News Release

New methodology reduces costs of monitoring contaminated sites

Peer-Reviewed Publication

University of Illinois at Urbana-Champaign, News Bureau

CHAMPAIGN, Ill. -- Researchers at the University of Illinois have developed an optimization tool that can reduce the costs of long-term sampling and monitoring at contaminated groundwater sites.

"There are hundreds of thousands of contaminated sites in the United States alone, and cleaning them up could take many years and cost billions of dollars," said Barbara Minsker, a professor of civil and environmental engineering at the U. of I. "Whether the goal is to contain the contaminant or to remediate it, wells must be drilled at the site and routinely sampled and monitored."

Large, complex sites may have hundreds of monitoring wells, and performing the necessary sampling to ensure regulatory compliance can cost millions of dollars a year, Minsker said. Determining the correct number and placement of wells to create the most cost-effective monitoring design by "trial and error" is complicated, tedious and time-consuming.

"We have developed a methodology that combines optimization, simulation and plume-interpolation tools to identify sampling locations that accurately quantify the mass of contaminant in the groundwater while minimizing monitoring costs," Minsker said. "Reliable estimation of the contaminant mass is critical to ensuring that remediation is proceeding sufficiently to protect human health and the environment."

The methodology developed by Minsker and her U. of I. colleagues -- civil and environmental engineering professor Albert Valocchi and graduate student Patrick Reed -- consists of three primary components: a groundwater fate-and-transport simulation, several plume-interpolation modules, and a monitoring plan selection process that uses a genetic algorithm.

"The genetic algorithm identifies optimal subsets of monitoring wells in a process analogous to the Darwinian concept of natural selection," Minsker said. "The algorithm searches the decision space and identifies subsets of wells that accurately describe the contaminant plume at minimal costs."

The genetic algorithm then assigns each candidate monitoring plan a "fitness" based on both cost and error, where the error is determined by the absolute difference between the best mass estimate and each candidate plan's estimate. The "fittest" plans are allowed to breed and survive into later generations, evolving into even better designs.

"Results have shown that our methodology is very effective at both reducing monitoring costs and accurately quantifying the mass of contaminant in the plume," said Minsker.

Patrick Reed presented the team's findings at the American Society of Civil Engineers meeting June 7 in Phoenix. The team's research is supported by the Illinois Water Resources Center and the U.S. Geological Survey.

###


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.