Mass-dependent sulfur isotope fractionation during reoxidative sulfur cycling: A case study from Mangrove Lake, Bermuda

Abstract

The multiple sulfur isotope composition of porewater sulfate from the anoxic marine sapropel of Mangrove Lake, Bermuda was measured in order to establish how multiple sulfur isotopes are fractionated during reoxidative sulfur cycling. The porewater-sulfate δ34S and Δ33S dataset exhibits the distinct isotopic signatures of microbial sulfate reduction and sulfur reoxidation. We reproduced the measurements with a simple diagenetic model that yielded fractionation factors for net sulfate removal of between −29.2‰ and −32.5‰. A new approach to isotopic modeling of the sulfate profiles, informed by the chemistry of sulfur intermediate compounds in Mangrove Lake, reveals that sulfate reduction produces a relatively small intrinsic fractionation and that an active reoxidative sulfur cycle increases the fractionation of the measured values. Based on the model results, the reoxidative cycle of Mangrove Lake appears to include sulfide oxidation to elemental sulfur followed by the disproportionation of the elemental sulfur to sulfate and sulfide. This model also indicates that the reoxidative sulfur cycle of Mangrove Lake turns over from 50 to 80% of the sulfide produced by microbial sulfate reduction. The Mangrove Lake case study shows how sulfur isotope fractionations can be separated into three different “domains” in Δ33S–δ34S space based on their ability to resolve reductive and reoxidative sulfur transformations. The first domain that differentiates reductive and reoxidative sulfur cycling is well illustrated by previous studies and requires 34S–32S fractionations more negative than ≈−70‰, beyond the fractionation limit of microbial sulfate reduction at earth surface temperatures. The second domain that distinguishes reductive and reoxidative processes is between 34S–32S fractionations of −40‰ and 0‰, where the 33S–32S fractionations of sulfate reduction and reoxidation are significantly different. In the remaining domain (between 34S–32S fractionations −70‰ and −40‰), the similarity of the multiple sulfur isotope signals from microbial sulfate reduction and disproportionation means that the two processes cannot be discriminated from each other.