A new form of regression is applied to the problem of modeling the flow of water and contaminants through soil. In a fashion analogous to nested ANOVA, the new method parametrizes global distributional structure separately from local structure. A blind study is conducted to assess the precision of mixing parameter estimation as a function of depth. It is shown that accurate estimates of the regression relationship can be obtained from a sample of size n1000 for mixing parameters and all other component parameters, with the exception of the standard deviation of small components which have large variances. It is shown that the hydraulic conductivity, transport, or infiltration of water borne contaminants through the vadose zone can be effectively modeled and simulated by the mixing parameter regression methods.
This article is from 'Computing Science and Statistics: Proceedings of the Symposium on the Interface Critical Applications of Scientific Computing: Biology, Engineering, Medicine, Speech Held in Seattle, Washington on 21-24 April 1991,' AD-A252 938, p94-101.