Probabilistic Risk Assessment of Pesticides of a rice field in Spain: A Bayesian Network Approach

The aquatic environment is constantly exposed to various chemicals caused by anthropogenic activities such as agricultural practices using plant protection products. Traditional Environmental Risk Assessment is based on calculated risk estimations usually representing a ratio of exposure to effects, in combination with assessment factors to account for uncertainty. In this study, we explore a more informative approach through probabilistic risk assessment, where probability distributions for exposure and effects are expressed and enable accounting for variability and uncertainty better. We focus on the risk assessment of various pesticides in a representative study area in the south east of Norway. Exposure data in this research was provided by the Norwegian Agricultural Environmental Monitoring Programme (JOVA)/ or predicted exposure concentration from a pesticide exposure model and effect data was derived from the NIVA Risk Assessment database (RAdb, www.niva.no/radb). A Bayesian network model is used as an alternative probabilistic approach to assess the risks of chemical. Bayesian Networks can serve as meta-models that link selected input and output variables from several separate project outputs and offer a transparent way of evaluating the required characterization of uncertainty for ERA. They can predict the probability of several risk levels, while facilitating the communication of estimates and uncertainties.