J. K. Mortensen, and M. Fanning (2005), Magmatic
and hydrothermal chronology of the giant Ro
Blanco porphyry copper deposit, central Chile:
Implications of an integrated UPb and 40Ar/39Ar
database, Econ. Geol., 100, 905934, doi:10.2113/
100.5.905.
DincAkdogan, N., M. Thorwart, Y. Dzierma, W. Rabbel,
E. R. Flh, J. Goler, W. Taylor, and G. Alvarado
(2007), Seismicity of southern Nicaragua and northern Costa Rica: A combined offshore and onshore
study, paper presented at EGU General Assembly,
Vienna, Austria.
England, P., and P. Molnar (1991), Inferences of deviatoric stress in actively deforming belts from simple
physical models, Philos. Trans. R. Soc. London,
Ser. A, 337, 151164, doi:10.1098/rsta.1991.0113.
Faras, M. (2007), Tectnica y erosin en la evolucin
del relieve de los Andes de Chile central durante
el Negeno/Tectonique, rosion et volution du
relief dans les Andes du Chili central au cours du
Nogene, Ph.D. thesis, 194 pp., Univ. de Chile,
Santiago, and Univ. de Toulouse III, Toulouse,
France, 30 Nov.
Faras, M., R. Charrier, D. Comte, J. Martinod, and
G. Hrail (2005), Late Cenozoic deformation and
uplift of the western flank of the Altiplano: Evidence
from the depositional, tectonic, and geomorphologic
evolution and shallow seismic activity (northern
Chile at 1 93 0S), Tectonics , 24, TC40 01,
doi:10.1029/2004TC001667.
Faras, M., D. Comte, and R. Charrier (2006), Sismicidad superficial en Chile central: Implicancias para el
estado cortical y crecimiento de los Andes central

Australes, paper presented at XI Congreso Geolgico
Chileno, Univ. Catlica del Norte, Antofagasta,
Chile.
Faras, M., R. Charrier, S. Carretier, J. Martinod,
A. Fock, D. Campbell, J. Cceres, and D. Comte
(2008a), Late Miocene high and rapid surface uplift
and its erosional response in the Andes of central
Chile (33 35S), Tectonics, 27, TC1005,
doi:10.1029/2006TC002046.
Faras, M., S. Carretier, R. Charrier, J. Martinod,
A. Tassara, A. Encinas, and D. Comte (2008b),
No subsidence in the development of the Central
Depression along the Chilean margin, paper presented at 7th International Symposium on Andean
Geodynamics, Inst. de Rech. pour le Dv., Nice,
France.
Flynn, J. J., A. R. Wyss, R. Charrier, and C. C. Swisher
(1995), An early Miocene anthropoid skull from the
Chilean Andes, Nature, 373, 603607, doi:10.1038/
373603a0.
Flynn, J. J., A. R. Wyss, D. A. Croft, and R. Charrier
(2003), The Tinguiririca fauna, Chile: Biochronology, paleoecology, biogeography, and a new earliest
Oligocene South American land mammal age,
Palaeogeogr. Palaeoclimatol. Palaeoecol., 195,
229259, doi:10.1016/S0031-0182(03)00360-2.
Fock, A., R. Charrier, M. Faras, and M. Muoz (2006),
Fallas de vergencia oeste en la Cordillera Principal
de Chile central: Inversin de la cuenca de Abanico
(3334S), Ser. Publ. Espec. 6, pp. 4855, Asoc.
Geol. Argent., Buenos Aires.
Fuentes, F. (2004). Petrologa y metamorfismo de muy
bajo grado de unidades volcnicas Oligoceno
Miocenas en la ladera occidental de los Andes de
Chile central (33S). Ph.D. thesis, 407 pp. Univ.
de Chile, Santiago.
Garca, M. (2002), volution oligonogne de lAltiplano Occidental (Arc et AvantArc du Nord du
Chili, Arica): Tectonique, volcanisme, sdimentation, gomorphologie et bilan rosionsdimentation,
Ph.D. thesis, Univ. Joseph Fourier, Grenoble, France.
Garca, M., and G. Hrail (2005), Faultrelated folding,
drainage network evolution and valley incision during the Neogene in the Andean Precordillera of
northern Chile, Geomorphology, 65(34), 279
300, doi:10.1016/j.geomorph.2004.09.007.
Giambiagi, L. B., and V. A. Ramos (2002), Structural
evolution of the Andes between 3330 and 3345
S, above the transition zone between the flat and
normal subduction segment, Argentina and Chile,
J. South Am. Earth Sci., 15, 101116,
doi:10.1016/S0895-9811(02)00008-1.
Giambiagi, L. B., V. A. Ramos, E. Godoy, P. P.
Alvarez, and S. Orts (2003a), Cenozoic deformation
and tectonic style of the Andes, between 33 and 34
south latitude, Tectonics, 22(4), 1041, doi:10.1029/
2001TC001354.
Giambiagi, L., P. P. Alvarez, E. Godoy, and V. A.
Ramos (2003b), The control of preexisting extensional structures in the evolution of the southern
sector of the Aconcagua fold and thrust belt, Tectonophysics, 369, 119, doi:10.1016/S0040-1951(03)
00171-9.
Giambiagi, L., F. Bechis, V. Garca, and A. H. Clark
(2008), Temporal and spatial relationships of thick
and thinskinned deformation: A case study from
the Malarge foldandthrust belt, southern central
Andes, Tectonophysics, 459(14), 123139,
doi:10.1016/j.tecto.2007.11.069.
Giambiagi, L., M. Ghighlione, E. Cristallini, and
G. Bottesi (2009), Caractersticas estructurales del
sector sur de la faja plegada y corrida de Malarge
(3536S): Distribucin del acortamiento e influencia de estructuras previas, Asoc. Geol. Argent.
Rev., 65(1), 140153.
Gilbert, H., S. Beck, and G. Zandt (2006), Lithospheric
and upper mantle structure of central Chile and
Argentina, Geophys. J. Int., 165, 383398,
doi:10.1111/j.1365-246X.2006.02867.x.
Godoy, E., G. Yez, and E. Vera (1999), Inversion of
an Oligocene volcanotectonic basin and uplift of
20 of 22
TC3006
its superimposed Miocene magmatic arc, Chilean

central Andes: First seismic and gravity evidence,
Tectonophysics, 306, 217236, doi:10.1016/
S0040-1951(99)00046-3.
Gripp, A. E., and R. G. Gordon (2002), Young tracks of
hotspots and current plate velocities, Geophys. J.
Int., 150, 321361, doi:10.1046/j.1365246X.2002.01627.x.
Gross, K., S. Buske, S. Shapiro, and P. Wigger (2007),
Seismic imaging of the subduction zone in southern
central Chile, paper presented at 20th Colloquium on
Latin American Earth Science, Dtsch. Forschungsgem., Kiel, Germany.
Gutscher, M. A., R. Maury, J. P. Eissen, and E. Bourdon
(2000), Can slab melting be caused by flat subduction?, Geology, 28, 535538, doi:10.1130/00917613(2000)28<535:CSMBCB>2.0.CO;2.
Hoke, G. D., B. L. Isacks, T. E. Jordan, N. Blanco, A. J.
Tomlinson, and J. Ramezani (2007), Geomorphic
evidence for post10 Ma uplift of the western flank
of the central Andes 183022S, Tectonics, 26,
TC5021, doi:10.1029/2006TC002082.
Introcaso, A., M. C. Pacino, and H. Fraga (1992),
Gravity, isostasy and Andean crustal shortening
between latitudes 30 and 35S, Tectonophysics,
205, 3148, doi:10.1016/0040-1951(92)90416-4.
Isacks, B. L. (1988), Uplift of the central Andean
Plateau and bending of the Bolivian Orocline,
J. Geophys. Res., 93, 32113231, doi:10.1029/
JB093iB04p03211.
Jordan, T. E., B. L. Isacks, R. W. Allmendinger, J. A.
Brewer, V. A. Ramos, and C. J. Ando (1983),
Andean tectonics related to geometry of subducted
Nazca plate, Geol. Soc. Am. Bull., 94, 341361,
doi:10.1130/0016-7606(1983)94<341:ATRTGO>
2.0.CO;2.
Kamiya, S., and Y. Kobayashi (2000), Seismological
evidence for the existence of serpentinized wedge
mantle, Geophys. Res. Lett., 27, 819822,
doi:10.1029/1999GL011080.
Kay, S. M., C. Mpodozis, V. A. Ramos, and F. Munizaga
(1991), Magma source variations for midlate Tertiary magmatic rocks associated with a shallowing
subduction zone and thickening crust in the central
Andes, in Andean Magmatism and Its Tectonic Setting, edited by R. S. Harmon and C. W. Rapela,
Spec. Pap. Geol. Soc. Am., 265, 113137.
Kay, S. M., E. Godoy, and A. Kurtz (2005), Episodic
arc migration, crustal thickening, subduction erosion, and magmatism in the southcentral Andes,
Geol. Soc. Am. Bull., 117, 6788, doi:10.1130/
B25431.1.
Kirby, S. H., and A. K. Kronenberg (1987), Rheology
of the lithosphere: Selected topics, Rev. Geophys.
Space Phys., 25, 12191244, doi:10.1029/
RG025i006p01219.
Klein, F. W. (1978), Hypocenter location program
HYPOINVERSE, U.S. Geol. Surv. Open File
Rep., 78694, 113 pp.
Kono, M., Y. Fukao, and A. Yamamoto (1989), Mountain building in the central Andes, J. Geophys. Res.,
94, 38913905, doi:10.1029/JB094iB04p03891.
Kurtz, A. C., S. M. Kay, R. Charrier, and E. Farrar
(1997), Geochronology of Miocene plutons and
exhumation history of the El Teniente region, central
Chile (3435S), Rev. Geol. Chile, 24(1), 7590.
Lamb, S. (2006), Shear stresses on megathrusts: Implications for mountain building behind subduction
zones, J. Geophys. Res., 111, B07401, doi:10.1029/
2005JB003916.
Lamb, S., and P. Davis (2003), Cenozoic climate
change as a possible cause for the rise of the Andes,
Nature, 425, 792797, doi:10.1038/nature02049.
Lamb, S., L. Hoke, L. Kennan, and J. Dewey (1997),
Cenozoic evolution of the central Andes in Bolivia
and northern Chile, in Orogeny Through Time, edited by J.P. Burg and M. Ford, Geol. Soc. Spec.
Publ., 121, 237264.
Lange, D., J. Cembrano, A. Rietbrock, C. Haberland,
T. Dahm, and K. Bataille (2008), First seismic record
for intraarc strikeslip tectonics along the Liquie
Ofqui fault zone at the obliquely convergent plate
TC3006
margin of the southern Andes, Tectonophysics,

455(14), 1424, doi:10.1016/j.tecto.2008.04.014.
Legrand, D., B. Delouis, L. Dorbath, C. David, J. Campos,
L. Marquz, J. Thompson, and D. Comte (2007),
Source parameters of the Mw = 6.3 Aroma crustal
earthquake of July 24, 2001 (northern Chile), and
its aftershock sequence, J. South Am. Earth Sci.,
24, 5868, doi:10.1016/j.jsames.2007.02.004.
Levi, B., L. Aguirre, J. Nystrm, H. Padilla, and
M. Vergara (1989), Lowgrade regional metamorphism in the MesozoicCenozoic volcanic sequences
of the central Chile, J. Metamorph. Petrol., 7, 487
495, doi:10.1111/j.1525-1314.1989.tb00611.x.
Maksaev, V., F. Munizaga, M. McWilliams, M. Fanning,
R. Mathur, J. Ruz, and M. Zentilli (2004), New
chronology for El Teniente, Chilean Andes, from
UPb, 40Ar/39Ar, ReOs, and fissiontrack dating:
Implications for the evolution of supergiant porphyry
CuMo deposits, in Andean Metallogeny: New Discoveries, Concepts and Updates, edited by R. H.
Sillitoe et al., Spec. Publ. SEPM Soc. Sediment.
Geol., 11, 1554.
Maksaev, V., F. Munizaga, M. Zentilli, and R. Charrier
(2009), Fission track thermochronology of Neogene plutons in the Principal Andean Cordillera
of central Chile (3335S): Implications for tectonic
evolution and porphyry CuMo mineralization,
Andean Geol., 36(2), 153171, doi:10.4067/S071871062009000200001.
McClay, K. R. (1995), The geometries and kinematics
of inverted fault systems: A review of analogue
model studies, in Basin Inversion, edited by P. G.
Buchanana, Geol. Soc. Spec. Publ., 88, 97118,
doi:10.1144/GSL.SP.1995.088.01.07.
McQuarrie, N. (2002), The kinematic history of the central Andean foldthrust belt, Bolivia: Implications
for building a high plateau, Geol. Soc. Am. Bull.,
114, 950963, doi:10.1130/0016-7606(2002)
114<0950:TKHOTC>2.0.CO;2.
McQuarrie, N., J. B. Barnes, and T. A. Ehlers (2008),
Geometric, kinematic, and erosional history of the
central Andean Plateau, Bolivia (1517S), Tectonics, 27, TC3007, doi:10.1029/2006TC002054.
Mishra, O. P., D. Zhao, N. Umino, and A. Hasagawa
(2003), Tomography of northern Japan forearc and
its implications for interplate seismic coupling,
Geophys. Res. Lett., 30(16), 1850, doi:10.1029/
2003GL017736.
Montecinos, P., U. Chrer, M. Vergara, and L. Aguirre
(2008), Lithospheric origin of OligoceneMiocene
magmatism in central Chile: UPb ages and Sr
PbHf isotope composition of minerals, J. Petrol.,
49(3), 555580, doi:10.1093/petrology/egn004.
Mpodozis, C., and V. Ramos (1989), The Andes of
Chile and Argentina, in Geology of the Andes and
Its Relation to Hydrocarbon and Mineral
Resources, Earth Sci. Ser., vol. 11, edited by
G. E. Ericksen et al., pp. 5990, CircumPac.
Counc. for Energy and Miner. Resour., Houston,
Tex.
Muoz, M., F. Fuentes, M. Vergara, L. Aguirre, J. O.
Nystrm, G. Fraud, and A. Demant (2006), Abanico East Formation: Petrology and geochemistry
of volcanic rocks behind the Cenozoic arc front in
the Andean Cordillera, central Chile (3350S),
Rev. Geol. Chile, 33, 109140, doi:10.4067/
S0716-02082006000100005.
Muoz, N., and R. Charrier (1996), Uplift of the western border of the Altiplano on a westvergent thrust
system, northern Chile, J. South Am. Earth Sci., 9,
171181, doi:10.1016/0895-9811(96)00004-1.
Nur, A., and Z. BenAvraham (1981), Volcanic gaps
and the consumption of aseismic ridges in South
America, in Nazca Plate: Crustal Formation and
Andean Convergence, edited by L. D. Kulm,
Mem. Geol. Soc. Am., 154, 729740.
Oleskevich, D. A., R. D. Hyndman, and K. Wang
(1999), The updip and downdip limits to great subduction earthquakes: Thermal and structural models
of Cascadia, south Alaska, SW Japan, and Chile,
J. Geophys. Res., 104, 14,96514,992, doi:10.1029/

1999JB900060.
Oliveros, V., L. Aguirre, D. Morata, A. Simonetti,
M. Vergara, M. Belmar, and S. Calderon (2008),
Geochronology of very lowgrade Mesozoic
Andean metabasites; an approach through the
KAr, 40Ar/39A and UPb LAMCICPMS methods, J. Geol. Soc., 165, 579584, doi:10.1144/
0016-76492007-113.
Pardo, M., D. Comte, and T. Monfret (2002), Seismotectonic and stress distribution in the central Chile
subduction zone, J. South Am. Earth Sci., 15, 11
22, doi:10.1016/S0895-9811(02)00003-2.
Pardo, M., E. Vera, T. Monfret, and G. Yez (2008),
Crustal seismicity and 3D seismic wave velocity
models in the Andes cordillera of central Chile
(3334.5S) from local earthquakes, paper presented at 7th International Symposium on Andean
Geodynamics, Inst. de Rech. pour le Dv., Nice,
France.
PrezGussiny, M., A. R. Lowry, and A. B. Watts
(2007), Effective elastic thickness of South America
and its implications for intracontinental deformation, Geochem. Geophys. Geosyst., 8, Q05009,
doi:10.1029/2006GC001511.
Pilger, R. H. (1981), Plate reconstruction, aseismic
ridges, and low angle subduction beneath the
Andes, Geol. Soc. Am. Bull., 92, 448456,
doi:10.1130/0016-7606(1981)92<448:
PRARAL>2.0.CO;2.
Pose, F. A., M. Spagnuolo, and A. Folguera (2005),
Modelo para la variacin del volumen orognico
andino y acortamientos en el sector 2046S, Asoc.
Geol. Argent. Rev., 60(4), 724730.
Ramos, V. A., M. Cegarra, and E. Cristallini (1996),
Cenozoic tectonics of the High Andes of west
central Argentina (3036S latitude), Tectonophysics, 259, 185200, doi:10.1016/0040-1951(95)
00064-X.
Ramos, V. A., E. O. Cristallini, and D. J. Prez (2002),
The Pampean flatslab of central Andes, J. South
Am. Earth Sci., 15(1), 5978, doi:10.1016/S08959811(02)00006-8.
Ramos, V. A., T. Zapata, E. Cristallini, and A. Introcaso
(2004), The Andean thrust systemLatitudinal variations in structural styles and orogenic shortening,
in Thrust Tectonics and Hydrocarbon Systems,
edited by K. R. McClay, AAPG Mem., 82, 3050.
Rauld, R., G. Vargas, R. Armijo, A. Ormeo, C. Valderas,
and J. Campos (2006), Cuantificacin de escarpes de
falla y deformacin reciente en el frente cordillerano
de Santiago, paper presented at XI Congreso Geolgico Chileno, Univ. Catlica del Norte, Antogafasta,
Chile.
Riquelme, R., G. Hrail, J. Martinod, R. Charrier, and
J. Darrozes (2007), Late Cenozoic geomorphologic
signal of Andean forearc deformation and tilting
associated with the uplift and climate changes of
the southern Atacama Desert (26S28S), Geomorphology, 86, 283306, doi:10.1016/j.geomorph.2006.
09.004.
Roecker, S.W., T. M. Sabitova, L. P. Vinnik, Y. A.
Burmakov, M. I. Golvanov, R. Mamatkanova,
and L. Munirova (1993), Threedimensional elastic
wave velocity structure of the western and central
Tien Shan, J. Geophys. Res., 98, 15,77915,795,
doi:10.1029/93JB01560.
Servicio Geolgico Minero Argentino (1997), Mapa
geolgico de la Repblica Argentina, scale
1:2,500,000, Buenos Aires, Argentina.
Servicio Nacional de Geologa y Minera (2002), Mapa
geolgico de Chile, scale 1:1,000,000, Map M61,
Santiago, Chile.
Smith, W. H. F., and D. T. Sandwell (1997), Global sea
floor topography from satellite altimetry and ship
depth soundings, Science, 277, 19561962,
doi:10.1126/science.277.5334.1956.
Suarez, G., and D. Comte (1993), Comment on Seismic
coupling along the Chilean subduction zone by
21 of 22
TC3006
B. W. Ticherlaar and L. R. Ruff, J. Geophys. Res.,

98, 15,82515,828, doi:10.1029/93JB00234.
Syracuse, E. M., G. A. Abers, K. Fischer, L. MacKenzie,
C. Rychert, M. Protti, V. Gonzlez, and W. Strauch
(2008), Seismic tomography and earthquake locations in the Nicaraguan and Costa Rican upper mantle, Geochem. Geophys. Geosyst., 9, Q07S08,
doi:10.1029/2008GC001963.
Tassara, A. (2005), Interaction between the Nazca and
South American plates and formation of the Altiplano
Puna plateau: Review of a flexural analysis along
the Andean margin (1534S), Tectonophysics,
399, 3957, doi:10.1016/j.tecto.2004.12.014.
Tassara, A., H.J. Gtze, S. Schmidt, and R. Hackney
(2006), Threedimensional density model of the
Nazca plate and the Andean continental margin,
J. Geophys. Res., 111, B09404, doi:10.1029/
2005JB003976.
Tassara, A., C. Swain, R. Hackney, and J. Kirby (2007),
Elastic thickness of South America estimated using
wavelets and satellitederived gravity data, Earth
Planet. Sci. Lett., 253, 1736, doi:10.1016/j.epsl.2006.
10.008.
Thiele, R. (1980), Hoja Santiago, Regin Metropolitana, Carta Geol. Chile 39, 51 pp., Inst. de Invest.
Geol., Santiago, Chile.
Thierer, P. O., E. R. Flh, H. Kopp, F. Tilmann,
D. Comte, and S. Contreras (2005), Local earthquake
monitoring offshore Valparaiso, Chile, Neues Jahrb.
Geol. Palaeontol. Abh., 236, 173183.
Thomas, H. (1958), Geologa de la Cordillera de la Costa
entre el Valle de la Ligua y la Cuesta de Barriga, Bol.
Inst. Invest. Geol. Chile, 2, 86 pp.
Uliana, M., K. Biddle, and J. Cerdn (1989), Mesozoic
extension and the formation of Argentina sedimentary basins, in Extensional Tectonics and Stratigraphy of the North Atlantic Margin, edited by A. J.
Tankard and H. R. Balkwill, AAPG Mem., 46,
599613.
Uyeda, S., and H. Kanamori (1979), Backarc opening
and the mode of subduction, J. Geophys. Res., 84,
10491061, doi:10.1029/JB084iB03p01049.
Vergara, M., B. Levi, and R. Villarroel (1993), Geothermal
type alteration in a burial metamorphosed volcanic
pile, central Chile, J. Metamorph. Geol., 11, 449
454, doi:10.1111/j.1525-1314.1993.tb00161.x.
Vergs, J., V. A. Ramos, A. Meigs, E. Cristallini, F. H.
Bettini, and J. M. Corts (2007), Crustal wedging
triggering recent deformation in the Andean thrust
front between 31S and 33S: Sierras Pampeanas
Precordillera interaction, J. Geophys. Res., 112,
B03S15, doi:10.1029/2006JB004287.
Victor, P., O. Oncken, and J. Glodny (2004), Uplift of
the western Altiplano plateau: Evidence from the
Precordillera between 20 and 21S (northern
Chile), Tectonics, 23, TC4004, doi:10.1029/
2003TC001519.
Wessel, P., and W. H. F. Smith (1998), New, improved
version of the Generic Mapping Tools released, Eos
Trans. AGU, 79(47), 579, doi:10.1029/98EO00426.
Wyss, A. R., M. A. Norell, J. J. Flynn, M. J. Novacek,
R. Charrier, M. C. McKenna, D. Frassinetti, P. Salinas,
and J. Meng (1990), A new early Tertiary mammal
fauna from central Chile: Implications for stratigraphy and tectonics, J. Vertebr. Paleontol., 10, 518
522.
Yez, G., and J. Cembrano (2004), Role of viscous
plate coupling in the late Tertiary Andean tectonics,
J. Geophys. Res., 109, B02407, doi:10.1029/
2003JB002494.
Yez, G., C. R. Ranero, R. von Huene, and J. Daz
(2001), Magnetic anomaly interpretation across the
southern central Andes (3234S): The role of
the Juan Fernndez Ridge in the late Tertiary evolution of the margin, J. Geophys. Res., 106, 6325
6345, doi:10.1029/2000JB900337.
Zhao, D., K. Asamori, and H. Iwamori (2000), Seismic
structure and magmatism of the young Kyushu subduction zone, Geophys. Res. Lett., 27, 20572060,
doi:10.1029/2000GL011512.
TC3006
Zhao, D., K. Wang, G. C. Rogers, and S. M. Peacock

(2001), Tomographic image of low P wave velocity
anomalies above slab in northern Cascadia subduction zone, Earth Planets Space, 53(4), 285293.
R. Charrier, C. David, M. Faras, and F. Tapia,
Departamento de Geologa, FCFM, Universidad de
Chile, Plaza Ercilla 803, Casilla 13518, Correo 21,

Santiago, Chile. (mfarias@dgf.uchile.cl)
D. Comte, Departamento de Geofsica, FCFM,
Universidad de Chile, Blanco Encalada 2002, Casilla
2777, Correo 21, Santiago, Chile.
A. Fock, SQM Salar S.A., Los Militares 4290,
7550081, Las Condes, Chile. (andres.fock@sqm.com)
22 of 22
TC3006
J. Martinod, LMTG, CNRSIRDUniversit de

Toulouse, 14 av. Edouard Belin, F31400 Toulouse,
France.
A. Tassara, Departamento de Ciencias de la Tierra,
Universidad de Concepcin, Campus Concepcin,
Casilla 160C, Concepcin, Chile. (andres.tassara@
udec.cl)

Farias Etal Tectonics 2010

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Farias Etal Tectonics 2010

Uploaded by

Copyright:

Available Formats

Click

TECTONICS, VOL. 29, TC3006, doi:10.1029/2009TC002480, 2010

Crustalscale structural architecture in central Chile based

[1] We document a crustalscale structural model for

Copyright 2010 by the American Geophysical Union.

geology: Implications for Andean mountain building, Tectonics,

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

Charrier [1996], Victor et al. [2004], Faras et al. [2005],

Andes (Figure 1): the Aroma earthquake (24 July 2001, Mw =

2. Tectonic and Geological Settings

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

of the slab is nearly 30E, being the typical example of the

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

Principal Cordillera, place in which the presentday volcanic

[17] We present two cross section (Figure 2b) in which we

3.1. Eastern Central Depression and Western Principal

3.2. Central Principal Cordillera

3. Structural Features of the Mountain Belt in

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

structures would be rooted in a detachment fault located at

Figure 4. Initial 1D P wave velocity models.

highangle west verging ChacayesYesillo fault (see profile

4.2. Tomographic Inversion and Hypocentral

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

[32] In order to use the most reliable hypocenters, we only

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

[38] In spite of the former features related to the ramp or

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

Yez and Cembrano [2004]) (Figure 8b), and experimental

correlates well with the western edge of this lowviscosity

6. Integrating Seismologic Data and Surface

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

constructed two upper crustal cross sections integrating

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

ticular faults may be debated, these cross sections show the

[53] After circa 16 Ma, shortening migrated from the

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

magmatic activity ended in the porphyry copper deposits

7. Discussion: Implications for Mountain

thickening should have occurred in order to compensate the

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

vergent thrusts, in which west vergent thrusts correspond to

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

al., 2004; Faras et al., 2005]) and its accommodated

coincides with the deepest limit of the seismogenic contact

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

[66] On the basis of the evidence presented in this work,

area at 60 km depth and controls the structuring of the

Cretaceous, Tectonics, 25, TC1008, doi:10.1029/

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

Cloetingh, S., P. A. Ziegler, F. Beekman, P. A. M.

estado cortical y crecimiento de los Andes central

its superimposed Miocene magmatic arc, Chilean

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

margin of the southern Andes, Tectonophysics,

J. Geophys. Res., 104, 14,96514,992, doi:10.1029/

B. W. Ticherlaar and L. R. Ruff, J. Geophys. Res.,

FARAS ET AL.: CENTRAL CHILE ANDES ARCHITECTURE

Zhao, D., K. Wang, G. C. Rogers, and S. M. Peacock