MOISTURE DEPENDENT ANISOTROPY IN UNSATURATED FLOW: THEORY AND APPLICATION

Spectral stochastic analytical methods developed at New Mexico Tech in the late 1970s and early 1980s were applied to the Richards equation for unsaturated flow by Yeh and Gelhar (1981). One of the key findings of this analysis was that the effective hydraulic conductivity tensor will exhibit a variable anisotropy, in other words the anisotropy will vary depending on the moisture content (or similarly, the matric potential) of the soil. Soil water tracer experiments were designed and implemented to test this theory for finely stratified sandy deposits. A numerical model was developed to simulate the tracer experiments, and the model could reproduce the observed results only when state-dependent anisotropy was included, essentially confirming the theory for those deposits (McCord, Stephens, and Wilson, 1989). Recognition of moisture dependent anisotropy has proved vital in analysis of many practical problems, from contaminated site assessment to design of capillary barriers for waste cover systems. This paper provides a summary of the theory, the experiments, and the modeling approach, and well as a sampling of practical situations where state dependent anisotropy has been an important factor to consider.