Title

Critical Issues Related to a Combined Probabilistic Numerical Analysis of Contaminant Transport in Porous Media

Abstract

Reliability algorithms provide attractive alternatives to Monte Carlo simulation and stochastic perturbation approaches for probabilistic analysis of contaminant transport in porous media. We describe results and issues for further study based on a first-order reliability analysis in conjunction with a two-dimensional finite element model of transport in porous media. Probabilistic sensitivity measures provide valuable information on the importance of the uncertain variables - particularly the spatially discretized permeability values. An example flow region is analyzed and longitudinal dispersivity, fluid density, solid grain density, and distribution coefficient have the greatest influence on the probabilistic outcome. These are all global uncertain variables, as opposed to the discretized permeability uncertain variables. The authors' previous work with analytical transport solutions indicated dispersivity had little influence on the probabilistic outcome; rather, those models were dominated by the uncertainty in flow velocity and the reaction terms.

Meeting Name

6th ASCE Specialty Conference on Probabilistic Mechanics, and Structural and Geotechnical Reliability (1992: Jul. 8-10, Denver, CO)

Department(s)

Geosciences and Geological and Petroleum Engineering

Keywords and Phrases

Finite Element Method; Monte Carlo Methods; Numerical Analysis; Probability; Sensitivity Analysis; Combined Probabilistic Numerical Analysis; First-Order Reliability Analysis; Porous Media Contaminant Transport; Probabilistic Sensitivity Measures; Spatially Discretized Permeability Values; Two-Dimensional Finite Element Model; Porous Materials

International Standard Book Number (ISBN)

872628736

Document Type

Article - Conference proceedings

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 1992 American Society of Civil Engineers (ASCE), All rights reserved.


Share

 
COinS