: The weight-of-evidence approach to the development of sediment quality guidelines (SQGs) was modified to support the derivation of biological effects-based SQGs for Florida coastal waters. Numerical SQGs were derived for 34 substances, including nine trace metals, 13 individual polycyclic aromatic hydrocarbons (PAHs), three groups of PAHs, total polychlorinated biphenyls (PCBs), seven pesticides and one phthalate ester. For each substance, a threshold effects level (TEL) and a probable effects level (PEL) was calculated. These two values defined three ranges of chemical concentrations, including those that were (1) rarely, (2) occasionally or (3) frequently associated with adverse effects. The SQGs were then evaluated to determine their degree of agreement with other guidelines (an indicator of comparability) and the percent incidence of adverse effects within each concentration range (an indicator of reliability). The guidelines also were used to classify (using a dichotomous system: toxic, with one or more exceedances of the PELs or non-toxic, with no exceedances of the TELs) sediment samples collected from various locations in Florida and the Gulf of Mexico. The accuracy of these predictions was then evaluated using the results of the biological tests that were performed on the same sediment samples. The resultant SQGs were demonstrated to provide practical, reliable and predictive tools for assessing sediment quality in Florida and elsewhere in the southeastern portion of the United States.

No abstract

The relative sensitivity, analytical precision, discriminatory power and concordance among endpoints and with sediment chemistry were compared among five sediment toxicity tests. The tests were performed with aliquots of 15 composited, homogenized sediment samples. Survival and a variety of sublethal endpoints were determined in tests performed with the amphipods Rhepoxynius abronius and Ampelisca abdita, embryos of the mussel Mytilus edulis, embryos of the urchin Strongylocentrotus purpuratus, and the polychaete Dinophilus gyrociliatus. Each sample was also tested for trace metal and organic compound concentrations, organic carbon content and texture. Two of the five tests (survival among M. edulis and survival among R. abronius) were highly sensitive to the samples and had relatively high precision, but the results were correlated most highly with sedimentological variables. One of the tests (survival among A. abdita) was relatively insensitive, but the results were highly correlated with only the concentrations of toxic chemicals. The test with S. purpuratus indicated mutagenicity in several samples that had high hydrocarbon concentrations. The test of pore water with D. gyrociliatus was intermediate in sensitivity and precision and not correlated highly with the results from the other tests.