Hydraulic conductivity is a vital parameter in the rainfall - runoff modeling. It displays high spatial variability, as it has been demonstrated by field experimentation in the past. Spatial variability of hydraulic conductivity can be interpreted through linear unbiased estimator, kriging. In the present study, a geo - statistical technique, kriging has been utilized for regionalized variables to exhibit spatial auto correlation structure. The universal kriging was applied, in which observations consist of an estimate at any point based on a weighted average of the data where weights were determined by using the semi - variogram and an assumed drift in the data. The weights were calculated by kriging algorithm, which involves a linear system of equations based on data. The observed data of hydraulic conductivity in a field located in Guelph, Ontario, Canada has been used to elucidate spatial structure through a semi variogram by applying kriging. Infiltration rates were measured with double ringinfiltrometer on grid basis. Two parameter Philip's equation was used to estimate saturated hydraulic conductivity. The resulting semi variograms were isotropic and fit well into several linear and spherical models. The verification of the kriging estimates was done by removing known data points & kriging an estimate at same location. The acceptable results of verification procedure demonstrated that geostatics could be used to describe spatial variability of hydraulic conductivity.
展开▼
