Modelling pesticide leaching in a sandy soil with the VARLEACH model

Within this paper the ability of the VARLEACH model to simulate field results is presented. The evaluation was carried out in the framework of a European modelling validation exercise, adopting a standardised modelling protocol. Simulations were performed with and without a calibrated data-set as identified by independent model users. Finally a simulation with a consensus parameter set was made. The model gave an accurate description of pesticide penetration in the soil profile, although occasionally with some overprediction, but it did not simulate the absolute level of soil residues. With bentazone, which is a weakly sorbed and moderately persistent compound, the laboratory data on degradation did not describe the observed field behaviour. Total residues of ethoprophos were poorly simulated because the VARLEACH model does not take account of losses by volatilisation. However, if a correction was applied for the potential vapour losses, the simulated results were in agreement with those measured. The modelling exercise with different users indicated how input value control is one of the most important aspects to increase validity and use of models to forecast pesticide behaviour.