Evaluation of a stepwise procedure for comparative validation of pesticide leaching models

Four pesticide leaching model codes (PELMO, PESTLA, MACRO and MIKE SHE) were evaluated and compared through a rigorous validation procedure combined with the application of statistical evaluation criteria. The validation procedure followed a strict stepwise approach based on suggestions put forward by a European work group on regulatory use of pesticide models (FOCUS). The experimental background comprised two different types of data sets. A laboratory and field lysimeter experiment were conducted on a Danish macroporous sandy loam soil. The aim of the study was to evaluate whether a priori model calibration on controlled laboratory data could improve the physical description of the flow and solute transport in the soil and hence the performance of uncalibrated models for predictions of field lysimeter data. The validation procedure proved to be valuable in terms of ensuring process-based evaluation of model performances and consistent model comparisons. Controlled laboratory experiments and lysimeter experiments consistently showed very significant influence of preferential flow on water and solute transport. Model codes including a description of preferential flow processes (MACRO and MIKE SHE) required less calibration efforts to meet the selected performance criteria on the investigated soil type than those without such description (PELMO and PESTLA).