Many lake studies have found good correlations between the mercury (Hg) concentrations in fish and the relative trophic positions of fish as determined by analyses of gut contents and (or) stable isotopes. But is this true in dynamic marsh ecosystems such as the Everglades? Based on the extensive REMAP (Regional Environmental Monitoring and Assessment Program) assessment of the Everglades ecosystem, Stober et al. (2001) concluded that spatial (especially N-S) differences in “food web structure” are a likely cause of spatial differences in Hg concentrations in mosquitofish. Food web structure is a general term that encompasses differences in food chain length, food web complexity, and food web base. The extensive isotope and gut contents datasets assembled by several agencies can be used to evaluate this hypothesis.

Over 5000 algae, macrophyte, invertebrate, and fish samples were collected at 15 sites studied by the USGS ACME (Aquatic Cycling of Mercury in the Environment) team during multiple trips 1995-1999. These samples were analyzed for ¹³C and ¹⁵N to determine temporal and spatial differences in food web structure; a smaller subset was also analyzed for ³⁴S. Samples consisted of both composites and individuals. Most samples were analyzed in bulk; however, muscle tissue was analyzed for most large fish. This large dataset was used to calculate the differences in ¹⁵N and ¹³C between all possible consumer-diet pairs. The difference in ¹⁵N between consumer and diet (¹⁵N) is a proxy for the length of the food chain between the species at a particular site and date, and the difference in ¹³C (¹³C) reflects differences in the base of the food webs of the species. An evaluation was then made of how often the ¹⁵N and ¹³C values differed between types of sites. There were no statistically significant differences in ¹⁵N values between high and low nutrient sites, suggesting that there was no evidence for shorter food chains at high nutrient sites, as found by Stober et al. (2001). However, there are statistically significant differences in ¹³C values between low and high nutrient sites. One likely explanation for the ¹³C differences is a difference in the dominant base of the food webs at such sites. Food webs at low nutrient sites appear to be more algal dominated while macrophyte debris appears to be a significant contribution to the base of the food webs at high nutrient sites.

Figure 1. Graph showing significant correlation of 13C with latitude (n = 68; p < 0.01; R2adj = 0.13). [larger image]

Samples of periphyton, mosquitofish, and sediment samples collected during September 1996 at about 100 REMAP marsh sites were analyzed for

C,

N, and

S. The

N calculated for periphyton and mosquitofish pairs (n = 68) showed no significant correlation with region (Above Alligator Alley, Between Alligator Alley and Tamiami Trail, Below Tamiami Trail) (p = 0.67) or latitude (p = 0.20), and no correlation with mosquitofish Hg levels (p = 0.55), suggesting a lack of evidence for spatial differences in food chain length. However, the

One likely explanation for the patterns is that they reflect spatial differences in the relative importance of algae versus macrophyte debris to the mosquitofish food web. In general, the spatial patterns of ¹³C values are very similar to spatial patterns interpreted by Stober et al. (2001) as differences in food web structure. Hence, two very different sets of samples and data from the same regional assessment agree that there are similar spatial differences in some aspect of the food web structure, but disagree about whether this difference is mostly due to trophic variations or food base differences. An evaluation of the gut contents of mosquitofish collected in 1996 and 1999 at the same REMAP sites, indicated that these data did not support the hypothesis that trophic position could be used to explain spatial differences in Hg concentrations (Stober et al., 2001). We are currently re-evaluating the isotope data in conjunction with the gut contents data to see if this will provide new insights into causes of spatial variations in food web characteristics.

This work was made possible by contracts from and collaborations with the US Geological Survey, US Environmental Protection Agency, and Florida Fish and Wildlife Conservation Commission.

References

Stober, Q.J., Thornton, K., Jones, R., Richards, J., Ivey, C., Welch, R., Madden, M., Trexler, J., Gaiser, E., Scheit, D., and Rathbun, S., 2001, South Florida Ecosystem Assessment: EPA Report 904-R-01-003.

Contact: Carol, Kendall, U.S. Geological Survey, 345 Middlefield Road, MS 434, Menlo Park, CA, 94025, Phone: 650-329-4576, Fax: 650-329-5590, ckendall@usgs.gov

(This abstract was taken from the Greater Everglades Ecosystem Restoration (GEER) Open File Report 03-54)

Back to Project Homepage