ABSTRACT
A module to simulate the volatilization of pesticides from soils and plants was implemented in the air quality model CHIMERE in order to simulate spatiotemporal distribution of pesticide atmospheric concentrations. Pesticide applications are spatially distributed according to the quantities of pesticides sold per municipality in France (recorded in the French BNVD-S database) and are temporally distributed according to the application periods determined with enquiries. The model was applied to S-metolachlor and folpet. In the first stage of the study, pesticide emissions simulated by the CHIMERE and Volt'Air models are compared. In the second stage, measured concentrations of S-metolachlor and folpet from mid-April to the end of June are compared to the simulation results at the French and PACA (Southeastern region of France) scales. The model can reproduce the spatial distribution of S-metolachlor concentrations (spatial correlation over France of 0.79) with a bias ranging from -50 to 50% for most stations during the application period. The simulation of folpet concentrations remains challenging with a lack of correlation between model results and measurements, that could possibly be due to a lack of precision in the temporalization of applications.
