INTERCALATED SEDIMENTS IN A HETEROGENEOUS BASALT AQUIFER, EASTERN SNAKE RIVER PLAIN, IDAHO: GEOSTATISTICAL ANALYSIS AND SPATIAL MODELING

The distribution of sediment in the eastern Snake River Plain aquifer was evaluated to improve the parametrization of hydraulic conductivity (K) for a subregional-scale ground-water flow model developed by the U.S. Geological Survey. The aquifer is hosted within a layered series of permeable basalts within which intercalated beds of fine-grained sediment comprise localized aquitards. These sediments have K values as much as six orders of magnitude lower than the most permeable basalt, and calibration of the flow model has shown that bulk aquifer K is sensitive to the proportion of intercalated sediment.

Stratigraphic data in the form of basalt and sediment thicknesses from 333 boreholes beneath the Idaho National Laboratory (INL) were analyzed and modeled geostatistically. The data were first evaluated in composited subsets of lithologic units corresponding to their time-stratigraphic position. The analysis demonstrated that statistical homogeneity characterizes sediment abundance in the stratigraphic units below the water table but not in the youngest unit spanning the past 250 ka; the data also exhibit spatial stationarity in a geographic sense. These determinations allowed the data to be kriged as a set of two-dimensional layers (cross-cutting aquifer stratigraphy) that correspond to the layers used to discretize the ground-water flow model.

Multiple indicator kriging was used to model sediment abundance within each layer by defining the cumulative frequency distribution (CFD) of sediment percentage based on local borehole data. This approach is superior to ordinary kriging because it provides a statistically "best" estimate of sediment proportion (the local median or other measure of central tendency) as well as the dispersion about the median (reflecting the degree of uncertainty in the estimate). A methodology is proposed for constraining the assignment of model K values based on local CFDs as an alternative to simply scaling K against median sediment abundance.