Comparison of two US environmental protection agency species sensitivity distribution methods for calculating ecological risk criteria

Two major methods used by the U. S. Environmental Protection Agency (USEPA) to calculate ecological risk criteria using "species sensitivity distributions" (SSD) are compared using identical datasets. One method is the current USEPA Office of Water method for deriving acute numeric water quality criteria (EPA-FAV method). The 95% protection level generated by this method is the Final Acute Value (FAV). The other method is the USEPA Office of Pesticide Programs log normal distribution regression method that uses all available toxicological data in performing ecological risk assessments (OPP-Ecorisk method). Of the 50 comparisons made, 38% of the 95% protection levels calculated by the OPP-Ecorisk method are within a factor of 1.5 of the EPA FAV, 68% are within a factor of 2.0, 86% are within a factor of 2.5 and 92% are within a factor of 3.5. For pollutants where the difference between the OPP-Ecorisk 95% protection level and the EPA FAV are greatest, the OPP-Ecorisk 95% protection level is more protective than the EPA FAV. Where there are major differences, an analysis of the toxicity data indicates that the majority of the differences can be attributed to outliers, especially on the least sensitive ends of the toxicity distribution. In the one example dataset presented with one very sensitive genus, the OPP-Ecorisk method gave a protection level that was much closer to the lowest GMAV than the EPA-FAV method.