Nonparametric Stratigraphic Interpretation from Drill Log Data

Near surface groundwater contamination is a widespread problem. The potential for contamination of deep aquifers in such areas depends on the vertical hydraulic gradient as well as the extent and location of interconnection between the upper and lower aquifers. In alluvial~ sedimentary environments, the geologic units are typically weakly connected lenses or layers of high or low conductivity with variable size~ geometry and orientation. Drill logs provide qualitative, local information on such aquifer heterogeneities. A binary (high or low conductivity) indicator function is used to describe the aquifer system. A nonparametric statistical methodology for assessing the probability that a particular location in the aquifer has high or low conductivity using drill log infonnation is developed. The stochastic behavior of the sedimentary process in the vertical is of particular interest. Example applications using data from Lake Bonneville deposits in Salt Lake County, Utah are presented.