Description: Using the sea ice proxy IP25 and phytoplankton-derived biomarkers (brassicasterol and dinosterol) Arctic sea-ice conditions were reconstructed for Marine Isotope Stage (MIS) 3 to 1 - with special emphasis on the Last Glacial Maximum (LGM) - in sediment cores from the northern Barents Sea continental margin across the Central Arctic Ocean to the Southern Mendeleev Ridge. Our results suggest more extensive sea-ice cover than present-day during latter part of MIS 3, increasing sea-ice growth during MIS 2 and decreased sea-ice cover during the last deglacial. The summer ice edge remained north of the Barents Sea even during extremely cold (i.e., Last Glacial Maximum (LGM)) as well as warm periods (i.e., Bølling-Allerød). During the LGM, the western Svalbard margin and the northern Barents Sea margin areas were characterized by high concentrations of both IP25 and phytoplankton biomarkers, interpreted as a productive ice-edge situation, caused by the inflow of warm Atlantic Water. In contrast, the LGM Central Arctic Ocean (north of 84°N) was covered by thick permanent sea ice throughout the year with rare break-up, indicated by zero or near-zero biomarker concentrations. The spring/summer sea-ice margin significantly extended southwards to the Laptev Sea shelf (southern Lomonosov Ridge) and southern Mendeleev Ridge during the LGM. Our proxy reconstructions are very consistent with published model results based on the North Atlantic/Arctic Ocean Sea Ice Model (NAOSIM).

Description: Here, we present a first (low-resolution) biomarker sea-ice proxy record from the High Arctic (southern Lomonosov Ridge), going back in time to about 60 ka (MIS 3 to MIS 1). Variable concentrations of the sea-ice diatom specific highly branched isoprenoid (HBI) with 25 carbon atoms (“IP25“), in combination with the phytoplankton biomarker brassicasterol, suggest variable seasonal sea-ice coverage and open-water productivity during MIS 3. During most of MIS 2, the spring to summer sea-ice margin significantly extended towards the south, resulting in a drastic decrease in phytoplankton productivity. During the Early Holocene Climate Optimum, brassicasterol reached its maximum, interpreted as signal for elevated phytoplankton productivity due to a significantly reduced sea-ice cover. During the mid-late Holocene, IP25 increased and brassicasterol decreased, indicating extended sea-ice cover and reduced phytoplankton productivity, respectively. The HBI diene/IP25 ratios probably reached maximum values during the Bølling-Allerød warm period and decreased during the Holocene, suggesting a correlation with sea-surface temperature.

Description: This review paper focusses on reconstructions of the long- and short-term history of past Arctic Ocean sea-ice cover. Based on commonly used sedimentological, geochemical and micropaleontological proxies (ice-rafted debris (IRD), mineralogical composition of terrigenous sediment frac¬tion, and abundances of specific diatoms and foraminifers), three examples of reconstructions of glacial history, sea-ice cover and surface-water character¬istics are presented and discussed: (1) the onset Arctic Ocean sea-ice cover near 47 Ma and its long-term variability through Cenozoic times; (2) the Quaternary glacial/interglacial variability in Arctic Ocean ice-rafting and its relationship to sea-ice and ice-sheet history; and (3) Last Glacial Maximum (LGM), Deglacial to Holocene changes in Arctic Ocean sea-ice cover and ice-sheet decay. In the second part of this paper we concentrate on Arctic Ocean sea-ice reconstructions, using a recently developed biomarker approach that is based on the determination of sea-ice diatom-specific highly-branched isoprenoids with 25 carbon atoms (IP25; BELT et al. 2007) and IP25 in combination with phytoplankton biomarkers (PIP25; MÜLLER et al. 2009, 2011). The diene/ IP25 ratio might give additional information about sea-surface temperature (SST) in the low temperature Arctic environment (FAHL & STEIN 2012). The high potential of these novel biomarker proxies to improve reconstructions of paleo-sea-ice cover and its variability through time is demonstrated in three examples: (a) the sea-ice variability in Fram Strait over the last 30 ka, (b) the deglacial/Holocene variability of central Arctic sea-ice cover with special emphasis on the Younger Dryas Cooling Event, and (c) a comparison of historical sea-ice observations off northern Iceland over the last millennium and a corresponding high-resolution IP25 record. In a pilot study carried out in a sediment core from the Barents Sea conti-nental slope we were able to prove for the first time that IP25 is even pre-served in sediments as old as 130 to 150 ka (MIS 6), i.e., IP25 can be used for reconstruction of sea-ice variability during older glacial/interglacial intervals (MIS 6/MIS 5). In order to establish the IP25 approach as a key proxy for reconstruction of past Arctic Ocean sea-ice conditions, more basic information about produc-tion, degradation and preservation/burial of the IP25 signal is still needed. Furthermore, the hypothesis that the diene/IP25 ratio might be used as reliable proxy for SST reconstructions in the low temperature Arctic environments has to be verified by a ground-truth study including the IP25 and diene data as well as independent SST proxies like alkenone-derived SST. All these data should be obtained in future investigations of sea-ice, water column, and sediment-trap samples as well as surface sediments and sediment cores with large spatial coverage from different environments of the entire Arctic Ocean.

Description: To evaluate the present sea-ice changes in a longer-term perspective the knowledge of sea-ice variability on pre-industrial and geological time scales is essential. For the interpretation of proxy reconstructions it is necessary to understand the recent signals of different sea-ice proxies from various regions. We present 260 new sediment surface samples collected in the (sub-) Arctic Oceans that were analysed for specific sea-ice (IP25) and open-water phytoplankton biomarkers (brassicasterol, dinosterol, HBI III). This new biomarker dataset was combined with 615 previously published biomarker surface samples into a pan-Arctic database. The resulting pan-Arctic biomarker and sea-ice index (PIP25) database shows a spatial distribution correlating well with the diverse modern sea-ice concentrations. We find correlations of PBIP25, PDIP25 and PIIIIP25 with spring and autumn concentrations. Similar correlations with modern sea-ice concentrations are observed in Baffin Bay. However, the correlations of the PIP25 indices with modern sea-ice concentrations differ in Fram Strait from those of the (sub-) Arctic dataset, which is likely caused by region-specific differences in sea-ice variability, nutrient availability and other environmental conditions. The extended (sea-ice) biomarker database strengthens the validity of biomarker sea-ice reconstructions in different Arctic regions and shows how different sea-ice proxies combined may resolve specific seasonal sea-ice signals.