Groundwater Vulnerability to Pesticide Contamination on a Regional Scale

Abstract

AbstractRegional groundwater vulnerability maps that indicate the impact of leaching uncertainties, should be accompanied by maps expressing the probability of surpassing that index value. This case study presents first steps in a methodology for developing such an index and the probability of its value. Regionalization of piezometric data provided an estimate of the lacking water table depth at soil profiles. Monte‐Carlo simulations of pesticide fate on these soil profiles produced local distributions of a groundwater vulnerability index. Soil hydraulic properties were randomly generated using pedotransfer functions and experimental probability density functions of particle‐ size distribution and organic matter content. Medians of locally simulated distributions were then mapped to show the spatial distribution of groundwater vulnerability. It appeared that the resulting groundwater vulnerability map might be significantly affected by soil spatial variability, although soils are quite similar throughout the area. Deciles of the locally simulated distributions seem to show intrinsic coregionalization. This property could be used to build, over the whole area, an estimated groundwater vulnerability index distribution.