| This dissertation describes the modeling efforts that devoted to understand the fate and transport processes of agricultural chemicals (pesticides and fertilizers) in aquatic environments and their impacts on surface and subsurface water quality. The main tasks were to develop and apply a two-dimensional (2D) reservoir toxic model for the fate and transport of toxic substances and to evaluate a field-scale non- point source model, the Erosion Productivity Impact Calculator (EPIC), as a tool for agricultural policy analysis.; The 2D toxic model was developed using finite difference method to the laterally integrated hydrodynamics, mass transport, and transformation equations. It was applied to the Shasta Reservoir, California to investigate the effects of reservoir flow regime on the persistence and behavior of a spilled toxicant, methyl isothiocyanate (MITC). The results demonstrated that flow regime can substantially affect the persistence and transport of contaminant in the late stage of the spill. The model was also tested using field data, atrazine [2-chloro-4-ethylamino-6-isopropylamino-1,3,5-triazine], collected from the Saylorville Reservoir, Iowa. Time-variable half-life of atrazine was estimated using a mass balance concept in the reservoir. The half-life varied monthly from 2 to 58 days depending upon environmental conditions (i.e., sunlight, temperature, and microorganisms). The 2D model accurately simulated the temporal and spatial variations of atrazine concentrations in the reservoir.; The EPIC model was tested at two field sites in Iowa. It is concluded that standard curve number values should be adequately reduced to represent the impacts of residue cover on the partition of precipitation between surface runoff and infiltration. The results showed that EPIC is sensitive to variations in tillage and cropping practices and can be used to estimate environmental indicators in response to different management systems. However, clear discrepancies occurred between some model estimates and corresponding measured values, e.g., under-prediction of peak flows and nitrogen losses during storm events. Two potential sources of these errors include: (1) the lack of a preferential flow component, and (2) nitrogen transformation routines that may not adequately reflect all of the processes that occur in the field. EPIC showed a limited capability to reproduce tillage and crop rotation effects on crop yield. |
