Variability and Scale Dependence of Hydraulic Conductivity for Hanford Site Sand and Gravel Aquifers.

Abstract

This study presents a comprehensive compilation of a hydraulic conductivity (K) database (over 800 measurements) collected over the past seven decades, encompassing test volumes ranging from laboratory to field scales for two principal sedimentary units at the Hanford site in south-central Washington State. Despite both units being gravel-dominated, the geometric mean K of the Hanford formation is orders of magnitude higher than that of the Ringold Formation for the permeameter and pumping test data. In contrast, the lnK variance across test volumes shows only moderate variation between the two units. Analysis of K values across different support scales reveals a clear scale dependence for the Hanford formation, contrasting with Ringold, which exhibits scale-invariant behavior at the field scale. These differences arise from their distinct depositional processes; while the Ringold Formation was deposited gradually over geologic time by fluvial systems, producing consistent K, the Hanford formation was deposited abruptly by catastrophic glacial floods, leading to scale-dependent K variability. The study underscores that scale dependence in unconsolidated sand and gravel aquifers is common but not universal. Calibrated inverse modeling of regional groundwater flow yields high K estimates, with the average for the Hanford formation paleochannel being ~15,000 m/d, ranging from 1002 to 21,514 m/d. Multiple lines of evidence, including pumping tests, support these model-calibrated high K estimates for the Hanford formation paleochannel comprised of open framework gravels. For both sedimentary units, the upscaled K estimates align with the inverse model-calibrated estimates for non-channel portions of the Hanford and Ringold formations. While previous studies examined scale dependence using data from multiple sites, this study focuses on a single site with two sedimentary units analyzed across multiple support scales. To the best of our knowledge, this represents the most extensive compilation of K data for two gravel-dominated formations at the same site, incorporating both laboratory and field test results across varied scales.