Evaluation of Sesoil, Przm and Pestan in a laboratory column leaching experiment

Four 2-m columns filled with sandy soil were loaded with six organic chemicals (dicamba, 2,4-dichlorophenoxyacetic acid, atrazine, diazinon, pentachlorophenol, and lindane) and watered for 30 d in a controlled laboratory experiment. Chemical migration was monitored by analyses of daily effluent samples and soil sample cores, the latter collected on day 30 from 5-cm compartments throughout each column. Observed chemical migration patterns were compared with the predictions of three vadose zone fate-and-transport models: SESOIL, PRZM and PESTAN. Two sets of comparisons with the observed data were conducted using input data either obtained from the literature or derived from laboratory measurements using the actual experimental chemicals and soil. Evaluations of model performance revealed substantial variability depending on the chemical type and model tested. PRZM and PESTAN showed greater improvement in prediction of observed chemical migration patterns with the use of measured input data than did SESOIL. The limitations of these models, if used for screening purposes without a priori measurement of chemical- and site-specific model input data or post hoc calibration with existing field data, should be recognized.