Premium
Effects of Soil Heterogeneity on Pesticide Leaching to Groundwater
Author(s) -
Zee Sjoerd E. A. T. M.,
Boesten Jos J. T. I.
Publication year - 1991
Publication title -
water resources research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.863
H-Index - 217
eISSN - 1944-7973
pISSN - 0043-1397
DOI - 10.1029/91wr01990
Subject(s) - spatial variability , spatial correlation , soil science , skewness , spatial heterogeneity , groundwater , decoupling (probability) , spatial dependence , dispersion (optics) , monte carlo method , environmental science , surface runoff , hydrology (agriculture) , mathematics , geology , geotechnical engineering , statistics , physics , engineering , ecology , optics , control engineering , biology
Pesticide leaching was simulated numerically assuming Freundlich adsorption, first‐order transformation and passive plant uptake, taking transient flow, hydrodynamic dispersion, and depth as well as temperature dependence of (bio)chemical parameters into account. The dependency of the leached fraction on sorption and transformation parameters appeared to be in good general agreement with the model developed by Jury et al. (1987). We incorporated spatial variability of )bio)chemical parameters into the piston flow model and showed how spatial variability may be accounted for without having to resort to demanding Monte Carlo techniques. Such spatial variability affects the leached fraction significantly. The data requirement of the simple stochastic model is small and the versatility relatively high. For parameter values often not of prime practical interest for pesticide screening this model fails. Considering transport by convection and dispersion in the semi‐infinite domain, this spatial variability can be easily accounted for using parameter ranges for which our original analytical model for spatially variable piston transport failed. In this revised model, hydrodynamic dispersion, preferential flow, and spatial variability of transformation, sorption and soil thickness are dealt with in a similar fashion, while both correlation or the absence of correlation of parameters can be incorporated through an approximation of apparent residence time variance.