New concepts in ecological risk assessment: where do we go from here?

Through the use of safety factors, the use of single-species test data has been adequate for use in protective hazard assessments and criteria setting but, because hazard quotients do not consider the presence of multiple species each with a particular sensitivity or the interactions that can occur between these species in a functioning community, they are ill-suited to environmental risk assessment. Significant functional redundancy occurs in most ecosystems but this is poorly considered in single-species tests conducted under laboratory conditions. A significant advance in effects assessment was the use of the microcosm as a unit within which to test interacting populations of organisms. The microcosm has allowed the measurement of the environmental effect measures such as the NOAEC(community) under laboratory or field conditions and the application of this and other similarly derived measures to ecological risk assessment (ERA). More recently, distributions of single-species laboratory test data have been used for criteria setting and, combined with distributions of exposure concentrations, for risk assessment. Distributions of species sensitivity values have been used in an a priori way for setting environmental quality criteria such as the final acute value (FAV) derived for water quality criteria. Similar distributional approaches have been combined with modeled or measured concentrations to produce estimates of the joint probability of a single species being affected or that a proportion of organisms in a community will be impacted in a posteriori risk assessments. These techniques have not been widely applied for risk assessment of dredged materials, however, with appropriate consideration of bioavailability and spatial and nature of the data these techniques can be applied to soils and sediments. (C) 2002 Elsevier Science Ltd. All rights reserved.