Isotopic constraints on the sulfur cycle of the Bottom Convective Layer in the Black Sea

Abstract

The Black Sea is the largest euxinic seawater basin with H2S at depths of below 90 m. Previously found that the presence of H2S in water was caused by microbial sulfate reduction. Herein, we report the sulfur isotopic composition of sulfate and sulfide in seawater at stations in the central part of the Black Sea, especially from the Bottom Convective Layer (BCL) located below a water depth of 1750 m. Our study shows that the sulfur isotopic composition of sulfate (δ34S(SO4)) in the surface layer of the oxic zone is +21.1‰ ± 0.1‰ relative to the Vienna Canyon Diablo troilite (VCDT), and does not differ from sulfate in the Mediterranean Sea. In the BCL, the average δ34S(SO4) value is +23.0‰ ± 0.2‰, which does not change vertically or laterally owing to effective convective mixing. The sulfur isotopic composition of H2S varies from −40.0‰ to −41.9‰ between 200 m water depth down to the seafloor with an average δ34S(H2S) value of −40.6‰ ± 0.4‰ for the BCL.Based on the water mass balance and sulfur isotopic composition of sulfate in the relevant water masses affecting the BCL, we have formulated a model to calculate the sulfate reduction rate in the BCL. After evaluating the sulfate and sulfide sources in the BCL, we assumed that the only apparent sulfide source is a bacterial sulfate reduction in the water column, and the main sulfate source is a mixture of water from the Black Sea and the lower Bosporus current with an initial sulfate isotopic composition of +21‰. The weak water exchange in the BCL leads to the largest observed isotopic difference ΔSR = δ34S(SO4) − δ34S(H2S) of 63.6‰ between sulfate and sulfide in the sulfidic water column of the Black Sea. To achieve a steady state with respect to the sulfate isotopic composition +23‰ in the BCL, an average sulfate reduction (SR) rate of 1.1 ± 0.1 nmol L−1 day−1 is required. A steady state relative to sulfur isotopic composition of sulfate in the BCL water was established 4200 years ago. The estimated residence time of H2S was 1.5 times shorter (946 years) than that of sulfate (1390 years). The latter is close to the water renewal time of the BCL (1432 years). Different residence times of sulfide and sulfate in the BCL are explained using their different fluxes from the BCL. Sulfate is mainly removed from the BCL via vertical advection, while a considerable part of the sulfide generated in the BCL (1/3 of total annual production) is probably spent on the sulfidation (FeS → FeS2) of muddy turbidites. The estimated annual H2S production in the BCL is 0.78 ± 0.07 Tg. The developed model predicts that the sulfide concentration in the BCL remains practically unchanged (380 ± 5 μM) over the last 1800 years (after 5700 years from the onset of anoxia).