HR: 09:30h
AN: H11E-07    [Abstracts]
TI: An Adaptive Response Surface Modeling Method for Uncertainty Quantification with Application to a Transient Regional Groundwater Model
AU: * Scheibe, T D
EM: tim.scheibe@pnl.gov
AF: Pacific Northwest National Laboratory, P.O. Box 999, Richland, WA 99352 United States
AU: Engel, D W
EM: dave.engel@pnl.gov
AF: Pacific Northwest National Laboratory, P.O. Box 999, Richland, WA 99352 United States
AU: Jarman, K D
EM: kj@pnl.gov
AF: Pacific Northwest National Laboratory, P.O. Box 999, Richland, WA 99352 United States
AB: Identification of regions of an uncertain parameter space that correspond to predictions that are consistent with observed system responses is the focus of model calibration / inverse modeling efforts. Many systematic model inversion methods utilize gradient search methods to find an optimal solution based on minimization of differences between predicted and observed responses. Associated uncertainty estimates are commonly based on local model sensitivities at the optimal solution parameter set. We have developed an alternative approach, based on iterative approximation of the complete model response surface, that provides a means of assessing global parameter uncertainty. We apply the methodology to a regional transient groundwater model of the U. S. Department of Energy's Hanford Site and compare our approach with results of parameter estimation and uncertainty assessment obtained by a conventional inverse modeling approach.
DE: 1805 Computational hydrology
DE: 1829 Groundwater hydrology
DE: 1846 Model calibration (3333)
DE: 1847 Modeling
DE: 1873 Uncertainty assessment (3275)
SC: Hydrology [H]
MN: Fall Meeting 2005