Probabilistic risk assessment of pesticides under present and future agricultural and climate scenarios using a Bayesian Network

In Northern Europe, future changes in land-use and weather patterns are expected to result in increased precipitation and temperature this may cause an increase in plant disease and insect pests. In addition, predicted population increase will change the production demands and in turn alter agricultural practices such as crop types and with that the use pattern of pesticides. Considering these variabilities and magnitudes of pesticide exposure to the aquatic environment still needs to be accounted for better in current probabilistic risk assessment. In order to improve ecological risk assessment, this study explores an alternative approach to probabilistic risk assessment using a Bayesian Network, as these can serve as meta-models that link selected input and output variables from other models and information sources. The developed model integrates variability in both exposure and effects in the calculation of risk estimate. We focus on environmental risk of pesticides in two Norwegian case study region representatives of northern Europe. Using pesticide fate and transport models (e.g. WISPE), environmental factors such as soil and site parameters together with chemical properties and climate scenarios (current and predicted) are linked to the exposure of a pesticide in the selected study area. In the long term, the use of tools based on Bayesian Network models will allow for a more refined assessment and targeted management of ecological risks by industry and policy makers.