STUDIES ON COMBINATION EFFECTS OF ENVIRONMENTALLY RELEVANT TOXICANTS Validation of prognostic concepts for assessing the algal toxicity of realistic aquatic pesticide mixtures

This thesis aims to analyse whether the joint toxicity of environmentally relevant toxicants can be predicted based on the known toxicity of the single mixture components. Two concepts have been successfully applied so far for the prediction of mixture toxicity: concentration addition (CA) for mixtures of similarly acting substances and independent action (IA) for mixtures of dissimilarly acting substances. In environmentally relevant mixtures though, mechanism of action patterns are more complex. The results of this thesis show, that if (i) the mechanisms of action of the mixture components are known and (ii) the components can be assigned to discrete mechanism of action groups, the combination of CA and IA to a two-stage prediction (TSP), allows for accurate predictions. The application of the TSP on a realistic pesticide mixture led to accurate predictions of the experimentally observed mixture toxicity. However, the application of CA and IA also led to good mixture toxicity estimations, since the quantitative differences between the predictions according to CA, IA and TSP were rather small. Two conclusions were drawn: (i) the combination of CA and IA to a TSP is a valid approach, which provides accurate predictions of the mixture toxicity in case that the components can be assigned to discrete mechanism of action groups. (ii) Even in case that mechanistic knowledge of the mixture components is not provided, the use of CA as a precautionary default assumption is a justifiable approach for estimating the mixture toxicity of realistic pesticide mixtures.