The significance of radiolarian isotopic signals for paleoceanographic reconstructions
Zusammenfassung der Projektergebnisse
Central to the project was the establishment of a method to generate δ30Si data sets of radiolarian opal. Together with δ18O measurements and the measurements of δ30Si and δ18O of diatom opal such data are crucial for the proper understanding of the variability of physical surface ocean conditions and related nutrient utilization at glacial/interglacial time scales that are prerequisite of the understanding of past carbon cycling and related global climate development. The project successfully accomplished 1) the establishment of a step-wise preparation technique for the separation of diatom and radiolarian opal into different size fractions, 2) the set up of a line allowing for the measurement of δ18O and δ30Si measurements on the same aliquot of biogenic opal, 3) the combined isotopic measurements of radiolarian and diatoms from the same sample and with this 4) the first Pleistocene glacial/interglacial δ30Si and δ18O records obtained from radiolarian opal. For measuring Si isotopes we are following two approaches: 1) we established a method to measure δ18O and δ30Si on the same aliquot of biogenic opal. For this, we used the high-temperature laser fluorination of samples in a BrF5 atmosphere to produce oxygen and SiF4 gas, followed by analysis o f a gas source isotope ratio massspectrometer (IRMS) similar to the method established by Leng and Sloane (2008); 2) we are going to develop a method for measuring very small sample sizes (of < 1 mg for plankton and surface sediment samples) where the silicon dioxide is converted to cesium hexafluorosilicate by dissolution in hydrogen fluoride and the addition of cesium chloride similar to the analytical method described by Brezezinskt et al. (2006). The development and implementation of the methods has been done in close cooperation with H. Friedrichsen (Berlin), B. Chapligin and H. Meyer (AWI Potsdam). To our knowledge, the AWl is the only laboratory apart from NERC (Isotope Geosciences Laboratory, British Geological Survey, Nottingham, UK) able to analyse δ18O and δ30Si of the same aliquot from biogenic samples but with a smaller sample amount (1.5 mg instead of 7 mg). The development and implementation of the methods has been done in close cooperation with H. Friedrichsen (Berlin), B. Chapligin and H. Meyer, (AWI-Potsdam). During the project, we have separated diatom and radiolarian opal for δ30SI andδ18O measurements from two sediment cores PS1768-8 and PS1778-5 from in the Antarctic and Polar Front zones of the Atlantic sector, respectively and 20 surface sediments samples recovered from the Atlantic, Indian and Pacific sectors of the Southern Ocean. We have accomplished 164 silicon and oxygen Isotopic measurements of radiolarian fractions >250 μm and >125 μm and of the diatom fraction 10-20 μm from both sediment cores, including replicates and triplicates to test the accuracy of the method. The obtained δ18Odiat and δ18Orad values are in the range of diatom silicon and oxygen isotope values presented by other studies. In Core PS1768-8 a pronounced glacial offset between silicon Isotope values of radiolarians and diatoms that occurs during the last glacial when the site was seasonal covered by sea ice may point to a seasonal developed sea surface stratification. This is supported by the δ18Odiat and δ18Orad records of core PS1768-8, which exhibit relative depleted glacial values. The comparison of the δ30Sidiat signals of Core PS1768-8 and PS1778-5 shows a distinct offset between the δ30Sidiat values during the last glacial and Termination I, which points to a northward shift of the biogenic silica belt to the area of the Polar Front Zone.
