Submitted to: Journal of Soil Contamination
Publication Type: Peer reviewed journal
Publication Acceptance Date: 7/1/1999
Publication Date: N/A
Citation: Interpretive Summary: Pesticide RootZone Model (PRZM), is a popular environmental fate model developed by EPA. It is designed to assess the effects of various soil and climate factors on pesticide movement in field. Atrazine is a herbicide commonly used to control weeds in corn production systems. Recent studies have shown that for the untilled fields, PRZM tended to over predict atrazine residue levels in the near surface soil and under estimates the residue levels in the deeper soil profile. The purpose of this work was to observe the effect of variations in partition coefficient (ratio of the amount of atrazine adsorbed to soil particles vs the amount in soil solution) on the model predictions, since this parameter is one of the most sensitive input to the PRZM model. This work was accomplished by comparing the measured atrazine distribution in the soil profile to several different PRZM simulated atrazine distributions. For each simulation only the value of the partition coefficient was modified and all of the other variables remained constant and their values were chosen to best represent the field conditions during the time the measured values were obtained. Results suggest that better quantification of the partition coefficient of atrazine for different layers in the soil profile may improve our ability to model the distribution of atrazine within and below the crop root zone in field.
Technical Abstract: PRZM has been used to compare predicted herbicide soil concentrations to measured values in conventional (CT) and no-till (NT) fields. Typically, predictions are reasonably close to measured values for the upper soil layer for both plots. However, in lower layers, predictions are closer to the measured values for CT and under-predicted in NT plots. Comparison of predictions for atrazine distribution with other chemicals indicates PRZM effectively transports water and many other solutes. This suggests that prediction of atrazine distribution may be improved by improving the description of atrazine characteristics. Our objective was to observe the effect of variation of the coefficient of distribution (Kd) in soil layers on PRZM predicted atrazine concentrations in the profile of CT and NT plots, and compare the predictions to field measured values. Results indicate that reducing Kd in the upper soil horizon improved predictions for the upper horizon but significantly over predicted atrazine concentrations in the second horizon. Reducing Kd throughout the profile reduced predicted atrazine concentrations in the upper horizon and increased predicted concentrations in the lower horizons. Improvement in the correlation of predicted atrazine concentrations with measured concentrations was greatest for NT plots; however, improvement was small compared to the improvement still possible.