HR: 1330h
AN: H12A-0960    [PDF]
TI: First-Order Sensitivity Analysis of the Variation in Temperature History at Yucca Mountain due to Thermal-hydrologic Parameter Uncertainty
AU: * Glascoe, L
EM: glascoe@llnl.gov
AF: Lawrence Livermore National Laboratory, PO Box 808, L-646, Livermore, CA 94550 United States
AU: Loosmore, G
EM: loosmore1@llnl.gov
AF: Lawrence Livermore National Laboratory, PO Box 808, L-646, Livermore, CA 94550 United States
AU: Buscheck, T
EM: buscheck1@llnl.gov
AF: Lawrence Livermore National Laboratory, PO Box 808, L-646, Livermore, CA 94550 United States
AU: Sun, Y
EM: sun4@llnl.gov
AF: Lawrence Livermore National Laboratory, PO Box 808, L-646, Livermore, CA 94550 United States
AU: Gansemer, J
EM: gansemer1@llnl.gov
AF: Lawrence Livermore National Laboratory, PO Box 808, L-646, Livermore, CA 94550 United States
AU: Lee, K
EM: lee23@llnl.gov
AF: Lawrence Livermore National Laboratory, PO Box 808, L-646, Livermore, CA 94550 United States
AB: Sensitivity analysis can be used for both uncertainty quantification and model simplification. Presented here is a thermal-hydrologic example with a first-order sensitivity analysis addressing the response of temperature to uncertainty in heat-transfer and hydrologic parameters. The Multi-Scale Thermal-Hydrologic Model (MSTHM) employs LLNL's Non-isothermal Unsaturated/saturated Flow and Transport (NUFT) code to simulate mountain and drift-scale heat transfer and hydrology for the proposed waste repository at Yucca Mountain. For MSTHM predictions of thermal-hydrologic conditions within the emplacement drifts and in the adjoining host rock, the key uncertain parameters include thermal-conductivity and percolation flux. Discussed here is a first-order sensitivity analysis of temperature response to the uncertainty of these two parameters. To address the impact of percolation-flux uncertainty, MSTHM simulations are run for three (mean, upper-bound, and lower-bound) infiltration-flux cases; to address the impact of thermal-conductivity uncertainty, MSTHM simulations are run for three (mean, +1 standard deviation, -1 standard deviation) cases. The sensitivity analysis anticipates the combined impact of thermal-conductivity and percolation uncertainty for the temperature response. MSTHM simulation illustrates that the first-order approach yields accurate prediction; the prediction is more accurate for perturbations in thermal-conductivity than for perturbations in percolation flux. It is anticipated that contours of predictive accuracy could be determined by such a sensitivity analysis as both parameters are monotonic with respect to temperature. Such contours of predictive accuracy may be useful in reducing the overall number of simulations necessary for bracketing the effect of parameter uncertainty. This work was performed under the auspices of the U.S. Department of Energy by University of California Lawrence Livermore National Laboratory under contract No. W-7405-Eng-48.
DE: 1803 Anthropogenic effects
DE: 1829 Groundwater hydrology
DE: 1833 Hydroclimatology
DE: 1878 Water/energy interactions
SC: Hydrology [H]
MN: 2003 Fall Meeting