ABSTRACT
Hazard assessment is a two-component exercise involving separate exposure assessment, in which Predicted Environmental Concentrations (PECs) are determined, and effect assessment, in which No Observed Effect Concentrations (NOECs) are determined. According to the Organisation for Economic Cooperation and Development (OECD) various stages can be distinguished in effect assessment: preliminary, refined and comprehensive effect assessment. Field studies, including mesocosm and enclosure studies, provide the basis for comprehensive effect assessment. In current national and international chemical programmes of the European Communities (EC) and the OECD, field tests play only a minor role in determining these effect concentrations. Effect assessments are generally made on the basis of single-species toxicity data. These data are extrapolated in order to derive “safe” levels for the environment using either fixed assessment factors (preliminary effect assessment) or calculated extrapolation factors on the basis of NOECs from a sample of species using statistical extrapolation procedures (refined effect assessment). Uncertainties arising from these preliminary or refined hazard assessments may be many and varied. They may concern the potential of a chemical to cause indirect or secondary effects on populations or the ecological interaction between species, but from many field tests it can be concluded that the real variable in hazard assessment is exposure assessment. Field tests may play a role in reducing these fate and effect-related uncertainties, but one should bear in mind that field tests also have a great number of disadvantages. From an extensive literature analysis, which was carried out to validate extrapolated aquatic singlespecies toxicity data on the basis of field studies, it appeared that only a few reliable field studies are available. It was also shown that the refined extrapolation methods of Aldenberg and Slob (1993) and Wagner and Løkke (1991) seem to be a good basis for determining ‘safe’ values for aquatic ecosystems, provided that at least 4 NOECs, and preferably many more are available for different sensitive taxonomic groups.
