Late Quaternary deep sea sedimentation off the Niger delta

The oxygen isotopes ratios of benthic foraminifera and detailed radiocarbon ages of the organic matter of an over 15 m long sediment core from the outer Niger delta allow us to date the oxygen isotope stage boundaries 1/2 to 11500 (+/- 650) years BP, 2/3 to approximately 23000 (+/- 2000) years BP. The composition of the predominantly terrigenous clays and accessory pelagic fossils reflects the evolution of the climate over the southwestern Sahel zone and the response of the Eastern Tropical Atlantic to these climatic fluctuations during the Late Quaternary. The dilution of the pelagic fossil concentrations by the terrigenous material and the oxygen isotopes ratios of planktonic foraminifera indicate large fluctuations in the freshwater discharge from the Niger, with high precipitations over the drainage area of this river from 4500 (+/- 300) to 11500 (+/- 650) years BP and from 11800 (+(- 600) to 13000 (+/- 600) years BP while the time intervals in between were as dry as today. Relative increase of kaolinite during wet phases and the association of smectite, chlorite and attapulgite during dry ones characterize the response of the weathering in the Niger drainage basins to the climatic fluctuations. The occurrence of 10-14 A mixed-layers prior to 26000 years BP is correlated with moderate alteration of the crystalline substratum outcrops from the middle-lower part of the Niger Basin. High quartz concentrations are particularly typical for the transition between oxygen isotope stages 1 and 2 at the inception of heavy precipitations in the southern Sahel zone. Sedimentation rates were quite constant, 30-35 cm/1000 years; they became unusually large at the beginning of the Holocene from 10900 (+/- 650) to 11500 (+/- 650) years BP where they reached more than 600 cm/1000 years. Bottom waters around 1100 m depth in the Gulf of Guinea responded to changes in paleo-oceanography of the entire Atlantic Ocean as well as to local influences. Abnormal carbon isotopes ratios and the drastic changes from a highly diversified fauna (during stages 2 and 3. and during the last part of stage 1 after approx. 7000 years BP) to a poorly diversified fauna in the intervenin time span point to the development of a local benthic environment which cannot easily be compared with the corresponding continental and slope environments of the entire Atlantic Ocean.

Strontium isotope analysis of seawater and at a mixture of planktic foraminifera during the Paleocene and Early Eocene from different holes of ODP Leg 2008

We refined the strontium isotope seawater curve for the Paleocene and early Eocene by analysis of samples recovered from the Walvis Ridge during Ocean Drilling Project (ODP) Leg 208. The highest 87Sr/86Sr values occurred in the earliest Paleocene at 65 Ma and generally decreased throughout the Paleocene, reaching minimum values between 53 and 51 Ma in the early Eocene before beginning to increase again at 50 Ma. A plausible explanation for the 87Sr/86Sr decrease between 65 and 51 Ma is increased rates of hydrothermal activity and/or the eruption and weathering of large igneous provinces (e.g., Deccan Traps and North Atlantic). Strontium isotope variations closely parallel sea level and benthic d18O changes during the late Paleocene and early Eocene, supporting previous studies linking tectonic reorganization and increased volcanism to high sea level, high CO2, and warm global temperatures.

(Appendix) Stable carbon and oxygen isotopes, and Mg/Ca ratios of foraminifera from ODP Hole 154-925A

To better understand the links between the carbon cycle and changes in past climate over tectonic timescales we need new geochemical proxy records of secular change in silicate weathering rates. A number of proxies are under development, but some of the most promising (e.g. palaeoseawater records of Li and Nd isotope change) can only be employed on such large samples of mono-specific foraminifera that application to the deep sea sediment core archive becomes highly problematic. "Dentoglobigerina" venezuelana presents a potentially attractive target for circumventing this problem because it is a typically large (> 355 ?m diameter), abundant and cosmopolitan planktic foraminifer that ranges from the early Oligocene to early Pliocene. Yet considerable taxonomic and ecological uncertainties associated with this taxon must first be addressed. Here, we assess the taxonomy, palaeoecology, and ontogeny of "D." venezuelana using stable isotope (oxygen and carbon) and Mg/Ca data measured in tests of late Oligocene to early Miocene age from Ocean Drilling Program (ODP) Site 925, on Ceara Rise, in the western equatorial Atlantic. To help constrain the depth habitat of "D." venezuelana relative to other species we report the stable isotope composition of selected planktic foraminifera species within Globigerina, Globigerinoides, Paragloborotalia and Catapsydrax. We define three morphotypes of "D." venezuelana based on the morphology of the final chamber and aperture architecture. We determine the trace element and stable isotope composition of each morphotype for different size fractions, to test the validity of pooling these morphotypes for the purposes of generating geochemical proxy datasets and to assess any ontogenetic variations in depth habitat. Our data indicate that "D." venezuelana maintains a lower thermocline depth habitat at Ceara Rise between 24 and 21 Ma. Comparing our results to published datasets we conclude that this lower thermocline depth ecology for the Oligo-Miocene is part of an Eocene-to-Pliocene evolution of depth habitat from surface to sub-thermocline for "D." venezuelana. Our size fraction data advocate the absence of photosymbionts in "D." venezuelana and suggest that juveniles calcify higher in the water column, descending into slightly deeper water during the later stages of its life cycle. Our morphotype data show that d18O and d13C variation between morphotypes is no greater than within-morphotype variability. This finding will permit future pooling of morphotypes in the generation of the "sample hungry" palaeoceanographic records.

(Table 1) Radiolarian accumulation rates and geochemistry of ODP Site 184-1143

Late Miocene-Recent micropaleontological and geochemical records from Ocean Drilling Program (ODP) Site 1143 in the southern South China Sea (SCS) indicate that increase and decrease in abundance of siliceous plankton may be controlled mainly by the input of nutrients derived from land and provided by upwelling. A high export production event - a "biogenic bloom" event - occurred in the southern SCS between 12 and 6 Ma. During this period, high ratios of smectite/(illite + chlorite), smectite/quartz and Al/K indicate a high weathering intensity of the Asian continent, possibly due to the intensification of the East Asian Summer Monsoon (EASM), which may have increased the net flux of nutrients to the ocean, both directly through terrestrial input and indirectly through upwelling activity. A drop in Ba/Ti, Al/Ti and Ca/Ti values around 6 Ma may indicate a lowering of productivity, possibly due to the large consumption of sea surface nutrients by the "biogenic bloom". Alternatively, it may indicate a shift in terrigenous input source area. At about 5.4 Ma, a decrease in weathering intensity, as indicated by a sudden decrease in the values of smectite/(illite + chlorite), smectite/quartz and Al/K, might have led to a sudden decrease of terrestrial nutrient input to the SCS. We suggest that the biogenic bloom ended when nutrients in surface waters were exhausted, because of a decrease in supply as well as a decrease in upwelling intensity due to weakening of the EASM. As a result, radiolarians were absent in the studied area between ~6 and 3.2 Ma. At ~3.2 Ma, radiolarians began to recover, possibly because the start of Northern Hemispheric glaciation and the rapid uplift of the Tibet Plateau led to intensification of the East Asian monsoon. After the Mid-Pleistocene Climate Transition at 0.9 Ma, the abundance and mass accumulation rates of radiolarians increased, probably as a result of increased upwelling activity driven by the increasing intensity of the summer monsoon.

Microbialites and lipid biomarker from coral reefs off the volcanic islands Tahiti and Vanuatu and from the nonvolcanic sites Belize and the Maldives

The occurrence of microbialites in post-glacial coral reefs has been interpreted to reflect an ecosystem response to environmental change. The greater thickness of microbialites in reefs with a volcanic hinterland compared to thinner microbial crusts in reefs with a non-volcanic hinterland led to the suggestion that fertilization of the reefal environment by chemical weathering of volcanic rocks stimulated primary productivity and microbialite formation. Using a molecular and isotopic approach on reef-microbialites from Tahiti (Pacific Ocean), it was recently shown that sulfate-reducing bacteria favored the formation of microbial carbonates. To test if similar mechanisms induced microbialite formation in other reefs as well, the Tahitian microbialites are compared with similar microbialites from coral reefs off Vanuatu (Pacific Ocean), Belize (Caribbean Sea, Atlantic Ocean), and the Maldives (Indian Ocean) in this study. The selected study sites cover a wide range of geological settings, reflecting variable input and composition of detritus. The new lipid biomarker data and stable sulfur isotope results confirm that sulfate-reducing bacteria played an intrinsic role in the precipitation of microbial carbonate at all study sites, irrespective of the geological setting. Abundant biomarkers indicative of sulfate reducers include a variety of terminally-branched and mid chain-branched fatty acids as well as mono-O-alkyl glycerol ethers. Isotope evidence for bacterial sulfate reduction is represented by low d34S values of pyrite (-43 to -42 per mill) enclosed in the microbialites and, compared to seawater sulfate, slightly elevated d34S and d18O values of carbonate-associated sulfate (21.9 to 22.2 per mill and 11.3 to 12.4 per mill, respectively). Microbialite formation took place in anoxic micro-environments, which presumably developed through the fertilization of the reef environment and the resultant accumulation of organic matter including bacterial extracellular polymeric substances (EPS), coral mucus, and marine snow in cavities within the coral framework. ToF-SIMS analysis reveals that the dark layers of laminated microbialites are enriched in carbohydrates, which are common constituents of EPS and coral mucus. These results support the hypothesis that bacterial degradation of EPS and coral mucus within microbial mats favored carbonate precipitation. Because reefal microbialites formed by similar processes in very different geological settings, this comparative study suggests that a volcanic hinterland is not required for microbialite growth. Yet, detrital input derived from the weathering of volcanic rocks appears to be a natural fertilizer, being conductive for the growth of microbial mats, which fosters the development of particularly abundant and thick microbial crusts.

Extractions and analyses of reactive phosphorus in the Bering Sea from IODP Holes 323-U1341A and 323-U1341B

To reconstruct the cycling of reactive phosphorus (P) in the Bering Sea, a P speciation record covering the last ~ 4 Ma was generated from sediments recovered during Integrated Ocean Drilling Program (IODP) Expedition 323 at Site U1341 (Bowers Ridge). A chemical extraction procedure distinguishing between different operationally defined P fractions provides new insight into reactive P input, burial and diagenetic transformations. Reactive P mass accumulation rates (MARs) are ~ 20-110 µmol/cm2/ka, which is comparable to other open ocean locations but orders of magnitude lower than most upwelling settings. We find that authigenic carbonate fluorapatite (CFA) and opal-bound P are the dominant P fractions at Site U1341. An overall increasing contribution of CFA to total P with sediment depth is consistent with a gradual "sink switching" from more labile P fractions (fish remains, Fe oxides, organic matter) to stable authigenic CFA. However, the positive correlation of CFA with Al content implies that a significant portion of the supposedly reactive CFA is non-reactive "detrital contamination" by eolian and/or riverine CFA. In contrast to CFA, opal-bound P has rarely been studied in marine sediments. We find for the first time that opal-bound P directly correlates with excess silica contents. This P fraction was apparently available to biosiliceous phytoplankton at the time of sediment deposition and is a long-term sink for reactive P in the ocean, despite the likelihood for diagenetic re-mobilisation of this P at depth (indicated by increasing ratios of excess silica to opal-bound P). Average reactive P MARs at Site U1341 increase by ~ 25% if opal-bound P is accounted for, but decrease by ~ 25% if 50% of the extracted CFA fraction (based on the lowest CFA value at Site U1341) is assumed to be detrital. Combining our results with literature data, we present a qualitative perspective of terrestrial CFA and opal-bound P deposition in the modern ocean. Riverine CFA input has mostly been reported from continental shelves and margins draining P-rich lithologies, while eolian CFA input is found across wide ocean regions underlying the Northern Hemispheric "dust belt". Opal-bound P burial is important in the Southern Ocean, North Pacific, and likely in upwelling areas. Shifts in detrital CFA and opal-bound P deposition across ocean basins likely occurred over time, responding to changing weathering patterns, sea level, and biogenic opal deposition.

(Table 1) Types of giant magnetofossils in ODP holes

Magnetotactic bacteria biomineralize magnetic minerals with precisely controlled size, morphology, and stoichiometry. These cosmopolitan bacteria are widely observed in aquatic environments. If preserved after burial, the inorganic remains of magnetotactic bacteria act as magnetofossils that record ancient geomagnetic field variations. They also have potential to provide paleoenvironmental information. In contrast to conventional magnetofossils, giant magnetofossils (most likely produced by eukaryotic organisms) have only been reported once before from Paleocene-Eocene Thermal Maximum (PETM; 55.8 Ma) sediments on the New Jersey coastal plain. Here, using transmission electron microscopic observations, we present evidence for abundant giant magnetofossils, including previously reported elongated prisms and spindles, and new giant bullet-shaped magnetite crystals, in the Southern Ocean near Antarctica, not only during the PETM, but also shortly before and after the PETM. Moreover, we have discovered giant bullet-shaped magnetite crystals from the equatorial Indian Ocean during the Mid-Eocene Climatic Optimum (~40 Ma). Our results indicate a more widespread geographic, environmental, and temporal distribution of giant magnetofossils in the geological record with a link to "hyperthermal" events. Enhanced global weathering during hyperthermals, and expanded suboxic diagenetic environments, probably provided more bioavailable iron that enabled biomineralization of giant magnetofossils. Our micromagnetic modelling indicates the presence of magnetic multi-domain (i.e., not ideal for navigation) and single domain (i.e., ideal for navigation) structures in the giant magnetite particles depending on their size, morphology and spatial arrangement. Different giant magnetite crystal morphologies appear to have had different biological functions, including magnetotaxis and other non-navigational purposes. Our observations suggest that hyperthermals provided ideal conditions for giant magnetofossils, and that these organisms were globally distributed. Much more work is needed to understand the interplay between magnetofossil morphology, climate, nutrient availability, and environmental variability.

Concentrations of brominated and chlorinated contaminants in adipose tissue of polar bears (Ursus maritimus) between 1991-2007

Adipose tissue was sampled from the western Hudson Bay (WHB) subpopulation of polar bears at intervals from 1991 to 2007 to examine temporal trends of PCB and OCP levels both on an individual and sum-contaminant basis. We also determined levels and temporal trends of emerging polybrominated diphenyl ethers (PBDEs), hexabromocyclododecane (HBCD), polybrominated biphenyls (PBBs) and other current-use brominated flame retardants. Over the 17-year period, sum DDT (and p,p'-DDE, p,p'-DDD, p,p'-DDT) decreased (-8.4%/year); alpha-hexachlorocyclohexane (alpha-HCH) decreased (-11%/year); beta-HCH increased ( + 8.3%/year); and sum PCB and sum chlordane (CHL), both contaminants at highest concentrations in all years (>1 ppm), showed no distinct trends even when compared to previous data for this subpopulation dating back to 1968. Some of the less persistent PCB congeners decreased significantly (-1.6%/year to -6.3%/year), whereas CB153 levels tended to increase (+ 3.3%/year). Parent CHLs (c-nonachlor, t-nonachlor) declined, whereas non-monotonic trends were detected for metabolites (heptachlor epoxide, oxychlordane). sum chlorobenzene, octachlorostyrene, sum mirex, sum MeSO2-PCB and dieldrin did not significantly change. Increasing sum PBDE levels (+13%/year) matched increases in the four consistently detected congeners, BDE47, BDE99, BDE100 and BDE153. Although no trend was observed, total-(alpha)-HBCD was only detected post-2000. Levels of the highest concentration brominated contaminant, BB153, showed no temporal change. As long-term ecosystem changes affecting contaminant levels may also affect contaminant patterns, we examined the influence of year (i.e., aging or "weathering" of the contaminant pattern), dietary tracers (carbon stable isotope ratios, fatty acid patterns) and biological (age/sex) group on congener/metabolite profiles. Patterns of PCBs, CHLs and PBDEs were correlated with dietary tracers and biological group, but only PCB and CHL patterns were correlated with year. DDT patterns were not associated with any explanatory variables, possibly related to local DDT sources. Contaminant pattern trends may be useful in distinguishing the possible role of ecological/diet changes on contaminant burdens from expected dynamics due to atmospheric sources and weathering.

Geochemical and stable isotope composition of Middle Campanian marl-limestone rhythmites of the Lehrte West Syncline near Hannover (Lower Saxony, northern Germany)

A cyclic marl-limestone succession of Middle-Late Campanian age has been investigated with respect to a Milankovitch-controlled origin of geochemical data. In general, the major element geochemistry of the marl-limestone rhythmites can be explained by a simple two-component mixing model with the end-members calcium carbonate and 'average shale'-like material. Carbonate content varies from 55 to 90%. Non-carbonate components are clay minerals (illite, smectite) and biogenic silica from sponge spicules, as well as authigenically formed zeolites (strontian heulandite) and quartz. The redox potential suggests oxidizing conditions throughout the section. Trace element and stable isotopic data as well as SEM investigations show that the carbonate mud is mostly composed of low-magnesium calcitic tests of planktic coccolithophorids and calcareous dinoflagellate cysts (calcispheres). Diagenetic overprint results in a decrease of 2% d18O and an increase in Mn of up to 250 ppm. However, the sediment seems to preserve most of its high Sr content compared to the primary low-magnesium calcite of co-occurring belemnite rostra. The periodicity of geochemical cycles is dominated by 413 ka and weak signals between 51 and 22.5 ka, attributable to orbital forcing. Accumulation rates within these cycles vary between 40 and 50 m/Ma. The resulting cyclic sedimentary sequence is the product of (a) changes in primary production of low-magnesium calcitic biogenic material in surface waters within the long eccentricity and the precession, demonstrated by the CaCO3 content and the Mg/Al, Mn/Al and Sr/Al ratios, and (b) fluctuations in climate and continental weathering, which changed the quality of supplied clay minerals (the illite/smectite ratio), demonstrated by the K/Al ratio. High carbonate productivity correlates with smectite-favouring weathering (semi-arid conditions, conspicuously dry and moist seasonal changes in warmer climates). Ti as the proxy indicator for the detrital terrigenous influx, as well as Rb, Si, Zr and Na, shows only low frequency signals, indicating nearly constant rates of supply throughout the more or less pure pelagic carbonate deposition of the long-lasting third-order Middle-Upper Campanian sedimentary cycle.

Mineralogical (bulk rock and clay fraction) and stable isotope composition (d13C, d18O) of Aptian deposits of the Bir Oum Ali succession, central Tunisia

Alternations between siliciclastic, carbonate and evaporitic sedimentary systems, as recorded in the Aptian mixed succession of southern Tunisia, reflect profound palaeoceanographic and palaeoclimatic changes in this area of the southern Tethyan margin. The evolution from Urgonian-type carbonates (Berrani Formation, lower Aptian) at the base of the series, to intervals dominated by gypsum or detrital deposits in the remainder of the Aptian is thought to result from the interplay between climate change and tectonic activity that affected North Africa. Based on the evolution of clay mineral assemblages, the early Aptian is interpreted as having been dominated by slightly humid conditions, since smectitic minerals are observed. Near the early to late Aptian boundary, the onset of a gypsiferous sedimentation is associated with the appearance of palygorskite and sepiolite, which supports the installation of arid conditions in this area of the southern Tethyan margin. The evaporitic sedimentation may have also been promoted by the peculiar tectonic setting of the Bir Oum Ali area during the Aptian, where local subsidence may have been tectonically enhanced linked to the opening of northern and central Atlantic. Stress associated with the west and central African rift systems may have triggered the development of NW-SE, hemi-graben structures. Uplifted areas may have constituted potential new sources for clastic material that has been subsequently deposited during the late Aptian. Chemostratigraphic (d13C) correlation of the Bir Oum Ali succession with other peri-Tethyan regions complements biostratigraphic findings, and indicates that a potential expression of the Oceanic Anoxic Event (OAE) 1a may be preserved in this area of Tunisia. Although the characteristic negative spike at the base of this event is not recognized in the present study, a subsequent, large positive excursion with d13C values is of similar amplitude and absolute values to that reported from other peri-Tethyan regions, thus supporting the identification of isotopic segments C4-C7 of the OAE1a. The absence of the negative spike may be linked to either non preservation or non deposition: the OAE1a occurred in a global transgressive context, and since the Bir Oum Ali region was located in the innermost part of the southern Tethyan margin during most of the Aptian, stratigraphic hiatuses may have been longer than in other regions of the Tethys. This emphasizes the importance of integrating several stratigraphic disciplines (bio-, chemo- and sequence stratigraphy) when performing long-distance correlation.

(Table 1) Nd isotope ratios of ODP Leg 208 samples

We present new Nd isotope records from Walvis Ridge Ocean Drilling Program (ODP) sites 1262-1264 (southeastern Atlantic) spanning the past 24 Ma to investigate the Neogene evolution of Atlantic thermohaline circulation. The new data indicate that deepwater epsilon-Nd(t) values from ODP Site 1262 decrease from -11.0 at 10.6 Ma to -12.5 by 7.3 Ma. This decrease parallels the Nd isotope trends contained in Fe-Mn crust records from the northwestern Atlantic; however, the shift at ODP Site 1262 (4755 m water depth) occurred up to ~6 Ma earlier than the corresponding changes in crust records from the Atlantic and earlier than any Atlantic site shallower than 2700 m paleowater depth. Recent interpretations of the rapid decrease in Fe-Mn crust Nd epsilon-Nd(t)ss values invoke changes in weathering inputs to the Labrador Sea region rather than a fundamental change in deepwater convection in the Labrador Sea. However, the new evidence for significant depth stratification of the Nd isotope signal in the southeastern Atlantic between 10.6 and 7.3 Ma suggests that the onset of deepwater convection in the Labrador Sea may have played a role in the deepwater decrease in Nd isotopic composition. Climatic conditions during the middle to late Miocene likely favored an increase in the importance of glacially induced mechanical weathering, while at the same time promoting deep convection in the Labrador Sea.

(Table DR1) Argon isotope data and dating for individual >63 µm amphibole grains from Labrador Sea sediments, ODP Hole 105-647A

Variations of global and regional silicate weathering rates and paleo-ocean circulation patterns are estimated by using radiogenic isotope records, but the effects of changes in provenance are generally ignored. Here sediment provenance has been constrained through the use of Ar-Ar ages for individual detrital minerals from the Labrador Sea, which can be compared directly to the radiogenic isotope compositions from the same core material. Dramatic changes in the radiogenic isotope composition of North Atlantic Deep Water through the Quaternary Period are shown to reflect discrete changes in both sources and weathering processes accompanying Northern Hemisphere glaciation. Changes in the different radiogenic isotope systems reflect the influence of source, physical weathering, and chemical weathering, and not simply changes in the underlying weathering rate or ocean circulation patterns that are typically inferred.

Alkenone stable carbon isotopes, carbonate stable carbon and oxygen isotopes, planktic foraminifera boron isotopes and atmospheric CO2 calculations and box model results from Ras il-Pellegrin section, Malta

The development of a permanent, stable ice sheet in East Antarctica happened during the middle Miocene, about 14 million years (Myr) ago. The middle Miocene therefore represents one of the distinct phases of rapid change in the transition from the "greenhouse" of the early Eocene to the "icehouse" of the present day. Carbonate carbon isotope records of the period immediately following the main stage of ice sheet development reveal a major perturbation in the carbon system, represented by the positive d13C excursion known as carbon maximum 6 ("M6"), which has traditionally been interpreted as reflecting increased burial of organic matter and atmospheric pCO2 drawdown. More recently, it has been suggested that the d13C excursion records a negative feedback resulting from the reduction of silicate weathering and an increase in atmospheric pCO2. Here we present high-resolution multi-proxy (alkenone carbon and foraminiferal boron isotope) records of atmospheric carbon dioxide and sea surface temperature across CM6. Similar to previously published records spanning this interval, our records document a world of generally low (~300 ppm) atmospheric pCO2 at a time generally accepted to be much warmer than today. Crucially, they also reveal a pCO2 decrease with associated cooling, which demonstrates that the carbon burial hypothesis for CM6 is feasible and could have acted as a positive feedback on global cooling.

Geochemistry and geomicrobiology profiles of ferruginous sediment, Lake Towuti, Indonesia

Lake Towuti is a tectonic basin, surrounded by ultramafic rocks. Lateritic soils form through weathering and deliver abundant iron (oxy)hydroxides but very little sulfate to the lake and its sediment. To characterize the sediment biogeochemistry, we collected cores at three sites with increasing water depth and decreasing bottom water oxygen concentrations. Microbial cell densities were highest at the shallow site - a feature we attribute to the availability of labile organic matter and the higher abundance of electron acceptors due to oxic bottom water conditions. At the two other sites, OM degradation and reduction processes below the oxycline led to partial electron acceptor depletion. Genetic information preserved in the sediment as extracellular DNA provides information on aerobic and anaerobic heterotrophs related to Actinobacteria, Nitrospirae, Chloroflexi and Thermoplasmatales. These taxa apparently played a significant role in the degradation of sinking organic matter. However, extracellular DNA concentrations rapidly decrease with core depth. Despite very low sulfate concentrations, sulfate-reducing bacteria were present and viable in sediments at all three sites, as confirmed by measurement of potential sulfate reduction rates. Microbial community fingerprinting supported the presence of taxa related to Deltaproteobacteria and Firmicutes with demonstrated capacity for iron and sulfate reduction. Concomitantly, sequences of Ruminococcaceae, Clostridiales and Methanomicrobiales indicated potential for fermentative hydrogen and methane production. Such first insights into ferruginous sediments show that microbial populations perform successive metabolisms related to sulfur, iron and methane. In theory, iron reduction could reoxidize reduced sulfur compounds and desorb OM from iron minerals to allow remineralization to methane. Overall, we found that biogeochemical processes in the sediments can be linked to redox differences in the bottom waters of the three sites, like oxidant concentrations and the supply of labile OM. At the scale of the lacustrine record, our geomicrobiological study should provide a means to link the extant subsurface biosphere to past environments.

Geochemistry of ODP Hole 108-658C sediments

During the African Humid Period (AHP), much of the modern hyperarid Saharan desert was vegetated and covered with numerous lakes. In marine sediments off northwestern Africa, the AHP is represented by markedly reduced siliciclastic sediment flux between ~ 12.3 and 5.5 ka. Changes in the origin of this terrigenous sediment fraction can be constrained by sediment chemistry and radiogenic isotope tracers. At Ocean Drilling Program (ODP) Site 658, Hole C (20°44.95'N, 18°34.85'W, 2263 mbsl), the neodymium (Nd) isotope composition of terrigenous detritus shows little variability throughout the last 25 kyr, indicating that the contributing geological terranes have not changed appreciably since the last glacial period. In contrast, there were large and abrupt changes in strontium (Sr) isotope ratios and chemical compositions associated with the AHP, during which 87Sr/86Sr ratios were markedly less radiogenic, and sediments show higher chemical indices of alteration. We show that sediment geochemical changes during the AHP cannot be attributed to changes in the source terranes, physical sorting, or intensity of chemical weathering. The low 87Sr/86Sr and high Sr concentrations of AHP-age samples also conflict with the interpretation of increased fine-grained, fluvially derived sediments. We propose that the most significant compositional changes at ODP 658C are due to the addition of an aluminosilicate component that has a highly altered major element signature but is enriched in soluble elements like Sr and magnesium (Mg) compared to aluminum (Al) and has low 87Sr/86Sr relative to local terrigenous source areas. We interpret these characteristics to reflect authigenic sediment supply from extensive North African paleolake basins that were prevalent during the AHP.