GEOSTATISTICAL CHARACTERIZATION OF APPARENT HYDRAULIC CONDUCTIVITY DISTRIBUTIONS WITHIN THREE HIGHLY HETEROGENEOUS GLACIAL TERRAINS OF INDIANA

The highly heterogeneous nature of sediments within glacial aquifer systems are rarely understood enough to be properly represented within groundwater flow modeling. Geostatistical methods can be applied to existing datasets in order to create idealized three-dimensional representations of apparent hydraulic conductivity distributions that should provide more realistic presentation of subsurface conditions than are typically deployed in layered models. In an effort to characterize typical glacial aquifer systems, well logs from three glacial terrains within Indiana (glacial fan, buried bedrock valley, and till plain) were classified into apparent hydraulic conductivity values (based on material types) and subjected to geostatistical analysis. A total of 2,477 well logs were subjected to apparent conductivity classification resulting in 190,751 individual units. Frequency distributions were developed, and variograms were created for each study area in order to evaluate the relative abundance and spatial correlation distribution of apparent conductivity values. Parameters fit to the experimental variograms were then used with statistical kriging methods to develop three-dimensional visualizations of the apparent conductivity distributions in each of the study terrains. The resulting visualizations seem to be consistent with conceptual models of glacial depositional systems. A vast network of high conductivity outwash material makes up the bulk of the glacial fan. Continuous zones of high conductivity materials follow surface drainage pathways within the till plain. Large amounts of low conductivity material with a few isolated (confined) aquifers are found within the buried bedrock valley. The results of this study underscore the importance of three-dimensional material variations in glacial aquifers and the value of edited well logs in characterizing hydraulic conductivity distributions.