GIS COMPARISON OF METHODS FOR PREDICTING PESTICIDE CONTAMINATION POTENTIAL OF GROUND WATER IN COLORADO

Ground-water reservoirs are an important resource for drinking water in the state of Colorado. Many human activities release chemicals to the environment that have the potential to contaminate these ground-water resources. Pesticides are organic chemicals of concern that are applied to the land surface over spatially extensive agricultural areas. Due to the large areas of application, pesticides can diminish ground water quality for a large percentage of Colorado residents. Protecting Colorado’s ground-water reservoirs from contamination by agricultural pesticides requires that we understand and evaluate, on a macroscale, the transport of pesticides through the unsaturated subsurface. Evaluation of pesticide transport must produce a spatial model of pesticide contamination potential for the entire state. The best method to evaluate unsaturated subsurface pesticide transport, and predict the spatial distribution of pesticide contamination potential of ground water is by applying predictive models using a Geographic Information System.

This study compared the results of three different methodologies and spatial models for predicting pesticide contamination potential of ground water in the state of Colorado. The first and simplest method used depth to ground water data, infiltration capacity of soil, and recharge availability to calculate a relative sensitivity index of ground water to pesticide contamination. The second method incorporated soil characteristics, depth to ground water data, and pesticide properties into a vulnerability index equation derived from the one-dimensional advection-dispersion equation. The third method incorporated soil characteristics, depth to ground water data, average annual ground water recharge, and pesticide properties into an attenuation factor derived as a simple solution to the one-dimensional advection-dispersion equation. Validation of the predictive models indicated that the second method produced the most statistically valid results and was the most practical to implement. The results of this study will be valuable for agronomists developing the pesticide management plan and regulatory requirements for pesticide use in the state of Colorado.