Systematic Evaluation of the Costs and Benefits of Field Conditioning in Robust Remediation Design

Uncertainties in input parameters, such as hydraulic conductivities, may result in inconsistencies between simulated and observed groundwater flows and contaminant plumes. If a groundwater management model for aquifer remediation design is based upon incorrect input, non-optimal remediation policies may result. Typically, aquifer uncertainties have been simulated by generating multiple realizations of unconditional hydraulic conductivity fields. To reflect a truer picture of the actual hydraulic conductivity field, thereby improve on the performance of the optimal remediation design, random conductivity fields have been conditioned to honor the values of the field measurement. However, the potential benefits of conditioning the random conductivity fields within a robust groundwater remediation optimizer have never been systematically demonstrated. Furthermore, the additional measurements of conductivity needed for conditioning come at a financial cost. This study expands upon previous robust optimal design studies through incorporation of random fields conditioned on localized field observations. In the study, the impacts of conditioning on both the costs and the reliability of the optimal remediation designs are explored by comparison to robust optimization using unconditional random fields and by varying the number of data points utilized for conditioning. An example cost-benefit analysis of the value of additional samples for conditioning will be presented.