2008 Joint Meeting of The Geological Society of America, Soil Science Society of America, American Society of Agronomy, Crop Science Society of America, Gulf Coast Association of Geological Societies with the Gulf Coast Section of SEPM

Paper No. 10
Presentation Time: 10:55 AM

Sensitivity Analysis of Contaminant Transport Uncertainty in Unsaturated Zone


PAN, Feng, Division of Hydrologic Science, Desert Research Institute, 755 E. Flamingo Road, Las Vegas, NV 89119, ZHU, Jianting, Division of Hydrologic Sciences, Desert Research Institute, 755 E Flamingo Road, Las Vegas, NV 89119, YE, Ming, School of Computational Science & Department of Geological Sciences, Florida State University, 441 Dirac Science Library, Tallahassee, FL 32306 and YU, Zhongbo, State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Hohai Univ, Nanjing 210098, China, Geoscience, Univ of Nevada at Las Vegas, 4505 Maryland Pkwy, Las Vegas, NV 89154-4010, Feng.Pan@dri.edu

The parametric uncertainty in hydrologic parameters (e.g., permeability, porosity, and water retention parameters) is significant in uncertainty of unsaturated flow and contaminant transport predictions in the unsaturated zone (UZ). While there have been many investigations addressing parametric uncertainty in the UZ of Yucca Mountain (YM), NV, the contribution of uncertainties in individual parameters to the uncertainty of flow and transport is yet to be systematically evaluated. The objective of this study is to identify the most influential parameters on unsaturated flow and contaminant transport uncertainty by conducting global sensitivity analysis. The probabilistic distributions of permeability and porosity for each hydrogeologic layer are identified using Lilliefors Test based on the site measurements. The distributions of the water retention parameters with sparse samples are estimated by the maximum likelihood estimation and Fisher information matrix. The random values of these parameters are generated using Latin Hypercube Sampling based on the determined probabilistic distributions. The Monte Carlo simulations of flow and transport are conducted using a 3-dimensional numerical model for the UZ of YM. The stepwise regression analysis is applied to investigate the relationships between the input parameters and the output variables. The importance of uncertainty in individual parameters to the flow and transport uncertainty is then ranked using the standardized rank regression coefficients. The outcome of this study provides important information which helps facilitate future data collection and monitoring efforts.