Groundwater contamination potential by pesticide was evaluated in Albemarle and Louisa counties of the Thomas Jefferson Planning District in Virginia. A qualitative method was developed to perform an assessment of pesticide contamination potential, using an existing pesticide screening model. The Attenuation Factor (AF), was selected for assessment of groundwater vulnerability to pesticide contamination in Albemarle and Louisa counties. Input data availability, consideration of the major transport processes, and ease of its linkage with a suitable geographic information system (GIS) were the main factors considered for selection of the AF model.

The input data requirement of the AF model includes soil, hydrogeologic, and pesticide chemical characteristics. An extensive database was developed to perform AF model simulations within a GIS. The database developed for this study included map databases (resolution = 1/9 ha) for landuse, soils, groundwater recharge, and groundwater depth, and non-spatial (relational tables) databases for pesticide chemical characteristics, SCS curve number, and soil properties. A total of 12 landuse categories were identified for Albemarle and Louisa counties. Groundwater recharge, an input to the AF model, was estimated using a water balance model. Runoff and evapotranspiration components of the water balance model were estimated using SCS curve number (CN), and Thornthwaite’s methods, respectively. Forty years of climatological data records were used for estimating groundwater recharge. Two types of groundwater depths, spatially varying and a constant depth of 2 m, were used for computing AF, The groundwater depth was mapped using the information available in groundwater well completion reports. All the data layers were overlaid within a GIS for spatial computation of AF for actual and 2m groundwater depth. This spatial (map) database was categorized into five categories of pollution potential namely, high, medium, low, very low, and unlikely, based on the numerical values of the AF. For evaluating the contamination potential of pesticides, three pesticide leaching potential scenarios were considered in order to facilitate the evaluation of pesticide leaching under maximum, average and minimum cases of degradation and sorption in the soil. A combination of high, average, and low values of half life and sorption coefficient were selected for three leaching scenarios. A total of six simulations were performed (two groundwater depths and three leaching scenarios) for each pesticide. Toxicity of the pesticides was not considered in the contamination potential assessment in this study.

A total of 11 relatively mobile pesticides were identified in Albemarle and Louisa counties, based on the results for various leaching scenarios. Groundwater contamination potential maps were produced for mobile pesticides and the results were discussed. Picloram was identified to be the most mobile pesticide in the two counties. Atrazine, carbofuran, metolachlor, simazine and triclopyr were found to have considerable potential to move to the groundwater. Contamination potential of three herbicides, atrazine, simazine and metolachlor, was predicted to be higher than other pesticides in light of the fact that they are often used in combination (tank mixed) for a wide variety of weed control. Dicamba was found to be the most heavily-used pesticide with regard to its area of application. In light of dicamba's moderate contamination potential and its higher usage amount this herbicide was identified to have a considerable potential to move to the groundwater, especially, in Albemarle County. Other pesticides such as fenarimol, lindane, metalaxyl, and metsulfuron methyl were shown to be relatively more mobile than some other pesticides. However, in light of small application areas for these pesticides, the contamination potential of these pesticides was predicted to be relatively small. In addition to 11 mobile pesticides, a few low mobility pesticides were also identified. Among the low mobility pesticides, diazinon was found to have relatively high contamination potential. Comparison between the contamination potential maps and the groundwater recharge map revealed that most of the high contamination potential areas coincided with the higher groundwater recharge regions in the two counties. Soil characteristics such as organic matter and percent sand and clay were also observed to affect the contamination potential of pesticides.

The performance of the AF model was evaluated by using six years of groundwater monitoring data from the Nomini Creek watershed study. Two types of rankings were made, one using the AF model simulation results, while the other ranking was based on the frequency of detection of pesticides in the Nomini Creek watershed. A comparison of the rankings revealed that the AF model performed fairly well in identifying the top few mobile pesticides. Sensitivity analysis was performed to identify the important parameters affecting the contamination potential of pesticides. Results of the sensitivity analysis revealed that considerable uncertainty in the model prediction can be invoked due to the variability in the soil and chemical data.

To make a better use of results of this study, it was recommended that groundwater monitoring be performed in the two counties to verify the so that results of this study. The results of this study will provide information about the potential threat to groundwater by pesticides to the citizens and policymakers in the two counties.