Validation of the exposure assessment for veterinary medicinal products.

Under the EU Directive 2004/28/EC, an environmental risk assessment of new veterinary medicinal products is required. Given the nature of risk assessment for new applications, there is a need to model exposure concentrations. Critical evaluations are essential to ensure that the use of models by regulators does not result in the propagation of misleading information. The empirical validations of soil exposure models, previously discussed in this journal, indicate that it is impossible to analyse the contribution of every model parameter to the variability in the predictions. In particular, the prediction of the slurry concentration is challenged by uncertainties concerning dilution, mixing and dissipation of residues. Surface water and groundwater models generated highly deviating results compared to the field results, questioning the usefulness of the available screening models. Animal husbandry, slurry handling and environmental conditions throughout Europe are considered in order to define realistic worst case scenarios, to be used in conjunction with distribution models for the environmental risk assessment of veterinary medicinal products at registration. Given the variability in manure management practice throughout Europe, a deterministic approach for the manure-to-soil model was selected. Both worst case and best case scenario were developed. Several modelling assumptions applied in the surface water exposure model for fish nursery effluent were validated against newly available data. Since the available data give no proof that a settling tank contributes to the removal of pesticides from waste water, it is recommended for risk assessment purposes to consider the contribution of the settling tank to removal of pesticides and medicines to be negligible. Surface water dilution factors may be considered to be rather small, a factor of 2, for low flow situations.