Author : Xiaoyi Liu
Publisher : Stanford University
ISBN 13 :
Total Pages : 215 pages
Book Rating : 4.F/5 ( download)
Book Synopsis Estimation, Optimization, and Value of Information in Groundwater Remediation by : Xiaoyi Liu
Download or read book Estimation, Optimization, and Value of Information in Groundwater Remediation written by Xiaoyi Liu and published by Stanford University. This book was released on 2011 with total page 215 pages. Available in PDF, EPUB and Kindle. Book excerpt: Solving groundwater problems involves a system of methods in characterization and optimization. However, no matter how theoretically sound a method may be, when it is applied in the field, uncertainty is always an important factor that cannot be neglected. Indeed, a good theory or method has to be validated in field applications, and uncertainty propagates from one stage (e.g. characterization) of the application to the next (e.g. remediation optimization). Thus, it is essential for such methods not only to include uncertainty but also to quantify uncertainty in a practical sense. This dissertation covers three important topics in groundwater remediation: site characterization, remediation optimization, and value of information. First, groundwater contamination site characterization with Monte Carlo methods, specifically, Markov chain Monte Carlo (MCMC) methods, is introduced. Then, another subsurface characterization method, stochastic inverse modeling, is covered with an emphasis on solving large-scale characterization problems, meaning resolving the subsurface heterogeneity with a high-resolution. Both methods provide estimation of the site and uncertainty about the estimation. When these characterization results are used for site management such as remediation management, effects of uncertainty on optimization need to be quantified under the specific context of the remediation problem, and traditional measures of uncertainty such as variance and correlation coefficients cannot handle this job well because they do not necessarily depict the severity of uncertainty. In this dissertation, value of information (VOI), defined as the expected cost with the present state of uncertainty minus the expected cost if uncertainty were fully or partially removed, is proposed as a context-specific measure of uncertainty, i.e., it is dependent on site conditions and remediation strategies as well as specific remediation objectives and unit costs. Several laboratory and numerical applications on hydraulic tomography and DNAPL contamination remediation are included in this dissertation to show the efficacy of the methods proposed.
