Temporal and spatial variation of sulfide invasion in eelgrass (Zostera marina) as reflected by its sulfur isotopic composition

Temporal and spatial variation of d34S, total sulfur (TS) concentration, and elemental sulfur concentration (S0) in leaves, roots, and rhizomes of Zostera marina was followed between June 2002 and May 2003 at four locations in Roskilde Fjord and Øresund, Denmark. These were related to temporal changes in sediment sulfide concentrations, sulfur pool size, and sulfur pool d34S. The d34S of Z. marina was most negative in the roots, followed by rhizomes and leaves, indicating that roots were mostly affected by sulfide. A significant relationship between decreasing d34S and increasing TS in the plant tissues indicated that sulfide accumulated in the plant and, furthermore, a positive relation between TS and S0 in the plant suggests that part of the sulfide is reoxidized to S0. There were marked temporal changes in all variables at all sites, but the pattern of change varied between sites. The temporal and spatial heterogeneity in plant d34S, TS, and S0 depended on a variety of factors, such as sediment sulfide concentrations and the below : aboveground biomass ratio of the plants. This suggests that mechanisms of sulfide invasion are complex, and several factors (plant morphology, environmental variables) acting in concert or against each other need to be considered to successfully predict sulfide invasion in seagrasses. Hydrogen sulfide (H2S) is highly toxic to plants in concentrations as low as 1023 to 1025 mol L21 (Howarth and Teal 1979; Raven and Scrimgeour 1997), and sulfide invasion may therefore pose a serious problem to the growth and survival of plants rooted in anaerobic and sulfide-rich sediments. Invasion of sulfide has been directly measured in seagrasses under unfavorable environmental conditions with low water column oxygen concentrations characteristic of eutrophic or special weather conditions (Pedersen et al. 2004; Borum et al. 2005). The isotopic composition of sulfur in seagrasses and wetland plants has been reported to reflect uptake or invasion of sulfide into plant tissues (Carlson and Forrest 1982; Fry et al. 1982). Accordingly, growth under different environmental conditions may be reflected in the plant’s stable sulfur isotope composition (d34S). The overall goal of the present study was to examine differences in d34S of eelgrass (Zostera marina) tissues from different seasons and locations to determine whether environmental conditions were reflected by differences in d34S. Submerged rooted macrophytes growing in marine environments may be supplied with sulfur from three different sources with different ratios of the stable isotopes, 34S and 32S. Leaves can take up sulfate directly from the water column, which has a constant d34S of around +21% (Rees et al. 1978). In the sediment, sulfur will be present as sulfate and as sulfide formed by sulfate reducing bacteria. Owing to bacterial fractionation, the produced sulfide is isotopically lighter (lower d34S) than seawater sulfate (Kaplan et al. 1963). Consequently the remaining, nonreduced sulfate is heavier than seawater sulfate (Kaplan et al. 1963), but reoxidation processes and exchange of sulfate across the sediment–water interface mask this effect in natural systems to various degrees (Böttcher et al. 2004). Sediment sulfide may have d34S as low as 227% in the forms of free and acid volatile sulfide and as low as 242% in the form of pyrite (Kaplan et al. 1963). Owing to this large range in d34S among the different sulfur sources, it should be possible to identify invasion and movement of sulfide in the plant tissues by analyzing the sulfur isotopic composition of the different tissues. Mesocosm experiments have shown elemental sulfur (S0) to accumulate in seagrasses exposed to high sulfide concentrations in the sediment, suggesting that oxidation of sulfide takes place in the plants (Holmer et al. 2005). Hence, sulfide invasion, in addition to changes in d34S, may also be reflected in increased concentrations of compounds derived from the oxidation of sulfide. 1 Present address: Greenland Institute of Natural Resources, Kivioq 2, Box 570, DK-3900 Nuuk, Greenland. Acknowledgments We thank Charlotte Andersen, Thomas Binzer, Peter Larsen, Tritep Vichkovitten, Chris Sørensen, and Ole Ilsøe for help during field work. The isotope analyses were funded through a Natural Environment Research Council Isotope Geosciences Facility grant, IP/778/0902. This research was funded by the European Union projects MedVeg (effects of nutrient release from Mediterranean fish farms on benthic vegetation in coastal ecosystems) (Q5RS-2001-02456) and Monitoring & Managing of European Seagrass Ecosystems (EVK3-CT-2000-00044). Limnol. Oceanogr., 51(5), 2006, 2308–2318 E 2006, by the American Society of Limnology and Oceanography, Inc.