Alkenones and SST of surface sediments (Appendix)

We have analysed alkenones in 149 surface sediments from the eastern South Atlantic in order to establish a sediment-based calibration of the U37K' paleotemperature index. Our study covers the major tropical to subpolar production systems and sea-surface temperatures (SST's) between 0° and 27°C. In order to define the most suitable calibration for this region, the U37K' values were correlated to seasonal, annual, and production-weighted annual mean atlas temperatures and compared to previously published culture and core-top calibrations. The best linear correlation between U37K' and SST was obtained using annual mean SST from 0 to 10 m water depth (U37K' = 0.033 T + 0.069, r**2 = 0.981). Data scattering increased significantly using temperatures of waters deeper than 20 m, suggesting that U37K' reflects mixed-layer SST and that alkenone production at thermocline depths was not high enough to significantly bias the mixed-layer signal. Regressions based on both production-weighted and on actual annual mean atlas SST were virtually identical, indicating that regional variations in the seasonality of primary production have no discernible effect on the U37K' vs. SST relationship. Comparison with published core-top calibrations from other oceanic regions revealed a high degree of accordance. We, therefore, established a global core-top calibration using U37K' data from 370 sites between 60°S and 60°N in the Atlantic, Indian, and Pacific Oceans and annual mean atlas SST (0-29°C) from 0 m water depth. The resulting relationship (U37K' = 0.033 T + 0.044, r**2 = 958) is identical within error limits to the widely used E. huxleyi calibrations of and attesting their general applicability. The observation that core-top calibrations extending over various biogeographical coccolithophorid zones are strongly linear and in better accordance than culture calibrations suggests that U37K' is less species-dependent than is indicated by culture experiments. The results also suggest that variations in growth rate of algae and nutrient availability do not significantly affect the sedimentary record of U37K' in open ocean environments.

Alkenone data of sediment traps off Cape Blanc, NW Africa

We analysed long-chain alkenones in sinking particles and surface sediments from the filamentous upwelling region off Cape Blanc, NW Africa, to evaluate the transfer of surface water signals into the geological record. Our study is based on time-series sediment trap records from 730 m (1990-1991) to 2195-3562 m depth (1988-1991). Alkenone fluxes showed considerable interannual variations and no consistent seasonality. The average flux of C37 and C38 alkenones to the deep traps was 1.9 µg/m**2/d from March 1988 to October 1990 and sevenfold higher in the subsequent year. Alkenone fluxes to the shallower traps were on average twice as high and showed similar temporal variations. The alkenone unsaturation indices UK'37, UK38Me and UK38Et closely mirrored the seasonal variations in sea-surface temperature (weekly Reynolds SST). Time lags of 10-48 days between the SST and unsaturation maxima suggest particle sinking rates of about 80 and 280 m/d for the periods of low and high alkenone fluxes, respectively. The average flux-weighted UK'37 temperature for the 4-year time series of the deeper traps was 22.1°C, in perfect agreement with the mean weekly SST for the same period. This and the comparison with seasonal temperature variations in the upper 100 m of the water column suggests that UK'37 records principally the yearly average of the mixed-layer temperature in this region. A comparison between the average annual alkenone fluxes to the lower traps (2400 µg/m**2/yr) and into the underlying sediments (4 µg/m**2/yr) suggests that only about 0.2% of the alkenones reaching the deep ocean became preserved in the sediments. The flux-weighted alkenone concentrations also decreased considerably, from 2466 µg/gC in the water column to 62 µg/gC in the surface sediments. Such a low degree of alkenone preservation is typical for slowly accumulating oxygenated sediments. Despite these dramatic diagenetic alkenone losses, the UK'37 ratio was not affected. The average UK'37 value of the sediments (0.796±0.010 or 22.3±0.3°C) was identical within error limits to the 4-year average of the lower traps. The unsaturation indices for C38 alkenones and the ratio between C37 and C38 alkenones also revealed a high degree of stability. Our results do not support the hypothesis that UK'37 is biased towards higher values during oxic diagenesis.

Shell preservation of Limacina inflata in surface sediments of the Central and South Atlantic

Over 300 surface sediment samples from the Central and South Atlantic Ocean and the Caribbean Sea were investigated for the preservation state of the aragonitic test of Limacina inflata. Results are displayed in spatial distribution maps and are plotted against cross-sections of vertical water mass configurations, illustrating the relationship between preservation state, saturation state of the overlying waters, and overall water mass distribution. The microscopic investigation of L. inflata (adults) yielded the Limacina dissolution index (LDX), and revealed three regional dissolution patterns. In the western Atlantic Ocean, sedimentary preservation states correspond to saturation states in the overlying waters. Poor preservation is found within intermediate water masses of southern origin (i.e. Antarctic intermediate water (AAIW), upper circumpolar water (UCDW)), which are distinctly aragonite-corrosive, whereas good preservation is observed within the surface waters above and within the upper North Atlantic deep water (UNADW) beneath the AAIW. In the eastern Atlantic Ocean, in particular along the African continental margin, the LDX fails in most cases (i.e. less than 10 tests of L. inflata per sample were found). This is most probably due to extensive "metabolic" aragonite dissolution at the sediment-water interface combined with a reduced abundance of L. inflata in the surface waters. In the Caribbean Sea, a more complex preservation pattern is observed because of the interaction between different water masses, which invade the Caribbean basins through several channels, and varying input of bank-derived fine aragonite and magnesian calcite material. The solubility of aragonite increases with increasing pressure, but aragonite dissolution in the sediments does not simply increase with water depth. Worse preservation is found in intermediate water depths following an S-shaped curve. As a result, two aragonite lysoclines are observed, one above the other. In four depth transects, we show that the western Atlantic and Caribbean LDX records resemble surficial calcium carbonate data and delta13C and carbonate ion concentration profiles in the water column. Moreover, preservation of L. inflata within AAIW and UCDW improves significantly to the north, whereas carbonate corrosiveness diminishes due to increased mixing of AAIW and UNADW. The close relationship between LDX values and aragonite contents in the sediments shows much promise for the quantification of the aragonite loss under the influence of different water masses. LDX failure and uncertainties may be attributed to (1) aragonite dissolution due to bottom water corrosiveness, (2) aragonite dissolution due to additional CO2 release into the bottom water by the degradation of organic matter based on an enhanced supply of organic matter into the sediment, (3) variations in the distribution of L. inflata and hence a lack of supply into the sediment, (4) dilution of the sediments and hence a lack of tests of L. inflata, or (5) redeposition of sediment particles.

Calcareous dinoflagellates in the eastern equatorial Atlantic

Sediments of the Equatorial Atlantic (core GeoB 1105-4) have been investigated for both calcareous dinoflagellates and organic-walled dinoflagellate cysts. In order to determine the ecological affinity of calcareous dinoflagellates the statistical methods of Detrended Correspondence Analysis (DCA) and Redundancy Analysis (RDA) were used. Utilising DCA, distribution patterns of calcareous dinoflagellates have been compared with those of the ecologically much better known organic-walled dinoflagellate cysts. This method was also used to determine which environmental gradients have a major influence on the species composition. By using existing environmental information based on benthic and planktic foraminifera, such as Sea Surface Temperature (SST) and stable oxygen and carbon isotopes, as well as information on the amount of Calcium Carbonate and Total Organic Carbon (TOC) in bottom sediments, these gradients could be interpreted in terms of productivity and glacial-interglacial trends. Using RDA, the direct relationships between the distribution patterns of calcareous dinoflagellates with the above mentioned external variables could be determined. For the studied region and time interval (141-6.7 ka) the calcareous dinoflagellates show enhanced abundances in periods with reduced productivity most probably related to decreased divergence and relatively stratified, oligotrophic oceanic conditions.

Cholesterol and C30 of sediment profile MD90-963

In a deep-sea sediment core recovered from a site lying well above the local lysocline, several organic geochemical proxies, and two different calcite dissolution indicators, are compared in order to evaluate the relationship between calcite dissolution and paleoproductivity over the past three glacial-interglacial cycles. The degree of foraminiferal break-up, and the CaCO3 particle size distribution, both point to significant periods of dissolution every 22 kyr during glacial stages and substages. These dissolution events are concomitant with periods of enhanced primary productivity, as indicated by the abundance of several biomarkers (alkenones, cholesterol, brassicasterol, keto-ol), used here to indicate changes in paleoproductivity. Dissolution fluctuations are highly coherent and in phase with the estimated paleoproductivity variations providing strong evidence that the observed dissolution is due to organic matter remineralization within the sediments rather, than to changes in CO32? concentration in the overlying water column.

Radiolarian-based bransfer function for the estimation of sea surface temperatures in the Southern Ocean (Atlantic Sector)

A new radiolarian-based transfer function for sea surface temperature (SST) estimations has been developed from 23 taxa and taxa groups in 53 surface sediment samples recovered between 35° and 72°S in the Atlantic sector of the Southern Ocean. For the selection of taxa and taxa groups ecological information from water column studies was considered. The transfer function allows the estimation of austral summer SST (December-March) ranging between -1 and 18°C with a standard error of estimate of 1.2°C. SST estimates from selected late Pleistocene squences were sucessfully compared with independend paleotemperature estimates derived from a diatom transfer function. This shows that radiolarians provide an excellent tool for paleotemperature reconstructions in Pleistocene sediments of the Southern Ocean.

Benthic foraminifera off West Africa (1°N to 32°S)

Recent benthic foraminifera (> 125 µm) were investigated from multicorer samples on a latitudinal transect of 20 stations between 1°N and 32°S along the upper slope off West Africa. Samples were selected from a narrow water depth interval, between 1200 and 1500 m, so that changes in water masses are minimized, but changes in surface productivity are important and the only significant environmental variable. Live (Rose Bengal stained) benthic foraminifera were counted from the surface sediment down to a maximum of 12 cm. Dead foraminifera were investigated in the top 5 cm of the sediment only. Five live and five dead benthic foraminiferal assemblages were identified using Q-mode principal component analysis, matching distinct primary productivity provinces, characterized by different systems of seasonal and permanent upwelling. Differences in seasonality, quantity, and quality of food supply are the main controlling parameters on species composition and distribution of the benthic foraminiferal faunas. To test the sensitivity of foraminiferal studies based on the uppermost centimeter of sediment only, a comparative Q-mode principal component analysis was conducted on live and dead foraminiferal data from the top 1 cm of sediment. It has been demonstrated that, on the upper slope off West Africa, most of the environmental signals as recorded by species composition and distribution of the 'total' live and dead assemblages, i.e., including live and dead foraminifera from the surface sediment down to 12 cm and 5 cm, respectively, can be extracted from the assemblages in the top centimeter of sediment only. On the contrary, subsurface abundance maxima of live foraminifera and dissolution of empty tests strongly bias quantitative approaches based on the calculation of standing stocks and foraminiferal numbers in the topmost centimeter.