Predicting effects of pesticides in freshwater aquatic ecosystems using field data validation

The plausibility of using single-species acute laboratory toxicity data to predict the effects of pesticide contamination of freshwater aquatic ecosystems was investigated, using single application studies. Toxicity units were developed using the geometric mean Daphnia spp. values or hazard concentrations for 5% of species derived from species sensitivity distributions for crustacea, insecta or algae, and peak pesticide water concentration. Pesticide physico-chemical (Kow, Koc) and fate (hydrolysis, water photolysis, aerobic aquatic biotransformation and aerobic soil biotransformation half-lives) properties were used along with the toxicity units and structural properties of the experimental systems (volume, surface area to volume ratio) to produce models capable of explaining the effects seen in the experiments. Akaike's Information Criterion was used to select the best model combinations, which were then generated using linear regressions. The hazard concentration is a better predictor than Daphnia spp. and fate parameters are necessary to produce better predictions.