T265 Groundwater Model Calibration and Uncertainty Analysis

Wednesday, 7 November 2012
1:30 PM - 5:30 PM
207BC

TESTING HYDROLOGIC UTILITY OF GEOLOGIC FRAMEWORKS

Paper number: T265-4
Program: Technical Sessions
Time: 2:20 PM – 2:35 PM

Authors

Abstract

Hydraulic properties frequently are distributed in groundwater flow and transport models with geologic frameworks. Creation of complex frameworks that support hydrologic investigations can be a reasonable endeavor where the hydraulic significance of geologic units and fault structures are substantiated with water-level and hydraulic property data. Proposed connections between geology and hydraulic properties seem more tenuous when justified with scatter plots that lack correlation. Weak correlations between hydraulic property estimates and geologic structures have been developed primarily with hydraulic property estimates from physical property measurements at length scales of less than 10 m. Water levels primarily are the hydraulic data that constrain groundwater-flow model calibration and typically are sampled at length scales of more than 1,000 m. Differences in length scales between weak correlations and hydraulic data commonly are bridged with assumed geologic relations. Differences in length scales between hydraulic properties and hydraulic data commonly are bridged with assumed geologic relations.

The utility of geologic frameworks for extrapolating hydraulic properties at length scales that are commensurate with hydraulic data has been assessed at the Nevada National Security Site in highly faulted, volcanic rocks. Observed drawdowns from multiple aquifer tests provided the necessary constraints. Bulk hydraulic properties in a volume of more than 20 cubic miles were defined primarily by drawdowns that were detected across fault structures and located more than 2 mi from pumping wells. Hydraulic properties were estimated by simultaneously interpreting all aquifer tests with three geologic frameworks. Each geologic framework was incorporated and tested as prior information that assumed homogeneity in each geologic structure. Hydraulic differences between geologic units and fault structures were less than the inherent uncertainty of the hydraulic conductivity estimates. Geologic frameworks that conceptualized fault structures as unique hydraulic features degraded extrapolation of hydraulic properties because uncertainty was increased in the two hydraulically significant geologic units.