Biogenic opal isotopes - new tool for evaluation of past nutrient cycling and hydrographic structure in the Pacific Southern Ocean in relation to climate and Antarctic cryosphere evolution
Palaeontology
Final Report Abstract
It was the purpose of the study to generate a calibration dataset for oxygen (δ18O) and silicon (δ30Si) isotopes from diatom and radiolarian (radiol.) silica from the polar Pacific Southern Ocean (SO) and use the established calibrations to investigate the past SO hydrography and nutrient dynamics by measuring and evaluating diatom and radiol. δ18O and δ30Si records documenting the last 30 ka, including the LGM, T-I and the Holocene, as well as the interval 160-110 ka, including MIS6, T-II and MIS5.5. It was planned to study altogether five sediment cores obtained from latitudinal cross-frontal transects across the eastern and western polar South Pacific. However, due to large technical problems with the required mass spectrometers the project´s progress was delayed by several months, and it was only possible to prepare and measure diatom and radiol. isotope samples from one core, PS75-072-4, from the Antarctic Zone. However, to compensate for some of project´s delay, some material of the prepared diatom and radiol. isotope samples was used to measure diatom-bound and radiol.-bound trace element concentrations to test their ability for paleoclimatic reconstructions. For the O and Si isotope calibration 12 diatom (10-20µm fraction) and 56 radiol. samples (5 fractions) were prepared from PS75 and SO225 MUC core top bulk samples, and checked for non-biogenic silicate contamination and species compositions, before being isotopically analyzed. An additional check of the diatom dehydration protocol for O isotope analysis using FTIR and DTA-TG demonstrated that the current dehydration protocol is equally applicable to radiol. silica. The calibration study indicates that the >250µm fraction is the best suitable fraction for radiol. O and Si isotope studies in the polar South Pacific, and that no species-related O and Si isotope effects exist between the main diatom or radiol. species in the isotope samples. Furthermore, study results suggest that absolute diatom δ18O values can be directly compared to absolute radiol. δ18O values, and provide the first temperature-dependent equation describing the relationship between radiol. O isotopes and temperature. The resulting radiol. temperature coefficient of ca. -0.2‰/°C is in close agreement with previous coefficients from diatom O isotope calibration studies. Radiol. and diatom δ30Si values both show a significant inverse correlation to ambient dissolved silica (DSi) concentrations. However, study results indicate that more than one DSi source is required to describe the diatom Si fractionation throughout the whole polar Pacific Ocean. For the O and Si isotope study of sediment core PS75-072-4 altogether 146 diatom (10-20µm) and 145 radiolarian samples (>250µm) were prepared. Non-biogenic silicate contamination and species composition was checked on 31 diatom samples and 24 radiol. samples. Isotopic results show an offset between the diatom and radiol. δ30Si values during the last glacial, which may indicate seasonal surface water stratification due to melting sea-ice. This is supported by elevated and overlapping diatom and radiol. δ18O values. During T-1 overlapping diatom and radiol. δ18O and δ30Si records suggest a combined DSi pool for diatoms and radiol. at that time, possibly related to a deepening of the seasonal mixed-layer. Elevated glacial and distinctly low diatom δ30Si in the mid-Holocene correspond to elevated db-Fe57 concentrations, suggesting an increased input of iron to the core site during these times, thereby reducing the Si:N uptake ratio during diatom growth. While this results in low diatom δ30Si values during the mid-Holocene, characterized by increased upwelling, a strong glacial surface water stratification might limit the DSi supply to such an extent that iron stays the limiting factor in last glacial surface waters. MIS6 diatom δ30Si values are lower than last glacial diatom δ30Si values before they increase sharply at the onset on T-II. This increase in DSi utilization might be related to a decrease in dust-borne iron input, which counteracts the increased upwelling during T-II. Increased upwelling can also explain the overlapping diatom and radiolarian δ30Si records, indicating a shared DSi source. Similar to T-I the radiol. δ18O record shows a continuous increase of ~1.5‰ over T-II. While MIS6 diatom δ18O values are close to radiol. δ18O values, diatom δ18O increases at ca. 130 ka, possibly indicating a decrease in SST, in accordance with temperature records from the North Atlantic realm. While it is possible to explain main features of the diatom and radiol. δ18O and δ30Si records, some features, particularly of the radio. isotope records are not yet fully understood, highlighting the need for more calibration and sediment core studies analyzing both radiolarian and diatom δ18O and δ30Si.
