STRIKING THE BALANCE: LONG TERM GROUNDWATER MONITORING DESIGN FOR MULTIPLE CONFLICTING OBJECTIVES

摘要

The National Research Council (NRC) (1997) estimates that between 300,000 and 400,000 contaminated sites within the United States will require a potential remedial cost of up to a trillion dollars. The tremendous magnitude of these costs has motivated a shift in the environmental industry toward risk-based approaches to remedial design. Risk-based approaches often involve leaving some contamination in place, which can require significant long-term monitoring (LTM). Long-term monitoring at many sites can involve decades of expensive sampling at tens or even hundreds of existing monitoring wells, resulting in hundreds of thousands or millions of dollars for sampling and data management per year. Developing efficient and effective LTM sampling plans can be difficult when numerous options exist. In any given monitoring period, the number of possible sampling plans is 2n, where n is the product of the number of wells and the number of possible constituents to be measured. A site with 10 wells where up to 3 constituents can be measured (n = 30) requires the decision maker to identify a sampling plan among the more than one billion that exist while also trying to balance the cost and other performance objectives for each sampling scheme. Assume that the practitioner can eliminate 99 percent of these designs using intuition and further assume that it only takes him or her 1 second to evaluate each of the remaining designs. A comprehensive analysis of the remaining designs would still require more than a full work year.