High levels of toxic methylmercury in fish and other biota in the Everglades is a serious threat to the vitality of this ecosystem, and poses a potential human health concern. The biogeochemistry of sulfur in the freshwater Everglades is important, because the reduction of sulfate to sulfide in anoxic marsh sediments is linked to methylmercury production through processes mediated by sulfate-reducing bacteria. The objectives of this project, in collaboration with the Aquatic Cycling of Mercury in the Everglades group were to: (1) examine the distribution of sulfur species in the Everglades spatially and temporally, (2) determine the relation, if any, in the observed distribution of sulfur to mercury methylation, and (3) examine the source(s) of sulfur to the Everglades using sulfur isotope geochemistry as a tracer tool. Our results to date indicate that excess sulfate enters the northern Water Conservation Areas (WCA) in water discharged from canals draining the Everglades Agricultural Area (EAA). This excess sulfate probably originates from agricultural sulfur used in the EAA. Studies of dated cores from WCA-2A show that the influx of excess sulfur began in the early part of this century, concomitant with an influx of excess phosphorus (Craft and Richardson, 1993). The excess sulfate stimulates bacterial sulfate reduction over large areas, and has a significant but complex relation to the extent and distribution of methylmercury in the ecosystem (Gilmour and others, 1998).

Studies of sediment cores collected in WCA-2A indicate trends of increasing concentrations of total sulfur (TS) in sediments during the last century (Bates and others, 1998). At a site near the discharge of water from the Hillsboro Canal, TS concentrations in sediments below 25-30 cm depth held fairly steady at about 0.7 percent (dry wt.). Above this depth, corresponding to the early part of this century, TS concentrations in the core increase sharply to concentrations of about 1.5 percent. This increase in TS correlates closely with increases in total phosphorus observed at this site, and suggests that the load of sulfur and phosphorus to the sediments at this location have increased during the last century. A similar increase in TS near the surface is also observed at a site in the center of WCA-2A, although no concomitant increase in total phosphorus was observed. This suggests that phosphorus entering the ecosystem is quickly sequestered by aquatic macrophytes near canal discharge sites; in contrast, sulfur is an element having greater mobility in this wetland environment. Accumulation rates of TS in surface sediments are about 5 times higher at the Hillsboro Canal site compared to the site in the center of WCA-2A. The principal form of TS in sediments at all marsh sites is organic sulfur, reflecting the reaction of bacterially produced sulfide with organic matter in the sediments. Relatively low levels of iron in the ecosystem is apparently limiting the formation of disulfide minerals in the sediments.

Sulfate concentrations in surface water from WCA-2A are typically 20 to 60 mg/L, with values up to 120 mg/L, near points of water discharge from the Hillsboro Canal. In contrast, surface-water sulfate concentrations from the center of pristine WCA-1A are typically < 1 mg/L. Porewater sulfide concentrations are also elevated in WCA-2A (concentrations often > 2,500 µg/L) compared to concentrations of < 0.1 µg/L in the center of WCA-1A. The high levels of sulfide in porewaters from WCA-2A reflect high rates of sulfate reduction in the anoxic sediments (porewater Eh values typically in the range of -200 mv), stimulated by the high levels of sulfate in the wetland. High levels of surface-water sulfate and porewater sulfide are prevalent throughout WCA-2A, attesting to the general mobility of sulfur within the wetland ecosystem. In contrast to phosphorus contamination, which is confined at present to a broad swath of WCA-2A bordering the Hillsboro Canal, sulfur contamination extends to the center of WCA-2A. Sulfate concentrations in surface water and sulfide concentrations in porewater in the center of WCA-2A are nearly as high as those observed at sites located near points of water discharge from the Hillsboro Canal. Sulfur contamination has also been observed in WCA-3A, but to a lesser degree than in WCA-2A. At a site adjacent to the Miami Canal in WCA-3A, surface-water sulfate concentrations of 30 mg/L, and porewater sulfide concentrations of about 500 µg/L were observed. At a site farther south near the center of WCA-3A (site 3A-15), surface-water sulfate concentrations ranged from 2 to 12 mg/L, and porewater sulfide concentrations varied between 1 and 100 µg/L. Still farther south at several sites in WCA-3A and 3B near Tamiami Trail, surface-water sulfate concentrations were typically < 1 mg/L and porewater sulfide concentrations were < 0.1 ppb. The following pattern emerges from these results: (1) heavy sulfur contamination in all of WCA-2A, but with the highest levels near sites of water discharge from the Hillsboro Canal; (2) generally lower levels of sulfur contamination in WCA-3A compared to WCA-2A, and with the levels of sulfur contamination decreasing toward the south, away from the Miami Canal; and (3) background levels of sulfur for the Everglades at sites in the center of WCA-1A and in the far southern reaches of WCA-3A and B.

An extensive survey of sulfate concentrations and isotopic composition (³⁴S) in: (1) surface-water samples from Lake Okeechobee, canals draining the EAA, and marshes of the WCA, (2) ground-water samples from the WCA-and the EAA, and (3) rainfall. The survey was to determine the major source(s) of sulfur contamination to the northern Everglades (Bates and others, 1999). Rainwater has far to low a concentration (< 1 mg/L) and isotopic composition (³⁴S = +2 to +11) to account for the sulfur contamination in the marshes (sulfate concentrations of 20-60 mg/L and typical ³⁴S values of +20 to +25). Similarly, ground-water samples from sites in WCA-2A and the EAA (except for ground water deeper than 8 m) have been eliminated as sources of sulfur contamination, either because of very low sulfate concentrations, and/or sulfur isotope compositions incompatible with those observed in the marshes. In contrast, canal water draining the EAA has variable, but usually high sulfate concentrations (35-180 mg/L), and an isotopic composition compatible with sulfate present in the marshes. Although these canals originate at Lake Okeechobee, the lake water has sulfate concentrations too low (< 10 mg/L) to account for most of the sulfate in the canals. Furthermore, the sulfate in the canals of the EAA has an isotopic composition similar to that of agricultural sulfur used on fields in the EAA (³⁴S +16). Deep groundwater (> 8 m) below the EAA also has sulfate concentrations and a sulfur isotopic composition consistent with the sulfur contamination in the WCA. It may be that both agricultural sulfur and deep ground-water contribute to the very high sulfur concentrations observed in the EAA canals and the northern WCA.

In conclusion, high concentrations of sulfate originating from canals draining the EAA are resulting in extensive sulfur contamination of the WCA. Sulfur contamination in WCA-2A appears to have started during the early part of this century, concomitant with phosphorus contamination here. High concentrations of sulfate in WCA-2A are stimulating bacterial sulfate reduction, but the high levels of sulfide in porewaters here apparently inhibit mercury methylation (Hayes and others, 1998). Lower levels of sulfate contamination in the center of WCA-3A apparently increase sulfate reduction and mercury methylation without the inhibitory effects of excess sulfide on mercury methylation.

REFERENCES

Bates, A.L., Orem, W.H., Harvey, J.W., and Spiker, E.C., 1999, Sources of sulfate to the northern Everglades, Florida, USA: Nature, in review.

Bates, A.L., Spiker, E.C., and Holmes, C.W., 1998, Speciation and isotopic composition of sedimentary sulfur in the Everglades, Florida, USA: Chemical Geology, v. 146, p. 155-170.

Craft, C.B, and Richardson, C.J., 1993, Peat accretion and phosphorus accumulation along a eutophication gradient in the northern Everglades: Biogeochemistry, v. 22, p. 133-156.

Gilmour, C.C., Riedel, G.S., Ederington, M.C., Bell, J.T., Benoit, J.M., Gill, G.A., and Stordal, M.C., 1998, Methylmercury concentrations and production rates across a trophic gradient in the northern Everglades: Biogeochemistry, v. 40, p. 327-345.

Hayes, A., Gilmour, C.C., and Benoit, J.M., 1998, Controls on the distribution of methylmercury in the Florida Everglades. AGU Meeting: Boston, Program with Abstracts.