Abundance of copepods from multinet samples during POLARSTERN cruise ANT-XX/1

Abundance and species composition of copepods were studied during the expedition ANT XXI/1 on a latitudinal transect in the eastern Atlantic from 34°49.5' N to 27°28.1' S between 2-20 November 2002. Stratified zooplankton tows were carried out at 19 stations with a multiple opening-closing net between 300 m water depth and the surface. Cyclopoid and calanoid copepods showed similar patterns of distribution and abundance. Oithona was the most abundant cyclopoid genus, followed by Oncaea. A total of 149 calanoid copepod species were identified. Clausocalanus was by far the most abundant genus, comprising on average about 45% of all calanoids, followed by Calocalanus (13%), Delibus (9%), Paracalanus (6%), and Pleuromamma (5%). All other genera comprised on average less than 5% each, with 40 genera less than 1%. The calanoid copepod communities were distinguished broadly in accordance with sea surface temperature, separating the subtropical from the tropical stations, and were largely determined by variation in species composition and species abundance. Nine Clausocalanus species were identified. The most numerous Clausocalanus species was C. furcatus, which on average comprised half of all adult of this genus. C. pergens, C. paululus, and C. jobei, contributed an average of 19%, 9%, and 9%, respectively. The Clausocalanus species differed markedly in their horizontal and vertical distributions: C. furcatus, C. jobei, and C. mastigophorus had widespread distributions and inhabited the upper water layers. Major differences between the species were found in abundance. C. paululus and C. arcuicornis were biantitropical and were absent or occurred in very low numbers in the equatorial zone. C. parapergens was found at all stations and showed a bimodal distribution pattern with maxima in the subtropics. C. pergens occurred in higher numbers only at the southern stations, where it replaced C. furcatus in dominance. In contrast to the widespread species, the bulk of the C. paululus, C. arcuicornis, C. parapergens, and C. pergens populations was concentrated in the colder, deeper water layers below the thermocline, thereby avoiding the warm surface waters. C. lividus was found only at the most northern and C. ingens only at the most southern stations. Both species were found almost exclusively in the upper 50 m. The distinct differences in abundance and horizontal and vertical distribution suggest a strong ecological differentiation among the Clausocalanus species.

Estimation of sea surface temperature from sediment core PS1768-8

The quantitative diatom analysis of 218 surface sediment samples recovered in the Atlantic and western Indian sector of the Southern Ocean is used to define a base of reference data for paleotemperature estimations from diatom assemblages using the Imbrie and Kipp transfer function method. The criteria which justify the exclusion of samples and species out of the raw data set in order to define a reference database are outlined and discussed. Sensitivity tests with eight data sets were achieved evaluating the effects of overall dominance of single species, different methods of species abundance ranking, and no-analog conditions (e.g., Eucampia Antarctica) on the estimated paleotemperatures. The defined transfer functions were applied on a sediment core from the northern Antarctic zone. Overall dominance of Fragilariopsis kerguelensis in the diatom assemblages resulted in a close affinity between paleotemperature curve and relative abundance pattern of this species downcore. Logarithmic conversion of counting data applied with other ranking methods in order to compensate the dominance of F. kerguelensis revealed the best statistical results. A reliable diatom transfer function for future paleotemperature estimations is presented.

Diatom abundance in surface sediments of the northern North Pacific

In order to map the modern distribution of diatoms and to establish a reliable reference data set for paleoenvironmental reconstruction in the northern North Pacific, a new data set including the relative abundance of diatom species preserved in a total of 422 surface sediments was generated, which covers a broad range of environmental variables characteristic of the subarctic North Pacific, the Sea of Okhotsk and the Bering Sea between 30° and 70°N. The biogeographic distribution patterns as well as the preferences in sea surface temperature of 38 diatom species and species groups are documented. A Q-mode factor analysis yields a three-factor model representing assemblages associated with the Arctic, Subarctic and Subtropical water mass, indicating a close relationship between the diatom composition and the sea surface temperatures. The relative abundance pattern of 38 diatom species and species groups was statistically compared with nine environmental variables, i.e. the summer sea surface temperature and salinity, annual surface nutrient concentration (nitrate, phosphate, silicate), summer and winter mixed layer depth and summer and winter sea ice concentrations. Canonical Correspondence Analysis (CCA) indicates 32 species and species groups have strong correspondence with the pattern of summer sea surface temperature. In addition, the total diatom flux data compiled from ten sediment traps reveal that the seasonal signals preserved in the surface sediments are mostly from spring through autumn. This close relationship between diatom composition and the summer sea surface temperature will be useful in deriving a transfer function in the subarctic North Pacific for the quantitative paleoceanographic and paleoenvironmental studies. The relative abundance of the sea-ice indicator diatoms Fragilariopsis cylindrus and F. oceanica of >20% in the diatom composition is used to represent the winter sea ice edge in the Bering Sea. The northern boundary of the distribution of F. doliolus in the open ocean is suggested to be an indicator of the Subarctic Front, while the abundance of Chaetoceros resting spores may indicate iron input from nearby continents and shelves and induced productivity events in the study area.

Diatom abundance in surface sediments of the Southern Ocean

Based on the quantitative analysis of diatom assemblages preserved in 274 surface sediment samples recovered in the Pacific, Atlantic and western Indian sectors of the Southern Ocean we have defined a new reference database for quantitative estimation of late-middle Pleistocene Antarctic sea ice fields using the transfer function technique. The Detrended Canonical Analysis (DCA) of the diatom data set points to a unimodal distribution of the diatom assemblages. Canonical Correspondence Analysis (CCA) indicates that winter sea ice (WSI) but also summer sea surface temperature (SSST) represent the most prominent environmental variables that control the spatial species distribution. To test the applicability of transfer functions for sea ice reconstruction in terms of concentration and occurrence probability we applied four different methods, the Imbrie and Kipp Method (IKM), the Modern Analog Technique (MAT), Weighted Averaging (WA), and Weighted Averaging Partial Least Squares (WAPLS), using logarithm-transformed diatom data and satellite-derived (1981-2010) sea ice data as a reference. The best performance for IKM results was obtained using a subset of 172 samples with 28 diatom taxa/taxa groups, quadratic regression and a three-factor model (IKM-D172/28/3q) resulting in root mean square errors of prediction (RMSEP) of 7.27% and 11.4% for WSI and summer sea ice (SSI) concentration, respectively. MAT estimates were calculated with different numbers of analogs (4, 6) using a 274-sample/28-taxa reference data set (MAT-D274/28/4an, -6an) resulting in RMSEP's ranging from 5.52% (4an) to 5.91% (6an) for WSI as well as 8.93% (4an) to 9.05% (6an) for SSI. WA and WAPLS performed less well with the D274 data set, compared to MAT, achieving WSI concentration RMSEP's of 9.91% with WA and 11.29% with WAPLS, recommending the use of IKM and MAT. The application of IKM and MAT to surface sediment data revealed strong relations to the satellite-derived winter and summer sea ice field. Sea ice reconstructions performed on an Atlantic- and a Pacific Southern Ocean sediment core, both documenting sea ice variability over the past 150,000 years (MIS 1 - MIS 6), resulted in similar glacial/interglacial trends of IKM and MAT-based sea-ice estimates. On the average, however, IKM estimates display smaller WSI and slightly higher SSI concentration and probability at lower variability in comparison with MAT. This pattern is a result of different estimation techniques with integration of WSI and SSI signals in one single factor assemblage by applying IKM and selecting specific single samples, thus keeping close to the original diatom database and included variability, by MAT. In contrast to the estimation of WSI, reconstructions of past SSI variability remains weaker. Combined with diatom-based estimates, the abundance and flux pattern of biogenic opal represents an additional indication for the WSI and SSI extent.

(Table S1) Diatom abundance according to ice cycles in ANDRILL AND1-1B drill core

The influence of Antarctica and the Southern Ocean on Late Pliocene global climate reconstructions has remained ambiguous due to a lack of well-dated Antarctic-proximal, paleoenvironmental records. Here we present ice sheet, sea-surface temperature, and sea ice reconstructions from the ANDRILL AND-1B sediment core recovered from beneath the Ross Ice Shelf. We provide evidence for a major expansion of an ice sheet in the Ross Sea that began at ~3.3 Ma, followed by a coastal sea surface temperature cooling of ~2.5°C, a stepwise expansion of sea ice, and polynya-style deep mixing in the Ross Sea between 3.3 and 2.5 Ma. The intensification of Antarctic cooling resulted in strengthened westerly winds and invigorated ocean circulation. The associated northward migration of Southern Ocean fronts has been linked with reduced Atlantic Meridional Overturning Circulation by restricting surface water connectivity between the ocean basins, with implications for heat transport to the high latitudes of the North Atlantic. While our results do not exclude low-latitude mechanisms as drivers for Pliocene cooling, they indicate an additional role played by southern high-latitude cooling during development of the bipolar world.