Description: During the Pleistocene glaciations, Arctic ice sheets on western Eurasia, Greenland and North America terminated at their continental margins. In contrast, the exposed continental shelves in the Beringian region of Siberia are thought to have been covered by a tundra landscape. Evidence of grounded ice on seafloor ridges and plateaux off the coast of the Beringian margin, at depths of up to 1,000 m, have generally been attributed to ice shelves or giant icebergs that spread oceanwards during glacial maxima. Here we identify marine glaciogenic landforms visible in seismic profiles and detailed bathymetric maps along the East Siberian continental margin. We interpret these features, which occur in present water depths of up to 1,200 m, as traces from grounding events of ice sheets and ice shelves. We conclude that the Siberian Shelf edge and parts of the Arctic Ocean were covered by ice sheets of about 1 km in thickness during several Pleistocene glaciations before the most recent glacial period, which must have had a significant influence on albedo and oceanic and atmospheric circulation.

Description: We determined d18OCib values of live (Rose Bengal stained) and dead epibenthic foraminifera Cibicidoides wuellerstorfi, Cibicides lobatulus, and Cibicides refulgens in surface sediment samples from the Arctic Ocean and the Greenland, Iceland, and Norwegian seas (Nordic Sea). This is the first time that a comprehensive d18OCib data set is generated and compiled from the Arctic Ocean. For comparison, we defined Atlantic Water (AW), upper Arctic Bottom Water (uABW), and Arctic Bottom Water (ABW) by their temperature/salinity characteristics and calculated mean equilibrium calcite d18Oequ from summer sea-water d18Ow and in situ temperatures. As a result, in the Arctic environment we compensate for Cibicidoides- and Cibicides-specific offsets from equilibrium calcite of -0.35 and -0.55 per mil, respectively. After this taxon-specific adjustment, mean d18OCib values plausibly reflect the density stratification of principle water masses in the Nordic Sea and Arctic Ocean. In addition, mean d18OCib from AW not only significantly differs from mean d18OCib from ABW, but also d18OCib from within AW differentiates in function of provenience and water mass age. Furthermore, in shallow waters brine-derived low d18Ow can significantly lower the d18OCib of Cibicides spp. and thus d18OCib may serve as a paleobrine indicator. There is no statistically significant difference, however, between deeper water masses mean d18OCib of the Nordic Sea, and of the Eurasian and Amerasian basins, and no influence of low-d18Ow brines is recorded in Recent uABW and ABW d18OCib of C. wuellerstorfi. This may be due to dilution of a low-d18Ow brine signal in the deep sea, and/or to preferential incorporation of relatively high-d18Ow brines from high-salinity shelves. Although our data encompass environments with seasonal sea-ice and brine formation supposed to ultimately ventilate the deep Arctic Ocean, d18OCib from uABW and ABW do not indicate negative excursions. This may challenge hypotheses that call for enhanced Arctic brine release to explain negative benthic d18O spikes in deep-sea sediments from the late Pleistocene North Atlantic Ocean.

Description: In this study, we present new detailed biomarker-based sea ice records from two sediment cores recovered in the Chukchi Sea and the East Siberian Sea. These new biomarker data may provide new insights on processes controlling recent and past sea ice changes. The biomarker proxy records show (i) minimum sea ice extent during the Early Holocene, (ii) a prominent Mid-Holocene short-term high-amplitude variability in sea ice, primary production and Pacific-Water inflow, and (iii) significantly increased sea ice extent during the last ca. 4.5k cal a BP. This Late Holocene trend in sea ice change in the Chukchi and East Siberian Seas seems to be contemporaneous with similar changes in sea ice extent recorded from other Arctic marginal seas. The main factors controlling the millennial variability in sea ice (and surface-water productivity) are probably changes in surface water and heat flow from the Pacific into the Arctic Ocean as well as the long-term decrease in summer insolation. The short-term centennial variability observed in the high-resolution Middle Holocene record is probably related to solar forcing. Our new data on Holocene sea ice variability may contribute to synoptic reconstructions of regional to global Holocene climate change based on terrestrial and marine archives.