Accounting for Toxicity Test Variability in Evaluating Test Results

摘要

Limits on whole effluent toxicity (WET) now have been part of the U.S. EnvironmentalProtection Agency’s (EPA) National Pollutant Discharge Elimination System (NPDES) for morethan a decade. While it is generally agreed that such limits are an important component of effortsto improve the quality of our nation’s surface waters, implementation of WET limits in NPDESpermits has been controversial since its inception. Two major issues at the source of thecontroversy have been (1) “how well do WET test results predict the actual toxic effects of aneffluent to the aquatic communities present in receiving waters?”and (2) “how reliable are thetests for measuring effluent toxicity?” Over the past decade a large number of studies have beenconducted to address both of these issues, but the issues have not been resolved satisfactorily.This paper presents the results of a research project addressing the precision of WET testprocedures, and develops methods for incorporating variance estimators into EPA decisionmakingframeworks. In addition, a national data base of reference toxicity tests, a national database of WET compliance test results, and a series of case studies are developed in support of thestatistical analyses.