Author's personal copygrupo179/pdf/Gallego Torres 2014.pdf · 2014-03-26 · Author's personal copy Rapid bottom-water circulation changes during the last glacial cycle in the coastal

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/authorsrights

http://www.elsevier.com/authorsrights

Author's personal copy

Rapid bottom-water circulation changes during the last glacial cycle inthe coastal low-latitude NE Atlantic

David Gallego-Torres a,b,⁎, Oscar E. Romero a, Francisca Martínez-Ruiz a, Jung-Hyun Kim c,Barbara Donner d, Miguel Ortega-Huertas b

a Instituto Andaluz de Ciencias de la Tierra (CSIC-UGR), Avenida de las Palmeras 4, 18100 Armilla, Granada, Spainb Departamento de Mineralogia y Petrologia (UGR), Facultad de Ciencias, Campus Fuentenueva, 18002 Granada, Spainc NIOZ Royal Netherlands Institute for Sea Research, Department of Marine Organic Biogeochemistry, PO Box 59, AB Den Burg, 1790 Texel, The Netherlandsd MARUM—Center for Marine Environmental Sciences, University of Bremen, P.O. Box 330440, 28334 Bremen, Germany

a b s t r a c ta r t i c l e i n f o

Article history:Received 7 November 2012Available online 7 January 2014

Keywords:Redox proxiesNE AtlanticPaleoceanographyAMOC collapse

Previous paleoceanographic studies along the NW African margin focused on the dynamics of surface and inter-mediate waters, whereas little attention has been devoted to deep-water masses. Currently, these deep watersconsist mainly of North Atlantic Deep Waters as part of the Atlantic Meridional Overturning Circulation(AMOC). However, this configuration was altered during periods of AMOC collapse. We present a high-resolution reconstruction of bottom-water ventilation and current evolution off Mauritania from the last glacialmaximum into the early Holocene. Applying redox proxies (Mo, U and Mn) measured on sediments from offMauritania, we describe changes in deep-water oxygenation andwe infer the evolution of deep-water conditionsduring millennial-scale climate/oceanographic events in the area. The second half of Heinrich Event 1 and theYounger Dryas were recognized as periods of reduced ventilation, coinciding with events of AMOC reduction.We propose that these weakening circulation events induced deficient deep-water oxygenation in theMauritanian upwelling region, which together with increased productivity promoted reducing conditions andenhanced organic-matter preservation. This is the first time the effect of AMOC collapse in the area is describedat high resolution, broadening the knowledge on basin-wide oceanographic changes associated with rapid cli-mate variability during the last deglaciation.

© 2013 University of Washington. Published by Elsevier Inc. All rights reserved.

Introduction

Greenland ice-core records show that climate amelioration duringthe last deglaciation in the North Atlantic and surrounding regionswas not smooth but rather included a series of abrupt changes and re-versals (e.g., Dansgaard et al., 1993; Grootes and Stuiver, 1997). Our un-derstanding of rapid climate changes, and the role played by oceancirculation through heat and salinity redistribution, still remains incom-plete. This is partly due to the scarcity of suitable sedimentary materialexhibiting high sedimentation rate and containing appropriate proxyvariables that are able to resolve the timing and phasing of atmosphericand ocean responses during abrupt climate reversals, such as theHeinrich Events (HEs), Bølling–Allerød (B–A) and Younger Dryas(YD), thus undermining efforts to document the role played by oceancirculation during these events. The significance of oceans in global re-distribution of heat and their ability to switch between different statesof equilibrium have suggested that changes in the Atlantic MeridionalOverturning Circulation (AMOC), and associated modes of deep-water

formation,were responsible for the sharp climate shifts seen inGreenlandice-core records (e.g., Rahmstorf, 2002). Several of these rapid AMOCshifts since the last glacialmaximum(LGM)had a global imprint detectedat localities at different latitudes, such as the Southern Iceland Rise(Thornalley et al., 2011), the Bermuda Rise (Hall et al., 2011; McManuset al., 2004), or the SouthernOcean (Rutberg et al., 2000), though their ef-fect in the eastern North Atlantic is still poorly described (e.g., Chapmanand Shackleton, 1998; Skinner and Shackleton, 2006), particularly fordeep-water masses. Furthermore, a decreased AMOC could facilitate theintrusion of deep Antarctic BottomWaters (AABW) flowing northwardsalong the eastern margins (e.g., Labeyrie et al., 2005; Skinner andShackleton, 2006).

Due to its location on the easternmost border of the North AtlanticSubtropical Gyre, the upwelling system off NW Africa is very sensitiveto variations of the AMOC (Kim et al., 2012). In addition to surface-water processes (Romero et al., 2008; Kim et al., 2012), variationsof deep-water conditions underlying this highly productive area arealso recorded in down-core sediments (Filipsson et al., 2011). Thus, itis important to distinguish between the effects of oxygen consumptionrelated to export of organic matter and the result of decreased deep-water ventilation. Hence, interpreting the record of deep-water condi-tions linked to deep currents, compared to the possible effect of primary

Quaternary Research 81 (2014) 330–338

⁎ Corrresponding author: Instituto Andaluz de Ciencias de la Tierra (CSIC-UGR).Avenida de las Palmeras, 4. 18100 Armilla, Granada, Spain.

E-mail address: [email protected] (D. Gallego-Torres).

0033-5894/$ – see front matter © 2013 University of Washington. Published by Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.yqres.2013.11.004

Contents lists available at ScienceDirect

Quaternary Research

j ourna l homepage: www.e lsev ie r .com/ locate /yqres


productivity on the geochemical conditions at the sediment–water in-terface, can deliver important insights for the interpretation of thelocal paleoceanographic setting and its possible extrapolation to otherregions of the North Atlantic Ocean.

Much attention has been paid to processes andmechanisms that oc-curred during short-term Quaternary climate variations in the lower at-mosphere in NW Africa (de Menocal et al., 2000a,b; Kuhlmann et al.,2004; Kim et al., 2007; Mulitza et al., 2008), and the adjacent upperwa-ters (Romero et al., 2008). However, much less is known about bottom-water processes affecting the paleoceanographic signal preserved in thesedimentary record. To assess variations of deep-sea conditions offMauritania, we build on Romero et al. (2008), Filipsson et al. (2011)and Kim et al. (2012) and reconstruct redox conditions and their signif-icance for deep-water circulation during the last glacial cycle in thegravity core GeoB7926. We use a set of geochemical proxies for oxygenavailability, sedimentary regime and paleoproductivity. Mo/Al and U/Alratios and their relative co-variation were used to infer down-core var-iations in redox conditions. The Mn/Al profile allowed the reconstruc-tion of oxidation fronts in the sedimentary column. K/Al and Ca/Alwere used as proxies for variations in the sedimentary regime, andCorg and Corg Accumulation Rate (AR) served as paleoproductivityindicators.

Materials and methods

Core GeoB7926 and study area

Our high-resolution record was obtained from the gravity coreGeoB7926, (20°13′N, 18°27′W, 2500 m water depth, Fig. 1, 1328 cmlength). We present results for the depth interval between 113 and803 cm (~10–25 ka). The dominant lithology is foramifer-bearingnannofossil or diatom ooze, with short intervals marked by clay andquartz-bearing beds and a few turbidite layers in the upper 820 cm(Romero et al., 2008). This core was recovered below themajor upwell-ing region along the northwestern African continent (Romero et al.,2008). The Mauritanian shelf is generally narrow (about 45–55 kmwide) and the continental slope is about 45 km wide with an averageinclination of 2–3°.

Site GeoB7926 is ideally situated to monitor past variations in theclimate and hydrography of northwest Africa, being on the conflu-ence of different oceanic currents and within the latitudinal migra-tion of the Intertropical Convergence Zone (ITCZ). The mainoceanic currents affecting the area at present (Fig. 1) are the CanaryCurrent (CC) flowing southwards at the sea surface, and the SouthAtlantic Central Waters (SACW) immediately below (from 200 to

A

B

Figure 1.A) Localitymap showing the position of site GeoB7926 and location of sites studied byGherardi et al. (2005), Lippold et al. (2012),McManus et al. (2004), Praetorius et al. (2008),Schmidt and Lynch-Stieglitz (2011) and Thornalley et al. (2011). North Atlantic Deep Water (NADW); Antarctic BottomWater (AABW); North Atlantic Central Water (NACW); SouthAtlantic CentralWater (SACW); Canary Current (CC); North Equatorial Counter Current (NECC). B) Present-day water column dissolved-oxygen profile. Red vertical line indicates the po-sition of site GeoB7926.

331D. Gallego-Torres et al. / Quaternary Research 81 (2014) 330–338


600 m depth), which is the main source of upwelling waters duringcold, high-productivity phases (Filipsson et al., 2011). North AtlanticCentral and Deep Waters (NACW–NADW) flow beneath SACW,eventually influenced by Mediterranean Outflow Waters (MOW)(Mittelstaedt, 1991; Zhao et al., 2000; Eberwein and Mackensen,2008). Currently, AABW are restricted to water depths below4000 m (Eberwein and Mackensen, 2008).

Stratigraphy

The age control for the entire core is based on 16 Accelerator MassSpectrometry (AMS) 14C dates from specimens of the planktonic fora-minifera Globigerina inflata (Leibniz Laboratory for Radiometric Datingand Stable Isotope Research, Kiel University, Germany), as well as theoxygen isotope record of the planktonic foraminifera Globigerinabulloides (Romero et al., 2008; Kim et al., 2012). The conversion of 14Cages into calendar years was carried out using the CALIB REV6.0.0 pro-gram with the marine calibration dataset (MARINE09) (Stuiver et al.,1998). Calendar ages between dated levels were obtained by linear in-terpolation between the nearest AMS dating points. The mean oceanreservoir age of 400 yr (Bard, 1988) was applied, although the upwell-ing of older subsurface watersmight have produced larger regional res-ervoir ages (ΔR) (de Menocal et al., 2000b; Kim et al., 2012). Similarchanges in radiocarbon reservoir age have been assumed throughoutthe studied interval. Nevertheless, we acknowledge that the ΔR mighthave varied through time, as is the case for the Younger Dryas, whenmarine 14C ΔR was nearly twice the modern value (de Menocal et al.,

2000a,b; Staubwasser et al., 2002; Kim et al., 2012). Based on this strat-igraphicmodel, themain climatic phases are defined as follows: the lastglacial maximum (LGM), from 25 and 19 ka; Heinrich Event 1 (HE1),17.5 to 15.5 ka; Bølling–Allerød (B–A), 15.5 to 13.5 ka; Younger Dryas(YD), 13.5 to 11.5 ka; and the Holocene, from 11.5 ka onwards. Agemodel and sedimentation rates are shown on Fig. 2. Absolute dates aresummarized on Table 1.

Geochemical proxies

With the aimof providing a high-resolution reconstruction of the os-cillations of deep-water oxygenation, productivity variations, andchanges in detrital input and depositional environment, redox sensitiveelements (Mo and U) (e.g., Calvert and Pedersen, 2007; Algeo andTribovillard, 2009; Piper and Calvert, 2009 and references therein), de-trital proxies (K/Al ratio) and Ca/Al (e.g., Martinez et al., 1999; Haslettand Davies, 2006), and total organic carbon (TOC) were determinedby bulk sediment geochemical analyses. Before geochemical analyses,samples were dried and ground in an agate mortar, homogenized andaliquots portions were used to determine TOC and major and trace ele-ment concentrations. TOC content was calculated as the difference be-tween carbon content of the total sample and carbon content afteracidification (Romero et al., 2008).

Total dissolution of the sample was carried out using HF and HNO3

following the standard procedures described in Gallego-Torres et al.(2007). Al, Ca, K and Mn contents were determined by Atomic Absorp-tion spectrometry at the Analytical Facilities of the University of

0 100 200 300 400 500 600 700 800

Core depth (cm)

0

5

10

15

20

25

30

Age (ka)

0

40

80

120

280

320

(cm

ka-1

)S

edim

enta

tion

rate

Figure 2. Age model (black line) obtained from 13 calibrated δ14C dates (vertical bars represent standard deviations), and accumulation rates (cmka–1) from LGM to present for coreGeoB7926 (modified from Romero et al., 2008). Absolute values are summarized on Table 2.

Table 1Age control points used for the age model for core GeoB7926 from LGM to present. Uncorrected ages were converted to calendar ages using MARINE09 (Kim et al., 2012).

Lab no. Depth in core (cm) Uncorrected AMS14C ages (yr BP) Analytical error (± ) (yr) Ages (Δ = 0 yr) (±26) (cal yr BP) Ages (cal ka BP)

KIA 24287 8 1170 35 649–787 0.718KIA 25812 48 4455 35 4519–4779 4.649KIA 25810 98 8970 45 9515–9798 9.6565KIA 24286 173 10830 70 12054–12560 12.307KIA 22417 248 11020 70 12357–12677 12.517KIA 24285 378 12220 70 13450–13824 13.637KIA 29030 418 13050 70 14487–15247 14.867KIA 27310 483 13720 100 15568–16813 16.1905KIA 29029 508 14290 60 16777–17166 16.9715KIA 22416 553 14670 80 16990–17686 17.338KIA 27309 688 16600 130 18903–19571 19.237KIA 29028 803 20450 160 23484–24388 26.936KIA 24283 868 23650 220–230 27562–28622 28.092

332 D. Gallego-Torres et al. / Quaternary Research 81 (2014) 330–338


Granada (CIC) using Re and Rh as internal standards. The trace elements(U and Mo) were measured with an ICP-MS Perkin-Elmer Sciex Elan5000 spectrometer (CIC). Coefficients of variation calculated by dissolu-tion and subsequent analyses of 10 replicates of powdered sampleswere better than 3% and 8% for analyte concentrations of 50 and5 ppm, respectively. In order to correct for detrital variations, all ele-ments were normalized to Al content (Van Der Weijden, 2002).

Mo and U have been traditionally used as redox proxies (e.g.,Brumsack, 1980; de Lange and Ten Haven, 1983; Calvert and Pedersen,1993; Jones and Manning, 1994; McManus et al., 2005), since theychange their solubility and stability depending on their redox stateprecipitating under reduced oxygen concentration and/or in the pres-ence of HS− (e.g., Tribovillard et al., 2006; Algeo and Tribovillard, 2009;Morford et al., 2009). U canprecipitate under hypoxic conditions forminguraninite (e.g., Morford et al., 2009), whereas Mo precipitates mainly assulfides in the presence of HS− (e.g., Morford et al., 2005, 2009), thus im-plying anoxic–euxinic conditions. Thus, the combined study of the twoelements allows for the identification of relatively minor to intense oxy-gen depletion in deep water masses overlying the sedimentary column.In addition, their co-variation, expressed as enrichment factors (Arthuret al., 1990), provides information about oxygenation in the water col-umn and/or the evolution of redox conditions and the chemistry of thebasin (see Algeo and Tribovillard, 2009 for details).

At an open-ocean setting (where GeoB7926 is located), the averagesea-water Mo/U ratio can be considered the “limiting budget” of Mo–Uco-variation assuming that the water column is neither anoxic nor stag-nant throughout. Under these circumstances, average seawater Mo/Uratio represents the maximum rate of fixation, meaning an anoxic envi-ronment at the sediment–water interface and the redoxcline being slight-ly above the seafloor. Above thisMo/U ratio somedegree ofwater columnanoxia must be invoked. Additionally, Mn precipitates as oxy-hydroxidesin the presence of free O2 (Thomson et al., 1993). Diagenetic Mn concen-tration in the sediments implies oxygen penetration in the sediment andprecipitation of oxy-hydroxideswhendownward diffusingO2 encountersupwards diffusing dissolvedMn inporewaters fromanunderlying anoxicsediment column, allowing the reconstruction of re-oxygenation events(van Santvoort et al., 1996; Gallego-Torres et al., 2010). K and Ca areused to detect changes in the sedimentary regime. Ca can be quantifiedas CaCO3, thus related to calcareous organisms, although a calcite preser-vation effect must be considered, and K ismainly derived from fluvial de-tritalmaterial (illite, other clayminerals; Bloemsma et al., 2012). The totalprimary productivity is interpreted from Corg content and Corg (AR)(Romero et al., 2008).

Results

Inorganic geochemistry records as a whole, and trace element re-cords in particular, are scarce in the area off Mauritania. Martinez et al.(1999) provided themost reliableMo record tracing long-term changesin redox conditions below this upwelling system. However, no high-resolution studies have been carried out to detect abrupt changes inoceanographic conditions in the region. We present a set of high-resolution geochemical data that allows detailed reconstruction.

Corg content and Corg AR are plotted on Fig. 3A and B. Corg shows highconcentrations in the LGM and YD, and minimum values during earlyHE1. Corg AR is at a maximum during the YD and displays high valuesduring the early deglaciation (~19–17 ka) and earlyHE1 (~13–11.5 ka).

A

B

C

D

E

F

G

Figure 3. Corg (A) and Corg accumulation rate (B) profiles. K/Al (C) and Ca/Al (D) used asa proxy for detrital vs. carbonate sediment supply. Redox sensitive elements Mo/Al (E) andU/Al (F) are evidence of reducing environments. Yellow boxes highlight AMOC collapseevidenced by McManus et al. (2004). Yellow-gray bar indicates the re-ventilation eventdescribed by Thornalley et al. (2011). Mn/Al (G) marks diagenetic oxidation fronts.LGM: last glacialmaximum;HE1:HeinrichEvent 1; B–A:Bolling-Allerod; YD:YoungerDryas.



The K/Al ratio shows high values during the LGM, with a sharp de-crease during the onset of the deglaciation (Fig. 3C). A progressive in-crease until the early HE1 is followed by low K/Al ratios from 17.0 to15.5 ka. The early B–A (15–14 ka) is characterized by high K/Al follow-ed by a steep decrease, whereas high values define the YD. Ca/Al ratiospresent generally low values during LGM (Fig. 3D) progressively in-creasing until the early HE1. The later part of HE1 has relatively lowvalues while the warm B–A period exhibits high Ca/Al ratios. Similarto the LGM, the YD shows low Ca/Al.

Redox sensitive element profiles show low values of Mo/Al (Fig. 3 E)andU/Al (Fig. 3F) throughout the LGM(Fig. 3). Aminor increase inMo/Aland U/Al is recorded around 22.3 ka. Mo/Al and U/Al progressively in-crease until the HE1, when a sharp peak of both ratios occurs. Low Moconcentration are observed during the B–A. TheU/Al ratiofluctuates dur-ing this period, though displaying generally lower values than during theHE1. The YD event is again characterized by high Mo/Al and U/Al with asharp Mn/Al increase around 12.3 ka. After the YD, low Mo/Al, U/Al andMn/Al values characterize the transition into the early Holocene. Prefer-ential Mn precipitation characterizes the period between 19 and 17 ka(Fig. 3G).

Fig. 3 shows Mo to U co-variation expressed as enrichment factors.Most of the values remain below the average oceanic Mo/U ratio (grayline, Algeo and Tribovillard, 2009) and this ratio reaches the seawaterthreshold only during the YD.

Discussion

Bottom water oxygenation off Mauritania during the LGM and the lastdeglaciation

Low Corg AR (Fig. 3B) evidences that productivity was relatively lowduring the early part of the LGM (Romero et al., 2008). The dominanceof silicate-poor NACW bathing the upwelling area off Mauritania mighthave limited diatom production at the time, whereas calcareous micro-organisms, as shown by high Ca/Al values (Fig. 3D), contributed themost to productivity in surface waters (Romero et al., 2008). Duringthe LGM, Ca content decreased and an increase in Corg occurred, butCorg AR remained constant. More corrosive bottom waters, such asAABW (Lippold et al., 2012), might have enhanced carbonate dissolu-tion. Very low Mo/Al and U/Al values (Fig. 3) indicate the presence ofdissolved oxygen in deep-water masses as well as downward oxygendiffusion into the uppermost section of the sediment column. This im-plies efficient deep-water ventilation maintaining oxic conditions atthe sediment–water interface at 2500 m off Mauritania throughoutthe LGM.

The changes in North Atlantic circulation during the LGM involveddecreased intermediate water flow (Lynch-Stieglitz et al., 2007),allowing for the upwelling of southern-sourced silica-rich waters andenhancing diatom productivity during the latest LGM and the transitioninto the glacial termination (Romero et al., 2008). As shown by lowtrace-metal content (Fig. 3 G–F), deep-water circulation did not slowdown throughout the late LGM nor at the onset of the deglaciation(McManus et al., 2004), allowing for efficient oxygenation of the seabottom at site GeoB7926. We may deduce that NADW flow remainedactive or, alternatively, that deep-water masses might have alsochanged from a northern to southern source (AABW) (e.g., Adkinset al., 2002; Skinner and Shackleton, 2006; Martrat et al., 2007).

A strong decrease in Corg content (Fig. 3A) characterizes the earlystage of deglaciation. Higher Corg AR, however, suggests a possible dilu-tion effect related to higher calcareous production (Romero et al., 2008).Carbonate deposition and delivery of detrital material progressively in-creased, reflected by increasingK/Al and Ca/Al ratios (Fig. 3C andD). Be-tween the early deglaciation and the onset of HE1 (~18.8–17.8 ka), theenhancedMnprecipitation (increase inMn/Al) indicates the occurrenceof an oxidation front within the sediment. This might have partiallydegraded organic matter, similar to the process described for eastern

Mediterranean sapropels (e.g., de Lange and Ten Haven, 1983; Gallego-Torres et al., 2010). At site GeoB7926, however, this phenomenonoccurred in a hemipelagic environment with much higher accumulationrates than those recorded in the eastern Mediterranean Sea.

Two conclusions can be extracted from the occurrence of this oxida-tion front: (1) the amount of Corg accumulated must have been greaterthan those presently measured at site GeoB7926 for this interval(~18.8–17.8 ka) (e.g., Thomson et al., 1995); (2) deep-water ventilationmust have been strong enough to keep oxic conditions at the sediment–water interface, thus promoting organic-matter oxidation. This issupported by high Mn and low U and Mo concentrations (Fig. 3). Afterthe LGM and at the onset of HE1, primary productivity was high and or-ganic matter might have been more efficiently transported downwardsand accumulated as observed by high Corg AR at site GeoB7926 (Fig. 3B,Romero et al., 2008). Simultaneously, as suggested by the spiky signal ofCorg, U/Al andMn/Al between 18.0 ka and 17.3 ka (Fig. 3), high bottom-water ventilation allowed for partial oxidation of the deposited organiccarbon. Low Mo fixation supports the oxic seafloor scenario, eventhough organic-matter input was intense.

The first shift in redox conditions corresponds to the late HE1(~16 ka). Intensified NE trade winds characterized the end of HE1 offMauritania (e.g., Tjallingii et al., 2008; Filipsson et al., 2011), which fa-vored upwelling and strong diatom productivity in surface waters(Romero et al., 2008). Although Corg content preserved in the sedimentsincreases during the late HE1 (~16 ka), Corg AR was actually lower thanimmediately before (19–17 ka), resembling LGM values. The origin ofthe organic matter accumulated in the sediments and the compositionof the diatom assemblage (Romero et al., 2008) might have played animportant role in Corg preservation in the sediment, although it wasnot the only factor inducing high Corg values. Besides that, high valuesMo/Al and U/Al (Fig. 3E–F) indicate that sub-oxic to anoxic conditionsoccurred at the sediment-water interface during second half of HE1.Water column stratification is very unlikely in this oceanic setting, albeitboth Mo/Al and U/Al profiles evidence reducing conditions at the sea-floor and surface sediments from 16.5 through 15 ka. Additional sup-port to this scenario is provided by the dominance of the benthic fora-minifera Bulimina exilis at GeoB7926 between 17 and 15 ka, which istypical for reduced oxygen conditions at the seafloor (Filipsson et al.,2011). Intense export productivity could have led to the consumptionofmost of the dissolved oxygen in bottomwaters, hence inducing anox-ia at the sediment–water interface and the surface sediments. However,this was not the case as suggested by relatively low Corg AR during thelate HE1 (b0.5 g/cm2/ka) compared with previous intervals. For in-stance, during the interval between 19 ka–17.3 ka, Corg AR increasedup to 1.0 g/cm2/ka (Romero et al., 2008), although the lack of Mo or Uindicates oxic conditions. The opposite is true for the HE1, with no dra-matic increase in Corg AR and highMo/Al andU/Al. Hence, we invoke re-duced ventilation, thus slower deep-water currents, during this period.

The relatively warm B–A (15.5–13.5 ka, Kim et al., 2012) experi-enced an important shift in the composition of the primary producersin surface waters, from the production of siliceous to calcareous micro-organisms (Romero et al., 2008). Corg AR remained constant throughoutthe B–A (Fig. 3), although Corg content clearly fluctuated during this pe-riod, suggesting varying degrees of organic-matter preservation. Mo/Alvalues indicate mostly well oxygenated bottom-water conditions, al-though the “background” values are higher than during the LGM orpresent day (Martinez et al., 1999). On the other hand, U fixation atsite GeoB7926 was relatively high and fluctuated intensively (Fig. 3F).Since U precipitates beforeMo in the transition from oxic to anoxic con-ditions (Tribovillard et al., 2006; Algeo and Tribovillard, 2009), the com-parison of both parameters indicates only reduced oxygen availability atsite GeoB7926 (though not complete depletion) or oxygenation onlylimited to the sediment–water interface since no HS− was present forMo fixation. Accordingly, restricted ventilation occurred during short-duration cold spells during the B–A (Thornalley et al., 2011). We arguethat the resumption of deep-water circulation (and ventilation) was



not complete during the B–A off Mauritania. Thus, the redox boundarywas, at least intermittently, very close to the sediment–water interfaceat site GeoB7926. This scenario is additionally supported by lowMn values (Fig. 3G), which indicate that neither was Mn diffusing(upwards) nor was O2 penetrating deep enough into the sediment.

The YD (13.5–11.5 ka) represented a return to bottom-water condi-tions similar to those recorded during the HE1, though during this timedissolved-oxygen deficiency was more severe than during the HE1.Maxima in trace-element fixation (Fig. 3E and F) (hence very low oxy-gen content at the sea bottom) were also accompanied by high Corg

AR, due to increased diatom productivity (Romero et al., 2008). Addi-tionally, the sharp decrease in carbonate sedimentation (Ca/Al) andthe increase in detrital material (K/Al) at the early YD led to high sedi-mentation rates at site GeoB7926; thus, low Corg content during thisearly phase is possibly influenced by dilution. However, the progressiverise in diatom productivity in surface waters between ~13–12 ka led toan increase in Corg content (Romero et al., 2008) and Corg AR, and re-duced oxygen availability at the sea floor (Filipsson et al., 2011). Thisis further demonstrated bymaximum values of Mo/Al and U/Al indicat-ing minimum bottom-water oxygenation. Mo to U co-variation datapoints corresponding to the YD (Fig. 4, black diamonds) are locatedalong or slightly above the mean seawater Mo/U average, which indi-cates a maximum Mo and U rate of fixation. In fact, Mo enrichment ishigher than the mean Mo/U seawater (above the mean Mo/U line)suggesting direct precipitation from the water column, which in turnimplies water-column anoxia. This scenario is similar to the one de-scribed by Algeo and Tribovillard (2009) for other upwelling areasexhibiting bottom-water anoxia. These geochemical signals provide ev-idence that restricted deep-water circulation was established duringthis cold interval.

In addition to the weakening of deep-water circulation, strongly in-creased diatom productivity in surface waters and rapid sinking ofvalves promoted bottom-water anoxia. Enhanced export productivityimplied intense oxygen consumption, accelerating Mo and U fixationin the sediment (Fig. 3E–F). Additionally, the Mn/Al record (Fig. 3G) in-dicatesMn precipitation typical for oxidation fronts in the sediment col-umn (Froelich et al., 1979; de Lange, 1992), and also indicative of aninterval of oxygen enrichment in bottom waters flowing over the area.

While high productivity occurred in surface waters overlying siteGeoB7926, deep-water ventilationwas temporarily resumed at the sea-floor during the YD. The water depth at this site is around the area ofmaximum Mn precipitation (e.g., de Lange et al., 2008) and thus short

increases of oxygen concentration in bottom waters allowed penetra-tion of the oxidation front, inducing Mn precipitation. At the sametime, high sedimentation rates in the area during this period rapidlyburied this oxic front, “freezing” the geochemical signal (e.g., de Lange,1998). This process can also partially oxidize deposited organicmaterial,explaining the spiky signal of the Corg profile during the YD. Thus, apulse of highly oxygenated deep water flowed over site GeoB7926, al-though a poorly ventilated seafloorwas the normduring the YD. Greatlyenhanced Corg AR also helped to overwhelmed oxygen replenishmenton the seafloor. After this cold phase, deep-water oxic conditions werereestablished off Mauritania, remaining stable throughout the Holocene(Martinez et al., 1999).

Deep-water masses circulation in the coastal northeastern Atlantic

The discussed scenarios of bottom-water oxygenation and ventila-tion conditions are compared with 231 Pa/230Th records (McManuset al., 2004; Gherardi et al., 2005; Lippold et al., 2012; Fig. 5). The sitesused for this comparison are scattered across the North Atlantic(Fig. 1). Although the use of 231Pa/230Th record has been debated forthe North Atlantic eastern margin, Lippold et al. (2012) proved its reli-ability on core GeoB9508-5, located closely to the south of our studiedsite. This comparison allows us to propose five distinct phases ofdeep-water circulation off of Mauritania, described below and summa-rized on Table 2:

1) During the LGM, deep-water circulation must have been constantsince ventilation was constant according to the trace-elements re-cord. Marchal et al. (2000) suggested a decrease of almost 40% inAMOC intensity, although McManus et al. (2004) pointed towardsa weak slowdown. On the other hand, Zhao et al. (2000) proposedan intensified effect of oxygenated AABW flowing northwards. Ithas been proposed that intense brine rejection in the SouthernOcean was the motor for enhanced deep, dense water flow north-wards during the LGM (e.g., Adkins et al., 2002; Labeyrie et al.,2005; Skinner and Shackleton, 2006). Piotrowski et al. (2012) re-cently highlighted the importance of southern-sourced deep watersin the eastern North Atlantic basins during periods of decreasedspilling of NADW from the west into the eastern basins. Our inter-pretation of constant ventilation, oxygen availability and low Cacontent is consistent with northward-flowing and more corrosiveAABW (Lippold et al., 2012). Glacial conditions induced restrictedNorth Atlantic advection, allowing AABW to flow northwards (e.g.,Martrat et al., 2007; Roberts et al., 2012) to fill the gap left byNADW, which was constrained to shallower depths. AlthoughAABW is presently relatively oxygen-poor compared to NADW, theconstant flow must have been enough to maintain oxic bottom-water conditions at this site.

2) The first stage of the deglaciation (~19.0 to 17.5 ka) recorded no no-ticeable change in redox conditions (Fig. 3E–F). Circulation condi-tions at the transition between the LGM and HE1 were very similarto those of the present day (Zhao et al., 2000; McManus et al.,2004) with NADW flowing between 1700 and 2500 m depth(Mittelstaedt, 1991; Gherardi et al., 2005). Thus, the water depthof site GeoB7926 is within this range of NADW circulation.

3) Although the HE1 event was a cold spell, it is unlikely that AABW(presently flowing below 4000 m depth; Sarnthein et al., 1981;Gherardi et al., 2005) shifted shallower than 2500 m depth in such ashort time interval. Even if an AABW northward incursion is evi-denced farther north (Skinner and Shackleton, 2006), sub-oxic condi-tions observed during this period at GeoB7926 (Fig. 3F) are notconsistent with a constant flow over our study area. Instead, de-creased intensity of the North Atlantic circulation cell down to2500 m depth is a more plausible explanation. The collapse of theAMOC between 17.5 and 15 ka (McManus et al., 2004) matcheswith striking precisionwith the top and bottom of theMo/Al increase

Figure 4.Mo to U co-variation for three time intervals relative to average seawater com-position at GeoB7926 (off Mauritania, NW Africa). Average seawater ratio by Algeo andTribovillard (2009). LGM: last glacial maximum; HE1: Heinrich Event 1; B–A: Bolling–Allerod; YD: Younger Dryas.



at GeoB7926 (yellow highlight, Figs. 2E, 4C). Therefore, we infer thatsuboxic–anoxic conditions at this site are due to the AMOC collapsethat weakened or even prevented normal circulation on the seafloor of the coastal NE Atlantic (Table 2) and decreased the bottom-water ventilation rate (Filipsson et al., 2011), in combination withhigh productivity of surface waters (Romero et al., 2008).

4) Warm SST during the B–A period at site GeoB7926 (Kim et al., 2012;Romero et al., 2008) coincides with relatively well oxygenatedbottom-water conditions. The 231 Pa/230Th record shows slightlyhigher values than present day (McManus et al., 2004), pointing to aless efficient AMOC during the B–A than at present. Likewise,Gherardi et al. (2005) proposed slower deep-water circulation duringthe B–A than at present, at both eastern and western margins of theNorth Atlantic. Our results agree with these observations and supportour interpretation of intermittent suboxic bottom waters at siteGeoB7926 (Fig. 5 and Table 2). Warmer conditions during the B–Aoff Mauritania were accompanied by the resumption of NADW circu-lation and enhanced ventilation, though intermittent cold spells(Thornalley et al., 2011) induced less efficient bottom-wateroxygenation.

5) The deep-water low oxygen availability at site GeoB7926 during theYD is linked to the slowdown of the AMOC along theNWAfricanmar-gin. Decreased NADW circulation caused oxygen deficient conditionsthat were enhanced by very intense surface productivity. The slow-down of the AMOC during the YD was less intense than during HE1(McManus et al., 2004; Fig. 5). Roberts et al. (2012) also detected anoceanic-scale decrease in overturning rate based on εNd records. Abenthic foraminifera record from south of Iceland shows that the YD,although globally characterized by a decrease in deep ocean

ventilation (McManus et al., 2004; Gherardi et al., 2005), wasinterrupted by a moderate ventilation event between 12.6 and12.1 ka (Thornalley et al., 2011). This ventilation event correspondswith a Mn oxidation front in between the suboxic phase marked byMo and U. Thus, the signals of AMOC slowdown and resumption dur-ing the YD recorded inGeoB7926 arewell correlatedwith different re-cords across the North Atlantic, supporting our interpretation andextending the reconstruction of AMOC circulation towards the EastAfrican margin.

Conclusions

We highlight the importance of the geochemical record offMauritania for the reconstruction of deep-water paleoceanographyalong the NW African margin, its correlation with other North Atlanticrecords, and its implications for North Atlantic Ocean circulation.Deep-water redox conditions at site GeoB7926 are interpreted by com-bining paleoproductivity and bottom-water ventilation to reconstruct abalance between oxygen consumption due to enhanced export produc-tivity and decreased deep-water circulation.

Two main phases of decreased AMOC ventilation are recognized offMauritania, HE1 and YD, both corresponding to cold spells andfreshwater-input events that prevented normal AMOC (McManus et al.,2004; Gherardi et al., 2005). We confirm that the phases of decreasedAMOC recorded for the western part of the low-latitude Atlantic(McManus et al., 2004) also occurred in the eastern part of the NorthAtlantic, further supporting shifts of North Atlantic circulation in the east-ern basin.

A B C D

Figure 5. Comparison of redox proxies at site GeoB7926 (A), and circulation proxies in thewestern-most Atlantic (B), the central and easternNorth Atlantic (C), and the eastern equatorialAtlantic (D).

Table 2Summary of local and hemispheric-scale deep-water circulation events. (1) Zhao et al. (2000); (2): McManus et al. (2004); 3): Gherardi et al. (2005); (4): Thornalley et al. (2011); (5):Romero et al. (2008).

Period Age (ka) Deep-water ventilation at site GeoB7926 Paleoceanographic setting in the North Atlantic

LGM 25–19 Oxic, well ventilated Restricted AMOC. AABW flow northward up to 2 km water depth (1,2)Early deglaciation 19–17.5 Oxic Reestablished AMOC below 2 kmwater depth (3)HE1 17.5–15.5 Sub-oxic AMOC collapse (2)B–A 15.5–13.5 Oxic to suboxic, slightly reduced circulation AMOC resumption, reduced ventilation events (2,3)YD 13.5–12.6 Suboxic to anoxic AMOC almost collapsed (2)

12.6–12.1 Oxic to suboxic Moderate resumption of deep-water circulation south of Iceland and southward (5)12.1–11.5 Suboxic to anoxic AMOC almost collapsed and intense export productivity in NW-Africa upwelling system (2)



Acknowledgments

We thank funding by Projects CGL2012-32659 and CGL2009-07603(Secretaria de Estado de Investigación), RNM-5212 and the ResearchGroup RNM-179 (Junta de Andalucía). We also thank the Centro deInstrumentación Científica (UniversidaddeGrananda, Spain) for analyt-ical support. We are grateful to Dr. Leah Schneider (IODP-USIO, TexasA&M University, USA) for a helpful review of the manuscript. O.E.R.was partially supported by the Spanish Council of Scientific Research.D.G-T acknowledges the funding from the CSIC JAE-Doc" postdoctoralprogram. Finally, we thank Dr. Marchitto for his effort as AssociateEditor and 3 anonymous reviewers for their helpful comments.

References

Adkins, J.F., McIntyre, K., Schrag, D.P., 2002. The salinity, temperature, and δ18O of the gla-cial deep ocean. Science 298, 1769–1773.

Algeo, T.J., Tribovillard, N., 2009. Environmental analysis of paleoceanographic sys-tems based on molybdenum–uranium covariation. Chemical Geology 268,211–225.

Arthur, M.A., Brumsack, H.-J., Jenkyns, H.C., Schlanger, S.O., 1990. Stratigraphy, geochem-istry, and paleoceanography of organic carbon-rich Cretaceous sequences. In:Ginsburg, R.N., Beauoin, B. (Eds.), Cretaceous Resources, Events and Rhythms. Kluwer,Dordrecht, pp. 75–119.

Bard, E., 1988. Correction of accelerator mass spectrometry 14C ages measured inplanktonic foraminifera: paleoceanographic implications. Paleoceanography 3,635–645.

Bloemsma, M.R., et al., 2012. Modelling the joint variability of grain size and chemicalcomposition in sediments. Sedimentary Geology 280, 135–148.

Brumsack, H.J., 1980. Geochemistry of Cretaceous black shales from the Atlantic Ocean(DSDP Legs 11, 14, 36 and 41). Chemical Geology 31, 1–25.

Calvert, S.E., Pedersen, T.F., 1993. Geochemistry of recent oxic and anoxic marine sedi-ments: implications for the geological record. Marine Geology 113, 67–88.

Calvert, S.E., Pedersen, T.F., 2007. Chapter fourteen elemental proxies for palaeoclimaticand palaeoceanographic variability in marine sediments: interpretation and applica-tion. Developments in Marine Geology, pp. 567–644.

Chapman, M.R., Shackleton, N.J., 1998. Millennial-scale fluctuations in North Atlanticheat flux during the last 150 000 years. Earth and Planetary Science Letters 159,57–70.

Dansgaard, W., Johnsen, S.J., Clausen, H.B., Dahl-Jensen, D., Gundestrup, N.S., Hammer,C.U., Hvidberg, C.S., Steffensen, J.P., Sveinbjörnsdottir, A.E., Jouzel, J., Bond, G., 1993.Evidence for general instability of past climate from a 250-ka ice-core record. Nature364, 218–220.

de Lange, G.J., 1992. Distribution of exchangeable, fixed, organic and total nitrogen in in-terbedded turbiditic/pelagic sediments of the Madeira Abyssal Plain, eastern NorthAtlantic. Marine Geology 109, 95–114.

de Lange, G.J., 1998. Oxic vs. anoxic diagenetic alteration of turbiditic sediments in theMadeira Abyssal Plain, eastern North Atlantic. Proceedings of the Ocean Drilling Pro-gram: Scientific Results 157, 573–580.

de Lange, G.J., Ten Haven, H.L., 1983. Recent sapropel formation in the Eastern Mediterra-nean. Nature 305, 797–798.

de Lange, G.J., Thomson, J., Reitz, A., Slomp, C.P., Principato, M.S., Erba, E., Corselli, C., 2008.Synchronous basin-wide formation and redox-controlled preservation of a Mediter-ranean sapropel. Nature Geoscience 1, 606–610.

de Menocal, P., Ortiz, J., Guilderson, T., Adkins, J., Sarnthein, M., Baker, L., Yarusinsky,M., 2000a. Abrupt onset and termination of the African Humid Period: rapid cli-mate responses to gradual insolation forcing. Quaternary Science Reviews 19,347–361.

deMenocal, P., Ortiz, J., Guilderson, T., Sarnthein, M., 2000b. Coherent high- and low-latitudeclimate variability during the Holocene warm period. Science 288, 2198–2202.

Eberwein, A., Mackensen, A., 2008. Last Glacial Maximum paleoproductivity and watermasses off NW-Africa: evidence from benthic foraminifera and stable isotopes. Ma-rine Micropaleontology 67, 87–103.

Filipsson, H.L., Romero, O.E., Jan-Berend, W.S., Donner, B., 2011. Relationships betweenprimary productivity and bottom-water oxygenation off northwest Africa duringthe last deglaciation. Journal of Quaternary Science 26, 448–456.

Froelich, P.N., Klinkhammer, G.P., Bender, M.L., Luedtke, N.A., Heath, G.R., Cullen, D.,Dauphin, P., Hammond, D., Hartman, B., Maynard, V., 1979. Early oxidation of organicmatter in pelagic sediments of the eastern equatorial Atlantic: suboxic diagenesis.Geochimica et Cosmochimica Acta 43, 1075–1090.

Gallego-Torres,D.,Martinez-Ruiz, F., Paytan, A., Jimenez-Espejo, F.J., Ortega-Huertas,M., 2007.Pliocene–Holocene evolution of depositional conditions in the eastern Mediterranean:role of anoxia vs. productivity at time of sapropel deposition. Palaeogeography,Palaeoclimatology, Palaeoecology 246, 424–439.

Gallego-Torres, D., Martinez-Ruiz, F., De Lange, G., Jimenez-Espejo, F.J., Ortega-Huertas,M., 2010. Trace-elemental derived paleoceanographic and paleoclimatic conditionsfor Pleistocene EasternMediterranean sapropels. Palaeogeography PalaeoclimatologyPalaeoecology 293, 76–89.

Gherardi, J.M., Labeyrie, L., McManus, J.F., Francois, R., Skinner, L.C., Cortijo, E., 2005. Evi-dence from the Northeastern Atlantic basin for variability in the rate of the

meridional overturning circulation through the last deglaciation. Earth and PlanetaryScience Letters 240, 710–723.

Grootes, P.M., Stuiver, M., 1997. Oxygen 18/16 variability in Greenland snow and ice with10-3- to 105-year time resolution. Journal of Geophysical Research 102,26455–26470.

Hall, I.R., Evans, H.K., Thornalley, D.J.R., 2011. Deep water flow speed and surface oceanchanges in the subtropical North Atlantic during the last deglaciation. Global andPlanetary Change 79, 255–263.

Haslett, S.K., Davies, C.F.C., 2006. Late Quaternary climate-ocean changes inwestern NorthAfrica: offshore geochemical evidence. Transactions of the Institute of British Geogra-phers 31, 34–52.

Jones, B., Manning, D.A.C., 1994. Comparison of geochemical indices used for the interpre-tation of palaeoredox conditions in ancient mudstones. Chemical Geology 111,111–129.

Kim, J.H., Meggers, H., Rimbu, N., Lohmann, G., Freudnethal, T., Müller, P.J., Schneider, R.R.,2007. Impacts of the North Atlantic gyre circulation on Holocene climate off north-west Africa. Geology 35, 387–390.

Kim, J.H., Romero, O.E., Lohmann, G., Donner, B., Laepple, T., Haam, E., Sinninghe Damste,J.S., 2012. Pronounced subsurface cooling of North Atlantic waters off NorthwestAfrica during Dansgaard–Oeschger interstadials. Earth and Planetary Science Letters339–340, 95–102.

Kuhlmann, H., Meggers, H., Freudenthal, T., Wefer, G., 2004. The transition of the mon-soonal and the N Atlantic climate system off NW Africa during the Holocene. Geo-physical Research Letters 31, 1–4.

Labeyrie, L., Waelbroeck, C., Cortijo, E., Michel, E., Duplessy, J.C., 2005. Changes in deepwater hydrology during the last deglaciation. Changements de l'hydrologie profondependant la dernière déglaciation 337, 919–927.

Lippold, J., Mulitza, S., Mollenhauer, G., Weyer, S., Heslop, D., Christl, M., 2012.Boundary scavenging at the East Atlantic margin does not negate use of231 Pa/230Th to trace Atlantic overturning. Earth and Planetary Science Letters333–334, 317–331.

Lynch-Stieglitz, J., Adkins, J.F., Curry, W.B., Dokken, T., Hall, I.R., Herguera, J.C., Hirschi,J.J.M., Ivanova, E.V., Kissel, C., Marchal, O., Marchitto, T.M., McCave, I.N., McManus,J.F., Mulitza, S., Ninnemann, U., Peeters, F., Yu, E.F., Zahn, R., 2007. Atlantic merid-ional overturning circulation during the last glacial maximum. Science 316,66–69.

Marchal, O., François, R., Stocker, T.F., Joos, F., 2000. Ocean thermohaline circulation andsedimentary 231PA/230Th ratio. Paleoceanography 15, 625–641.

Martinez, P., Bertrand, P., Shimmield, G.B., Cochrane, K., Jorissen, F.J., Foster, J., Dignan,M., 1999. Upwelling intensity and ocean productivity changes off Cape Blanc(northwest Africa) during the last 70,000 years: geochemical andmicropalaeontological evidence. Marine Geology 158, 57–74.

Martrat, B., et al., 2007. Four climate cycles of recurring deep and surface water destabili-zations on the Iberian margin. Science 317 (5837), 502–507.

McManus, J.F., Francois, R., Gherardl, J.M., Kelgwin, L., Drown-Leger, S., 2004. Collapse andrapid resumption of Atlantic meridional circulation linked to deglacial climate chang-es. Nature 428, 834–837.

McManus, J., Berelson, W.M., Klinkhammer, G.P., Hammond, D.E., Holm, C., 2005.Authigenic uranium: relationship to oxygen penetration depth and organic carbonrain. Geochimica et Cosmochimica Acta 69, 95–108.

Mittelstaedt, E., 1991. The ocean boundary along the northwest African coast: circulationand oceanographic properties at the sea surface. Progress in Oceanography 26,307–355.

Morford, J.L., Emerson, S.R., Breckel, E.J., Kim, S.H., 2005. Diagenesis of oxyanions (V, U, Re,and Mo) in pore waters and sediments from a continental margin. Geochimica etCosmochimica Acta 69, 5021–5032.

Morford, J.L., Martin, W.R., François, R., Carney, C.M., 2009. Amodel for uranium, rhenium,andmolybdenum diagenesis in marine sediments based on results from coastal loca-tions. Geochimica et Cosmochimica Acta 73, 2938–2960.

Mulitza, S., et al., 2008. Sahel megadroughts triggered by glacial slowdowns of Atlanticmeridional overturning. Paleoceanography 23 (4).

Piotrowski, A.M., Galy, A., Nicholl, J.A.L., Roberts, N., Wilson, D.J., Clegg, J.A., Yu, J.,2012. Reconstructing deglacial North and South Atlantic deep water sourcingusing foraminiferal Nd isotopes. Earth and Planetary Science Letters 357–358,289–297.

Piper, D.Z., Calvert, S.E., 2009. A marine biogeochemical perspective on black shale depo-sition. Earth-Science Reviews 95, 63–96.

Praetorius, S.K., McManus, J.F., Oppo, D.W., Curry, W.B., 2008. Episodic reduc-tions in bottom-water currents since the last ice age. Nature Geoscience1, 449–452.

Rahmstorf, S., 2002. Ocean circulation and climate during the past 120,000 years. Nature419, 207–214.

Roberts, N.L., Piotrowski, A.M., McManus, J.F., Keigwin, L.D., 2012. Synchronous deglacialoverturning and water mass source changes. Science 327, 75–78.

Romero, O.E., Kim, J.H., Donner, B., 2008. Submillennial-to-millennial variability of diatomproduction offMauritania, NWAfrica, during the last glacial cycle. Paleoceanography 23.

Rutberg, R.L., Hemming, S.R., Goldstein, S.L., 2000. Reduced North Atlantic Deep Waterflux to the glacial Southern Ocean inferred from neodymium isotope ratios. Nature405, 935–938.

Sarnthein, M., Tetzlaff, G., Koopmann, B., Wolter, K., Pflaumann, U., 1981. Glacial and in-terglacial wind regimes over the eastern subtropical Atlantic and North-WestAfrica. Nature 293, 193–196.

Schmidt, B.C., Lynch-Stieglitz, J., 2011. Florida Straits deglacial temperature and salinitychange: implications for tropical hydrologic cycle variability. Paleoceanography 26,PA4205.



Skinner, L.C., Shackleton, N.J., 2006. Deconstructing terminations I and II: revisiting theglacioeustatic paradigmbased on deep-water temperature estimates. Quaternary Sci-ence Reviews 25, 3312–3321.

Staubwasser, M., Sirocko, F., Grootes, P.M., Erlenkeuser, H., 2002. South Asian monsoonclimate change and radiocarbon in the Arabian Sea during early and middle Holo-cene. Paleoceanography 17, 15-1.

Stuiver, M., Reimer, P.J., Bard, E., Beck, J.W., Burr, G.S., Hughen, K.A., Kromer, B., McCormac,G., Van Der Plicht, J., Spurk, M., 1998. INTCAL98 radiocarbon age calibration, 24,000-0 cal BP. Radiocarbon 40, 1041–1083.

Thomson, J., Higgs, N.C., Croudace, I.W., Colley, S., Hydes, D.J., 1993. Redox zonation of el-ements at an oxic/postoxic boundary in deep-sea sediments. Geochemical andComochimical Acta 57, 579–595.

Thomson, J., Higgs, N.C., Wilson, T.R.S., Croudace, I.W., de Lange, G.J., van Santvoort, P.J.M.,1995. Redistribution and geochemical behaviour of redox-sensitive elements aroundS1, the most recent eastern Mediterranean sapropel. Geochimica et CosmochimicaActa 59, 3487–3501.

Thornalley, D.J.R., Elderfield, H., McCave, I.N., 2011. Reconstructing North Atlanticdeglacial surface hydrography and its link to the Atlantic overturning circulation.Global and Planetary Change 79, 163–175.

Tjallingii, R., et al., 2008. Coherent high- and low-latitude control of the northwest Africanhydrological balance. Nature Geoscience 1 (10), 670–675.

Tribovillard, N., Algeo, T.J., Lyons, T., Riboulleau, A., 2006. Trace metals as paleoredox andpaleoproductivity proxies: an update. Chemical Geology 232, 12–32.

van Der Weijden, C.H., 2002. Pitfalls of normalization of marine geochemical data using acommon divisor. Marine Geology 184, 167–187.

van Santvoort, P.J.M., de Lange, G.J., Thomson, J., Cussen, H.,Wilson, T.R.S., Krom,M.D., Strohle,K., 1996. Active post-depositional oxidation of the most recent sapropel (S1) in sedi-ments of the eastern Mediterranean Sea. Geochimica et Cosmochimica Acta 60,4007–4024.

Zhao, M., Eglinton, G., Haslett, S.K., Jordan, R.W., Sarnthein, M., Zhang, Z., 2000. Marineand terrestrial biomarker records for the last 35,000 years at ODP site 658C off NWAfrica. Organic Geochemistry 31, 919–930.


Documents

Author's personal copygrupo179/pdf/Gallego Torres 2014.pdf · 2014-03-26 · Author's personal copy Rapid bottom-water circulation changes during the last glacial cycle in the coastal