(Appendix 1) Age, Ge/Si ratio, aluminium and opal contents in ODP Sites 175-1081 and 175-1084

As a result of both culture and sediment core studies, the ratio of germanium (Ge) to silicon (Si) in diatom shells has been proposed as a proxy for monitoring whole-ocean changes in seawater Ge/Si, a ratio affected by changes in continental weathering. However, because of the difficulties of extracting and cleaning diatom frustules from deep-sea sediments, only samples from highly pure diatom oozes in the Antarctic region have been previously analyzed. Here we present data on diatom Ge/Si ratios, (Ge/Si)opal, for the time interval between 3.1 and 1.9 Ma from a mid-latitude, coastal upwelling area where significant terrigenous sediment input complicated the sample processing and analyses. In general, our (Ge/Si)opal values show the same decreasing trend after 2.6 Ma than previously measured in Antarctic sediments (Shemesh et al., 1989. Paleoceanography 4, 221-231), but with a noisier background that may reflect the local imprint of proximal continental input superimposed upon global changes in the ocean reservoir. The time of initiation of large-scale North Hemisphere glaciation at ~2.6 Ma is characterized by a declining pattern of diatom Ge/Si ratios, which could have resulted from a global increase in the input of riverine Si due to enhanced silica weathering and/or equatorward (northward) intrusions of subantarctic waters enriched in silica. High (Ge/Si)opal ratios are associated with high opal contents from the same sediment samples and with warm climate as indicated by depleted benthic foraminiferal d18O values from the North and Equatorial Atlantic. Cold periods signified by enriched benthic d18O values, on the contrary, are associated with lower (Ge/Si)opal ratios. We interpret diatom Ge/Si values to reflect the prevailing weathering state on the continents, with greater chemical weathering during warm and wet periods of the Pliocene and less during cooler and drier intervals.

Pb isotopes in Cenozoic sediments of ODP Site 177-1090

Nd and Pb isotopic compositions extracted from bulk deep sea sediments have been shown to be robust proxies for deep water circulation as well as weathering provenance and intensity over geologically young time scales. In this study we evaluated ten deep sea samples from Ocean Drilling Program (ODP) site 1090 ranging in age from mid Eocene to early-Miocene to test whether Pb isotopic compositions extracted from geologically older sediments record reliable seawater isotopic ratios and to evaluate the source of the extracted Pb. The sequential extraction protocol used in this study is similar to protocols reported for previous studies and produces acetic acid, hydroxylamine hydrochloride (HH) and residue fractions. Each extracted fraction was analyzed for Pb isotopes, rare earth elements (REEs), and a suite of major elements. Similar 206Pb/204Pb, 207Pb/204Pb, and 208Pb/204Pb ratios are recorded from the acetic acid and HH fractions for ~70-80% of the samples, suggesting that either the acetic acid dissolves Fe-Mn oxides or multiple phases are recording the same seawater isotopic value. Several indirect tests, such as Al mass balance, comparison of Sr isotopes in HH extracts to contemporaneous seawater Sr isotopes, and comparison of Nd isotopic compositions in HH extracts to published fossil fish teeth values, provide evidence that Pb isotopic compositions measured in our bulk HH extracts record bottom water values. The relationship between Pb, Mn and Ca concentrations in HH fractions indicates that Fe-Mn oxides and a Mn-bearing carbonate are the dominant phases contributing seawater Pb. Comparison of REE patterns derived from the HH fraction and total digestions of Fe-Mn nodule standards reveals that the trivalent REEs exhibit patterns consistent with the parent archive, but Ce can be fractionated during extraction. Ratios of REEs also produce unique fields for each fraction and can be used to test the purity of the seawater signal of the extraction protocol. Finally, an initial evaluation of Pb isotopic compositions in fossil fish indicates that this archive is not suitable for bottom water Pb isotope studies.

A time-slice study in Heinrich Stadial 1 and Mid Holocene sediment cores off northern Senegal

Sediment dynamics in limnic, fluvial and marine environments can be assessed by granulometric and rock-magnetic methodologies. While classical grain-size analysis by sieving or settling mainly bears information on composition and transport, the magnetic mineral assemblages reflect to a larger extent the petrology and weathering conditions in the sediment source areas. Here, we combine both methods to investigate Late Quaternary marine sediments from five cores along a transect across the continental slope off Senegal. This region near the modern summer Intertropical Convergence Zone is particularly sensitive to climate change and receives sediments from several aeolian, fluvial and marine sources. From each of the investigated five GeoB sediment cores (494-2956 m water depth) two time slices were processed which represent contrasting climatic conditions: the arid Heinrich Stadial 1 (~ 15 kyr BP) and the humid Mid Holocene (~ 6 kyr BP). Each sediment sample was split into 16 grain-size fractions ranging from 1.6 to 500 µm. Concentration and grain-size indicative magnetic parameters (susceptibility, SIRM, HIRM, ARM and ARM/IRM) were determined at room temperature for each of these fractions. The joint consideration of whole sediment and magnetic mineral grain-size distributions allows to address several important issues: (i) distinction of two aeolian sediment fractions, one carried by the north-easterly trade winds (40-63 µm) and the other by the overlying easterly Harmattan wind (10-20 µm) as well as a fluvial fraction assigned to the Senegal River (< 10 µm); (ii) identification of three terrigenous sediment source areas: southern Sahara and Sahel dust (low fine-grained magnetite amounts and a comparatively high haematite content), dust from Senegalese coastal dunes (intermediate fine-grained magnetite and haematite contents) and soils from the upper reaches of the Senegal River (high fine-grained magnetite content); (iii) detection of partial diagenetic dissolution of fine magnetite particles as a function of organic input and shore distance; (iv) analysis of magnetic properties of marine carbonates dominating the grain-size fractions 63-500 µm.

Chert intervals in DSDP and ODP sites (Table 1)

Radiolarian cherts in the Tethyan realm of Jurassic age were recently interpreted as resulting from high biosiliceous productivity along upwelling zones in subequatorial paleolatitudes the locations of which were confirmed by revised paleomagnetic estimates. However, the widespread occurrence of cherts in the Eocene suggests that cherts may not always be reliable proxies of latitude and upwelling zones. In a new survey of the global spatio-temporal distribution of Cenozoic cherts in Deep Sea Drilling Project (DSDP) and Ocean Drilling Program (ODP) sediment cores, we found that cherts occur most frequently in the Paleocene and early Eocene, with a peak in occurrences at ~50 Ma that is coincident with the time of highest bottom water temperatures of the early Eocene climatic optimum (EECO) when the global ocean was presumably characterized by reduced upwelling efficiency and biosiliceous productivity. Cherts occur less commonly during the subsequent Eocene global cooling trend. Primary paleoclimatic factors rather than secondary diagenetic processes seem therefore to control chert formation. This timing of peak Eocene chert occurrence, which is supported by detailed stratigraphic correlations, contradicts currently accepted models that involve an initial loading of large amounts of dissolved silica from enhanced weathering and/or volcanism in a supposedly sluggish ocean of the EECO, followed during the subsequent middle Eocene global cooling by more vigorous oceanic circulation and consequent upwelling that made this silica reservoir available for enhanced biosilicification, with the formation of chert as a result of biosilica transformation during diagenesis. Instead, we suggest that basin-basin fractionation by deep-sea circulation could have raised the concentration of EECO dissolved silica especially in the North Atlantic, where an alternative mode of silica burial involving widespread direct precipitation and/or absorption of silica by clay minerals could have been operative in order to maintain balance between silica input and output during the upwelling-deficient conditions of the EECO. Cherts may therefore not always be proxies of biosiliceous productivity associated with latitudinally focused upwelling zones.

Geochemistry at DSDP Leg 75 Holes

Samples of sediments and rocks collected at DSDP Sites 530 and 532 were analyzed for 44 major, minor, and trace elements for the following purposes: (1) to document the downhole variability in geochemistry within and between lithologic units; (2) to document trace-element enrichment, if any, in Cretaceous organic-carbon-rich black shales at Site 530; (3) to document trace-element enrichment, if any, in Neogene organic-carbon-rich sediments at Site 532; (4) to document trace-element enrichment, if any, in red claystone above basalt basement at Site 530 that might be attributed to hydrothermal activity or weathering of basalt. Results of the geochemical analyses showed that there are no significant enrichments of elements in the organic-carbon-rich sediments at Site 532, but a number of elements, notably Cd, Co, Cr, Cu, Mo, Ni, Pb, V, and Zn, are enriched in the Cretaceous black shales. These elements have different concentration gradients within the black-shale section, however, which suggests that there was differential mobility of trace elements during diagenesis of interbedded more-oxidized and less-oxidized sediments. There is little or no enrichment of elements from hydrothermal activity in the red claystone immediately overlying basalt basement at Site 530, but slight enrichments of several elements in the lowest meter of sediment may be related to subsea weathering of basalt

Interstitial-water chemistry and stable isotope record of Prydz Bay sediments

Leg 119 of the Ocean Drilling Program (ODP) provided the first opportunity to study the interstitial-water chemistry of the eastern Antarctic continental margin. Five sites were cored in a northwest-southeast transect of Prydz Bay that extended from the top of the continental slope to within 30 km of the coastline. Geological studies of the cores reveal a continental margin that has evolved through terrestrial, glacial, and glacial-marine environments. Chemical and stable isotopic analyses of the interstitial-waters were performed to determine the types of depositional environments and the diagenetic and hydrologic processes that are operating in this unusual marine environment. Highly compacted glacial sediments provide an effective barrier to the vertical diffusion of interstitial-water solutes. Meteoric water from the Antarctic continent appears to be flowing into Prydz Bay sediments through the sequence of terrestrial sediments that lie underneath the glacial sediments. The large amounts of erosion associated with glacial advances appear to have had the effect of limiting the amount of marine organic matter that is incorporated into the sediments on the continental shelf. Although all of the sites cored in Prydz Bay exhibit depletions in dissolved sulfate with increasing depth, the greatest bacterial activity is associated with a thin layer of diatom ooze that coats the seafloor of the inner bay. Results of alkalinity modeling, thermodynamic calculations, and strontium analyses indicate that (1) ocean bottom waters seaward of Site 740 are undersaturated with respect to both calcite and aragonite, (2) interstitial waters at each site become saturated or supersaturated with respect to calcite and aragonite with increasing depth, (3) precipitation of calcium carbonate reduces the alkalinity of the pore waters with increasing depth, and (4) recrystallization of aragonite to calcite accounts for 24% of the pore-water strontium. Weathering of unstable terrestrial debris and cation exchange between clay minerals and pore fluids are the most probable chemical processes affecting interstitial water cation gradients.

Chemistry of smectite and volcanic glass of ODP Site 159-959

The Ocean Drilling Program (ODP) Site 959 was drilled in the northern border of the Côte d'Ivoire-Ghana Ridge at a water depth of 2100 m. Pleistocene total thickness does not exceed 20 m. Winnowing processes resulted in a low accumulation rate and notable stratigraphic hiatuses. During the Late Pleistocene, bottom circulation was very active and controlled laminae deposition (contourites) which increased the concentration of glauconitic infillings of foraminifera, and of volcanic glass and blue-green grains more rarely, with one or several subordinate ferromagnesian silicates. Volcanic glass generally was X-ray amorphous and schematically classified as basic to intermediate (44-60% SiO2). Opal-A or opal-CT suggested the beginning of the palagonitisation process, and previous smectitic deposits may have been eroded mechanically. The blue-green grains presented two main types of mineralogic composition: (1) neoformed K, Fe-smectite associated with zeolite (like phillipsite) and unequal amounts of quartz and anorthite; (2) feldspathic grains dominated by albite but including quartz, volcanic glass and smectites as accessory components. They were more or less associated with the volcanic glass. On the basis of their chemical composition, the genetic relationship between the blue-green grains and the volcanic glass seemed to be obvious although some heterogeneous grains seemed to be primary ignimbrite and not the result of glass weathering. The most reasonable origin of these pyroclastic ejecta would be explosive events from the Cameroon Volcanic Ridge, especially from the Sao Thome and Principe Islands and Mount Cameroon area. This is supported both by grain geochemistry and the time of volcanic activity, i.e. Pleistocene. After westward wind transport (some 1200 km) and ash fall-out, the subsequent winnowing by bottom currents controlled the concentration of the volcanic grains previously disseminated inside the hemipelagic sediment. Palagonitisation, and especially phillipsite formation, may result from a relatively rapid reaction during burial diagenesis (<1 m.y.), in deep-sea deposits at relatively low sedimentation rate. However, it cannot be excluded that the weathering had begun widely on the Cameroon Ridge before the explosive event.

Mineralogy and diagenesis of slope sediments at DSDP Leg 84 holes

The distribution and composition of minerals in the silt and clay fraction of the fine-grained slope sediments were examined. Special interest was focused on diagenesis. The results are listed as follows. (1) Smectite, andesitic Plagioclase, quartz, and low-Mg calcite are the main mineral components of the sediment. Authigenic dolomite was observed in the weathering zones of serpentinites, together with aragonite, as well as in clayey silt. (2) The mineralogy and geochemistry of the sediments is analogous to that of the andesitic rocks of Costa Rica and Guatemala. (3) Unstable components like volcanic glass, amphiboles, and pyroxenes show increasing etching with depth. (4) The diagenetic alteration of opal-A skeletons from etching pits and replacement by opal-CT to replacement by chalcedony as a final stage corresponds to the typical opal diagenesis. (5) Clinoptilolite is the stable zeolite mineral according to mineral stability fields; its neoformation is well documented. (6) The early diagenesis of smectites is shown by an increase of crystallinity with depth. Only the smectites in the oldest sediments (Oligocene and early Eocene) contain nonexpanding illite layers.

Element concentrations, and Sr and Os isotope ratios of DSDP Leg 92 metalliferous carbonates

The Os concentration and isotopic composition of metalliferous carbonates deposited on the East Pacific Rise over the past 28 Ma are reported with complimentary Sr isotope data. Variations in the Os isotopic composition of these samples are interpreted as a record of past changes in the Os isotopic composition of seawater. These results are consistent with isotopic analyses of leachable Os in pelagic clays which have also been interpreted as a record of the 187Os/186Os ratio of seawater through time (Pegram et al., 1992, doi:10.1016/0012-821X(92)90132-F). The metalliferous carbonate record clearly shows that seawater Os and Sr isotope systems are partially decoupled from one another over the past 28 Ma. Accelerated weathering of ancient organic-rich sediments is suggested as a possible mechanism to account for this decoupling and the rapid increase in the 187Os//186Os ratio of seawater over the past 15 Ma. This rapid increase suggests that the seawater Os record can potentially be used as a stratigraphic tool in some Neogene marine deposits.

Plant-wax, alkenone and major element analyses of ODP Site 659 for the Last Glacial cycle and associated coretop data

We present a hydrologic reconstruction of the Sahara-Sahel transition, covering the complete last glacial cycle (130 ka), based on a combination of plant-wax-specific hydrogen (dD) and carbon isotopes (d13C). The dD and d13C signatures of long-chain n-alkanes from ODP Site 659 off NW Africa reveal a significant anti-correlation. Complementary to published pollen data, we infer that this plant-wax signal reflects sensitive responses of the vegetation cover to precipitation changes in the Sahel region, as well as varying contributions from biomes north of the Sahara (C3 domain) by North-East Trade Winds (NETW). During arid phases, especially the northern parts of the Sahel likely experienced crucial water stress, which resulted in a pronounced contraction of the vegetation cover, thus reducing the amount of C4 plant waxes from the region. The increase in NETW strength during dry periods further promoted a more pronounced C3-plant-wax signal derived from the North African C3 plant domain. During humid periods, the C4-dominated Sahelian environments spread northward into the Saharan realm, in association with lower NETW inputs of C3 plant waxes. Arid-humid cycles deduced from plant-wax dD are in accordance with concomitant changes in weathering intensity reflected in varying major element distributions. Environmental shifts are generally linked to periods with large fluctuations in Northern Hemisphere summer insolation. During Marine Isotope Stages 2 and 3, when insolation variability was low, coupling of the hydrologic regime to alkenone-based estimates of NE Atlantic sea-surface temperatures becomes apparent.

Eocene sedimentary calcium carbonate contents and stable isotope composition of benthic foraminifera

'Hyperthermals' are intervals of rapid, pronounced global warming known from six episodes within the Palaeocene and Eocene epochs (~65-34 million years (Myr) ago) (Zachos et al., 2005, doi:10.1126/science.1109004; 2008, doi:10.1038/nature06588; Roehl et al., 2007, doi:10.1029/2007GC001784; Thomas et al., 2000; Cramer et al., 2003, doi:10.1029/2003PA000909; Lourens et al., 2005, doi:10.1038/nature03814; Petrizzo, 2005, doi:10.2973/odp.proc.sr.198.102.2005; Sexton et al., 2006, doi:10.1029/2005PA001253; Westerhold et al., 2007, doi:10.1029/2006PA001322; Edgar et al., 2007, doi:10.1038/nature06053; Nicolo et al., 2007, doi:10.1130/G23648A.1; Quillévéré et al., 2008, doi:10.1016/j.epsl.2007.10.040; Stap et al., 2010, doi:10.1130/G30777.1). The most extreme hyperthermal was the 170 thousand year (kyr) interval (Roehl et al., 2007) of 5-7 °C global warming (Zachos et al., 2008) during the Palaeocene-Eocene Thermal Maximum (PETM, 56 Myr ago). The PETM is widely attributed to massive release of greenhouse gases from buried sedimentary carbon reservoirs (Zachos et al., 2005; 2008; Lourenbs et al., 2005; Nicolo et al., 2007; Dickens et al., 1995, doi:10.1029/95PA02087; Dickens, 2000; 2003, doi:10.1016/S0012-821X(03)00325-X; Panchuk et al., 2008, doi:10.1130/G24474A.1) and other, comparatively modest, hyperthermals have also been linked to the release of sedimentary carbon (Zachos et al., 2008, Lourens et al., 2005; Nicolo et al., 2007; Dickens, 2003; Panchuk et al., 2003). Here we show, using new 2.4-Myr-long Eocene deep ocean records, that the comparatively modest hyperthermals are much more numerous than previously documented, paced by the eccentricity of Earth's orbit and have shorter durations (~40 kyr) and more rapid recovery phases than the PETM. These findings point to the operation of fundamentally different forcing and feedback mechanisms than for the PETM, involving redistribution of carbon among Earth's readily exchangeable surface reservoirs rather than carbon exhumation from, and subsequent burial back into, the sedimentary reservoir. Specifically, we interpret our records to indicate repeated, large-scale releases of dissolved organic carbon (at least 1,600 gigatonnes) from the ocean by ventilation (strengthened oxidation) of the ocean interior. The rapid recovery of the carbon cycle following each Eocene hyperthermal strongly suggests that carbon was resequestered by the ocean, rather than the much slower process of silicate rock weathering proposed for the PETM (Zachos et al., 2005; 2003). Our findings suggest that these pronounced climate warming events were driven not by repeated releases of carbon from buried sedimentary sources (Zachos et al., 2008, Lourens et al., 2005; Nicolo et al., 2007; Dickens, 2003; Panchuk et al., 2003) but, rather, by patterns of surficial carbon redistribution familiar from younger intervals of Earth history.