Evaluating the Reliability of Grain-size-Sorted Sediments for Organic Biomarker Analysis

Organic and inorganic geochemical proxies are fundamental for interpreting past climate variability. For inorganic analysis, marine sediments are often size sorted to isolate foraminifera, while organic analysis employs subsamples of bulk sediment. Although previously sieved sediments could be used for biomarker analysis, this approach and potential methodological bias has not been assessed in pre-Holocene sediments. Here we explore lipid biomarkers (isoGDGTs, brGDGTs, and n-alkanes) across grain-size fractions in Cenozoic-aged marine sediments, assessing whether sieved sediments can be used to reconstruct paleoenvironmental change. We analyse biomarkers in fine (<63 µm) and coarse (>63 µm) grain-size fractions and compare proxies (e.g. TetraEther indeX of tetraethers containing 86 carbon atom [TEX86], methylation of branched tetraethers [MBT5ME]) with those in bulk sediments. Our findings reveal a strong linear correlation between TEX86 in size-fractions and bulk sediment (r2 = 0.85-0.93), indicating sea surface temperature reconstructions are viable using previously sieved sediments. MBT5ME shows a weaker yet significant correlation between fine and bulk sediment (r2 = 0.70-0.81). In fine fractions, n-alkanes typically reflect the bulk sediment composition, whereas coarse sediments consistently underestimate bulk sediment chain length metrics, suggesting variable organo-mineral associations. Our results demonstrate GDGT-based temperature proxies can be analysed using sieved sediments, whilst n-alkanes can be influenced by organo-mineral associations and contamination during processing. Our study suggests that archives of sediment fractions can be resurrected for biomarker analysis, encouraging a re-examination of global stores of processed samples. We additionally assessed contamination from traditional processing techniques, determining sieving sediments into plastic beakers introduces contamination.