Iron isotope signature of magnetofossils and oceanic biogeochemical changes through the Middle Eocene Climatic Optimum

Magnetotactic bacteria (MTB) intracellularly precipitate magnetite (Fe3O4) crystals that can be preserved in the geological record. When MTB die, the so-called magnetofossils constitute valuable proxies for paleoenvironmental reconstructions and are suspected to represent some of the oldest traces of biomineralization on Earth. Yet, the biogenicity of putative magnetofossils found in ancient terrestrial and extra-terrestrial samples is still largely debated and their significance for past climate still holds uncertainties. Here we studied a sedimentary sequence from the Indian Ocean (ODP Hole 711A) recording the Middle Eocene Climatic Optimum (MECO) through which a magnetofossil-rich interval was deposited. We investigated for the first time the potential of Fe isotopes as a biosignature in magnetofossils and thoroughly describe MECO related paleoenvironmental disruptions based on major and trace element concentrations. Bulk sediment Fe isotopes showed limited variations, with delta Fe-56 around -0.13 +/- 0.04 parts per thousand (n = 24), linked to detrital iron rather than MTB activity. Hence, a sequential chemical extraction protocol was applied to determine the specific composition of magnetite. We discuss analytical biases related to this protocol (i.e. partial phyllosilicate and Mn-oxide leaching) and apply corrections to the data. Outside the magnetofossil-rich interval, Fe isotope compositions of oxides (mainly biotic and/or abiotic magnetites and possibly Fe coprecipitated with Mn-oxides) display a small range averaging -0.54 +/- 0.05 parts per thousand, and are interpreted as reflecting dominantly hydrothermal contribution, a conclusion also supported by prominent Eu anomaly. In contrast, the magnetofossil-rich interval shows larger delta(56)Fevariability in oxides, from -0.12 to -0.94 parts per thousand, decreasing upwards in the stratigraphic section. This interval likely records enhanced Fe supply from atmospheric fallout, increase in biological productivity (illustrated by increased Ba accumulation rate) and subsequent development of ferruginous conditions in the sediment porewater. Covariations of Fe isotope compositions and Mn/Fe ratios can be explained by a vertical migration of a redox front and associated diagenetic modifications. Precipitation of barite (BaSO4) in the sediments after organic matter decay probably favored the preservation of magnetofossils by decreasing SO42- concentration in porewaters and subsequent H2S production, which usually dissolve magnetite in the sulfidic zone. Finally, we model the evolution of porewater fluid and estimate Fe isotope fractionation between magnetofossils and fluid to Delta(56)Femag-Fe(II)aq 0.1-0.3 parts per thousand, a value significantly different from abiotic magnetite fractionation (similar to 1.5 parts per thousand). Contrasting with recent results on MTB laboratory culture, no mass independent fractionation of Fe isotopes was observed in the present study. Nevertheless, the diverse geochemical proxies presented here provide important constraints on paleoclimate and magnetofossil biogenicity evaluation. (C) 2021 Elsevier Ltd. All rights reserved.