You are on page 1of 28

Journal of South American Earth Sciences 14 (2001) 147174

www.elsevier.nl/locate/jsames

Cretaceous separation of Africa and South America: the view from the
West African margin (ODP Leg 159)
T. Pletsch a,*, J. Erbacher b, A.E.L. Holbourn a, W. Kuhnt a, M. Moullade c, F.E. Oboh-Ikuenobede d,
E. Soding e, T. Wagner f
a
Institut fur Geowissenschaften, Universitat Kiel, Olshausenstr. 40-60, Kiel D-24118, Germany
Bundesanstalt fur Geowissenschaften und Rohstoffe, Postfach 510153, Hannover D-30631, Germany
c
Laboratoire de Micropaleontologie et Geologie Marines, Universite de Nice-Sophia Antipolis, Parc Valrose, Nice cx 2 F-06108, France
d
Department of Geology and Geophysics, University of Missouri-Rolla, Rolla, MO 65409-0410, USA
e
Geomar/Forschungszentrum fur Marine Geowissenschaften, Universitat Kiel, Wischhofstr. 1-3, Kiel D-24148, Germany
f
Fachbereich 5, Universitat Bremen, Klagenfurter Str. 5, Bremen D-28359, Germany
b

Received 1 April 1999; revised 1 January 2000; accepted 1 January 2001

Abstract
The opening of the Equatorial Atlantic Gateway (EAG) during the Cretaceous was accompanied by the disruption of the sedimentary
basins that had developed on the conjugate margins of Africa and South America. Drilling along the Cote d'IvoireGhana Transform Margin
(ODP Leg 159) provided a transect across the northern rim of this gateway. The interplay of tectonic and oceanic processes along the gateway
created a complex continental margin that evolved in three stages interrupted by dramatic changes in sedimentary facies, waterdepths, and
subsidence rates. The earliest stage records the formation of small basins with restricted connection to the world ocean and rapid inll with
siliciclastic deposits in an Early Cretaceous intracratonic rift or wrench tectonic setting. This stage ended with an uplift event and the
formation of a regional unconformity. During the late Albian to middle Coniacian, the oceanward side of the margin subsided below the
calcite compensation depth (CCD) and a deepwater connection between Central and South Atlantic became established. Deepening of the
basement ridge and its landward slope, in contrast, were delayed and detrital limestones intercalated with carbonaceous shales accumulated
at shelf to slope depths. During the ensuing, latest Cretaceous to present stage, passive margin subsidence led to continuous deepening of the
basement ridge and on its landward slope. Condensation and gradually decreasing organic contents point to an intensied exposure to
deepwater circulation. The replacement of the zonal circulation system through the Mesozoic Tethys and Central Atlantic with a modern,
oxidizing meridional circulation system through the EAG appears to be intimately related to the changing depositional conditions over large
parts of the Cretaceous Atlantic. q 2001 Elsevier Science Ltd. All rights reserved.
Keywords: Cretaceous; Plate tectonics; Palaeoceanography; Equatorial Atlantic; Gateway; Transform margin; Sedimentary basin; Stratigraphic correlation

1. Introduction
The disruption of Africa and South America led to the
connection of the formerly separated Central and South
Atlantic oceans to produce one of the major deep marine
passages of the late Mesozoic and early Cenozoic. The area
of nal disruption, grossly located between the Guinea and
Ascension Fracture Zones, is called the Equatorial Atlantic
Gateway (EAG). The palaeobiological effects attributed to
the opening of this gateway, including the differentiation of
faunas and oras on the African and South American conti* Corresponding author. Present address: Geol Inst. Abt Sedimentgeologie, Universitat Koeln, Zuelpicher Str. 49a, D-50674 Koeln, Germany. Tel.:
149-221-470-7316; fax: 149-221-470-5149.
E-mail address: thomas.pletsch@uni-koeln.de (T. Pletsch).

nents and the exchange of marine biota between the Central


and South Atlantic, were established decades ago (e.g.,
Wegener, 1929; Reyment, 1969; Berggren and Hollister,
1978; Wiedmann and Neugebauer, 1978; Moullade and
Guerin, 1982). More recently, there has also been growing
interest in the oceanographic implications of this gateway.
Deepening and widening of the gateway had a signicant
impact on the current systems and the resulting sedimentary
and geochemical evolution in adjacent ocean basins (e.g.,
van Andel et al., 1977; Tucholke and Vogt, 1979; Summerhayes, 1981; Brass et al., 1982; Zimmermann et al., 1987;
Hu et al., 1988; Sarnthein and Faugeres, 1993; Jones et al.,
1995; Hay, 1995a,b, 1996).
Large-scale plate tectonic reconstructions have demonstrated that passive continental margins existed around the
Central Atlantic and the southern South Atlantic by the

0895-9811/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved.
PII: S 0895-981 1(01) 00020-7

148

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

Fig. 1. Plate movement of Africa with respect to xed South America.


CIGTM: Cote d'IvoireGhana Transform Margin, BPB: BarreirinhasPiauI Basin. Vectors display relative movement of ODP Leg 159 drill
sites. Based on a revised set of palaeomagnetic data and new calculations
of plate margin deformation by Hay et al. (1999).

earliest Cretaceous, when the present equatorial region still


formed a barrier between them (Larson and Ladd, 1973;
Rabinowitz and LaBreque, 1979; Mascle et al., 1986; Nurnberg and Muller, 1991). Unfortunately, an adequate consideration of the biotic, physical, and chemical effects of the
EAG is hampered by the incomplete knowledge of its
spatial evolution during the Cretaceous. Palaeomagnetic
methods can only provide limited constraints on the geometry of the EAG as a result of its low latitude position and the
lack of magnetic lineations in mid-Cretaceous ocean crust.
Plate tectonic reconstructions must rely on interpolation of
palaeomagnetic data from outside the area of interest and
from before and after the `Cretaceous Quiet Zone', when
opening occurred (Fig. 1; Hay et al., 1999). Previous interpretations of the fracture zone pattern (Le Pichon and Fox,
1971; Le Pichon and Hayes, 1971; Jones, 1987; Mascle and
Blarez, 1987) require assumptions about spreading rates that
are rarely backed by core-based age control. Consequently,
these large-scale plate tectonic reconstructions cannot
account for minor crustal slivers or transverse ridges so
common along the transform margins of the Equatorial
Atlantic (Sibuet and Mascle, 1978), and which probably
determined the effective depth and restriction of the gateway (Jones et al., 1995; Handoh et al., 1999).
Uncertainty concerning the geometry of the EAG is
further aggravated by the discrepancies of proposed biostratigraphic ages for the opening which differ widely according
to the palaeobiogeographic concepts and the favourite fossil
groups of the authors and as a result of the considerable
difculties in assigning numerical palaeo-waterdepths on
the basis of extinct fossil assemblages. Another problem is
the uneven record of the earliest sediments in equatorial

Atlantic basins. Terrestrial and shallow-marine sediments


of pre-Albian age occur in a number of proximal basins
around the Gulf of Guinea and on the conjugate margin
off northeastern Brazil, but hardly any information other
than geophysical data is available for the distal parts of
the gateway, where the earliest rift deposits are to be
expected. The scantiness of age-dated evidence on the
early evolution of the Equatorial Atlantic rendered the
scientic drilling campaign of Ocean Drilling Program
(ODP) Leg 159 in the northern Gulf of Guinea a particularly
worthwile effort. One of the major objectives of ODP Leg
159 was to recover sediments deposited during the opening
history of the EAG and to provide the critically needed age
control of its tectonic, sedimentary, and oceanographic
evolution. This was performed with a set of drill holes
along the continental basement ridge of the Cote
d'IvoireGhana Transform Margin (CIGTM; Fig. 2). This
margin bordered the EAG throughout its Cretaceous opening history and is thus in a favourable position to record the
separation of South America and Africa (Mascle et al.,
1996, 1997).
We rst present a compilation of new and recently
published sedimentologic, mineralogic, micropalaeontologic (foraminifers, radiolarians, palynomorphs) and organic
geochemical data on Aptian to Paleocene core material from
ODP Leg 159 that were placed in a completely revised
stratigraphy. In a second part, we discuss our combined
results on the early history of the CIGTM with a focus on
the evolution of Cretaceous waterdepths and depositional
environments along with a critical review of previous
studies on the EAG. Criteria for the recognition of different
opening stages along the gateway include palaeobathymetric estimates on the basis of lithofacies and benthic
foraminiferal assemblages, and on the presence of superregional, marine black shale events.
2. Geological setting
The CIGTM is located in the northern Gulf of Guinea
where the Romanche Fracture Zone meets the African
shelf. Geophysical surveys, dredges, and deep dives along
the CIGTM had shown that it consists of three main
morphostructural elements (Fig. 2; summarized in Mascle
et al., 1997; Basile et al., 1993, 1996): the most prominent
feature is a NE striking continental basement ridge (the
Marginal Ridge (MR)) that projects southwestward from
the steep Ghanaian slope into the Gulf of Guinea to line
up with the Romanche Fracture Zone. The MR separates
the southwestward sloping Deep Ivorian Basin (DIB) on
the landward side from the oceanic Gulf of Guinea Abyssal
Plain in the south. Both DIB and the MR are underlain by
Pan-African, continental crust. The MR is only covered by a
thin post-tectonic sediment blanket that makes the syntectonic sediments readily accessible to drilling.
During Leg 159, Cretaceous sediments were recovered at

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

149

Fig. 2. Location, stratigraphy, and generalized lithology of drill sites of ODP Leg 159 along the CIGTM. Stratigraphic column of the PETROCI well DX-1X
modied after Saint-Marc and N'da (1997) and Morrison et al. (1999).

four sites along the MR and its westward prolongation


(Figs. 2 and 3). Site 959, situated on a bank on the southern
margin of the DIB, was drilled in a waterdepth of 2100 m. It
was designed to study Cretaceous to Neogene sediments of
the DIB in a relatively undeformed and complete section.
About 200 m of Upper Cretaceous deposits were recovered
in Hole 959D (Fig. 4). The deepest target, strongly
deformed Lower Cretaceous sediments of seismic unit A
which comprises pre-rift to early post-rift sediments (Basile
et al., 1996), was not penetrated to the anticipated depth
because the hole was given up at 1158 mbsf for safety
reasons and poor penetration rate (Mascle et al., 1996).
Site 960, on the crest of the MR in 2000 m of water, and

only 5 km south of Site 959, was chosen to reach the


strongly deformed sediments at a shallower depth than at
Site 959. More than 100 m of Lower Cretaceous sediments
plus about 150 m of Upper Cretaceous deposits were recovered in Holes 960A and 960C (Fig. 5). At Site 961, located
at the westernmost tip of the MR in 3500 m waterdepth, two
holes (961A with continous coring and 961B with spot
coring) reached poorly dated and strongly deformed lower
Cretaceous sediments directly overlain by Paleogene
deposits (Fig. 6). Recovery was low (av. of 39%) and drilling disturbance often severe. Site 962 was placed on a
minor ridge along the southwestward prolongation of the
MR in a waterdepth of 4650 m. Cretaceous sediments

150

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

Fig. 3. Stratigraphic range and dominant lithofacies of sediments recovered during ODP Leg 159 according to the biostratigraphic synthesis of Moullade et al.
(1998). Time scale is by Gradstein et al. (1995). Fig. does not account for periods of non-deposition below the substage level.
Fig. 4. Summary of lithofacies, organofacies and biofacies of the Cretaceous section in Hole 959D. Lithologic Units after Janik et al. (1998). Sample positions
indicated by horizontal lines on left margins of respective columns. Width of the bars for clay minerals relates to estimated quantity. Averaged RockEval data
after Wagner and Pletsch (1999). SH: shelf, neritic; UB: upper bathyal; MB: middle bathyal; LB: lower bathyal; AB: abyssal.

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

151

152

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

were recovered from three out of four holes, but the site had
to be abandoned after drilling through more than 250 m of
Lower Cretaceous turbidites (Fig. 7).
3. Materials and methods
Most samples for this study were taken by the authors on
board the JOIDES Resolution or at the Bremen Core Repository. Where this was not possible, we inspected the respective intervals to determine the sedimentological context and
studied smear slides or thin sections. Shipboard analytical
techniques, an ample photographic documentation of the
core material, and additional data are presented in the Initial
Reports volume of Leg 159 (Mascle et al., 1996). Procedures for X-ray diffraction analysis, RockEval pyrolysis,
and for micropalaeontological and palynofacies studies are
given in Erbacher (1998), Holbourn and Moullade (1998),
Holbourn and Kuhnt (1998), Kuhnt et al. (1998), ObohIkuenobe et al. (1997, 1998), Pletsch (1998), and Wagner
and Pletsch (1999). Because clay mineral data were
produced in four laboratories with different instrumental
settings, results are given as graphic representations of the
relative intensities. `Smectite' and `IS clay' are loosely
used to describe expandable illitesmectite mixed-layer
and at smaller spacings,
minerals with a peak around 17 A
respectively. Numerical ages are from Gradstein et al.
(1995).
3.1. Sedimentary facies and correlation of the drill sites
The development of the CIGTM was provisionally subdivided into four seismically dened stages separated by
tectonic breaks, that correspond to the plate tectonic development of the margin (Blarez and Mascle, 1988; Mascle and
Blarez, 1987): (1) earliest Cretaceous oblique rifting stage,
(2) Aptian to Albian intracontinental wrench stage, (3) midCretaceous continentocean transform margin stage, and
(4) Late Cretaceous passive margin stage. Stages 1 and 2
have since been merged into one, intracontinental transform
stage (Mascle et al., 1996, 1997; Basile et al., 1998). We
have generally retained this tripartite subdivision, because
it closely matches the major changes in sedimentary
facies detected in the core material. With the higher
biostratigraphic resolution reached by shorebased studies,
we found it useful to further subdivide the Cretaceous
evolution into six depositional phases (DPs) to allow for a
better correlation between the drill sites (Fig. 3). Boundaries
between DPs were placed where signicant changes in
lithofacies, organofacies, or biofacies occur, but were
drawn on the basis of biostratigraphic ages. Wherever possible, we used the calcareous nannofossil data of Watkins et
al. (1998) for age assignments. The stratigraphic uncertainties of the Leg 159 drill sites are summarized and discussed

153

by Moullade et al. (1998). Further details and a few modications to the correlation presented by Wagner and Pletsch
(1999) are given below.

4. Early opening stage: Barremian to Middle Albian


(DP1)
Sediments attributed to DP1 were recovered at all sites
except Site 962 because Hole 962D was given up before
reaching them. Common lithologies of the early opening
stage are siliciclastic, texturally immature middle Albian
and older sediments that display intense tectonic deformation and variable, sometimes strong, diagenetic overprints
(Figs. 46). Moderately inclined to overturned strata
suggest that they probably belong to seismic unit A which
extends several hundred metres beneath the depths reached
in Holes 959D, 960A, 961A and 961B (Basile et al., 1996).
At Sites 960 and 961, these deposits are truncated by hardly
deformed calcareous sediments of signicantly younger
age. The age of DP1 is poorly constrained, since microfossils are scarce and have broad stratigraphic ranges. Sediments are completely barren of foraminifers and
radiolarians, and only two samples have yielded sparse,
very poorly preserved calcareous nannofossils of a broad,
late Mesozoic stratigraphic range (Mascle et al., 1996).
Palynological residues are devoid of dinoagellates, except
for single specimens of post-Santonian age in one sample,
which are most likely downhole contamination. Pollen
assemblages have been attributed to an Aptian to Cenomanian (Oboh-Ikuenobe et al., 1998) or late Barremian to
middle Albian age (Masure et al., 1998). Below this coarse
resolution, however, the ages at individual drill sites probably differ, as is suggested by discrepancies in sediment
composition. The very broad stratigraphic range, the
absence of marker beds, and the upward truncation of
these deposits do not warrant a further subdivision into DPs.
In Hole 959D, the oldest, faulted and folded sediments
comprise an irregular alternation of light to dark grey siliciclastic sandstones, siltstones and mudstones and of poorly
sorted mixtures of these lithologies. Diagenetic Mg- and Ferich carbonate nodules and cements occur, but no neritic or
hemipelagic carbonate rocks, either in place or reworked,
were found. Organic debris with abundant structured plant
tissue is a conspicuous component, mainly of clayey interbeds, but it also occurs as asers or dispersed in the sandstones. Sedimentary structures, including parallel and
discontinuous lamination, cross and convolute bedding,
small-scale slump folds, ripples, size-grading, and waterescape structures, point to high-energy conditions and
high sedimentation rates (Fig. 4). This is supported by the
low textural maturity, i.e. the generally poor sorting, of
these deposits. In contrast, the compositional maturity is

Fig. 4. (continued)

154

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

Fig. 5. Summary of lithofacies, organofacies and biofacies of the Cretaceous section in Holes 960A and 960C. For explanations see Fig. 4. P: palygorskite.

surprisingly high. According to the data of Strand (1998),


the sandstones are subarkoses and sublitharenites typical of
recycled sedimentary or metamorphic rocks from extensional basins or cratonic terrains.
Clay mineral assemblages consist of major illite, kaoli-

nite, IS mixed layer clay, and minor chlorite, whereas


smectite is conspicuously absent. Holmes (1998) interpreted
the lack of smectite as a result of its diagenetic replacement
by IS clay in the tectonically disturbed sediments. This
author invoked a pervasive late Albian heating event with

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

155

Fig. 6. Summary of lithofacies, organofacies and biofacies of the Cretaceous section in Holes 961A and 961B. For explanations see Fig. 4.

a geothermal gradient of 3508C/km or more to explain the


apparent thermal alteration at shallow burial depth. Such an
event would have important bearings with regard to the
subsidence history and the time of passing of the spreading
ridge along the CIGTM. Strikingly, however, the organic
maturity level, derived from RockEval pyrolysis and vitrinite reectance data, does not support a downward increasing thermal overprint. Sediments at the very base of Hole
959D are rich in amorphous organic constituents and bear
an immature thermal signature. This means that the bottom
of this hole was not heated to more than 808C (Wagner and
Pletsch, 1999). Considering that IS clays starts to form at
higher temperatures, we propose that the observed absence
of smectite is due to the emplacement of allochthonous,
previously altered material on top of less mature sediments
(Wagner and Pletsch, 1999). In accord with the low compositional maturity, this peculiar clay mineral composition
suggests rapid erosion of detrital material from a previously

buried source when little time was available to develop


smectite-rich soil proles. By contrast, kaolinite is at least
partly of diagenetic origin and partly displays vermiform
aggregates of euhedral platelets. Acidic and/or oxidizing
interstitial waters that are required for vermicular kaolinite
formation are capable of selectively destroying calcareous
and fragile organic remains and may be an alternative explanation for the complete lack of marine microfossils at Site
959.
In Hole 960A, on the MR of the CIGTM, a lower interval
was identied that consists of siliciclastic sandstones, siltstones and claystones part of which display very ne,
continuous laminations and graded beds in ning-upward
sequences (Fig. 5; Mascle et al., 1996). Based on the lack
of marine microfossils and pyrite, the scarcity of lebensspuren, and the abundance of siderite in the lower interval,
these sediments were interpreted to have formed in a
lacustrine environment (shipboard lithologic unit VB;

156

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

Fig. 7. Summary of lithofacies, organofacies and biofacies of the Cretaceous section in Holes 962B and 962D. For explanations see Fig. 4. P: palygorskite;
mineral, possibly odinite/berthierine.
question mark denotes an undetermined 7 A

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

Mascle et al., 1996). However, later studies have revealed


the presence of signicant quantities of pyrite in these
deposits (Hisada et al., 1998). Because of the ambiguity of
the evidence suggesting a freshwater lake environment for
these sediments, we tentatively attribute the lower interval
to a marginal marine, probably estuarine depositional
setting with a variable freshwater inuence. This lower
interval is overlain by siliciclastics with a similar sand and
clay composition, but without the ne lamination present
below. Instead, lamination is coarser, discontinuous or oblique and other sedimentary structures point to rapid, currentinduced sedimentation, probably in a deltaic environment.
Sandstones are slightly more feldspathic, but otherwise
similar to those in Hole 959D (Strand, 1998). Cementing
minerals include siderite and dolomite with variable Fe
contents in the sandstones, but in ne-grained sediments,
microquartz is a common and pervasive cement that imparts
a strong aggregate extinction to the rocks in thin section.
Clay mineral assemblages comprise abundant illite, and
variable concentrations of kaolinite, chlorite, and IS
clay, but no smectite. Nacrite, chlorite, and a regularly interstratied IS clay were identied in veins and as coatings
on slickensided fractures (Mascle et al., 1996; Holmes,
1998). Fluid inclusion data suggest the circulation of hot
uids at the base of Hole 960A (Lespinasse et al., 1998).
A higher level of thermal alteration in the DP1 siliciclastics
of Site 960 is also supported by RockEval data, most of
which show an overmature thermal signature. The discrepancy between the diagenetic levels of Sites 959 and 960
may be related either to a deeper burial or a higher heat ow,
or both, at Site 960.
At Site 961, siliciclastic sediments are similar to those at
the top of DP1 at Site 960, but there are rare intercalations of
carbonate-bearing laminae that have yielded very poorly
preserved calcareous nannofossils (Fig. 6; Mascle et al.,
1996). Palynological data indicate a middle Albian age for
these deposits. Aside from these rare laminae, carbonate
only occurs as sideritic cement and nodules. Sandstones
are quartz arenites or quartz-rich (.90%) subarkoses
(Strand, 1998). They are compositionally the most mature
(virtually no rock fragments and little feldspar) and contain
less carbonate cement than the subarkoses at Site 960. Only
chlorite, illite and IS clay, but no kaolinite were detected,
whereas the latter is abundant at the other sites.
4.1. Margin differentiation stage
The margin differentiation stage is characterized by an
extreme diversity of sedimentary structures and compositions, organofacies, and biofacies over a short time interval (late Albianearly Coniacian) and is subdivided into
three DPs.
4.1.1. Upper Albian to lower Cenomanian (DP2A)
The lower boundary of DP2 is dened by the rst occurrence of upper Albian foraminifers in Hole 959D. At Site

157

962, the base of DP2 corresponds to the bottom of Hole


962D, where the late Albian to early Cenomanian age is
supported by the occurrence of calcareous nannoplankton
(Watkins et al., 1998). An abrupt break in litho, bio, and
organofacies, as well as in tectonic deformation, occurs at
Sites 959 and 962 whithin the intervals corresponding to
DP2 (Fig. 3). Such a change would otherwise qualify for a
discrimination into different DPs, but since the biostratigraphic ages of the intervals above and below the break
overlap to a large extent), DP2 was split into sub-phases
(DP2A and DP2B), the underlying assumption being that
the difference in tectonic deformation occurred almost
synchronously during the latest Albian to early Cenomanian
(Holmes, 1998; Watkins et al., 1998).
At Site 959, DP1 sediments are overlain by similar siliciclastics without an obvious change in either tectonic
deformation, lithofacies, organofacies, or in the apparent
diagenetic overprint. The only difference is the occurrence
of upper Albian to lower Cenomanian planktonic foraminifers and upper Aptian to middle Albian benthic foraminiferal assemblages in the DP2A sediments. The lack of a clear
change between the two units may be related to the intense
erosion and resedimentation of DP1 material during DP2A,
but may alternatively be an artefact of the low recovery in
that interval (1555%, Fig. 4).
It is important to note that those intervals of DP2A that
have yielded the upper Albian to lower Cenomanian foraminifers indicative of a middle to lower neritic environment
are overlain by sediments containing older benthic foraminiferal assemblages. The latter benthic assemblages were
tentatively assigned to neritic conditions and provide the
only constraint on the palaeo-waterdepth prior to deposition
of DP2A (Holbourn and Moullade, 1998; Moullade et al.,
1998). The reversed stratigraphy in the DP2A sediments
implies that older marine sediments had become subject to
erosion by the late Albianearly Cenomanian in an environment that was not signicantly deeper than the one
recorded by the reworked benthic foraminifers.
No DP2 sediments were found at Sites 960 and 961 on the
crest of the MR, where the DP1 siliciclastics are unconformably overlain by signicantly younger deposits (upper Turonian and upper Paleocene, respectively). The higher level of
thermal alteration in DP1 sediments, and the absence of
upper Albian to middle Turonian sediments at these sites
suggest that they were deeply buried before the formation of
the unconformity and became uplifted during, or after, the
middle Albian.
In Hole 962D, a 270 m thick irregular alternation of tectonically deformed, mixed calcareous/siliciclastic turbidites
that belong to a single, late Albian to early Cenomanian
calcareous nannofossil and planktonic foraminifer zone
was attributed to DP2A (Watkins et al., 1998; Bellier,
1998; Figs. 3 and 7). Most lithologies are poorly sorted,
mixed calcareous and siliciclastic turbidite sandstones, siltstones or mudstones. Compositional maturity of the sandstones is high with both feldspar and rock fragment contents

158

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

below 5% (Strand, 1998). Euhedral dolomite is a common


authigenic component of the ne-grained lithologies. In the
upper part of the section, quartz becomes progressively
more important in the clay size fraction. In the SEM,
these samples commonly display quartz overgrowth on
detrital grains that occlude the porosity in this interval.
Clay mineral assemblages are dominated by illite, kaolinite,
and IS clays with variable degrees of expandability. Smectite and chlorite are absent at the base, but appear and
increase upsection. Autochthonous, uppermost Albian to
lower Cenomanian benthic foraminiferal assemblages indicate a restricted outer shelf or upper slope environment.
Similar to Site 959, reworked Albian benthic foraminifers
were found within DP2A sediments and their erosion and
intermixture with younger deposits was tentatively related
to the unroong of an adjacent uplifted source area
(Holbourn and Moullade, 1998). Alternatively, they may
have been reworked as a result of increasing surface current
activity accompanying the long-term rising sea level during
the middlelate Albian (Watkins, pers. comm.; Haq et al.,
1987).
4.1.2. Upper Albian to upper Cenomanian (DP2B)
At Site 959, the contact between the tectonically
deformed siliciclastic sediments of DP2A and the overlying,
hardly deformed coarse-grained detrital limestones of DP2B
was not recovered, but was interpreted as an unconformity
of late Albian or possibly younger age on the basis of differences in physical properties and of preliminary age assignments of the fragmentary record in the underlying and
overlying cores (Mascle et al., 1996; Janik et al., 1998).
Later biostratigraphic work has shown, however, that
there is little, if any, time gap between the sediments
above and beneath the unconformity as both have yielded
late Albian to early Cenomanian dates (Moullade et al.,
1998; Watkins et al., 1998).
DP2B sediments at Site 959 are coarse-grained detrital
limestones. No appropriate material was available for
microfossil washing, clay mineral, or RockEval analyses,
but lithofacies and colour suggest that clay and organic
matter contents are negligible. Most limestones are packstones and grainstones with abundant red algae, bivalve, and
echinoderm debris, and with occasional intercalations or
reworked clasts of deeper water, foraminifer wackestone.
Only very few clasts point to a subaerial weathering
(Marcano et al., 1998) suggesting that erosion took place
mainly below sea level. Overall, the component spectrum
bears similarities with Lower Cretaceous to lower Cenomanian carbonate platforms in Nigeria, Congo, eastern Brazil,
and in the Tethyan realm (Moullade et al., 1998; Masse,
pers. comm.). Rare and poorly preserved Lower Cretaceous
larger foraminifers (Moullade, unpublished data) may indicate recycling of signicantly older limestone formations
rather than shedding from a productive carbonate platform.
Besides the dominant biogenic components, there are
variable admixtures of silt- to granule-sized siliciclastic

grains. Although a minor component among the dominant


biogenic clasts, the siliciclastic grains are important,
because they comprise common feldspars and lithic grains
that testify to an ongoing erosion of siliciclastic sources. The
upward decreasing compositional maturity may be interpreted in terms of a lowering of the erosional level.
Among the lithic components are elongate, dark brown
rock fragments of silicied and laminated siliciclastic
mudstones whith a strong aggregate extinction, but without
calcareous biogenic remains. These silicied shale clasts are
identical to the characteristic lithofacies of ne-grained DP1
sediments from Site 960 and 961 and their occurrence at
Site 959 may indicate reworking of DP1 sediments from the
crest of the MR.
DP2B sediments at Site 962 are characterized by the
transition from calcareous hemipelagic deposits to siliceous
and clayey pelagic sediments with upward decreasing
carbonate contents. The amount of detrital particles
diminishes dramatically and seems to be limited to local
fracturing and reworking of chert clasts in the clayey porcellanitic matrix. The absence of carbonate in the uppermost
part of this section suggests that it was deposited below the
local calcite compensation depth (CCD). Only illite and
smectite were found in clay separates, but clay mineral
data from this interval are sparse and difcult to interpret
because of the poor recovery and partly intense silicication
of the sediment. In the uppermost part of this interval, additional palygorskite, a brous, Mg-rich clay mineral, appears
in the clayey matrix of a chaotic breccia with chert cobbles.
Sedimentologic interpretation of this interval (Core 962B8H) on board was problematic, because no decision was
reached whether the chaotic interval is a sedimentary breccia or whether deformation is drilling induced. The latter
was suggested by the incomplete stroke of the hydraulic
piston corer. Subsequent studies of the radiolarian and
clay mineral assemblages have shown that the chaotic interval is in stratigraphic sequence and that there is no major
intermixture with clay minerals from overlying intervals
(Erbacher, 1998; Pletsch, 1998). Palygorskite clay is a widespread component of upper Albian to lower Cenomanian
Atlantic and Tethyan marine sediments and is often interpreted as a detrital mineral which formed in terrestrial or
shallow marine basins under evaporative conditions (Chamley, 1979; Chamley and Debrabant, 1984; Robert, 1987;
Pletsch et al., 1996). At Site 962, however, detrital input
of palygorskite is very unlikely, because authigenic structures were observed in the SEM and palygorskite is absent
in coeval deposits at the more proximal sites. Similar, palygorskite-rich clays with authigenic structures were found in
the lower Eocene of Sites 960 and 961, and in an undated
interval above the one discussed here (Pletsch, 1998). These
clays must have formed by an early diagenetic reaction,
probably induced by warm, magnesium and silica-enriched,
alkaline deeper waters in settings with reduced sediment
accumulation rates. We propose that the uppermost Albian
to Cenomanian palygorskites at Site 962 grew under similar

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

conditions on, or shortly below, the sediment/sea-water


surface.
Organic carbon concentrations in the cherts and clayey
porcellanites are about 1.5% and decrease to below 0.5% in
the overlying clays. Low Tmax and moderate HI indicate
immature kerogen of type II to III. Foraminiferal assemblages are characterized by the virtual absence of benthic
taxa and by the upward increasing dissolution of planktonic
calcareous tests (Holbourn and Moullade, 1998). Toward
the top of DP2B, foraminifers disappear and only opal-CT
replaced radiolarians, sh debris, and phosphatic cephalopod casts were found in washing residues (Mascle et al.,
1996). Radiolarian assemblages are dominated by Tethyan
species but lack many forms that are typical of middle to
low latitudes of the the latest Albian to Cenomanian (Erbacher, 1998).
4.1.3. Lowerupper Turonian (DP3)
The base of DP3 is dened by the rst occurrences of
lower Turonian biostratigraphic markers in the lower black
shales of Site 959 (Bellier, 1998; Watkins et al., 1998;
Holbourn and Kuhnt, 1998). At Site 962, age assignments
are based solely on radiolarians (Erbacher, 1998), since the
interval is barren of calcareous microfossils.
At Site 959, lower to middle Turonian dark brown to
black carbonaceous marls (`black shales') alternating with
detrital limestone beds overly the organic carbon-lean detrital limestones of DP2B. The black shales are irregularly
laminated and contain abundant sand and pebble-sized
calcareous and phosphatic biogenic remains. In thin
sections, wavy laminae rich in clay and organic matter alternate with layers rich in foraminifers and calcareous debris.
The coarse-grained detrital limestone interbeds are compositionally similar to those of DP2B. Quartz and feldspar
occur throughout, and reach sand size in the coarser limestone beds. Authigenic minerals include calcite, ferrous
dolomite, gypsum, barite, and pyrite. Clay minerals in the
black shales differ from the underlying sediments by the rst
occurrence of minor smectite (Fig. 4). The presence of both
smectite and chlorite points to a detrital rather than diagenetic origin of both minerals, because the stability ranges of
these two minerals do not overlap under burial diagenetic
conditions. This is supported by Tmax values consistently
below 4208C, i.e. burial temperatures below 808C. Total
organic carbon contents vary between 2.7 and 6.1% and
the relationship of the hydrogen to oxygen indices indicates
type II kerogen. Benthic foraminiferal assemblages (lsquo;Buliminid Biofacies', Fig. 4) with close relationships to
assemblages known from coastal basins of Morocco, Senegal, and from Nigeria, and with a high level of endemic
forms indicate deposition in an oxygen-decient outer
shelf or upper slope setting (Holbourn and Kuhnt, 1998;
Holbourn and Kuhnt, 1999a,b).
In Hole 960C, black shales similar to those at Site 959,
with organic carbon contents reaching more than 5% and
elevated hydrogen indices indicating kerogen type II lie on

159

top of the DP1 siliciclastics. They have yielded a mixture of


upper Turonian calcareous nannofossils (Subzone CC12a;
Watkins et al., 1998) and lower Turonian planktonic foraminifers (Bellier, 1998). The different ages derived from the
two microfossil groups they likely result from discrepancies
in age calibration.
At Site 962, the interval corresponding to DP3 is made up
of drilling disturbed, dark green glauconite hardgrounds,
intercalated with lighter green, carbonate-free clays with
abundant brassy pyrite nodules. The clay fraction consists
almost exclusively of a smectitic mineral which, judging
from the elevated Fe contents in EDX measurements, probably is a smectiteglauconite mixed-layer clay. Opal-CT
and zeolite (clinoptilolite) were found to replace and to
overgrow radiolarian skeletons. Only traces of illite were
detected. The enrichment of glauconite and sh teeth, and
the reduced thickness of this interval point to extremely low
sediment accumulation rates. The radiolarian maximum
found in this interval was biostratigraphically attributed to
the time shortly before the onset of the CTBE, whereas
typical CTBE facies at this deeper site were probably eroded
by bottom currents (Erbacher, 1998). The thick glauconitic
hardground that occurs 60 cm below the rst purportedly
late TuronianSantonian radiolarian assemblage was interpreted to correspond to the CTBE. Glauconitic clays above
the lower Turonian glauconite hardground are characterized
by the same, monospecic occurrence of an expandable
mixed layer clay as in the underlying sediments. These
glauconitic and manganiferous hardgrounds may correlate
with those of DP4 or DP5 at the other sites, but correlation is
tenuous, since Site 962 hardgrounds are barren of age-diagnostic microfossils and no lithologic markers or logging
data are available. Consequently, the top of DP3 is not
dened at Site 962. However, bluish palygorskite clays
about 1.5 m above the glauconitic hardground were tentatively correlated with similar, palygorskite-rich deposits of
early Eocene age at the other sites (Pletsch, 1998). The
interval between the glauconitic hardground and the palygorskite clays may thus potentially comprise not only the
Turonian, but also the rest of the Upper Cretaceous and
Paleocene section (see below).
4.1.4. Uppermost Turonianmiddle Coniacian (DP4)
The base of DP4 is placed at the rst occurrence of uppermost Turonian to lower Coniacian calcareous nannofossils
(Subzone CC13a of the zonation used by Watkins et al.,
1998). At Site 959, DP4 corresponds to an interval of
about 10 m of generally ne-grained lithologies, whereas
more than 150 m of coarse-grained detrital limestones,
intercalated with black shales at the very base and capped
by a dolomitic breccia were recovered at Site 960.
The interval corresponding to DP4 at Site 959 consists of
partly dolomitized marlstones, sparry calcarenites, and
black shales. Benthic biofacies (`Bulbobaculites Biofacies')
indicate increased detrital ux in an oxygen-decient
environment. Clay mineral assemblages are similar to the

160

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

underlying black shales, but organic carbon contents are


signicantly lower (around 0.5%) at the bottom of the interval and increase upward to more than 6%. Similar moderate
TOC values in sediments overlying the CTBE are well
known from many Atlantic DSDP/ODP sites and from
sections in the Tarfaya, Senegal, Sergipe, and several Nigerian basins (Koutsoukos et al., 1991; Ly and Kuhnt, 1994; El
Albani et al., 1999; Holbourn et al., 1999a, b). Holbourn and
Kuhnt (1998) interpreted changes in the benthic foraminiferal assemblage at Site 959 as resulting from a shrinking
OMZ during the global sea level lowering at that time and a
concurrent increase in detrital ux. Common sand-sized
quartz and clay mineral assemblages typical for the reworking of rocky substrates point to the continued, but rapidly
decreasing erosion of the MR.
In spite of duplicate coring of the top and the bottom of
DP4 sediments, recovery at Site 960 was poor (Fig. 5). In
Hole 960C, upper Turonian black shales intercalated with
detrital limestones of DP3 in the core catcher of Core 26X
are overlain by a similar alternation of limestones and black
shales in Section 26X-2, but these are early Coniacian in age
(Subzone CC13b; Watkins et al., 1998). Sediments coeval
to the Subzone CC13a-dated dolomitized marlstones at Site
959 were not found at Site 960, which may be due to recovery problems. However, in contrast to the correlative dolomitic marlstones less than 10 m thick at Site 959, the black
shales of Site 960 are overlain by about 145 m of coarsegrained, lower to middle Coniacian detrital limestones.
Litho and biofacies of the limestones are similar to the
upper Albian-lower Cenomanian detrital limestones at Site
959 (Masse, pers. comm.). Like these, quartz, feldspar, and
silicied shale clasts are intermixed with the carbonate
grains, but silicilastic grains are more frequent and coarser,
especially toward the top. Also, ne-grained matrix with
abundant planktonic foraminifers appears to be more
common than in the lower detrital limestones of DP2B.
Some of the calcareous grains contain fragments reminiscent of the Lower Cretaceous larger foraminifer subfamily
Dictyoconinae that are indicative of a BarremianAptian
age (M. Moullade, unpubl. data). Although the evidence is
still sparse, this nd may imply that conditions favourable
for carbonate platform growth already existed during the
Early Cretaceous in the source area of these detrital limestones. Given the microfacial similarity to the DP2B carbonates, the calcareous debris of DP4 was probably locally
reworked from older carbonate formations during the
lower-middle Coniacian, rather than shed from an active,
carbonate-producing environment. Further studies have to
show whether or not the bulk of the calcareous debris is
signicantly older than the depositional age.
At Site 962, no precise boundaries can be given for the
interval corresponding to DP4 through DP6, because it is
barren of microfossils except for sh debis and radiolarians.
Radiolarian assemblages in the carbonate-free glauconite
clays are poorly preserved, but may indicate a post-early Turonian, probably Coniacian to Santonian age (Erbacher, 1998).

4.2. Oceanic opening stage


4.2.1. Upper Coniacianlower Campanian (DP5)
Sediments attributed to DP5 have in common that they
record times when sedimentation rates were intermittently
(959) or almost continously (960, 962) reduced. This is
documented in the frequent occurrence of hardground
surfaces and intercalations of reworked hardground debris
with abundant glauconite and phosphate (Fig. 4). The occurrence of upper Coniacian calcareous nannooras and foraminifers denes the base of DP5 (Watkins et al., 1998;
Holbourn et al., 1999b). At Site 959, the lower boundary
was placed in an interval of strongly carbonaceous, ssile
marls with abundant hardgrounds. These black shales correlate with a thin, phosphatic hardground at Site 960 that has
yielded calcareous nannofossils of late Coniacian Subzone
CC14 (Watkins et al., 1998) as well as upper Cretaceous
planktonic foraminifers (Mascle et al., 1996; Bellier, 1998).
Correlation of DP5 between these two sites is further indicated by a pronounced maximum in gamma-ray intensity in
the uranium spectral band (Mascle et al., 1996) that probably has its source in the phosphate of the hardgrounds and/
or in the organic matter of the black shales. Glauconitic and
manganiferous hardgrounds at Site 962 may belong to DP5,
but the correlation is tenuous, because they are barren of
age-diagnostic microfossils and because no logging data are
available.
Based on a marked downhole increase in sonic velocity,
Janik et al. (1998) proposed a lithologic subunit boundary at
975 mbsf, 10 m above the top of what we dened as DP5.
Interestingly, however, velocity decreases and porosity
increases to `normal' values downhole, beneath the subunit
boundary. The break in physical properties probably corresponds to intense cementation closely below the proposed
subunit boundary and may be related to the availability of
biogenic silica as evidenced by radiolarian-dominated
microfossil assemblages at the top of DP5. The abundance
of radiolarians and of diagenetically redistributed silica in
the upper 30 m of DP5 at Site 959 suggest that this interval
represents the `Early Campanian Biosiliceous Event' which,
according to Kuhnt et al. (1998), ended before the middle
Campanian. The Coniacian/Santonian stage boundary
occurs a few metres below this level (Watkins et al.,
1998). DP5 sediments also straddle the Santonian/Campanian boundary that is placed near the top of Core 959D-65R
on the basis of benthic foraminiferal assemblages and new
palynological results (Kuhnt et al., 1998; Yepes, pers.
comm.). The lower Coniacian to upper Santonian section
is thus represented by less than 20 m of sediment.
At Site 959, black shale deposition resumed, but lithofacies differ from the CTBE black shales (DP3) in that coarsegrained shallow-water carbonate and siliciclastic debris are
absent. The upper black shales consist of dark brownish
grey marlstones with glauconite and calcareous microfossils
at the base, but they are characterized by a pronounced
decrease in both organic and inorganic carbon contents

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

upsection. Above Core 959D-65R, carbonate only occurs as


diagenetic ankerite or in calcite veins. The lower part of
DP5 consists of an irregular alternation of at least ten laminated, very organic-rich intervals with non-laminated, less
organic-rich intervals that partly comprise pebbly facies.
Dark laminae are rich in organic matter (up to 16%) and
clay. They alternate on a mm-scale with light laminae rich
in calcareous nannofossils and ne-grained shelly debris.
Organic matter in the laminated facies is of immature type
I or type II kerogen, probably of marine algal origin.
The non-laminated intervals consist of a structureless,
ne-grained marly matrix that partly comprises phosphate,
barite, and glauconite grains that measure up to 1 cm across.
Despite the abundance of these diagenetic mineral pebbles,
cementation in the non-laminated facies is not pervasive and
mineralogically different from the pebbles. The pebbles are
thus thought to be allochthonous to the depositional site
(Wagner and Pletsch, 1999). They were probably reworked
from adjacent hardgrounds that formed during prolonged
times of non-deposition. Organic carbon in the pebbly facies
does not exceed 4%, and RockEval data indicate that the
kerogen (type II to III) is slightly more mature than in the
laminated facies. Relatively high proportions of woody
particles and non-uorescent amorphous organic matter in
palynological residues from the non-laminated intervals
suggest that this difference in the organic character is due
to a greater contribution of terrestrial kerogen (Oboh-Ikuenobe et al., 1997. Alternatively, it may result from a stronger
degradation of the organic matter in these intervals.
Benthic foraminiferal assemblages at the base of the
upper black shales DP5 are characterized by the Buliminid
Biofacies, adapted to oxygen-decient outer shelf or upper
slope conditions, similar to those of the lower black shales.
In the Coniacian assemblages, however, there is a signicantly lower number of endemic taxa and diversity is negatively correlated to TOC (Holbourn and Kuhnt, 1998). This
biofacies is progressively replaced upsection over an interval of less than 20 m by assemblages rich in cosmopolitan
deep-water agglutinated foraminifers (DWAFs). By the
early Campanian, assemblages consist exclusively of impoverished `Biofacies B' agglutinated forms together with
abundant radiolarians (Kuhnt et al., 1998). Clearly marine
kerogen and the occurrence of benthic foraminiferal assemblages adapted to low-oxygen conditions point to a renewed
expansion of the OMZ and its establishment in shelf depths
(Holbourn and Kuhnt, 1998, Holbourn et al., 1999b). Rhythmic alternation of laminated marls with bioturbated intervals rich in authigenic mineral pebbles may indicate cyclical
changes in the intensity of the OMZ or of the physical
restriction of the DIB, possibly brought about by higherorder sea-level uctuations.
Contrasting with the unusual bulk-rock composition, clay
mineral assemblages of the upper black shales are fairly
monotonous. Typical minerals of the underlying section,
such as illite, chlorite and mixed-layer clay, which indicate
the rapid erosion of rocky substrates are replaced by domi-

161

nant smectite with only small additions of illite and occasional kaolinite. No chlorite or ordered IS clays were
detected in DP5 sediments. As in the underlying section,
kaolinite appears to have grown in place, indicated by the
occurrence of vermicular kaolinite stacks in the laminated
intervals (Wagner and Pletsch, 1999).
At Site 960, a thin layer made up of phosphatic skeletal
grainstone with hardground crusts contains abundant phosphatic sh debris caps the detrital limestones of DP4. This
interval is characterized by very high uranium and potassium readings in downhole gamma-ray measurements.
Correlation with the highly radioactive phosphate-bearing
black shales at Site 959 is supported by middle to late
Coniacian calcareous nannofossils (Watkins et al., 1998).
Elevated potassium spectral intensities may originate from
the overlying glauconite clay.
4.2.2. Middle Campanianlower Paleocene (DP6)
Since the interval corresponding to DP6 contains
virtually no carbonate, it could not be dated with calcareous
microfossils. The base of DP6 was dened by the rst occurrence of middle Campanian agglutinated benthic foraminifer assemblages in Hole 959D (Kuhnt et al., 1998). It was
placed slightly below the palynologic early/middle Campanian boundary (Oboh-Ikuenobe et al., 1998) and about 10 m
below a signicant break in physical properties (Janik et al.,
1998). The exact position of the Cretaceous/Tertiary (K/T)boundary is problematic since calcareous microfossils are
completely absent and neither lithostratigraphic nor
chemostratigraphic markers were found (Ravizza, 1998).
Pervasive bioturbation has destroyed most bedding planes
in the respective interval. Palynological studies resulted in a
discrepancy of 35 m in the placement of the boundary. We
used the K/T-boundary position proposed by Kuhnt et al.
(1998) and Oboh-Ikuenobe et al. (1998) on the basis of
benthic foraminifers and palynological residues, respectively, which is between sections 959D-48R-5, 2831 cm
and 959D-48R-6, 2831 cm (Kuhnt et al., 1998; Moullade
et al., 1998). The upper boundary of DP6 was placed at the
early/late Paleocene boundary between Cores 959D-45R
and -44R as dened by benthic foraminifers (Kuhnt et al.,
1998). Marlstones from above Section 44R-3 have been
dated as late Paleocene with calcareous nannoplankton
(Shak et al., 1998).
At Site 959, the 150 m long section of DP6 is monotonous
at rst glance. It consists entirely of black, organic-rich, but
virtually carbonate-free claystones and siliciclastic
mudstones. However, besides some mineralized tectonic
fractures and veins, there are variations in the intensity of
bioturbation as well as compositional changes. Glauconite is
particularly abundant at the base and in the upper part of the
section where it is occasionally concentrated in pebble beds.
The discontinous trend of an upward decrease in organic
carbon content that started in the Santonian continues
throughout DP6. Organic carbon is 1.2% on average and
values never fall to below 0.6% with generally low

162

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

hydrogen indices. This overall trend is interrupted in the


early Maastrichtian by a rise in TOC to more than 3% and
elevated hydrogen indices. Several samples from the K/Tboundary interval display exceptionally high Tmax values
which may be due to input of reworked organic matter.
Pyrite and barite occur as nodules, burrow lls, or dispersed
throughout parts of the section. Toward the top of DP6,
there is a gradational change to lighter-coloured, carbonate-bearing mottled upper Paleocene marlstones.
Clay mineral assemblages continue to be dominated by
smectite with only minor quantities of illite. Kaolinite occurs
in two distinct intervals near the top of DP6. Biostratigraphic
resolution is limited for the lower kaolinite interval, but palynological data point to an early Maastrichtian age. Since the
presence of kaolinite corresponds to a decrease in clay-sized
quartz, we assume that the compositional change is related to
increasing humidity in the source area of the DP6 sediments.
Also in the lower Maastrichtian interval, TOC, HI, and the
proportion of non-uorescent organic matter rise signicantly,
and a maximum is observed in the abundance of the benthic
foraminiferal genus Rzehakina. This form is thought to be
adapted to enhance organic carbon ux to the sea oor
which is consistent with the elevated organic carbon content
of this interval.
In the upper kaolinite interval, which contains several
pebbly siliciclastic mudstone beds, kaolinite becomes the
dominant clay mineral 7 m below the proposed K/T-boundary
and drops sharply above that interval to reach zero at the top of
DP6, 30 m above it. Similar maxima in the relative abundance
of kaolinite are known from several K/T-boundary localities in
West Africa and southern Europe (e.g. Chamley et al., 1988;
Deconinck, 1992; Ortega-Huertas et al., 1995). The short-term
maximum in kaolinite below the K/T-boundary may be related
to the basinwide (Chierici, 1996) and probably global thirdorder sea-level fall at this time.
Another conspicuous feature in the lower and middle part
of DP6 is the abundance of clay-sized quartz (Fig. 4). Its
dominance in the clay size fraction is surprising because
quartz is usually enriched in the coarser size fractions of
marine sediments. Replicate analyses have shown that
ne-grained quartz is disproportionately abundant in both
the clay fraction and whole rock samples. In the lower part
of DP6 this quartz may be of diagenetic origin as suggested
by the abundance of biogenic silica in the underlying unit.
However, the lack of diagenetic overgrowths and high
microporosity seen in the SEM (Wagner and Pletsch,
1999) argue in favour of a detrital origin of the ne-grained
quartz for most of the section.

5. Tectono-sedimentary evolution
5.1. Basin formation in the equatorial area
Since none of the Leg 159 wells has penetrated the base
of the rifted sedimentary sequence, speculations about the

onset of rifting along the CIGTM and the initiation of an


EAG must rely on literature data and on the sparse evidence
found in the sedimentary components of the oldest deposits
(DP1). The latter includes reworked benthic foraminifers
and detrital chromian spinels, apatite ssion track data,
and the striking discrepancies in the thermal overprint of
the tectonically deformed sediments of the Sites 959 to 961.
Rifting of the Equatorial Atlantic and adjacent areas is
thought of as a polyphase process with the rst, oblique or
transform rift phase starting contemporaneous to South
Atlantic rifting in the latest Jurassicearliest Cretaceous
(Guiraud and Maurin, 1992). Throughout the Neocomian,
the developing plate margins along the Equatorial Atlantic
were dominated by the dextral strike-slip movement of the
African and South American plates past each other (Mascle
et al., 1988; Zalan et al., 1985; Basile et al., 1992). The
resulting shear tectonics probably created the characteristic
folded basement ridges, whereas small divergent basins
continued to subside (Mascle and Blarez, 1987; Blarez
and Mascle, 1988). This probably created an array of pullapart basins with highly variable depositional styles. Thick
upper Jurassic and Neocomian terrestrial deposits are
regarded as evidence for the early separation of western
Africa and South America (Guiraud et al., 1992).
Apatite ssion track cooling ages from the oldest sediments
recovered during Leg 159 fall into two groups. The rst group
of samples ranges from 118 to 105 Ma which closely matches
their Aptian to middle Albian depositional age. This group of
ages was interpreted to result from friction heating during
Aptian intracontinental faulting and subsequent rapid cooling.
Slightly reduced track lengths in the signicantly younger
cooling age found in two other samples (88 and 92 Ma, respectively), point to a later, probably localized middle Turonian to
middle Coniacian heating to above 608C (Bouillin et al., 1997,
1998; Clift et al., 1997, 1998).
A post-depositional thermal overprint of the oldest sediments was also proposed on the basis of overmature organic
matter, stable isotope and uid inclusion data, and on diagenetic clay minerals (Mascle et al., 1996; Oboh-Ikuenobe et
al., 1997; Holmes, 1998; Lespinasse et al., 1998), but
opinions diverge as to what extent this has affected the
host sediments (Bouillin et al., 1997; Clift et al., 1998;
Wagner and Pletsch, 1999). Our data relating to the reworking of older sediments, to the composition of clays, and to
organic maturity indicate that detrital input of previously
buried and thermally altered material, rather than a pervasive late Albian heating event, is responsible for the composition of these sediments. Moderately altered clay minerals
and marginally mature organic matter at the bottom of Holes
959D and 962D are incompatible with the proposed
temperatures. The irregular downhole distribution of clay
minerals and organic matter with variable thermal overprints appears to result from a uctuating input of strongly
altered material. Any later thermal alteration was insufcient to increase the diagenetic level/maturity recorded by
the least altered material present in these sediments.

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

Diagenetic alteration appears to be stronger at Sites 960


and 961, but instead of an exceedingly high geothermal
gradient we would ascribe this to deeper levels of unroong
of these sites, as is suggested by their position on top of the
MR. Detrital chromian spinels from these sediments have
been attributed to tholeiitic basalts known from offshore
Ghana and from the conjugate Barreirinhas-Piau basin in
NE Brazil (Fig. 1; Hisada et al., 1998). The latter basin
contains uvial and lacustrine sediments of a probable
Aptian age (Asmus and de Almeida Campos, 1983; Zalan
et al., 1985). The ages determined on the Brazilian volcanic
rocks (122115 Ma; Wilson and Guiraud, 1992) are earlier
than, or synchronous with, the Aptian to middle Albian
depositional age proposed for the Leg 159 deposits. This
suggests that these volcanics are related to the Aptian
Albian wrench stage of the Equatorial Atlantic region
(Blarez and Mascle, 1988; Guiraud and Maurin, 1992;
Kjemperud et al., 1992).
5.2. Earliest marine sediments: vestiges of an early
Cretaceous Gateway?
Although the age attributed the the oldest marine sediments recovered during Leg 159 is Aptian to middle Albian
(Moullade et al., 1998), the debate about marine incursions
into the equatorial area earlier in the Cretaceous has recently
received new attention through the discovery of Berriasian
(c. 140 Ma) deep-water limestones dredged at the
Romanche Fracture Zone (Bonatti et al., 1996). Unfortunately, outcrops of pre-Aptian sediments are extremely
patchy around the Gulf of Guinea (Guiraud et al., 1992).
Lower Cretaceous terrestrial sediments are well-known
from the marginal basins along the conjugate Brazilian
and African margins, including the southern scarp of the
CIGTM basement ridge where potentially correlative
deltaic sediments were collected, but they were interpreted
to be related to the second, AptianAlbian rift phase (e.g.
Machens, 1973; Allix et al., 1981; Zalan et al., 1985; Kesse,
1986; Ojoh, 1990; Kjemperud et al., 1992; Chierici, 1996;
Guiraud et al., 1997). Hardly any infomation is available for
the distal parts of the gateway, where the earliest and probably the most distal rift deposits are to be expected. Plate
tectonic reconstructions, however, provide little room for a
gateway, since there is general agreement that South America remained attached to Africa until about Aptian times
(Fig. 1; Nurnberg and Muller, 1991; Hay et al., 1999).
Thus in spite of the evidence for earliest Cretaceous subsidence in the equatorial area, it was probably not until the
second rift phase in the late Aptianearly Albian that a
permanent marine connection became established.
The beginning of the second rift (or rift-wrench) phase
has been related to the formation of a regional unconformity
on both African and South American sides of the future
gateway (Zalan et al., 1985; Basile et al., 1992; Chang et
al., 1992 ; Guiraud and Maurin, 1992). This event probably
corresponds to the onset of oceanic accretion in the equa-

163

torial area (Blarez and Mascle, 1988). After this time, the
continentcontinent transform zones became replaced by
continentocean transform margins, thus potentially
providing for a deeper-water connection between Central
and South Atlantic. During the same, AptianAlbian period,
subsidence in the distal basins of the EAG became sufcient
to allow marine incursions (de Klasz and Jan du Chene,
1978; Koutsoukos, 1992; Moullade et al., 1993; Holbourn
and Moullade, 1998). The occurrence of lower to middle
Albian Tethyan fossils in the South Atlantic and the relatively minor endemism found in late Aptian to Albian deep
neritic to upper bathyal benthic foraminiferal assemblages
of Brazilian coastal basins have been interpreted to result
from a shallow-water connection before the middle Albian
(Forster, 1978; Wiedmann and Neugebauer, 1978; de Klasz
and Jan du Chene, 1978; Moullade and Guerin, 1982; DiasBrito, 1987; Koutsoukos, 1992; Holbourn and Moullade,
1998).
No agreement was reached, however, as to whether these
incursions were driven solely by tectonic subsidence and
whether they provided for continuous marine connections
between the adjacent ocean basins. Notably, others have
claimed that Early Cretaceous marine incursions were
related to sea-level rise and that shoals separating the deep
oceanic environments in the equatorial area were not permanently breached until much later in the Cretaceous
(Reyment and Tait, 1972; van Andel et al., 1977; Berggren
and Hollister, 1978; Scheibnerova, 1981; Mabesoone and
Alheiros, 1993). In fact, compressional deformation and
uplift along the cross-cutting transform faults may have
continued to create shallow barriers or even landbridges
that intermittently restricted or interrupted the communication between the Central and South Atlantic (Le Pichon and
Fox, 1971; Bonatti et al., 1994; Jones et al., 1995; Hay et al.,
1999). However, the remarkable decline in tectonic deformation after the Cenomanian, and in particular the absence
of direct evidence for strike-slip deformation in post-Albian
sediments (Pickett and Allerton, 1998) provide evidence
that transform deformation had become inactive on the
meridian of the drill sites by the late Albian. The end of
intra-continental shear tectonics on the CIGTM and the
onset of rapid, quasi-oceanic subsidence on the oceanward
slope of the CIGTM in the latest Albian imply that
South America had cleared the southwestern tip of the
margin before the Cenomanian. This may have given rise
to a permanent marine connection that was to become the
EAG.
All evidence considered, the Early Cretaceous EAG
probably consisted of several small rifted or strikeslip basins that were aligned along the major fault zone
like a string of pearls (Fig. 8). The occurrence of reworked
late Aptian to middle Albian benthic foraminiferal assemblages at Sites 959 and 962, of calcareous nannofossils in
sediments of a similar age at Site 961, and of marine sedimentary structures at Site 960, testify to the existence of
marine basins on the CIGTM during or before the middle

164

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

Fig. 8. Simplied model of the opening EAG and the possible relationship between plate tectonic evolution and the development of a deep-water connection.
Modied after Mascle and Blarez (1987).

Albian (Fig. 9). Subsidence was probably rapid, but intervening basement uplifts and the massive siliciclastic input
seem to have prevented the formation of a deep, interconnected gateway. There may have been intermittent shallowwater connections, especially during periods of sea-level
highstand, but the deeper parts of these basins remained
isolated.

5.3. Middlelate Albian Uplift


The contact between the tectonically deformed detrital sediments of DP1 through DP2B with the overlying, hardly
deformed deposits was interpreted as a major unconformity
of late Albian or possibly younger age on the basis of differences in physical properties and preliminary age assignments

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

165

Fig. 9. Summary of the Cretaceous evolution of the CIGTM on an idealized transect from the African continent over the DIB and MR to the EAG. CCD:
calcite compensation depth, OMZ: oxygen minimum zone. Modied from Wagner and Pletsch (1999).

(Mascle et al., 1996). Texturally immature, but compositionally mature siliciclastic deposits below the unconformity point
to a recycled sedimentary source, with only minor input from
fresh metamorphic or plutonic rocks (Strand, 1998). Tilting
and uplift of the MR was seen as the cause for this unconformity (Basile et al., 1998). The only observable unconformities
were found in Holes 960A and 961A where thermally altered
Aptianmiddle Albian sediments are overlain by Coniacian
and Paleocene deposits, respectively, which helps little in
dating of the unconformity and of the potentially triggering
tectonic event.
Better constraints for the timing of the erosional event are

biostratigraphic ages at Sites 959 and 962. The time of


erosion is bracketed by the late Albianearliest Cenomanian
age determined at the top of the tectonically deformed detrital sediments indicative of unroong in the source area, and
by the late Cenomanianearly Turonian age at the bottom of
the overlying, hardly-deformed section (Erbacher, 1998;
Watkins et al., 1998). Considering that the reworked benthic
foraminifers at Sites 959 and 962 may also result from
tectonically induced erosion (Holbourn and Moullade,
1998) and that uplift must have preceded erosion at least
slightly leads us to propose that the uplift phase recorded at
Leg 159 started in the middle to late Albian. Ongoing uplift

166

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

throughout the Cenomanian may have brought about the


tectonic deformation of DP1 and DP2A and may have
given rise to the formation of an unconformity. This unconformity marked the end of the second rifting (or rift-wrench)
phase and probably corresponds to the widespread postmiddle Albian to pre-late Cenomanian erosional unconformity known from the Guinean, Cote d'Ivoire, and Ghanaian
margins (e.g., Clifford, 1986; de Caprona, 1992; Kjemperud
et al., 1992; Moullade et al., 1993, Chierici, 1996). This
post-rift unconformity has been related to the relaxation of
the African continental margin after crustal detachment of
Africa and South America and is regarded as an equivalent
to the post breakup unconformity of classic divergent
margins (Delteil et al., 1974; Scrutton, 1982; Clifford,
1986; de Caprona, 1992; Blarez and Mascle, 1988; Mascle
et al., 1988; Moullade et al., 1993; Morrison et al., 1999).
Early Cretaceous uplift and unroong of the MR explains
the lack of DP2 sediments on its top. Erosion of these and
the underlying deposits rich in thermally altered clay and
degraded organic matter is seen as the main cause for the
inherited diagenetic signature of the sediments in the
surrounding basins at Sites 959 and 962, located downslope
the crest of the MR (Wagner and Pletsch, 1999). In an
alternative model, a South American crustal sliver remained
attached to the southern margin of the CIGTM until Coniacian times and supplied its detritus to the drill sites
(Marcano et al., 1998), but the evidence at Site 962 does
not t with this model. The position of Site 962 at the southwestern edge of the CIGTM would make it an ideal site for
the deposition of sediment shed from a proximal southern
source area, but instead, detrital input strongly decreases
after the late Albian. Moreover, Bouillin et al. (1998)
noted the conspicuous lack of pre-Cretaceous apatite ssion
track ages that would be expected if the Brazilian margin
were a sediment source for the Leg 159 drill sites. Provenance from Africa is equally unlikely because seismic data
suggest that the intervening DIB separated the MR from the
African continent at least since Albian times (Mascle et al.,
1997; Basile et al., 1998). In agreement with Bouillin et al.
(1998) and Clift et al. (1998), we assume that the detrital
sediments were shed from the uplifted MR itself into the
fringing sedimentary basins from the Albian onwards.
5.4. Latest AlbianCenomanian subsidence of the
oceanward slope
Mixed calcareous and siliciclastic sediments at Site 962
that were deposited in a restricted outer shelf or upper slope
environment are replaced by carbonate-free siliceous and
clayey lithologies that show an increasing level of dissolution and subsequent disappearance of planktonic foraminiferal tests over a period of less than 5 myr. This dramatic
change in sediment composition on the oceanward side of
the CIGTM was interpreted to record rapid deepening to
sub-CCD depths during the latest Albian to middle Cenomanian (Mascle et al., 1996). It is important to note,

however, that the CCD was much shallower during the


mid-Cretaceous than it is today (25003000 m; van Andel
et al., 1977; Tucholke and Vogt, 1979; Kuhnt and Moullade,
1991). Near the continents, where surface productivity is
stronger, the CCD may be even shallower, so additional
evidence is required to corroborate the pronounced deepening.
In fact, the lithologic transition at Site 962 is characterized by a distinct upper Albianlower Cenomanian marine
black shale interval which bears indications of an enhanced
surface productivity (Watkins et al., 1998; Wagner and
Pletsch, 1999). The presence of marine kerogen, rich planktonic microfossil assemblages with foraminifers, calcareous
nannoplankton, diatoms, and radiolarians, as observed in
this interval, are well known from uppermost Albian series
in other parts of the Atlantic ocean (Koutsoukos et al., 1991;
Erbacher and Thurow, 1997). They were interpreted as signs
of a supposedly oceanwide productivity-controlled anoxic
event (OAE 1d; Erbacher et al., 1996; Erbacher, 1998). The
scarcity of benthic foraminifers and the high level of dissolution of planktonic tests in this interval suggest oxygen
decient conditions and an increased carbon ux to the sea
oor that would tend to raise the local CCD (Holbourn and
Moullade, 1998). Even though the benthic foraminiferal
assemblages in this interval are partly endemic, the putative
correlation of this rst marine black shale event with OAE
1d may indicate that a continuous gateway allowed an
exchange of deeper waters between the Central and South
Atlantic. Wagner and Pletsch (1999) proposed that the
presence of this black shale event implies at least a
midwater connection, that is, of 500 m or greater waterdepth, to the open ocean, since upwelling of nutrient-rich
deeper waters is generally required to produce marinederived organic matter (Parrish, 1995).
The occurrence of authigenic palygorskite clay at Site
962 may be supplementary evidence that the southern
slope of the CIGTM was open to midwater circulation by
the latest Albianearly Cenomanian. These clays probably
formed when the seaoor was swept by warm and saline
bottom waters (WSBW; Thiry and Jacquin, 1993; Pletsch,
1998). On the basis of independent evidence, WSBW inuence has been invoked for mid-Cretaceous intermediate and
deep waters of the Central and South Atlantic (Natland,
1978; Woo et al., 1992). The spatial and temporal relationship between WSBW and marine authigenic palygorskite
supports the hypothesis that the southern slope of the
CIGTM was bathed by these waters.
Strikingly, there is no indication of deepening in the detrital limestones at Site 959, although changing concentrations
of coarse, siliciclastic components and the occurrence of
planktonic foraminifers in a calcareous mudstone or wackestone matrix suggest that the carbonate debris was deposited outside the source area at somewhat greater waterdepth
to allow for mixing of siliciclastic and carbonate material.
The diversity of benthic foraminiferal morphotypes seen in
thin section supports deposition at greater depth, probably in

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

an outer shelf environment. The occurrence of unweathered


clasts of silicied shale, a typical facies of DP1 at Sites 960
and 961, within the detrital limestones of DP2B at Site 959,
and the lack of upper Albian and Cenomanian sediments on
the MR indicate that erosion of the ridge continued throughout the late Albian and Cenomanian. It is not clear, however,
whether the calcareous debris was reworked from older
carbonate formations or whether it was shed from an active,
carbonate producing environment.
5.5. The CenomanianTuronian boundary event
Turonian black shales recovered in the DIB were deposited in a setting not deeper than upper bathyal. They show
litho, organo, and biofacies characteristic of the CenomanianTuronian boundary event (CTBE or OAE 2, Schlanger
et al., 1987; Arthur et al., 1990). Benthic foraminiferal
assemblages in the black shales point to an expanded
oxygen minimum zone (OMZ) and an enhanced carbon
ux to the ocean oor, probably as a result of increased
surface water productivity (Holbourn and Kuhnt, 1998).
The geographic distribution of characteristic benthic foraminifer assemblages delineates a high-productivity belt
around the palaeo-equator (Kuhnt and Wiedmann, 1995;
Kuhnt et al., 1990; Holbourn et al., 1999a). The abundance
of endemic forms suggests a continued restriction from
vigorous oceanic circulation of these sites. Signicant
contributions of terrestrial organic matter and clay minerals
point to an additional role of terrigenous input and intensied nutrient leaching as a stimulus for enhanced productivity and organic carbon burial.
These data support models that explain the worldwide
occurrence of organic-rich CTBE deposits as a coincidence
of several interrelated processes. Among these, the early
Turonian second-order sea-level rise probably played a
crucial role as it allowed upwelling of nutrient-rich intermediate waters over the ooded shelves. There is also
general agreement that the CenomanianTuronian interval
was characterized by an increased deepwater production in
low latitude seas, and in particular in the South Atlantic
(Arthur and Natland, 1979; Tucholke and Vogt, 1979;
Brass et al., 1982; Arthur et al., 1987; Hay, 1988). Because
of their elevated nutrient contents and temperature, these
waters are thought to have increased the efciency of planktonic nutrient extraction (Herbert and Sarmiento, 1991).
With the ongoing separation of South America and Africa,
South Atlantic warm and saline deepwaters would have
eventually invaded the Central and North Atlantic
(Summerhayes, 1981, 1987). Dispersal and upwelling of
South Atlantic warm and saline deepwaters would have
stimulated biologic productivity in other parts of the
world ocean. Because of its marginal position with respect
to the world oceans, the Cretaceous Atlantic was particularly prone to the expansion of the OMZ, inducing anoxia of
the water column down to the sea oor during the CTBE
(Thurow et al., 1992). The combination of enhanced

167

productivity and of partial restriction favoured the widespread deposition and preservation of black shales with
labile marine organic matter (Arthur et al., 1987; de
Graciansky et al., 1987; Hay, 1995a,b; Erbacher et al.,
1996).
Although the CTBE black shales at Site 959 were
certainly not deposited in the deepest region of the conduit
between the South and Central Atlantic basins, the presence
of these black shales on the MR and in the DIB, in spite of
their restricted positions, suggests that they had open
midwater exchange with the world oceans (Fig. 9). Falling
sea level after the early Turonian highstand may have
limited the advection of nutrient-rich waters and led to a
reduction in thickness and/or intensity of the OMZ and to an
increase in detrital ux as is signalled by changing biofacies
and increasing quartz content (Holbourn and Kuhnt, 1998;
Holbourn et al., 1999b).
5.6. Deep-water connection between Central and South
Atlantic
CenomanianTuronian deposits at Site 962 on the southern, oceanward slope of the CIGTM lack the CTBE-black
shale facies, but are characterized by organic-lean, extremely condensed glauconitic sediments that were deposited
below the local CCD. The radiolarian maximum found in
this interval was attributed to the time shortly before the
onset of the CTBE, whereas sediments recording the event
itself were probably eroded by bottom currents (Erbacher,
1998). This situation is similar to many mid-Cretaceous
sections in the Atlantic where erosive hiatuses encompass
the CTBE interval (Event `E2', de Graciansky et al., 1987;
`ventilation event' of Zimmermann et al., 1987. Wagner and
Pletsch (1999) inferred that this current-related event results
from an opening to erosive deepwater currents that represents the deep connection between the Central and South
Atlantic during the Cenomanian (Fig. 9). As mentioned
before, the existence of such a deep, sub-CCD gateway in
the Equatorial Atlantic since the Cenomanian is supported
by several authors who suggested that deposition of CTBE
black shales in the Central and North Atlantic ocean resulted
from the establishment of a deepwater connection with the
South Atlantic (Tucholke and Vogt, 1979; Summerhayes,
1981, 1987). This contrasts markedly with the interpretation
of Handoh et al. (1999) who suggested that the effective
depth of the EAG was shallower than 300 m until the Turonian based on a comparison of the modelled palaeo-upwelling areas with occurrences of marine black shales.
However, their results suffer from stratigraphic inconsistencies and from model constraints which tend to ignore the
vast depth range from 300 to 4000 m (Handoh, pers.
comm.).
From the Leg 159 data it appears that deepwaters in the
EAG were strong enough to build up erosive currents and
that they possibly had free exchange between the Central
and South Atlantic by the early Turonian. In the marginal

168

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

seas of the gateway, however, restricted circulation


prevailed and allowed for the accumulation of marine
black shales in an expanded OMZ.
5.7. Submergence of the MR
The 140 m thick section of lower to middle Coniacian
detrital limestones recovered on the MR stands in stark
contrast to the coeval deposition of a few meters of calcarenites and carbonaceous marls at Site 959, only 5 km to the
north. Speculations that the detrital limestones came from a
South American source are based on erroneous correlations
with the older limestones at Site 959 (Marcano et al., 1998).
The striking differences in thickness and the complete lack
of coarse-grained detritus in coeaval deposits of Site 959
may indicate that deposition of the detrital limestones was a
very local phenomenon. Alternatively, Site 959 may have
been sheltered by a morphologic barrier or detrital material
may have bypassed Site 959 to become deposited at greater
depth (Wagner and Pletsch, 1999). However, seismic
section MT02 (Basile et al., 1996; Mascle et al., 1997)
shows no evidence for either the bypass deposits at greater
depth or for a barrier. We therefore assume that the calcareous of debris was eroded along faults and depositied only
locally, but in massive amounts, at Site 960. This is
supported by the similarity of apatite ssion track ages
from the Coniacian detrital limestone interval to those determined in the underlying siliciclastic units at Site 960 (118 to
108 Ma; Bouillin et al., 1997, 1998; Clift et al., 1997, 1998),
which favours a local supply of the detrital material along
strike from the MR until the middle Coniacian. This probably took place below sea level since indications of meteoric
diagenesis in the limestones are very rare (Marcano et al.,
1998).
After the middle Coniacian erosional event, clastic input,
most notably of siliciclastic material, became extremely
reduced. This is well illustrated in clay mineral assemblages. Typical minerals indicative of rapid erosion of
rocky substrates, such as chlorite and mixed-layer clay,
present below, are replaced by abundant smectite with
only small additions of illite and occasional kaolinite.
This change in clay mineral assemblages is thought to
record the drowning of proximal terrestrial sediment
sources and the onset of clay supply from distant source
areas where clay minerals formed through chemical weathering in soils. Similar clay mineral assemblages are known
from open ocean localities all over the Late Cretaceous
Atlantic and have been interpreted to result from the prevailing warm climate with seasonal changes in humidity
combined with long-term stability in subsidence of the
source areas (Chamley, 1979; Chamley and Debrabant,
1984; Robert, 1987). The abrupt reduction in detrital sediment supply is interpreted as resulting from the subsidence
of the MR, since no other source areas lay adjacent to the
drill sites by that time. Subsequently, the position of drill
sites either on the top or on the slopes of the submerged MR

prevented signicant input of detrital material from the


adjacent continents and deposition became governed by
oceanographic factors.
Faulting and submergence along the MR may be related
to the crustal readjustments caused by the extinction of
transform movements along the CIGTM in the Upper
Cretaceous (Blarez and Mascle, 1988). Rapid subsidence
in the surrounding basins and the evolving steep slope of
the basement ridge probably led to denudation and burial of
the former source area under the apron of debris that covers
the southern scarp of the MR (Bouillin et al., 1998).
5.8. Rapid subsidence of the Marginal Basin
Whereas Site 962, on the southern slope of the CIGTM,
was in the reach of deepwater circulation since the Turonian, the MR and its northern slope remained at outer shelf
to upper bathyal depth until the middle Coniacian. Evidence
for rapid deepening after the middle Coniacian comes from
deep current-related lithofacies and from benthic foraminiferal assemblages. The dramatic decline of siliciclastic accumulation rates was not counterbalanced by biogenic
sedimentation, even though the sites on the submarine MR
were probably above the CCD. Thus, winnowing by deepwaters may have contributed to the development of extensive hardground facies at Site 960 and to their reworking
and redeposition at Site 959 (Watkins et al., 1998). The
increasing abundance, after the middle Coniacian, of
cosmopolitan DWAFs and the exclusice occurrence of
organic-walled DWAF assemblages since the Early Campanian suggest that Site 959 had subsided sufciently to
become connected to oceanic deepwaters. Rapid deepening
was also proposed for the lower Campanian Abidjan Margin
based on commercial seismic and well data (Morrison et al.,
1999) and may indicate that this was a major subsidence
event throughout the DIB.
The discrepancy in the onset of rapid subsidence between
Site 959 and Site 962 may be related to the position of the
latter on the oceanward edge of the continental crust. Crustal detachment of the CIGTM in the late Early Cretaceous
may have induced quasi-oceanic subsidence at Site 962 due
to the mechanical coupling of the MR with the adjacent
oceanic crust (Lorenzo and Wessel, 1997). Increased subsidence in the DIB and the MR started about 5 myr later,
probably after the oceanic spreading ridge had passed the
CIGTM (Figs. 8 and 9). This is supported by a group of
apatite ssion track data whose ages closely matches the
time of the proposed drowning of the MR and the onset of
subsidence in the DIB (8892 Ma; Bouillin et al., 1998;
Clift et al., 1998). This annealing event was interpreted to
result from heat transfer at the contact to the relatively hot
oceanic crust to the south of the CIGTM (Bouillin et al.,
1998).
5.9. Upper Cretaceous OAE 3 black shales
Whereas Atlantic and other basinal sites generally

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

became oxygenated after the CTBE, several marginal basins


of the Atlantic maintained an expanded OMZ throughout
the Late Cretaceous and accumulated substantial sections of
marine black shales (Thiede and van Andel, 1977; van
Andel et al., 1977; Dean et al., 1984; Ly and Kuhnt, 1994;
Holbourn et al., 1999a, b; Wagner and Pletsch, 1999).
Although less widespread and stratigraphically less
constrained, these black shales were attributed to the Coniacian to early Campanian OAE 3 (Arthur et al., 1990).
Leg 159 provides evidence for both the deepwater,
oxygenated conditions with the post-Turonian lightcoloured glauconite clays at Site 962, and for the marginal
basin conditions with the ConiacianCampanian black
shales of Sites 959 and 960. In fact, the richest black shales
recovered from the DIB fall to within this time envelope. In
Hole 959D, this event seems to consist of three sub-events
that are separated from each other by lower organic carbon
contents (Fig. 4). The uppermost sub-event correlates with
the oceanwide `Early Campanian Biosiliceous Event' that is
characterized by elevated carbon ux even to the deep sea
oor (Kuhnt et al., 1989; Wightman and Kuhnt, 1992).
Upward decreasing organic carbon concentrations and the
reduction in the algal marine contribution to this component
after the Coniacian correlates with the continued deepening
in the DIB. Subsidence of the seaoor eventually brought
Site 959 below the OMZ, probably after the early Campanian, when the organic content decreased markedly and
when foraminiferal biofacies adapted to high carbon ux
(Biofacies B) were replaced by more oligotrophic assemblages (Kuhnt et al., 1998).
5.10. Uppermost Cretaceous hemipelagic sedimentation
The dark siliciclastic mudstones overlying the upper
black shales at Site 959 are peculiar for the complete
absence of calcium carbonate, for the abundance in ne
silt and clay-sized quartz and their scarcity in phyllosilicates
compared to oceanic clays. Fine-grained quartz is a frequent
component of air-born dust and is common in marine
surface sediments offshore arid areas. Lever and McCave
(1983) have shown that CampanianMaastrichtian deposits
in the Atlantic contain a considerable portion of aeolian
material, including quartz, derived from Africa. Interpretation of the clay-sized quartz as aeolian particles derived
from arid areas is supported by the accessory clay minerals
that are mainly smectite and kaolinite. Whereas smectite
dominates in the quartz-rich deposits, kaolinite comes in
where quartz intensities drop. Background values of quartz
are reached in the K/T-boundary interval, where kaolinite is
dominant. Smectite typically forms in soils under conditions
of seasonal humidity variations while kaolinite formation
usually requires more continuous humidity to achieve the
leaching of soluble cations (Chamley, 1989). Antithetical
changes in smectite and kaolinite may be interpreted, therefore, as changes in the humidity of the source areas with
smectite and abundant quartz indicating arid conditions

169

while kaolinite with less quartz would indicate a wetter


climate. Recent climate simulations for the Campanian
(80 Ma) suggest that extremely arid areas extended in the
southern tropical regions of Africa and South America (Hay
et al., 1997; DeConto et al., 1999). Conversely, precipitation
in a narrow, equatorial belt was more intense than in the
modern intertropical convergence zone (ITCZ). Aeolian
sediment transport via the southern trade winds or an
equivalent of the modern Harmattan may have supplied
the quartz-rich material from these areas until they became
precipitated where the winds reached the Late Cretaceous
ITCZ.
Another striking characteristics of the dark claystones is
their elevated organic matter content. Although benthic
foraminiferal assemblages point to an overall reduction in
the carbon ux to the sea oor and although the proportion
of marine organic matter had decreased strongly after the
early Campanian, most of these clays would still qualify as
black shales in that their organic carbon content exceeds 1%
on average (Arthur and Sageman, 1994). This is in marked
contrast to the coeval fully oxygenated deposits of the Plantagenet Formation in the open Central and North Atlantic
(Berger and von Rad, 1972; Jansa et al., 1977). It is generally accepted that the Plantagenet Formation represents a
period when the ocean oor was ventilated enough to
break down all incoming organic matter. Why did continuous deposition of organic-rich sediments in the DIB prevail
when the adjacent ocean was well oxygenated? We assume
that this difference is related to the proximal position of the
DIB with respect to the Guinean coast and to the damming
effect of the MR. During the Late Cretaceous, the Guinean
coast approached the inner tropics as Africa moved northward (Fig. 1). Palynological data suggest that rain forests
emerged at that time around the Gulf of Guinea (Maley,
1996). Luxurious vegetation may have given rise to a
substantial export of terrestrial organic matter into the sea.
Campanian to lower Paleocene dark shales, silty claystones,
sandstones, and limestones on the landward margin of the
DIB, offshore Cote d'Ivoire, are four times as thick as the
black claystones at Site 959 and attest to the intense terrestrial input at that time (Fig. 2; Saint-Marc and N'da, 1997;
Morrison et al., 1999). Only buoyant particles like clay
minerals and organic matter could be transported to the
opposite, southern margin of the DIB, where Site 959 is
located.
Uppermost Cretaceous hemipelagic sedimentation on the
northern slope of the MR Site 959 thus appears to be inuenced by two sediment sources. A proximal, northerly
source was located in the humid equatorial belt and supplied
abundant terrestrial organic material and a distant, arid
source that mainly provided aeolian quartz. We propose
that the position of Site 959 on the slope of the MR largely
excluded the dilution of both organic matter and aeolian
particles with siliciclastic sediment from the Guinean
coast. Partial damming of the DIB, despite its position
below the CCD, limited the inow of oxygenated waters

170

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

that would have mineralized organic matter coming from


the continent. In addition, damming by the MR probably
conned the dispersal of terrestrial organic matter to the
DIB thus leading to its concentration on the southern
slope of the basin.
A last return to black shale sedimentation is recorded in
the Maastrichtian in an interval with abundant marine
organic matter. Concurrent excursions in palynofacies,
kaolinite concentration, and in the quantitative composition
of benthic foraminferal assemblages (Kuhnt et al., 1998)
may indicate that marine productivity was enhanced by an
increasing terrestrial input. Interestingly, an apparently
oceanwide event is known to occur at about the same
time. The extinction of inoceramids and rudistids and deviations in the oxygen, carbon, and strontium isotope composition of sea water have been intepreted as a reection of an
increasing weathering intensity and of a pronounced change
in circulation leading to climatic cooling after the midMaastrichtian (Nelson et al., 1991; Barrera, 1994; MacLeod
and Huber, 1996).
According to Blarez and Mascle (1988), the spreading
ridges in the equatorial Atlantic had cleared the tips of the
African and South American continental margins only after
the Santonian, when the former active continentocean
transform zones became extinct. Although we propose
that this happened slightly earlier, in Coniacian times, at
the CIGTM, this would still imply that the effective depth
of the EAG was not greater than the maximum depth of the
spreading ridges (c. 22003000 m) until that time. As a
consequence, deepwater circulation remained restricted
even when the basins on the CIGTM had subsided to considerable depth (Fig. 9). Hu et al. (1988) tentatively related a
major break in the Rare Earth Element composition of pelagic carbonate in the South Atlantic at about the Paleocene/
Eocene-boundary (5558 Ma) to the deep subsidence of
barriers in the equatorial Atlantic. This may indicate that
the effective depth of the EAG continued to be affected by
the Equatorial fracture zones long after the volcanic ridges
had passed the adjacent continental margins in the Late
Cretaceous (Jones, 1987; Binks and Fairhead, 1992; Jones
et al., 1995).
6. Conclusions
The CIGTM is characterized by complex subsidence and
depositional histories, which differ markedly depending on
the position of the individual drill sites. The Cretaceous
depositional evolution of this margin can be subdivided
into three stages that correspond to the tectonic evolution
from an early Cretaceous intracratonic rift and transform
basin to a Late Cretaceous passive margin:
1. During the Early Cretaceous opening stage (middle
Albian and earlier), marine sedimentation had started,
both towards the north and the south of the basement

ridge. The existence of a shallow marine gateway cannot


be proven in these proximal environemnts, but literature
data suggest that the South Atlantic was invaded by Tethyan species since at least the middle Albian. This stage
ends with an erosional event that leads to the redeposition
of older marine sediments and, later, to the formation of a
regional unconformity.
2. The mid-Cretaceous margin differentiation stage (late
Albian to middle Coniacian) is characterized by the
strongly divergent, non-synchronous subsidence of the
basins north and south of the CIGTM. While the MR
continued to be uplifted during the Cenomanian, rapid
subsidence to below the local CCD started on its oceanward side. A regional deepening of distal environments
in the CenomanianTuronian led to the establishment of
a deep-water connection between Central and South
Atlantic. We assume that this deepening was related to
the end of continentcontinent wrench tectonics along
the CIGTM. In contrast to the distal margin, the landward
slope remained at neritic to bathyal depths until the
middle Coniacian. Black shales with benthic biofacies
indicative of an expanded OMZ became deposited in
this physically dammed basin.
3. During the oceanic opening stage (middle Coniacian and
later), rapid subsidence started to affect the landward,
DIB. Extreme condensation and erosion on the subsiding
MR and its northern and southern slopes point to intensied deep currents. Restricted conditions and black
shale deposition were only maintained in the landward
marginal basin. Rapid subsidence to below the local CCD
was delayed, there, by about 8 Ma, presumably induced
by the passing of the volcanic spreading ridge along the
margin. The subsequent subsidence and/or disruption of
fracture zones, spreading ridges, or other seaoor elevations engendered an increased meridional deepwater
circulation throughout the Atlantic.

Acknowledgements
We are indebted to the shipboard and shorebased participants of ODP Leg 159, including the staff on board, at
TAMU, and at the Bremen Core Repository, for their help
in making samples and informations available. J.P. Masse, J.
Morrison, and O. Yepes provided unpublished stratigraphic
data. Technical help by P. Recourt, P. Vanderaveroet, D.
Malengros, N. Pernot, A. Kirch, D. Lein, U. Schmolke, A.
Frantz, and R. Petschick is gratefully acknowledged. The
paper and earlier drafts of the manuscript beneted from the
reviews by P.D. Clift, A. Fleet, W.W. Hay, C. Robert, and
D.K. Watkins. Funding was provided by the German
Research Council (DFG-grant Ku 649/2) to TP and WK,
by the Royal Society to AELH, and by the French CNRS
to MM. This is a contribution to IGCP Project 381: South
Atlantic Mesozoic Correlations.

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

References
Allix, P., Grosdidier, E., Jardine, S., Legoux, O., Popoff, M., 1981. Decouverte d'Aptien superieur a Albien inferieur date par microfossiles dans
la serie detritique cretacee du fosse de la Benoue (Nigeria). Comptes
Rendus de l'Academie des Sciences de Paris 292, 12911294.
van Andel, T.H., Thiede, J., Sclater, J.G., Hay, W.W., 1977. Depositional
history of the South Atlantic Ocean during the last 125 million years.
Journal of Geology 85, 651698.
Arthur, M.A., Natland, J.H., 1979. Carbonaceous matter in the North and
South Atlantic: the role of salinity in stable stratication of Early
Cretaceous basins. Deep Drilling Results in the Atlantic Ocean: Continental Margins and Paleoenvironment, Talwani, M., Hay, W., Ryan,
W.B.F. (Eds.). American Geophysical Union, Maurice Ewing Series 3,
375401.
Arthur, M.A., Sageman, B.B., 1994. Marine black shales: depositional
mechanism and environments of ancient deposits. Annual Review of
Earth and Planetary Sciences 22, 499551.
Arthur, M.A., Schlanger, S.O., Jenkyns, H.C., 1987. The Cenomanian
Turonian oceanic anoxic event, II: palaeoceanographic controls on
organic-matter production and preservation. Marine Petroleum Source
Rocks, Brooks, J., Fleet, A.J. (Eds.). Geological Society of London,
Special Publication 26, 401420.
Arthur, M.A., Jenkyns, H.C., Brumsack, H.J., Schlanger, S.O., 1990. Stratigraphy, geochemistry, and oceanography of organic carbon-rich
Cretaceous sequences. In: Ginsburg, R.N., Beaudoin, B. (Eds.). Cretaceous Resources, Events and Rhythms: Background and Plans for
Research. NATO ASI Series C: Mathematical and Physical Sciences,
304. , pp. 75120.
Asmus, H.E., de Almeida Campos, D., 1983. Stratigraphic division of the
Brazilian continental margin and its paleogeographic signicance.
Zitteliana 10, 265276.
Barrera, E., 1994. Global environmental changes preceding the CretaceousTertiary boundary: earlylate Maastrichtian transition. Geology 22,
877880.
Basile, C., Brun, J.P., Mascle, J., 1992. Structure et formation de la marge
transformante de Cote d'IvoireGhana: apports de la sismique reexion
et de la modelisation analogique. Bulletin de la Societe Geologique de
France 163, 207216.
Basile, C., Mascle, J., Popoff, M., Bouillin, J.P., Mascle, G., 1993. The
Ivory CoastGhana Transform Margin: a marginal ridge structure
deduced from seismic data. Tectonophysics 222, 119.
Basile, C., Mascle, J., Sage, F., Lamarche, G., Pontoise, B., 1996. Pre-cruise
and site-surveys: a synthesis of marine geological and geophysical data
on the Cote d'IvoireGhana Transform Margin. Proceedings of the
Ocean Drilling Program, Part A: Initial Reports 159, 4760.
Basile, C., Mascle, J., Benkhelil, J., Bouillin, J.P., 1998. Geodynamic
evolution of the Cote d'IvoireGhana Transform Margin: an overview
from ODP Leg 159 results. Proceedings of the Ocean Drilling Program,
Scientic Results 159, 101110.
Bellier, J.P., 1998. Cretaceous planktonic foraminifers from Leg 159, eastern equatorial Atlantic. Proceedings of the Ocean Drilling Program,
Scientic Results 159, 335345.
Berger, W.H., von Rad, U., 1972. Cretaceous and Cenozoic sediments from
the Atlantic Ocean: Lisbon, Portugal to Jan Juan, Puerto Rico . Initial
Reports of the Deep Sea Drilling Project 14, 787954.
Berggren, W.A., Hollister, C.D., 1978. Paleogeography, paleobiogeography, and the history of circulation in the Atlantic Ocean. Studies in
Paleo-Oceanography, Hay, W.W. (Ed.). Society of Economic Paleontologists and Mineralogists, Special Publication 20, 126186.
Binks, R.M., Fairhead, J.D., 1992. A plate tectonic setting for Mesozoic
rifts of western and central Africa. Tectonophysics 213, 141151.
Blarez, E., Mascle, J., 1988. Shallow structures and evolution of the Ivory
Coast and Ghana Transform Margin. Marine and Petroleum Geology 5,
5464.
Bonatti, E., Ligi, M., Gasperini, L., Peyve, A., Rasnitzin, Y., Chen, Y.J.,

171

1994. Transform migration and vertical tectonics at the Romanche


Fracture Zone, equatorial Atlantic. Journal of Geophysical Research
99B, 21,77921,802.
Bonatti, E., Ligi, M., Borsetti, A.M., Gasperini, L., Negri, A., Sartori, R.,
1996. Lower Cretaceous deposits trapped near the equatorial MidAtlantic Ridge. Nature 380, 518520.
Bouillin, J.P., Poupeau, G., Labrin, E., Basile, C., Sabil, N., Mascle, J.,
Mascle, G., Gillot, F., Riou, L., 1997. Fission track study: heating and
denudation of marginal ridge of the Ivory CoastGhana Transform
Margin. Geo-Marine Letters 17, 5561.
Bouillin, J.P., Poupeau, G., Basile, C., Labrin, E., Mascle, J., 1998. Thermal
constraints on the Cote d'IvoireGhana Transform Margin: evidence
from Apatite ssion tracks. Proceedings of the Ocean Drilling Program,
Scientic Results 159, 4348.
Brass, G.W., Southam, J.R., Peterson, W.H., 1982. Warm saline bottom
water in the ancient ocean. Nature 296, 620623.
de Caprona, G.C., 1992. The continental margin of western Cote d'Ivoire:
Structural framework inherited from intra-continental shearing. Geologiska Institutionen, Publ. A 69, 1150.
Chamley, H., 1979. North Atlantic clay sedimentation and paleoenvironment since the Late Jurassic. Deep Drilling Results in the Atlantic
Ocean: Continental Margins and Paleoenvironments, Talwani, M.,
Hay, W., Ryan, W.B.F. (Eds.). American Geophysical Union, Maurice
Ewing Series 3, 342361.
Chamley, H., 1989. Clay Sedimentology. Springer, Berlin, p. 623.
Chamley, H., Debrabant, P., 1984. Mineralogical and geochemical investigations of sediments on the Mazagan Plateau, northwestern African
margin (Leg 79, Deep Sea Drilling Project). Initial Reports of the Deep
Sea Drilling Project 79, 497508.
Chamley, H., Debrabant, P., Flicoteaux, R., 1988. Comparative sedimentology of the Senegal and eastern central Atlantic Basins, from mineralogical and geochemical investigations. Sedimentology 35, 85103.
Chang, H.K., Kowsmann, R.O., Ferreira Figueiredo, A.M., Bender, A.A.,
1992. Tectonics and stratigraphy of the East Brazil Rift system: an
overview. Tectonophysics 213, 97138.
Chierici, M.A., 1996. Stratigraphy, palaeoenvironments and geological
evolution of the Ivory CoastGhana basin. Geologie de l'Afrique et
de l'Atlantique Sud: Actes des Colloques (Angers, 1994), Elf Aquitaine
Edition, Pau, France, pp. 293303.
Clifford, A.C., 1986. African oil past, present, and future. Halbouty,
M.T. (Ed.). Future Petroleum Provinces of the World (American Association of Petroleum Geologists, Memoir) 40, 339372.
ODP Leg 159 Scientic Party, Clift, P.D., Lorenzo, J., Carter, A., Hurford,
A.J., 1997. Transform tectonics and thermal rejuvenation on the Cote
d'IvoireGhana West Africa margin. Journal of the Geological Society
of London 154, 483489.
Clift, P.D., Carter, A., Hurford, A.J., 1998. Apatite ssion track analysis of
sites 959 and 960 on the transform continental margin of Ghana West
Africa. Proceedings of the Ocean Drilling Program, Scientic Results
159, 3541.
Dean, W.E., Arthur, M.A., Stow, D.A.V., 1984. Origin and geochemistry of
Cretaceous deep-sea black shales and multicolored claystones, with
emphasis on Deep Sea Drilling Project Site 530, southern Angola
Basin. Initial Reports of the Deep Sea Drilling Project 75, 819844.
Deconinck, J.F., 1992. Sedimentologie des Argiles dans le JurassiqueCretace d'Europe Occidentale et du Maroc. Habilitation Thesis,
Universite des Sciences et Technologie de Lille, Villeneuve d'Ascq
Cedex, France.
DeConto, R.M., Hay, W.W., Thompson, S.L., Bergengren, J., 1999. Late
Cretaceous climate and vegetation interactions: Cold continental interior paradox. Evolution of the Cretaceous Ocean-Climate System,
Barrera, E., Johnson, C.C. (Eds.). Geological Society of America,
Special Paper 332, 391406.
Delteil, J.R., Valery, P., Montadert, C., Fondeur, C., Patriat, P., Mascle, J.,
1974. Continental margin in the northern part of the Gulf of Guinea. In:
Burk, C.A., Drake, C.L. (Eds.). The Geology of Continental Margins.
Springer, New York, USA, pp. 297311.

172

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

Dias-Brito, D.A., 1987. A Bacia de Campos no mesocretaceo: uma contribucao a paleoceanograa do Atlantico Sul primitivo. Revista Brasileira
de Geociencias 17, 162167.
El Albani, A., Kuhnt, W., Luderer, F., Herbin, J.P., Caron, M., 1999.
Palaeoenvironmental evolution of the Late Cretaceous sequence in
the Tarfaya Basin (southwest of Morocco). In: Cameron, N.R., Bate,
R.H., Clure, V.S. (Eds.). The Oil and Gas Habitats of the South Atlantic.
Geological Society of London, Special Publication, 153. , pp. 223240.
Erbacher, J., 1998. Mid-Cretaceous radiolarians from the eastern equatorial
Atlantic and their oceanography: Leg 159 of the Ocean Drilling
Program. Proceedings of the Ocean Drilling Program, Scientic Results
159, 363373.
Erbacher, J., Thurow, J., 1997. Inuence of oceanic anoxic events on the
evolution of mid-Cretaceous radiolaria in the North Atlantic and
western Tethys. Marine Micropaleonology 30, 139158.
Erbacher, J., Thurow, J., Littke, R., 1996. Evolution of radiolaria and
organic matter variations: a new approach to identify sea-level changes
in mid-Cretaceous pelagic environments. Geology 24, 499502.
Forster, R., 1978. Evidence for an open seaway between northern and
southern proto-Atlantic in Albian times. Nature 272, 158159.
de Graciansky, P.C., Brosse, E., Deroo, G., Herbin, J.P., Montade, L.,
Muller, C., Sigal, J., Schaaf, A., 1987. Organic-rich sediments and
palaeoenvironmental reconstructions of the Cretaceous North Atlantic.
Marine Petroleum Source Rocks, Brooks, J., Fleet, A.J. (Eds.). Geological Society of London, Special Publication 26, 317344.
Gradstein, F.M., Agterberg, F.P., Ogg, J.G., Hardenbol, J., van Veen, P.,
Thierry, J., Huang, Z., 1995. A Triassic, Jurassic and Cretaceous time
scale. Geochronology, Time Scales and Global Stratigraphic Correlation, Berggren, W.A., Kent, D.V., Aubry, M.P., Hardenbol, J. (Eds.).
Society of Economic Paleontologists and Mineralogists, Special Publication 54, 95126.
Guiraud, R., Maurin, J.C., 1992. Early Cretaceous rifts of western and
central Africa: an overview. Tectonophysics 213, 153168.
Guiraud, R., Binks, R.M., Fairhead, J.D., Wilson, M., 1992. Chronology
and geodynamic setting of Cretaceous-Cenozoic rifting in western and
central Africa. Tectonophysics 213, 227234.
Guiraud, M., Mascle, J., Benkhelil, J., Basile, C., Mascle, G., Durand, M.,
1997. Early Creteceous deltaic sedimentary environment of the Cote
d'IvoireGhana Transform Margin as deduced from deep dive data.
Geo-Marine Letters 17, 7086.
Handoh, I., Bigg, G.R., Jones, E.J.W., Inoue, M., 1999. An ocean modelling
study of the Cenomanian Atlantic: Equatorial paleo-upwelling, organicrich sediments and the consequences for a connection between the
proto-North and South Atlantic. Geophysical Research Letters 26,
223226.
Haq, B.U., Hardenbol, J., Vail, P.R., 1987. Chronology of uctuating sea
levels since the Triassic (250 million years ago) to present. Science 235,
11581167.
Hay, W.W., 1988. Paleoceanography: A review for the GSA Centennial.
Geological Society of America Bulletin 100, 19341956.
Hay, W.W., 1995a. Cretaceous oceanography. Geologica Carpathica 46,
257266.
Hay, W.W., 1995. Paleoceanography of marine organic-carbon-rich sediments. Paleogeography, Paleoclimate, and Source Rocks, Huc, A.Y.
(Ed.). Americal Association of Petroleum Geologists, Studies in Geology 40, 2159.
Hay, W.W., 1996. Tectonics and climate. Geologische Rundschau 85, 409
437.
Hay, W.W., DeConto, R.M., Wold, C.N., 1997. Climate: Is the past the key
to the future?. Geologische Rundschau 86, 471491.
Hay, W.W., DeConto, R.M., Wold, C.N., Wilson, K.M., Voigt, S., Schulz,
M., Wold-Rossby, A., Dullo, W.C., Ronov, A.B., Balukhovsky, A.N.,
Soding, E., 1999. Alternative global Cretaceous paleogeography. In:
Barrera, E., Johnson, C.C. (Eds.). Evolution of the Cretaceous OceanClimate System. Geological Society of America, Special Paper, 332. ,
pp. 147.
Herbert, T.D., Sarmiento, J.L., 1991. Ocean nutrient distribution and

oxygenation: limits on the formation of warm saline bottom water


over the past 91 m.y. Geology 19, 702705.
Hisada, K., Arai, S., Yamaguchi, T., 1998. Detrital chromian spinels from
Site 960 in the Cote d'IvoireGhana Transform Margin. Proceedings of
the Ocean Drilling Program, Scientic Results 159, 133139.
Holbourn, A.E.L., Kuhnt, W., 1998. TuronianSantonian benthic foraminiferal assemblages from Site 959D (Cote d'IvoireGhana Transform
Margin Equatorial Atlantic): Indication of a Late Cretaceous oxygen
minimum zone . Proceedings of the Ocean Drilling Program, Scientic
Results 159, 375387.
Holbourn, A.E.L., Moullade, M., 1998. Lower Cretaceous benthic
foraminiferal assemblages from ODP Leg 159 (Cote d'IvoireGhana
Transform Margin Equatorial Atlantic): Biostratigraphic, paleoenvironmental and paleobiogeographic signicance . Proceedings of the Ocean
Drilling Program, Scientic Results 159, 347362.
Holbourn, A.E.L., Kuhnt, W., El Albani, A., Ly, A., Herbin, J.P., 1999a.
Palaeoenvironments and palaeobiogeography of the Late Cretaceous
Casamance Transect (Senegal, NW Africa): Distribution patterns of
benthic foraminifera, organic carbon and terrigenous ux. Neues Jahrbuch fur Geologie und Palaontologie, Abhandlungen 212, 335377.
Holbourn, A.E.L., Kuhnt, W., El Albani, A., Pletsch, T., Luderer, F.,
Wagner, T., 1999b. Upper Cretaceous palaeoenvironments and benthic
foraminiferal assemblages of potential source rocks from the western
African margin, Central Atlantic. The Oil and Gas Habitats of the South
Atlantic, Cameron, N.R., Bate, R.H., Clure, V.S. (Eds.). Geological
Society of London, Special Publication 153, 195222.
Holmes, M.A., 1998. Thermal diagenesis of Cretaceous sediment recovered
during ODP Leg 159 to the Cote d'IvoireGhana margin. Proceedings
of the Ocean Drilling Program, Scientic Results 159, 5370.
Hu, X., Wang, Y.L., Schmitt, R.A., 1988. Geochemistry of sediments on the
Rio Grande Rise and the redox evolution of the South Atlantic. Geochimica et Cosmochimica Acta 52, 201207.
Janik, A.G., Hood, J.A., Ask, M.V., 1998. Physical properties data at Hole
959D of Leg 159: comparison of core and log measurements and a
proposed revision of lithological units. Proceedings of the Ocean Drilling Program, Scientic Results 159, 241249.
Jansa, L.F., Gardner, J.V., Dean, W.E., 1977. Mesozoic sequences of the
central North Atlantic. Initial Reports of the Deep Sea Drilling Project
41, 9911031.
Jones, E.J.W., 1987. Fracture zones in the equatorial Atlantic and the
breakup of western Pangaea. Geology 15, 533536.
Jones, E.J.W., Cande, S.C., Spathopoulos, F., 1995. Evolution of a major
oceanographic pathway: the equatorial Atlantic. The Tectonics, Sedimentation and Palaeoceanography of the North Atlantic Regions, Scrutton, R.A., Stoker, M.S., Shimield, G.B., Tudhope, A.W. (Eds.).
Geological Society of London, Special Publication 90, 199213.
Kesse, G.O., 1986. Oil and gas possibilities on- and offshore Ghana. Future
Petroleum Provinces of the World, Halbouty, M.T. (Ed.). American
Association of Petroleum Geologists, Memoir 40, 427444.
Kjemperud, A., Agbesinyale, W., Agdestein, T., Gustafsson, C., Yukler, A.,
1992. Tectono-stratigraphic history of the Keta Basin, Ghana, with
emphasis on late erosional episodes. Geologie Africaine, Curnelle, R.
(Ed.). Bulletin des Centres de Recherche Exploration et Production Elf
Aquitaine, Memoire 13, 5569.
de Klasz, I., Jan du Chene, R., 1978. Presence of AlbianCenomanian in
southwestern Nigeria and its paleogeographic implications. Comptes
Rendus des Seances, SPHN Geneve, NS 13, 1015.
Koutsoukos, E.A.M., 1992. Late Aptian to Maastrichtian foraminiferal
biogeography and palaeoceanography of the Sergipe Basin Brazil.
Palaeogeography, Palaeoclimatology, Palaeoecology 92, 295324.
Koutsoukos, E.A.M., Mello, M.R., de Azambuja Filho, N.C., 1991. Micropalaeontological and geochemical evidence of mid-Cretaceous
dysoxic-anoxic palaeoenvironments in the Sergipe Basin, northeastern
Brazil. Modern and Ancient Continental Shelf Anoxia, Tyson, R.V.,
Pearson, T.H. (Eds.). Geological Society of London, Special Publication 58, 427447.
Kuhnt, W., Moullade, M., 1991. Quantitative analysis of Upper Cretaceous

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174


abyssal agglutinated foraminiferal distribution in the North Atlantic
paleoceanographic implications. Revue de Micropaleontologie 34,
313349.
Kuhnt, W., Wiedmann, J., 1995. Cenomanian-Turonian source rocks:
paleobiogeographic and paleoenvironmental aspects. Paleogeography,
Paleoclimate, and Source Rocks, Huc, A.Y. (Ed.). American Association of Petroleum Geologists, Studies in Geology 40, 213231.
Kuhnt, W., Kaminski, M.A., Moullade, M., 1989. Late Cretaceous deepwater agglutinated foraminiferal assemblages from the North Atlantic
and its marginal seas. Geologische Rundschau 78, 11211140.
Kuhnt, W., Herbin, J.P., Thurow, J., Wiedmann, J., 1990. Distribution of
CenomanianTuronian organic facies in the western Mediterranean and
along the adjacent Atlantic margin. Deposition of Organic Facies, Huc,
A.Y. (Ed.). American Association of Petroleum Geologists, Studies in
Geology 30, 133160.
Kuhnt, W., Moullade, M., Kaminski, M.A., 1998. Upper Cretaceous K/T
boundary, and lowermost Paleocene agglutinated foraminifera from
Hole 959D (ODP Leg 159, Cote d'IvoireGhana Transform Margin).
Proceedings of the Ocean Drilling Program, Scientic Results 159,
389411.
Larson, R.L., Ladd, J.W., 1973. Evidence for the opening of the South
Atlantic in the early Cretaceous. Nature 246, 209212.
Le Pichon, X., Fox, P.J., 1971. Marginal offsets, fracture zones, and the
early opening of the North Atlantic. Journal of Geophysical Research
76, 62946308.
Le Pichon, X., Hayes, D.E., 1971. Marginal offsets, fracture zones, and the
early opening of the South Atlantic. Journal of Geophysical Research
76, 62836293.
Lespinasse, M., Leroy, J.L., Pironon, J., Moiron, M.C., 1998. Paleouids
from the marginal ridge of the Cote d'IvoireGhana Transform Margin
(Hole 960A) as thermal indicators. Proceedings of the Ocean Drilling
Program, Scientic Results 159, 4952.
Lever, A., McCave, I.N., 1983. Eolian components in Cretaceous and
Tertiary North Atlantic sediments. Journal of Sedimentary Petrology
53, 811832.
Lorenzo, J.M., Wessel, P., 1997. Flexure across a continent-ocean fracture
zone: The northern Falkland/Malvinas Plateau. South Atlantic. GeoMarine Letters 17, 110118.
Ly, A., Kuhnt, W., 1994. Late Cretaceous benthic foraminiferal assemblages of the Casamance shelf (Senegal. NW Africa): Indication of a
Late Cretaceous oxygen minimum zone. Revue de Micropaleontologie
37, 4974.
Mabesoone, J.M., Alheiros, M.M., 1993. Evolution of the PernambucoParaba-Rio Grande do Norte Basin and the problem of the South
Atlantic connection. Geologie en Mijnbouw 71, 351362.
Machens, E., 1973. The geologic history of the marginal basins along the
north shore of the Gulf of Guinea. In: Nairn, A.E.M, Stehli, F.G. (Eds.).
The Ocean Basins and Margins, The South Atlantic, vol. 1. Plenum
Press, New York, NY, USA, p. 583.
MacLeod, K.G., Huber, B.T., 1996. Reorganization of deep ocean circulation accompanying a Late Cretaceous extinction event. Nature 380,
422425.
Maley, J., 1996. The African rain forest main characteristics of changes
in vegetation and climate from the Upper Cretaceous to the Quaternary.
Proceedings of the Royal Society of Edinburgh 104B, 3173.
Marcano, M.C., Lohmann, K.C., Pickett, E.A., 1998. Uplift and exposure
history of the Cote d'IvoireGhana Transform Margin: geochemistry of
pore-lling and fracture vein calcites. Proceedings of the Ocean Drilling Program, Scientic Results 159, 7179.
Mascle, J., Blarez, E., 1987. Evidence for transform margin evolution for
the Cote d'IvoireGhana continental margin. Nature 326, 378381.
Mascle, J., Marinho, M., Wannesson, J., 1986. The structure of the Guinean
continental margin: implications for the connection between the central
and the South Atlantic Oceans. Geologische Rundschau 75, 5770.
Mascle, J., Blarez, E., Marinho, M., 1988. The shallow structure of the
Guinea and Ivory CoastGhana Transform Margins: Their bearing on
the equatorial Atlantic evolution. Tectonophysics 155, 193209.

173

Mascle, J., Lohmann, G.P., Clift, P.D., et al., 1996. Introduction. Proceedings of the Ocean Drilling Program, Part A: Initial Reports 159, 516.
ODP 159 Scientic Party, Mascle, J., Lohmann, G.P., Clift, P.D., 1997.
Development of a passive transform margin: Cote d'IvoireGhana
Transform Margin - ODP Leg 159 preliminary results. Geo-Marine
Letters 17, 411.
Masure, E., Rauscher, R., Dejax, J., Schuler, M., Ferre, B., 1998. Cretaceous-Paleocene palynonology from Cote d'IvoireGhana Transform
Margin: Leg 159, Sites 959, 960, 961, 962. Proceedings of the Ocean
Drilling Program, Scientic Results 159, 253276.
Morrison, J., Tea, J., N'Zalasse, B., Boblai, V., 1999. A sequence stratigraphic approach to exploration and re-development in the Abidjan
margin, Cote d'Ivoire. Offshore West Africa '99: Conference and Exhibition, Abidjan, 2325 March, 1999, pp. 112.
Moullade, M., Guerin, S., 1982. Le probleme des relations de l'Atlantique
Sud et de l'Atlantique Central au Cretace moyen: Nouvelles donnes
microfauniques d'apres les forages DSDP. Bulletin de la Societe Geologique de France 24 (3), 511517.
Moullade, M., Mascle, J., Benkhelil, J., Cousin, M., Tricart, P., 1993.
Occurrence of marine mid-Cretaceous sediments along the Guinean
slope (Equamarge II cruise): their signicance for the evolution of
the central Atlantic African margin. Marine Geology 110, 6372.
Moullade, M., Watkins, D.K., Oboh-Ikuenobe, F.E., Bellier, J.P., Holbourn,
A.E.L., Erbacher, J., Kuhnt, W., Pletsch, T., Kaminski, M.A., Rauscher,
R., Shak, S., Yepes, O., Dejax, J., Gregg, J.M., Shin, I.C., Schuler, M.,
1998. ODP Leg 159, Equatorial Atlantic: Mesozoic biostratigraphic,
paleoenvironmental and paleobiogeographic synthesis. Proceedings of
the Ocean Drilling Program, Scientic Results 159, 481490.
Natland, J.H., 1978. Composition, provenance, and diagenesis of Cretaceous clastic sediments drilled on the Atlantic continental margin off
southern Africa, DSDP 361 implications for the early circulation
of the South Atlantic. Initial Reports of the Deep Sea Drilling Project
40, 10251061.
Nelson, B.K., MacLeod, K.G., Ward, P.D., 1991. Rapid change in strontium isotopic composition of the sea water before the Cretaceous/
Tertiary boundary. Nature 351, 644647.
Nurnberg, D., Muller, R.D., 1991. The tectonic evolution of the South
Atlantic Ocean from Late Jurassic to present. Tectonophysics 191,
2753.
ODP Leg 159 Scientic Party, Oboh-Ikuenobe, F.E., Yepes, O., 1997.
Palynofacies of sediments from the Cote d'IvoireGhana Transform
Margin: preliminary correlation with some regional events in the equatorial Atlantic. Palaeogeography, Palaeoclimatology, Palaeoecology
129, 291314.
Oboh-Ikuenobe, F.E., Yepes, O., Gregg, J.M., 1998. Palynostratigraphy,
palynofacies and thermal maturation of CretaceousPaleocene sediments from the Cote d'IvoireGhana Transform Margin. Proceedings
of the Ocean Drilling Program, Scientic Results 159, 277318.
Ojoh, K.A., 1990. Evolution geodynamique de bassins albo-santoniens du
sud-ouest du fosse de la Benoue (Nigeria). Bulletin des Centres de
Recherche Exploration et Production Elf Aquitaine 14, 419442.
Ortega-Huertas, M., MartInez-RuIz, F., Palomo, I., Chamley, H., 1995.
Comparative mineralogical and geochemical clay sedimentation in
the Betic Cordilleras and Basque-Cantabrian Basin areas at the CretaceousTertiary boundary. Sedimentary Geology 94, 209227.
Parrish, J.T., 1995. Paleogeography of Corg-rich rocks and the preservation
versus production controversy. Paleogeography, Paleoclimate, and
Source Rocks, Huc, A.Y. (Ed.). American Association of Petoleum
Geologists, Studies in Geology 40, 120.
Pickett, E.A., Allerton, S., 1998. Structual observations from the Cote
d'IvoireGhana Transform Margin. Proceedings of the Ocean Drilling
Program, Scientic Results 159, 311.
Pletsch, T., 1998. Origin of Lower Eocene palygorskite claystones on the
Cote d'IvoireGhana Margin (Leg 159, eastern equatorial Atlantic).
Proceedings of the Ocean Drilling Program, Scientic Results 159,
137152.
Pletsch, T., Daoudi, L., Chamley, H., Deconinck, J.F., Charroud, M., 1996.

174

T. Pletsch et al. / Journal of South American Earth Sciences 14 (2001) 147174

Palaeogeographic controls on palygorskite occurrence in mid-Cretaceous sediments of Morocco and adjacent basins. Clay Minerals 31,
403416.
Rabinowitz, P.D., LaBreque, J., 1979. The Mesozoic South Atlantic and
evolution of its continental margins. Journal of Geophysical Research
85, 59736002.
Ravizza, G., 1998. The osmium-isotope geochemistry of Site 959: implications for ReOs sedimentary geochronology, and reconstruction of past
variations in the Os-isotopic composition of seawater. Proceedings of
the Ocean Drilling Program, Scientic Results 159, 181186.
Reyment, R.A., 1969. Ammonite biostratigraphy, continental drift and
oscillatory transgressions. Nature 224, 6769.
Reyment, R.A., Tait, E.A., 1972. Biostratigraphical dating of the early
history of the South Atlantic Ocean. Philosophical Transactions of
The Royal Society, Series B 264, 5595.
Robert, C., 1987. Clay mineral associations and structural evolution of the
South Atlantic: Jurassic to Eocene. Palaeogeography, Palaeoclimatology, Palaeoecology 58, 87108.
Saint-Marc, P., N'da, V., 1997. Biostratigraphie et paleoenvironnements
des depots cretaces au large d'Abidjan (Golfe de Guinee). Cretaceous
Research 18, 545565.
Sarnthein, M., Faugeres, J.C., 1993. Radiolarian contourites record Eocene
AABW circulation in the equatorial East Atlantic. Sedimentary Geology 82, 145155.
Scheibnerova, V., 1981. Paleogeographical implications of Cretaceous
benthic foraminifera recovered by the Deep Sea Drilling Project in
the western Atlantic Ocean. Cretaceous Research 2, 118.
Schlanger, S.O., Arthur, M.A., Jenkyns, H.C., Scholle, P.A., 1987. The
CenomanianTuronian oceanic anoxic event, I. Stratigraphy and distribution of organic carbon-rich beds and the marine d 13C excursion.
Marine Petroleum Source Rocks, Brooks, J., Fleet, A.J. (Eds.). Geological Society of London, Special Publication 26, 371399.
Scrutton, R.A., 1982. Crustal structure of sheared passive continental
margins. Dynamics of Passive Margins, Scrutton, R.A. (Ed.). American
Geophysical Union, Geodynamics Series 6, 133140.
Shak, S., Watkins, D.K., Shin, I.C., 1998. Calcareous nannofossil Paleogene biostratigraphy, Cote d'IvireGhana Marginal Ridge, eastern
equatorial Atlantic. Proceedings of the Ocean Drilling Program, Scientic Results 159, 413431.
Sibuet, J.C., Mascle, J., 1978. Plate kinematic implications of equatorial
fracture zone trends. Journal of Geophysical Research 83, 34013421.
Strand, K., 1998. Sedimentary facies and sediment composition in response
to tectonics of the Cote d'IvoireGhana Transform Margin. Proceedings of the Ocean Drilling Program, Scientic Results 159, 113123.
Summerhayes, C.P., 1981. Organic facies of middle Cretaceous black
shales in the deep North Atlantic. American Association of Petroleum
Geologists Bulletin 65, 23642380.
Summerhayes, C.P., 1987. Organic-rich Cretaceous sediments from the
North Atlantic. Marine Petroleum Source Rocks, Fleet, A.J., Brooks,
J. (Eds.). Geological Society of London, Special Publication 26, 301
316.

Thiede, J., van Andel, T.H., 1977. The paleoenvironment of anaerobic


sediemnts in the Late Mesozoic South Atlantic Ocean. Earth and Planetary Science Letters 33, 301309.
Thiry, M., Jacquin, T., 1993. Clay mineral distribution related to rift activity, sea-level changes and oceanography in the Cretaceous of the Atlantic Ocean. Clay Minerals 28, 6184.
Thurow, J., Brumsack, H.J., Rullkotter, J., Meyers, P.A., 1992. The Cenomanian/Turonian Boundary Event in the Indian Ocean a key to
understand the global picture. Synthesis of Results from Scientic Drilling in the Indian Ocean, Duncan, R.A., Rea, D.K., Kidd, R.B., von Rad,
U., Weissel, J.K. (Eds.). American Geophysical Union, Geophysical
Monograph 70, 253273.
Tucholke, B.E., Vogt, P.R., 1979. Western North Atlantic sedimentary
evolution and aspects of tectonic history. Initial Reports of the Deep
Sea Drilling Project 43, 791825.
Wagner, T., Pletsch, T., 1999. Tectono-sedimentary controls on Cretaceous
black shale deposition along the opening Equatorial Atlantic Gateway
(ODP Leg 159). The Oil and Gas Habitats of the South Atlantic,
Cameron, N.R., Bate, R.H., Clure, V.S. (Eds.). Geological Society of
London, Special Publication 153, 241265.
Watkins, D.K., Shak, S., Shin, I.C., 1998. Calcareous nannofossils from
the Cretaceous of the Deep Ivorian Basin. Proceedings of the Ocean
Drilling Program, Scientic Results 159, 319333.
Wegener, A., 1929. Die Entstehung der Kontinente und Ozeane. Vieweg,
Braunschweig, Germany.
Wiedmann, J., Neugebauer, J., 1978. Lower Cretaceous ammonites from
the South Atlantic Leg 40 (DSDP), their stratigraphic value and sedimentologic properties. Initial Reports of the Deep Sea Drilling Project
40, 709734.
Wightman, W.G., Kuhnt, W., 1992. Biostratigraphy and paleoecology of
Late Cretaceous abyssal agglutinated foraminifers from the western
Pacic Ocean (Deep Sea Drilling Project Holes 196A and 198A and
Ocean Drilling Program Holes 800A and 801A). Proceedings of the
Ocean Drilling Program, Scientic Results 129, 247264.
Wilson, M., Guiraud, R., 1992. Magmatism and rifting in western and
central Africa, from Late Jurassic to Recent times. Tectonophysics
213, 203225.
Woo, K.S., Anderson, T.F., Railsback, L.B., Sandberg, P.A., 1992. Oxygen
isotope evidence for high-salinity surface seawater in the mid-Cretaceous Gulf of Mexico: Implications for warm saline deepwater formation. Paleoceanography 7, 673685.
Zalan, P.V., Nelson, E.P., Warme, J.E., Davis, T.L., 1985. The Piaui Basin:
rifting and wrenching in an equatorial Atlantic transform basin. StrikeSlip Deformation, Basin Formation, and Sedimentation, Biddle, K.T.,
Christie-Blick, N. (Eds.). Society of Economic Paleontologists and
Mineralogists, Special Publication 37, 177192.
Zimmermann, H.B., Boersma, A., McCoy, F.W., 1987. Carbonaceous sediments and palaeoenvironment of the Cretaceous South Atlantic Ocean.
Marine Petroleum Source Rocks, Brooks, J., Fleet, A.J. (Eds.). Geological Society of London, Special Publication, 271286.

You might also like