EFFECTIVE HYDRAULIC CONDUCTIVITY OF AN EXPERIMENTAL STRATIGRAPHY AND IMPLICATION FOR GROUNDWATER FLOW MODELING

A critical problem facing basin-scale groundwater flow models is the accuracy of geologic framework models and the selection of effective saturated hydraulic conductivity (K*). Based on an experimental stratigraphy, a two-dimensional, fully heterogeneous, hydraulic conductivity map provides the basis for the evaluation of K* using numerical and analytic methods. For selected deposits, K* changes with support in an asymptotic fashion. A power-law exponent fitted is found to increase for deposits of higher variability. Up-scaling irregular hydrostratigraphic units indicates that the more stratified deposits have higher ln(K) variance and higher anisotropy ratio of K*. For all deposit types, the arithmetic mean appears a good approximation for the maximum principal component of K*. The minimum principal component of K* fluctuates between the harmonic and geometric mean. Based on the heterogeneity map, two geologic framework models (GFM) are developed. Four sets of flow experiments are conducted in each model. We found that reproducing the mean flow is the most sensitive to the choice of GFM and the boundary condition.