The influence of variability of the mobility and persistence parameters of the PRZM 3.22A model for evaluation and leaching assessment of pesticides.

摘要

Leaching models are considered the best means for evaluating relative groundwater contamination potential in the absence of monitoring data. One important approach for minimizing disagreement between model predictions and observed field data is a model calibration (selected model inputs are modified to improve the model's predictability). The most commonly used model (particularly in the U.S.) has been the Pesticide Root Zone Model (PRZM). The objectives of this study were: first, to evaluate the performance of PRZM 3.22A in predicting the fate and transport of a range of mobile pesticides by refining the values of the following input parameters (dispersion coefficient, pesticide degradation rate with depth, temperature and moisture corrected degradation, and Freundlich Coefficient) second, to recommend the refinement of modeling procedures and subsequently improve the model's prediction capabilities, and third, to provide the U.S. EPA with a validated-Tier II leaching scenario for regulatory decision-making. The results are mixed in terms of the ability of PRZM to simulate tracer. Overall the timing of bromide movement at the four study sites was predicted well except for the Group B lysimeters (those with fine-textured soil layer) at the California site. With regard to pesticide leaching, in general peak and overall mass flux were about right or somewhat under predicted at CA Group A locations, under predicted at GA2L (except when the Freundlich coefficient was applied), over predicted at CA Group B locations, NC4L, and hugely over predicted at GA1L. Overall, our results show that there are too many unknowns to use PRZM to accurately predict the extent of pesticide leaching. However, while some specific model errors appear to have been identified (particularly with model's use of the DISP input parameter), it appears that most of the deviations of the predictions from the observed are due to lack of understanding of the behavior of solutes and of how specific site characteristics influence solute transport.