Millimetre-scale biomarker heterogeneity in lacustrine shale identifies the nature of signal-averaging and demonstrates anaerobic respiration control on organic matter preservation and dolomitization

Mm-scale and/or lm-scale in situ inorganic geochemical analyses of organic-rich, laminated fine-grained sediments increasingly offer decadal or even seasonal perspectives of ecological or environmental dynamics. Organic geochemical studies, in contrast, seldom explicitly consider mm-scale variability, so that it remains unclear to what extent organic geochemical proxy-based reconstructions are a product of signal-averaging of higher-resolution variability. Here we assess the 2-8 mm-scale variability in organic geochemical proxies for depositional environment and organic source, utilising an organic-rich, low maturity and well-laminated shale from the Eocene Shahejie Formation, Dongying Depression, Bohai Bay Basin. We identify mm-scale (approximately decadal-scale) compositional variability that is comparable in magnitude to vertical variability at the m-scale (approximately millennial-scale), test mul-tiple relationships between key biomarker and mineral proxies at the mm-scale, and assess likely impli-cations for palaeowater conditions, anaerobic heterotrophic microbial activity, early diagenetic dolomitization, and organic matter preservation. Our results suggest that the decadal-scale oscillation of organic carbon content is a result of varying degrees of anaerobic heterotrophic microbial reworking with periodic euxinia in the deep photic zone. In contrast to previous models based on analysis of bulk samples, our results suggest that anaerobic respiration during intervals of more intense/persistent eux-inia is likely accompanied by more organic matter loss. Variable low-light euxinia seems to be a modifier to the net OM content. The analysis of bulk samples may hide important relationships at shorter temporal scales which are key to reconstructing palaeoenvironments and the operation of key controls on, for example, the carbon cycle.(c) 2023 Elsevier Ltd. All rights reserved.