A Continuous Need To Determine What We Should Protect In Ecological Risk Assessments

摘要

What aspects of ecosystems should we protect from the adverse impacts of chemicals? How much of an effect on these end points would be acceptable? While a commonly advocated protection goal is viability of species,1 which is developed typically at the population (e.g., density) or community level (e.g., species richness), these are long-standing and fundamental questions that have been rather ambiguously dealt with in general ecological risk assessments and in the establishment of water quality criteria. Use of no observed effect concentration (NOEC) is a typical example. Using laboratory toxicity testing on individual- (or organismal-) level end points such as survival and reproduction, NOEC is determined solely by statistical significance which depends on experimental design, sample size, effect size, and data variability. Who knows what real ecological consequences have resulted from this reliance on statistical significance.