Description: The recent dramatic decline of Arctic sea over the last decades and its controlling processes are still poorly understood. In order to distinguish between natural and anthropogenic processes controlling these changes in sea ice, we have to look back to the past beyond the times of direct measurements. For this purpose, we carried out a multi-proxy approach combining organic-geochemical data (bulk parameters: C/N, TOC, δ13Corg; biomarkers: IP25, sterols, GDGTs) with sedimentological data (core lithology, physical properties, IRD counting, XRF scanning) determined in sediments of Yermak Plateau Core PS92/039-2. This core is situated close to the modern summer ice edge and thus very sensitive for environmental changes. Based on magnetostratigraphy and correlations with dated sediment cores, this core represents the time span from MIS 6 to 1 (ca. 180,000 years) and allows the reconstruction of sea ice variability and related changes in oceanic circulation patterns and the Svalbard Barents Ice Sheet (SBIS) fluctuations during glacial/interglacial changes. As sea ice and phytoplankton biomarkers occur throughout the entire sedimentary section but show some strong variability, a more seasonal sea ice cover was probably predominant during the entire time interval, superimposed by a distinct short-term variability in extent. Significant fluctuations in most of our proxy records indicate highly variable sea ice conditions over the Yermak Plateau during MIS 6. Based on our biomarker data, the SBIS could not have reached the Yermak Plateau during MIS 6. During MIS 4 and 2, coevally elevated concentrations of the sea ice proxy IP25 and the biomarkers for phytoplankton productivity and terrigenous input point to a stationary ice margin above the core position at that time. Strengthened Atlantic Water inflow possibly coupled with katabatic winds from the protruding SBIS may have created this stable ice edge situation and the related sedimentary regime.

Description: One of the main characteristics of the Arctic Ocean is its seasonal to perennial sea-ice cover. Variations of sea-ice conditions affect the Earth’s albedo, primary production, rate of deep-water etc.. During the last decades, a drastic decrease in sea ice has been recorded, and the causes of which, i.e. natural vs. anthropogenic forcings, and their relevance within the global climate system, are subject of intense scientific and societal debate. In this context, records of past sea-ice conditions going beyond instrumental records are of major significance. These records may help to better understand the processes controlling natural sea-ice variability and to improve models for forecasts of future climatic conditions. During RV Polarstern Cruise PS92 in summer 2015, a 860 cm long sediment core (PS92/039-2) was re- covered from the eastern flank of Yermak Plateau north of the Svalbard archipelago (Peeken, 2015). Based on a preliminary age model, this sediment core probably represents the time interval from MIS 6 to MIS 1. This core, located close to the modern summer ice edge, has been selected for reconstruction of past Arctic sea-ice variability based on specific biomarkers. In this context, we have determined the ice-algae-derived sea-ice proxy IP25 (Belt et al., 2007), in combination with other biomarkers indicative for open-water conditions (cf., Müller et al., 2009, 2011). Furthermore, organic carbon fluxes were differentiated using specific biomarkers indicative for marine primary production (brassicasterol, dinosterol) and terrigenous input (campesterol, β-sitosterol). In this poster, preliminary results of our organic-geochemical and sedimentological investigations are presented. Distinct fluctuations of these biomarkers indicate several major, partly abrupt changes in sea-ice cover in the Yermak Plateau area during the late Quaternary. These changes are probably linked to changes in the inflow of Atlantic Water along the western coastline of Svalbard into the Arctic Ocean. Furthermore, the repetitive advance and retreat of the Svalbard Barents Sea Ice Sheet might have influenced the terrigenous input and the environmental setting north of Svalbard, as reflected in the sediment composition of Core PS92/039-2.

Description: IODP site U1417 in the Gulf of Alaska provides a continuous sedimentary record of environmental changes (e.g., in sea surface temperature (SST), marine productivity, ice-rafting) in the subpolar NE Pacific through the Plio-Pleistocene time interval. Here, we present a multi-proxy data set, which allows us to discriminate between different fertilization mechanisms that promoted primary productivity events in the study area between 1.5 Ma and 0.5 Ma. Based on biomarker, micropaleontological, XRF, and sedimentological data, we find that diatom growth benefited from iron-fertilization from aeolian dust, iceberg, and volcanic ash input. Glacial-interglacial SST fluctuations were superimposed by a slight cooling trend with a first pronounced temperature drop during MIS 38 and significantly lowered SSTs persisting through MIS 30 and MIS 28. While the diatom productivity pulses were mainly independent from SST changes, they coincide with terrigenous organic matter input and their occurrence during both glacial and interglacial periods suggest that iron supply from glacigenic dust was mainly controlled by local ice-sheet dynamics. This data set highlights the complexity of fertilization mechanisms in areas affected by evolving ice-sheets and their potential control on the biological carbon pump in subpolar ocean environments.

Description: Past ice sheet conditions in the southern Weddell Sea remain poorly known. Previous studies have led to contradicting scenarios of maximum ice extent during the Last Glacial Maximum (LGM). Scenario A is mainly based on terrestrial data indicating limited ice sheet thickening in the hinterland and suggests a LGM grounding-line position on the inner shelf. Scenario B is based on marine geological/-physical data and concludes that the grounding line was located on the outer shelf (~650 km further offshore than in scenario A). In addition, studies suggest a complex history of ice retreat and drainage pattern since the LGM that needs further constraint. We investigated hydroacoustic data acquired during 17 expeditions. A key finding is a previously unknown stacked grounding zone wedge (GZW) located in Filchner Trough on the outer shelf showing that a palaeo-ice stream stabilized at this position at least twice. Radiocarbon dates from sediment cores indicate that (i) the GZW was formed in the early Holocene and (ii) grounded ice did not extend seaward at the LGM. Hence, the grounding line in Filchner Trough experienced dynamic changes in the Holocene and ice sheet retreat after the LGM was not linear. Ice-flow switches in the hinterland possibly explain this behaviour. Further interesting findings are made in Brunt Basin suggesting the existence of cold-based ice or impacts of large icebergs. In addition, new data will be acquired in the area with RV Polarstern in Jan-Mar 2018.

Description: IODP Expedition 341 succeeded in recovering a continuous sedimentary record of Miocene to Late Pleistocene climate history at drill Site U1417 in the Gulf of Alaska, NE Pacific. Site U1417 sediments provide an excellent opportunity to reconstruct North Pacific sea surface conditions during late Neogene large-scale (global) climate transitions. The Mid Pleistocene Transition (MPT) - one of the most prominent intervals of global Quaternary climate change - is clearly identifiable in Site U1417 sediments (Jaeger et al., 2014). To fully exploit the environmental information archived in U1417 sediments, a sampling strategy has been pursued that permits direct correlation of different (independent) proxy data obtained from biomarker, micropalaeontological, sedimentological and geochemical (XRF) analyses. Mid Pleistocene SSTs in the Gulf of Alaska are in good agreement with SST reconstructions for the North Atlantic and the NW Pacific. A general cooling at about 1 Ma supports earlier hypotheses of an overall Northern Hemisphere ocean cooling as a prerequisite for the increase in continental ice volume. While phytoplankton productivity seems rather independent from SST at Site U1417, it is strongly related to elevated TAR values depicting enhanced input of terrestrial leaf-wax lipids (Meyers, 1997). The transport of these lipids is supposed to be effected by strong winds carrying dust from Alaskan loess deposits to the open ocean as well as by icebergs released from Alaskan tidewater glaciers. The latter is supported by the occasional coincidence of high IRD contents and TAR values. The close relationship between the TAR record, Ba/Al values and the abundance of diatoms, however, strengthens that together with the leaf-wax lipids also iron-bearing dust was exported leading to high productivity events at Site U1417 throughout the Mid Pleistocene. The distinct "on-off" pattern in diatom productivity evolved with the onset of the MPT, which suggests that the expansion of the Northwest Cordilleran Ice Sheet lead to an effective production of glacigenic iron-rich dust that was exported i) by strong northwesterly winds and ii) by icebergs. The observation that productivity peaks in the Gulf of Alaska are not confined to glacial or interglacial periods points to a rather local feedback between the export of iron-bearing dust and an immediately responding ocean surface. The identification of these hitherto unconsidered fertilization mechanisms that potentially fostered ocean productivity and hence the sequestration of atmospheric carbon into the deep ocean are further detailed by Müller et al. (2018).

Description: Since about 15 years a growing number of evidence is found in water depth up to more than 1000 m of the Arctic Ocean that grounding of ice has occurred in various places including the "Beringian" continental margin north of the present Chukchi and East-Siberian seas and the Lomonosov Ridge. These landforms include moraines, drumlinized features, glacigenic debris flows, till wedges, mega-scale glacial lineations (MSGL), and iceberg plough marks (Polyak et al. 2001, Niessen et al. 2013, Dove et al. 2014, Jakobsson et al. 2014). They suggest that thick ice has occurred not only on nearly all margins of the Arctic Ocean but also covered pelagic areas. In a recent paper, Jakobsson et al. (2016) present more evidence of ice-shelf groundings on bathymetric highs in the central Arctic Ocean, thereby revitalising an old modelling concept of a kilometre-thick ice shelf extending over the entire central Arctic Ocean (Hughes et al. 1977) now dated to Marine Isotope Stage (MIS) 6. Other (including our) studies, however, suggest that the pattern, and, in particular, the timing of these glaciations is more complex. Most recent discoveries on the Lomonosov Ridge have not only gained different information on Pleistocene glaciations but also allowed for the first time to reconstruct upper Miocene Arctic Ocean sea-ice and SST conditions. This became possible since submarine sliding (likely associated with ice grounding) led to removal of younger sediments from steep headwalls and thus exhumation of Miocene to early Quaternary sediments close to the seafloor, allowing the retrieval and analysis of such old sediments by gravity coring (Stein et al. 2016). Submarine glacial landforms from the western and central Arctic Ocean were discovered and investigated during the cruises of RV "Polarstern" in 2008 and 2014, and RV "Araon" in 2012 and 2015. Orientations of some of these landforms suggest that thick ice has flown north into the deep Arctic Ocean from the continental margin of the East Siberian Sea repeatedly (Niessen et al. 2013), thereby grounded on plateaus and seamounts of the Medeleev Ridge. In addition, hydro-acoustic data is presented from the Lomonosov Ridge (Siberian side to close to the North Pole), which support the hypothesis of widespread grounding of ice in the Arctic Ocean, of which the sources are still difficult to determine. The data suggest that thick ice-shelves could have developed from continental ice sheets on a nearly circum-arctic scale, which disintegrated into large icebergs during glacial terminations. On the slopes of the East Siberian Sea and/or on the Arlis Plateau, three northerly-directed ice advances occurred, which are dated by sediment cores using the chronology of brown layers (B1 to B7) as suggested by Stein et al. (2010). According to our age model, the latest advance is slightly older than B2 (MIS-3/4), which has been interpreted as MIS-6 by Jakobsson et al. (2016). A larger well-constrained glaciation has occurred during MIS-4, of which an ice shelf grounded to 900 m on the Arlis Plateau. In the western Arctic Ocean, the oldest datable ice advance has an intra-MIS-5 age. In our data, the chronology of older ice advances along the East Siberian margin are not well constrained but may extend back as far as MIS-16. In contrast, cores from the southern and central Lomonosov Ridge indicate that the youngest ice grounding there has occurred during MIS-6. This grounding was less intense than previous ice-shelf groundings in the area, of which the chronology remains speculative until longer cores become available. Along the Lomonosov Ridge, detailed bathymetric mapping between 81° and 84°N exhibit numerous amphitheatre-like slide scars, under which large amounts of Cenozoic sediments were remobilized into mass-wasting features on both the Makarov and Amundsen sides of the ridge. In areas shallower than 1000 metres, slide scars appear to be associated with streamlined glacial lineations, whereby some of the bedforms have been removed by sliding. It appears that at least some of the mass-wasting events have been triggered by moving and/or loading of grounded ice. Sub-bottom seismic profiling discovered at least three generations of debris-flow deposits near the ridge, which were generated by the slides. In places, the nearly randomly distributed slide scars and debris-flow deposits make it hard to interpret past ice-flow directions from landforms and re-deposited sediments. The pattern allows interpretation of both directions off East Siberia (e.g. Jakobsson et al. 2016) and off Eurasia (e.g. Polyak et al. 2001) towards the central Arctic Ocean. Underneath the slide scars escarpments of up to 400 m in height were formed. Near the southern end of the Lomonosov Ridge the last exhumation of old sediments has occurred during MIS-6. Some of the old sediments recovered in 2014 were studied in more detail (Stein et al., 2016). We can show for the first time that the mid/late Miocene central Arctic Ocean was relatively warm (4-7°C) and ice-free during summer, but sea ice occurred during spring and autumn/winter. A comparison of our biomarker proxy data with Miocene climate simulations seems to favour relatively high late Miocene atmospheric CO2 concentrations. References Dove, D., Polyak, L. & Coakley, B., 2014. Widespread, multi-source glacial erosion on the Chukchi margin, Arctic Ocean. Quat. Sci. Rev. 92, 112–122 Hughes, T. J., Denton, G. H. & Grosswald, M. G., 1977. Was there a late-Würm Arctic ice sheet? Nature, 266, 596–602 Jakobsson, M. et al., 2014. Arctic Ocean glacial history. Quat. Sci. Rev. 92, 40-67 Jakobsson, M., et al., 2016. Evidence for an ice shelf covering the central Arctic Ocean during the penultimate glaciation. Nat. Comm., 7, 10365, DOI: 10.1038/ncomms10365, 1-10 Niessen, F. et al., 2013. Repeated Pleistocene glaciation of the East Siberian continental margin. Nat. Geosci. 6, 842–846 Polyak, L., Edwards, M. H., Coakley, B. J. & Jakobsson, M., 2001. Ice shelves in the Pleistocene Arctic Ocean inferred from glaciogenic deep-sea bedforms. Nature 410, 453–459 Stein, R., Matthiessen, J., Niessen, F., Krylov, A., Nam, S., Bazhenova, E., 2010. Towards a better (litho-) stratigraphy and reconstruction of Quaternary paleoenvironment in the Amerasian Basin (Arctic Ocean), Polarforschung, 79 (2), 97-121 Stein, R., K. Fahl, Schreck, M., Knorr, G., Niessen, F., Forwick, M., Gebhardt, C., Jensen, L., Kaminski, M., Kopf, A., Matthiessen, J., Jokat, W., and Lohmann, G., 2016. Evidence for ice-free summers in the late Miocene central Arctic Ocean. Nature Communications 7:11148, doi:10.1038/ncomms11148.