EVALUATION OF PRZM AND LEACHMP ON INTACT SOIL COLUMNS

Limiting factors towards the utilization of pesticide transport models are lack of both detailed data and cognizant personnel available to t est and validate model predictions. In this study, computer simulation s were carried out to test the performance of PRZM (a management model ) and LEACHMP (a research model) on leaching characteristics of atrazi ne in long intact soil columns. Also, in an attempt to evaluate model uncertainty, a sensitivity analysis of several parameters for each mod el was considered. LEACHMP was found to be superior in predicting hydr ological characteristics in the soil columns as compared to PRZM which required calibration. Both models underestimated levels of atrazine n ear the soil surface and in the leachate (115 cm depth) and over-estim ated them throughout most of the range in-between. Inaccurate predicti ons for the two models are attributed to simplistic linear adsorption equations and lack of a macropore flow subroutine. In increasing order , PRZM was most sensitive to rate of application, distribution coeffic ient, bulk density and field capacity, whereas LEACHMP was sensitive t o rate of application, organic carbon distribution coefficient, air en try value, and bulk density.