Groundwater Flow Model Development at the Nevada Test Siteby Gregory J. Ruskauff,
Document Type: Proceeding Paper
Part of: World Environmental and Water Resources Congress 2008: Ahupua'A
The Underground Test Area Project (UGTA) of the National Nuclear Security Administration.s Nevada Site Office is in the process of assessing and developing regulatory decision options based on modeling predictions of contaminant transport from underground testing of nuclear weapons at the Nevada Test Site (NTS). The modeling task presents multiple unique challenges to the hydrological sciences. The structural and stratigraphic framework of the NTS is complex and consists of Paleozoic and older carbonate and clastic rocks overlain by a diverse assemblage of mid-Cenozoic volcanic rocks derived from large coalesced calderas. Lateral facies changes in the volcanic rocks are significant and the carbonate and volcanic assemblages are disrupted by basin-range faulting that locally formed thick basins infilled with alluvial deposits. Groundwater flow through the geologic units is predominantly within carbonate, volcanic and alluvial aquifers with local interfingering aquitards that are all part of the internally drained Death Valley Regional Flow System (DVRFS) of the southern Great Basin. Characterization of the hydrogeologic system is difficult and expensive because of deep groundwater in the arid desert setting, thus high-level uncertainty remains in formulating conceptual and numerical models of the groundwater system. The UGTA Project attempts to address the uncertainty issue developing multiple possible models of the geologic framework that are then tested for their ability to match observed hydrologic conditions — hydraulic head, geologic unit characterization data, geochemical data, and estimated groundwater fluxes are all considered. The complex geologic geometries are handled by using a finite-element code, FEHM (developed by Los Alamos National Laboratories), to flexibly represent the geology. Model calibration is conducted on a Linux computing cluster using commodity-off-the-shelf equipment to achieve the high-speed calculations necessary for these complex models. Parameter estimation software, PEST (developed by Watermark Numerical Computing), is used to calibrate the models using singular value decomposition and Tikhonov regularization — necessary to get a stable result in these complicated models.
Services: Buy this book/Buy this article
Return to search