Diatom oxygen and silicon isotopes data from IODP Hole 323-U1341B in the Bering Sea through the late Pliocene and over the onset of major Northern Hemisphere Glaciation

The Bering Sea is one of the most biologically productive regions in the marine system and plays a key role in regulating the flow of waters to the Arctic Ocean and into the subarctic North Pacific Ocean. Cores from Integrated Ocean Drilling Program (IODP) Expedition 323 to the Bering Sea provide the first opportunity to obtain reconstructions from the region that extend back to the Pliocene. Previous research at Bowers Ridge, south Bering Sea, has revealed stable levels of siliceous productivity over the onset of major Northern Hemisphere Glaciation (NHG) (circa 2.85-2.73 Ma). However, diatom silica isotope records of oxygen (d18Odiatom) and silicon (d30Sidiatom) presented here demonstrate that this interval was associated with a progressive increase in the supply of silicic acid to the region, superimposed on shift to a more dynamic environment characterized by colder temperatures and increased sea ice. This concluded at 2.58 Ma with a sharp increase in diatom productivity, further increases in photic zone nutrient availability and a permanent shift to colder sea surface conditions. These transitions are suggested to reflect a gradually more intense nutrient leakage from the subarctic northwest Pacific Ocean, with increases in productivity further aided by increased sea ice- and wind-driven mixing in the Bering Sea. In suggesting a linkage in biogeochemical cycling between the south Bering Sea and subarctic Northwest Pacific Ocean, mainly via the Kamchatka Strait, this work highlights the need to consider the interconnectivity of these two systems when future reconstructions are carried out in the region.

Late Pliocene carbonate mineralogy of ODP Hole 133-818B

Shedding of shallow carbonate material toward the deep slopes and basin floors is clearly tied to the position of the carbonate bank tops relative to the photic zone. The onset of bank shedding in periplatform sediments can record either the flooding of the bank tops within the photic zone during a rise in sea level following a period of exposure, referred to in the literature as the "highstand shedding" scenario, or the reentry of the bank tops into the photic zone during a lowering of sea level following a period of drowning, referred to as the "lowstand shedding" scenario. Results from Leg 133 post-cruise research on the Pliocene sequences, drilled in six sites within different slope settings of the Queensland Plateau, seem to point out that the latter "lowstand shedding" scenario can be applied to this particular carbonate system. At the Queensland Plateau sites, the early Pliocene (5.2-3.5 Ma) and the earliest part of the late Pliocene (3.5-2.9 Ma) age sequences were characterized, especially in the ôdeepö Sites 811 and 817, by pelagic sediments (foraminifers and coccoliths) and by typically pelagic sedimentation rates not exceeding 20 mm/k.y. The earliest part of the late Pliocene age section was characterized by well-developed hardgrounds in the "shallow" Sites 812 and 814 and by normal pelagic sediments mixed with reworked phosphatized planktonic foraminifers in Site 813. Finally, the early part of the late Pliocene (2.9-2.4 Ma) section was characterized by high sedimentation rates, related to the shedding and admixture into the pelagic sediments of bank-derived materials. These bank-derived materials consist of either diagenetically unaltered fine aragonite with traces of dolomite in Site 818 or micritic calcite resulting from seafloor and/or shallow burial alteration in the deepest Sites 817 and 811. The highest sedimentation rates (163 mm/k.y.) were recorded in Site 818, drilled nearest the modern carbonate bank of Tregrosse Reef. The sedimentation rates decrease with increasing distance from Tregrosse Reef - 120 mm/k.y. in Site 817 and 47.5 mm/k.y. in Site 811. The initial appearance of fine aragonite in Site 818, corresponding to the transition from pelagic to periplatform sedimentation rates, has been dated at 2.9 Ma. This Pliocene sediment pattern on the Queensland Plateau is different from the pattern observed in sediments from two earlier ODP legs (i.e., Leg 101 in the Bahamas and in Leg 115 in the Maldives), where aragonite-rich sediments, characterized by high periplatform sedimentation rates, were observed in the lower Pliocene section (5.2-3.5 Ma), whereas the upper Pliocene (3.5-1.6 Ma) sediments are more pelagic in nature and are characterized by low sedimentation rates or major hiatuses. These Pliocene periplatform sequences in the Bahamas and in the Maldives and late Quaternary age periplatform sequences worldwide have pointed out that "highstand shedding" was the typical response of carbonate platforms to fluctuations in sea level, just opposite to a "lowstand shedding" response to sea-level fluctuations, typical of siliciclastic shelves. Assuming that the envelope of Haq et al.'s (1987) sea-level curve, showing a well-defined lowering of sea level between 3.5 and 2.9 Ma, can also be applied to the southwest Pacific Ocean, based on a high-resolution Pliocene d18O record from the Ontong Java Plateau recently published by Jansen et al. (1993, doi:10.2973/odp.proc.sr.130.028.1993), the Pliocene periplatform sequences on the Queensland Plateau would have recorded the reentry of the bank tops into the photic zone during a general lowering of sea level, following an interval characterized by high sea level, during which the shallow carbonate system on the Queensland Plateau was drowned. The early Pliocene age (5.2-3.5 Ma) sediments deposited on the Queensland Plateau, an established interval of eustatic sea-level highstand, are typically pelagic in character. In addition, relatively cold surface temperatures (estimated to have ranged from 18° to 20°C by Isern et al. [this volume]) might have also stressed the reefs during early Pliocene time and contributed to the drowning of the Queensland Plateau carbonate system during the late Miocene and early Pliocene. Differential and relatively high subsidence rates, inferred by variations in paleodepth of water (based upon benthic foraminifer assemblages; Katz and Miller, this volume) may also have influenced the drowning of the carbonate bank tops on the Queensland Plateau during the late Miocene and early Pliocene. The sediments of early late Pliocene age (2.9-2.4 Ma), a well-established interval of lowering of sea level, are clearly periplatform and cyclic in nature. High-frequency (~40 k.y.) aragonite cycles, well-developed between 2.9 and 2.45 Ma, correlate with the planktonic high-resolution Pliocene d18O record from the Ontong Java Plateau, a good sea-level proxy (Jansen et al., in press). Contrary to late Quaternary age aragonite cycles from the Bahamas, the Nicaragua Rise, the Maldives, and the Queensland Plateau, the late Pliocene aragonite cycles in Hole 818B display high levels of aragonite during glacial stages and, therefore, lowstands of sea level. In addition, sediments deposited during the earliest part of the late Pliocene (3.5-2.9 Ma), transition between the early Pliocene highstand and the late Pliocene lowering in sea level, have recorded the first evidence of a fall in sea level, by (1) the occurrence of synchronous submarine hardgrounds in the two shallowest sites (Sites 812 and 814), (2) the deposition of reworked material from the shallower part of the slope into the intermediate Sites 813 and 818, and (3) the deposition of pelagic sediments in the deepest Sites 817 and 817. In summary, contrary to previous findings, the Pliocene periplatform sediments on the Queensland Plateau appear to have recorded a regional shedding of shallow carbonate bank tops during an interval of sea-level lowering, a good illustration of the "carbonate lowstand shedding" scenario, occurring during the reentry of previously drowned carbonate bank tops into the photic zone related to a decrease in sea level.

Onset of long-term cooling of Greenland near the Eocene-Oligocene boundary as revealed by branched tetraether lipids

The Eocene-Oligocene (E-O) boundary interval is considered to be one of the major transitions in Earth's climate, witnessing the first major expansion of the East Antarctic Ice Sheet. However, the extent of the associated climatic cooling, especially for high northern latitude continental landmasses, is poorly constrained. In this study we reconstruct the first mean annual air temperature (MAAT) for the Greenland landmass during the late Eocene and early Oligocene by applying a new proxy based on the distribution of branched tetraether lipids derived from soil bacteria preserved in a marine sediment core from the Greenland Basin. The temperature estimates are compared with a composite continental temperature record based on bio-climatic analysis of pollen assemblages. Both proxies reveal comparable late Eocene MAATs of ~13-15 °C and a gradual long-term cooling of ~3-5 °C starting near the E-O boundary. These data are in agreement with other MAAT reconstructions from northern midlatitude continents and suggest a general cooling of the Northern Hemisphere during the E-O transition.

Late Neogene time scale for ODP Leg 138 sites

The sediments recovered during Leg 138 provide a remarkable opportunity to improve the geological time scale of the late Neogene. We have developed new time scales in the following steps. First, we constructed age models on the basis of shipboard magnetostratigraphy and biostratigraphy, using the time scale of Berggren, Kent, and Flynn (1985). Second, we refined these age models using shipboard GRAPE density measurements to provide more accurate correlation points. Third, we calibrated a time scale for the past 6 m.y. by matching the high-frequency GRAPE density variations to the orbital insolation record of Berger and Loutre (1991); we also took into account d18O records, where they were available. Fourth, we generated a new seafloor anomaly time scale using our astronomical calibration of C3A.n (t) at 5.875 Ma and an age of 9.639 Ma for C5n.1n (t) that is based on a new radiometric calibration (Baksi, 1992). Fifth, we recalibrated the records older than 6 Ma to this new scale. Finally, we reconsidered the 6- to 10-Ma interval and found that this could also be partially tuned astronomically.

Benthic foraminiferal assemblages of middle to late Pliocene sediments in the North Atlantic Ocean

Records of benthic foraminifera from North Atlantic DSDP Site 607 and Hole 610A indicate changes in deep water conditions through the middle to late Pliocene (3.15 to 2.85 Ma). Quantitative analyses of modem associations in the North Atlantic indicate that seven species, Fontbotia wuellerstorfi, Cibicidoides kullenbergi, Uvigerina peregrina, Nuttallides umboniferus, Melonis pompilioides, Globocassidulina subglobosa and Epistominella exigua are useful for paleoenvironmental interpretation. The western North Atlantic basin (Site 607) was occupied by North Atlantic Deep Water (NADW) until c. 2.88 Ma. At that time, N. umboniferus increased, indicating an influx of Southern Ocean Water (SOW). The eastern North Atlantic basin (Hole 610A) was occupied by a relatively warm water mass, possibly Northeastern Atlantic Deep Water (NEADW), through c. 2.94 Ma when SOW more strongly influenced the site. These interpretations are consistent with benthic delta18O and delta13C records from 607 and 610A (Raymo et al., 1992). The results presented in this paper suggest that the North Atlantic was strongly influenced by northern component deep water circulation until 2.90-2.95 Ma. After that there was a transition toward a glacially driven North Atlantic circulation more strongly influenced by SOW associated with the onset of Northern Hemisphere glaciation. The circulation change follows the last significant SST and atmospheric warming prior to c. 2.6 Ma.

Ice-rafted debris of Late Cenozoic sediments from ODP Leg 104 holes

The abundance and composition of the upper Cenozoic terrigenous coarse-sand fraction (250 µm-2 mm) at ODP Sites 642, 643, and 644 were investigated to date the onset of significant ice-rafting in the Norwegian Sea, establish the regional chronology of ice-rafting, and determine the relative importance of global vs. regional controls on ice-rafting in this area. The first input of ice-rafted debris (IRD) occurs at approximately 2.9 Ma, with significant ice-rafting beginning at about 2.5 Ma. IRD abundances increase significantly in sediments younger than 0.9 Ma at all three holes, indicating climatic deterioration in the late Pleistocene. Differences in the timing of this IRD increase between holes result from regional patterns of IRD supply and surface circulation. Variations in IRD sources and dispersal patterns may also explain the slightly higher background level of IRD abundance at Hole 642B, a seaward site. Major peaks in the generalized IRD records from the Norwegian Sea are tentatively correlated to glacial stages or glacial-to-interglacial transitions in the globally defined oxygen isotope record. This correlation indicates the effect of global conditions on the regional climate of the Norwegian Sea, although the detailed IRD records at these sites are also affected by local/regional processes (e.g., circulation patterns and source area differences).

Palynology of the Middle to Late Miocene from ODP Hole 175-1085A off the west coast of South Africa

Aim To test whether the radiation of the extremely rich Cape flora is correlated with marine-driven climate change. Location Middle to Late Miocene in the south-east Atlantic and the Benguela Upwelling System (BUS) off the west coast of South Africa. Methods We studied the palynology of the thoroughly dated Middle to Late Miocene sediments of Ocean Drilling Program (ODP) Site 1085 retrieved from the Atlantic off the mouth of the Orange River. Both marine upwelling and terrestrial input are recorded at this site, which allows a direct correlation between changes in the terrestrial flora and the marine BUS in the south-east Atlantic. Results Pollen types from plants of tropical affinity disappeared, and those from the Cape flora gradually increased, between 10 and 6 Ma. Our data corroborate the inferred dating of the diversification in Aizoaceae c. 8 Ma. Main conclusions Inferred vegetation changes for the Late Miocene south-western African coast are the disappearance of Podocarpus-dominated Afromontane forests, and a change in the vegetation of the coastal plain from tropical grassland and thicket to semi-arid succulent vegetation. These changes are indicative of an increased summer drought, and are in step with the development of the southern BUS. They pre-date the Pliocene uplift of the East African escarpment, suggesting that this did not play a role in stimulating vegetation change. Some Fynbos elements were present throughout the recorded period (from 11 Ma), suggesting that at least some elements of this vegetation were already in place during the onset of the BUS. This is consistent with a marine-driven climate change in south-western Africa triggering substantial radiation in the terrestrial flora, especially in the Aizoaceae.

Late Miocene to Pleistocene paleoceanographic records from the Feni and Gardar Drifts: Pliocene reduction in abyssal flow

Magnetic fabric analyses from two North Atlantic drift deposits provide proxies for determining relative variations in the strength of abyssal flow over the last 10 my. The data show a cessation of current-controlled sedimentation at the shallower Feni Drift (2417 m) at the time of onset of Northern Hemisphere glaciation (2.6 Ma). Drift formation ended nearly 2 my earlier (4.2 Ma) at the deeper Gardar Drift (3220 m), implying stepwise reduction in deep-water flow. Relatively light delta18O values at the deeper Gardar Drift indicate a warmer, thus also more salty, water mass site prior to 6 Ma. We interpret this as representing Mediterranean Sea water, which flowed north at depths greater than that of the Feni Drift Site. The supply of Mediterranean Water to the North Atlantic was shut off as the Gibraltar Straits closed, causing the Messinian salinity crisis, and never returned to that position in the water column after the Mediterranean opened again.

Benthic foraminiferal accumulation rates of the late Pliocene-Pleistocene in the northern Indian Ocean

During the late Pliocene-middle Pleistocene, 63 species of elongate, bathyal-upper abyssal benthic foraminifera (Extinction Group = Stilostomellidae, Pleurostomellidae, some Nodosariidae) declined in abundance and finally disappeared in the northern Indian Ocean (ODP Sites 722, 758), as part of the global extinction of at least 88 related species at this time. The detailed record of withdrawal of these species differs by depth and geography in the Indian Ocean. In northwest Indian Ocean Site 722 (2045 m), the Extinction Group of 54 species comprised 2-15% of the benthic foraminiferal fauna in the earliest Pleistocene, but declined dramatically during the onset of the mid-Pleistocene Transition (MPT) at 1.2-1.1 Ma, with all but three species disappearing by the end of the MPT (~0.6 Ma). In northeast Indian Ocean Site 758 (2925 m), the Extinction Group of 44 species comprised 1-5% of the benthic foraminiferal fauna at ~3.3-2.6 Ma, but declined in abundance and diversity in three steps, at ~2.5, 1.7, and 1.2 Ma, with all but one species disappearing by the end of the MPT. At both sites there are strong positive correlations between the accumulation rate of the Extinction Group and proxies indicating low-oxygen conditions with a high organic carbon input. In both sites, there was a pulsed decline in Extinction Group abundance and species richness, especially in glacial periods, with some partial recoveries in interglacials. We infer that the glacial declines at the deeper Site 758 were a result of increased production of colder, well-ventilated Antarctic Bottom Water (AABW), particularly in the late Pliocene and during the MPT. The Extinction Group at shallower water depths (Site 722) were not impacted by the deeper water mass changes until the onset of the MPT, when cold, well-ventilated Glacial North Atlantic Intermediate Water (GNAIW) production increased and may have spread into the Indian Ocean. Increased chemical ventilation at various water depths since late Pliocene, particularly in glacial periods, possibly in association with decreased or more fluctuating organic carbon flux, might be responsible for the pulsed global decline and extinction of this rather specialised group of benthic foraminifera.

Neodymium isotope record of late Paleocene to eraly Eocene fish teeth

High-resolution, fish tooth Nd isotopic records for eight Deep Sea Drilling Project and Ocean Drilling Program sites were used to reconstruct the nature of late Paleocene-early Eocene deep-water circulation. The goal of this reconstruction was to test the hypothesis that a change in thermohaline circulation patterns caused the abrupt 4-5°C warming of deep and bottom waters at the Paleocene/Eocene boundary - the Paleocene-Eocene thermal maximum (PETM) event. The combined set of records indicates a deep-water mass common to the North and South Atlantic, Southern and Indian oceans characterized by mean epsilon-Nd values of ~-8.7, and different water masses found in the central Pacific Ocean (epsilon-Nd ~-4.3) and Caribbean Sea (epsilon-Nd ~1.2). The geographic pattern of Nd isotopic values before and during the PETM suggests a Southern Ocean deep-water formation site for deep and bottom waters in the Atlantic and Indian ocean basins. The Nd data do not contain evidence for a change in the composition of deep waters prior to the onset of the PETM. This finding is consistent with the pattern of warming established by recently published stable isotope records, suggesting that deep- and bottom-water warming during the PETM was gradual and the consequence of surface-water warming in regions of downwelling.

Pb isotopes in individual sand-sized (>150 µm) ice-rafted feldspars deposited at subpolar North Atlantic Ocean Deep Sea Drilling Project Site 611 during the late Pliocene (Marine Isotope Stages, MIS, G6 through 100)

The onset of abundant ice-rafted debris (IRD) deposition in the Nordic Seas and subpolar North Atlantic Ocean 2.72 millions of years ago (Ma) is thought to record the Pliocene onset of major northern hemisphere glaciation (NHG) due to a synchronous advance of North American Laurentide, Scandinavian and Greenland ice-sheets to their marine calving margins during marine isotope stage (MIS) G6. Numerous marine and terrestrial records from the Nordic Seas region indicate that extensive ice sheets on Greenland and Scandinavia increased IRD inputs to these seas from 2.72 Ma. The timing of ice-sheet expansion on North America as tracked by IRD deposition in the subpolar North Atlantic Ocean, however, is less clear because both Europe and North America are potential sources for icebergs in this region. Moreover, cosmogenic-dating of terrestrial tills on North America indicate that the Laurentide Ice Sheet did not extend to ~39°N until 2.4 ±0.14 Ma, at least 180 ka after the onset of major IRD deposition at 2.72 Ma. To address this problem,we present the first detailed analysis of the geochemical provenance of individual sand-sized IRD deposited in the subpolar North Atlantic Ocean between MIS G6 and 100 (~2.72-2.52 Ma). IRD provenance is assessed using laser ablation lead (Pb) isotope analyses of single ice-rafted (>150 mm) feldspar grains. To track when an ice-rafting setting consistent with major NHG first occurred in the North Atlantic Ocean during the Pliocene intensification of NHG (iNHG), we investigate when the Pb-isotope composition (206Pb/204Pb, 207Pb/204Pb, 208Pb/204Pb) of feldspars deposited at DSDP Site 611 first resembles that determined for IRD deposited at this site during MIS 100, the oldest glacial for which there exists convincing evidence for widespread glaciation of North America. Whilst Quaternary-magnitude IRD fluxes exist at Site 611 during glacials from 2.72 Ma, we find that the provenance of this IRD is not constant. Instead, we find that the Pb isotope composition of IRD at our study site is not consistent with major NHG until MIS G2 (2.64 Ma). We hypothesise that IRD deposition in the North Atlantic Ocean prior to MIS G2 was dominated by iceberg calving from Greenland and Scandinavia. We further suggest that the grounding line of continental ice on Northeast America may not have extended onto the continental shelf and calved significant numbers of icebergs to the North Atlantic Ocean during glacials until 2.64 Ma.

Biogenic opal and carbonate concentration in late Neogene sediments of ODP Site 114-704

A preliminary composite depth section was generated for Site 704 by splicing Holes 704A and 704B together over the interval 0-350 mbsf (0-9 m.y.). High-resolution carbonate and opal data from the cores were correlated with the calcium and silicon signals from the GST logging run in Hole 704B to identify missing and disturbed intervals in the cores. Paleomagnetic and biostratigraphic age boundaries were then transferred to the composite depth records to obtain an age model, and sedimentation rates were calculated by linear interpolation between datums. Algorithms relating measured dry-bulk density to carbonate content and depth were generated to produce predicted values of density for every sample. Accumulation rates of bulk, carbonate, opal, and terrigenous sediment components were then computed to generate a record of sediment deposition on the Meteor Rise that has a resolution of better than 200,000 yr for the period from 8.6 to 1.0 m.y. From 8.6 to 2.5 m.y., bulk-accumulation rates on the Meteor Rise averaged less than 2 g/cm**2/1000 yr and were dominated by carbonate deposition. The first significant opal deposition (6.0 m.y.) punctuated a brief (less than 0.6 Ma) approach of the Polar Front Zone (PFZ) northward that heralded a period of increasing severity of periodic carbonate dissolution events (terrigenous maxima) that abruptly terminated at 4.8 m.y. (base of the Thvera Subchron), synchronous with the reflooding of the Mediterranean after the Messinian salinity crisis. From 4.8 to 2.5 m.y., carbonate again dominated deposition, and the PFZ was far south except during brief northward excursions bracketing 4.2-3.9, 3.3-2.9, and 2.8-2.7 m.y. At 2.5 m.y., all components of bulk-accumulation rates increased dramatically (up to 15 g/cm2/1000 yr), and by 2.4 m.y., a pattern of alternating, high-amplitude carbonate and opal cyclicity marked the initiation of rapid glacial to interglaci·l swings in the position of the PFZ, synchronous with the "onset" of major Northern Hemisphere glaciation. Both mass-accumulation rates and the amplitude of the cycles decreased by about 2 m.y., but opal accumulation rates remained high up through the base of the Jaramillo (0.98 m.y.). From 1.9 to 1 m.y., the record is characterized by moderate amplitude fluctuations in carbonate and opal. This record of opal accumulation rates is interpreted as a long-term "Polar Front Indicator" that monitors the advance and retreat of the opal-rich PFZ northward (southward) toward (away from) the Meteor Rise in the subantarctic sector of the South Atlantic Ocean. The timing of PFZ migrations in the subantarctic South Atlantic Ocean is remarkably similar to Pliocene-Pleistocene climate records deduced from benthic oxygen isotope records in the North Atlantic Ocean (Raymo et al., 1989, doi:10.1029/PA004i004p00413; Ruddiman et al., 1989, doi:10.1029/PA004i004p00353). These include northward migrations during "cold" intervals containing strong glacial isotope stages (2.4-2.3, 2.1-2.0, 1.95-1.55, 1.45-1.30 m.y. and at about 1.13 and 1.09 m.y.) and southward migrations during "warm" intervals containing weak glacial and/or strong interglacial stages (2.45-2.40, 2.30-2.10, 2.00-1.95, 1.52-1.45, 1.30-1.18, 1.11, and 1.06-0.93 m.y.). Although our preliminary composite record is not continuous (some stages are obviously missing), there is hope that future work will identify these missing intervals in the as yet incomplete Hole 704B and will extend this high-resolution Southern Hemisphere climate record back to 8.6 m.y.

Late glacial and deglacial age model and biomarker analyses of sediment core MSM05/5_712-2 from the western continental slope of Svalbard

The transition from last glacial to deglacial and subsequently to modern interglacial climate conditions was accompanied by abrupt shifts in the palaeoceanographic setting in the subpolar North Atlantic. Knowledge about the role that sea ice coverage played during these rapid climate reversals is limited since most marine sediment cores from the higher latitudes provide only a coarse temporal resolution and often poorly preserved microfossils. Here we present a highly resolved reconstruction of sea ice conditions that characterised the eastern Fram Strait - a key area for water mass exchange between the Arctic Ocean and the North Atlantic - for the past 30 ka BP. This reconstruction is based on the distribution of the sea ice biomarker IP25 and phytoplankton derived biomarkers in a sediment core from the continental slope of western Svalbard. During the late glacial (30 ka to 19 ka BP), recurrent advances and retreats of sea ice characterised the study area and point to a hitherto less considered oceanic (and/or atmospheric) variability. A long-lasting perennial sea ice coverage in eastern Fram Strait persisted only at the very end of the Last Glacial Maximum (i.e. from 19.2 to 17.6 ka BP) and was abruptly reduced at the onset of Heinrich Event 1 - coincident with or possibly even inducing the collapse of the Atlantic Meridional Overturning Circulation (AMOC). Further maximum sea ice conditions prevailed during the Younger Dryas cooling event and support the assumption of an AMOC reduction due to increased formation and export of Arctic sea ice through Fram Strait. A significant retreat of sea ice and sea surface warming are observed for the Early Holocene.

Water mass evolution of the Greenland Sea since late glacial times

Four sediment cores from the central and northern Greenland Sea basin, a crucial area for the renewal of North Atlantic deep water, were analyzed for planktic foraminiferal fauna, planktic and benthic stable oxygen and carbon iso- topes as well as ice-rafted debris to reconstruct the environ- mental variability in the last 23 kyr. During the Last Glacial Maximum, the Greenland Sea was dominated by cold and sea-ice bearing surface water masses. Meltwater discharges from the surrounding ice sheets affected the area during the deglaciation, influencing the water mass circulation. During the Younger Dryas interval the last major freshwater event occurred in the region. The onset of the Holocene interglacial was marked by an increase in the advection of Atlantic Wa- ter and a rise in sea surface temperatures (SST). Although the thermal maximum was not reached simultaneously across the basin, benthic isotope data indicate that the rate of overturn- ing circulation reached a maximum in the central Greenland Sea around 7ka. After 6-5ka a SST cooling and increas- ing sea-ice cover is noted. Conditions during this so-called "Neoglacial" cooling, however, changed after 3 ka, probably due to enhanced sea-ice expansion, which limited the deep convection. As a result, a well stratified upper water column amplified the warming of the subsurface waters in the central Greenland Sea, which were fed by increased inflow of At- lantic Water from the eastern Nordic Seas. Our data reveal that the Holocene oceanographic conditions in the Green- land Sea did not develop uniformly. These variations were a response to a complex interplay between the Atlantic and Polar water masses, the rate of sea-ice formation and melting and its effect on vertical convection intensity during times of Northern Hemisphere insolation changes.