Testing the efficacy of grain size-sorted sediment for biomarker analysis to reconstruct palaeoclimate

Abstract

Following the announcement of the retirement of the JOIDES Resolution drilling platform, it has become even more important to efficiently utilise the finite resource of marine sediment stored in IODP repositories. Marine sediments processed for inorganic geochemical analysis are often separated into fine (<0.63 µm) and coarse (>0.63 µm) fractions to help isolate benthic and planktonic foraminifera. However, organic matter can be associated with different particle size fractions and may have experienced different transport and diagenetic processes. Previous studies have suggested that sieving sediments into different size fractions does not affect the distribution of isoprenoidal [1,2] and branched glycerol dialkyl glycerol tetraethers (GDGTs) [3]. However, this has never been systematically investigated across a wide range of sample types (e.g., age, depositional environment, thermal maturity). It is also unclear whether size processing affects other lipid biomarker proxies (e.g., leaf waxes, alkenones). Here we test whether processing marine sediments into different size fractions influences lipid distributions by separating sediments into fine (<0.63 µm) and coarse (> 0.63 µm) fractions and comparing these to corresponding bulk un-sieved sediments. Temperature reconstruction using the marine sea surface temperature proxy TEX86 shows relatively minimal deviation (average ±0.12 TEX86 units, or ~2-3 °C) between the bulk un-sieved sediment (i.e,. control) and fine (<63 µm) fraction, suggesting isoGDGTs are well preserved in the fine fraction. In contrast, relatively more variation is seen in the coarse fraction (±0.25 TEX86 units, or >10 °C). We also analysed leaf wax derived n-alkanes extracted from the marine sediment to evaluate the impact on terrestrial biomarkers. The average chain length shows similar deviation in both the fine (±0.21 units) and coarse (±0.21 units) fractions relative to the bulk sediment, suggesting that either fraction is suitable for interpreting first-order changes in vegetation type.  Moving forward, our results suggest that the fine fraction of grain size-sorted sediment yield similar lipid distributions compared to the bulk un-washed sediment. However, coarse fractions often show large deviations from the bulk sediment across different proxies, perhaps making these unsuitable for biomarker-based climate reconstruction.