Because of their small size, ubiquity, and omnivorous habits, Gambusia have been proposed as a suitable indicator species for tracing the bioaccumulation of methyl mercury (MeHg) in Florida Everglades food webs. The USEPA REMAP program collected multiple samples of Gambusia in September 1996 at over 100 marsh sites throughout the Everglades, and analyzed them for MeHg and gut contents. As part of a USGS-USEPA collaboration, an additional subset of 5 Gambusia plus a representative periphyton sample were collected at each site to assess the local and regional ranges in their C, N, and S stable isotope ratios. Analysis of the C, N, and S stable isotope compositions of organisms can be used to trace nutrient sources and to establish relative trophic levels because at each ascending trophic level, there is an increase of about 1 permil in d13C and 3 permil in d15N in the organisms. These isotopic compositions can also provide valuable information about biogeochemical processes in the local environments.

The Gambusia were transported chilled to the lab, frozen, and later filleted to remove the gut contents, heads, scales, and vertebrae. The fillets were dried and processed according to the method described in the other Kendall and others abstract. At 6 sites, duplicate sets of 5 fish each were collected and processed whole; these samples were analyzed before and after being acidified to eliminate carbonate in the bones. The d13C and d15N values of the fillets and whole fish samples (untreated and acidified) are identical within experimental error, indicating that future filleting of Gambusia is unnecessary. All periphyton samples were acidified to remove calcite prior to isotopic analysis. Previous experiments have shown that acidification has no significant effect on the C or N isotopic composition.

Both Gambusia and periphyton had a wide range of isotopic compositions. Gambusia d13C values ranged from -35 to -20 permil, and d15N values ranged from +5 to +14 permil. Periphyton d13C values ranged from -32 to -17 permil, and d15N values ranged from about -2 to +5 permil. Despite the wide range of values among sites, there was extremely little variation among the Gambusia collected at the same site; the average standard deviation for sets of Gambusia at the same site was +/- 0.7 permil for C and +/- 0.5 permil for N. These narrow ranges indicate that all the Gambusia collected at the same site are within the same trophic level and presumably have similar diets. These data also suggest that marsh populations within a single site are sufficiently homogeneous in diet that few members need be sampled to adequately characterize trophic relations.

The d15N values of Gambusia and periphyton show a moderate correlation (r2 = 0.6). On average, the d15N values of Gambusia are 7 to 9 permil higher than those of the co-existing periphyton samples. If periphyton is an important source of N to Gambusia, this difference is equivalent to 2 to 3 trophic levels. However, at almost every site, the d13C values of omnivorous Gambusia are 1 to 4 permil lower than those of the co-existing periphyton samples. This apparent lack of carbon isotope enrichment at higher trophic levels suggests that carbon from bulk periphyton is not the base of the Gambusia and other consumer food chains. Likely alternative candidates for the base of the carbon food chain include plankton, detrital organic material, or some readily assimilated component within the periphyton community which has a much lower d13C value than bulk periphyton.

Contour plots of the isotopic compositions of periphyton and Gambusia show similar regional patterns. There is a broad area of lower d13C values (-28 to -32 permil for Gambusia) that extends southeast from the Everglades Agricultural Area (EAA) to L-67 and then southwest towards Shark Slough, with sharply increasing values to the west. Although the correlation coefficient between Gambusia and periphyton d13C values is less than 0.1, their general spatial patterns are very similar. It is presently unclear what processes are controlling the distribution of d13C values, but we note that the low d13C values roughly correlate with areas where thick peats are present. Several factors could explain these C isotope patterns, including regional differences in the contributions of C3 and C4 plants to local biomass, water turbulence, and redox conditions within the underlying peats which could cause oxidation of peats in some places and methane production in others.

The d15N values of Gambusia and periphyton are highest in the same general area where d13C values are lowest, extending southwest along L-67 towards Shark Slough, with lower d15N values to the northwest and southeast of this broad zone. Some of the areas with the highest d15N values (+3 permil for periphyton and +10 permil for Gambusia) correspond to locations with low sulfate and high MeHg values. One likely explanation for the high d15N values is denitrification in the sediments, resulting in dissolved inorganic N with higher d15N values. This 15N-enriched nitrate and ammonium is then incorporated into the biomass.

The similarities between the distributions of isotopic compositions and the spatial distributions of several parameters measured by the REMAP program are very encouraging, and hold the promise of providing a valuable linkage between the larger scale REMAP program and the more site-specific process-based research at USGS sites. The isotopic compositions may prove to be more cost-effective and reliable indicators of prevailing environmental conditions that generally favor MeHg production than other parameters currently being considered because biomass isotopic compositions are much more difficult to perturb than the more transient concentrations of aqueous species. Hence, the spatial isotope patterns are likely to provide a valuable integration of long-term environmental conditions in the Everglades.

Back to Project Homepage