MACROBENTHIC RESPONSES TO NATURAL AND CONTAMINANT-RELATED GRADIENTS IN NORTHERN GULF OF MEXICO ESTUARIES

Abstract

Effects of pollution on biotic integrity are difficult to identify when correlations occur between environmental gradients and contaminant effects, as they do in estuaries. In this broad-scale study, we used canonical correspondence analysis (CCA) to distinguish influences of natural and contaminant-related gradients on macrobenthic community structure among 319 sites from estuaries throughout the northern Gulf of Mexico. Natural gradients in salinity, depth, and sediment composition obscured the detection of macrobenthic responses to sediment contamination. After adjusting for natural environmental variability, however, partial CCA revealed important macrobenthic variation in relation to sediment contamination. A rotated principal component analysis (PCA) distinguished five composite environmental factors, each largely reflecting contaminant or natural variation. Two complex gradients in sediment contamination identified by the PCA diverged in partial CCA space and correlated with different macrobenthic indicator taxa. Contaminant gradients represented variation in two different classes of sediment contaminants: trace metals and organic chemicals. Dispersion patterns of CCA site coordinates enabled cross validation of implied contamination-related variation in community function and the utility of several interpretive or management metrics. Trophic diversity decreased with sediment contamination, linking shifts in macrobenthic community function and community structure along contaminant gradients. The CCA model complemented an earlier benthic index developed from these data to examine biotic integrity, but the benthic index could not discern macrobenthic responses to the different contaminant gradients. Neither was the benthic index useful for showing transitions in macrobenthic community structure commensurate with different levels of contamination. Ampelisca amphipod sediment bioassays were inadequate for identifying contaminant effects on biotic integrity, whereas Mysidopsis mysid sediment bioassays conservatively reflected sediment contamination and associated macrobenthic indicators.