Natural Attenuation: Using Statistics to Understand Ground Water Contamination

At Work on Groundwater Treatment Processes

Infrastructure Security Article December 28, 2000
Printer-friendly version

When ground water contamination is discovered at a site, a host of questions must be answered before it can be effectively removed or minimized. Battelle Memorial Institute, an international research organization headquartered in Columbus, Ohio, has recently developed a statistical approach for assessing natural attenuation of contaminants. This approach can be used to estimate how quickly contaminants can be remediated using biological methods.

The statistical approach, that utilizes geostatistical techniques such as kriging, is designed to address key questions such as how big is the area of the contamination (plume)? Is it growing, shrinking or stable? How fast is the size of the plume changing and how long will it take for the contamination levels to fall below a specified acceptable level? How quickly is contamination naturally lessened or thinned (natural attenuation)? What is the rate at which biological processes reduce contamination (biodegradation)?

Battelle's statistical tool was applied at a government military installation where trichloroethylene (TCE and chromium were discovered in groundwater. The Battelle team used data collected from 152 wells over a period of three years to analyze the rate at which contamination was decreasing in three levels of the aquifier. Although TCE appeared to be decreasing by 12 percent per year in the middle portion of the aquifier, no conclusions could be made about decreases in the upper and lower levels. On the other hand, the analysis concluded that chromium levels were decreasing in all layers of the aquifier at rates that ranged from 25 to 90 percent per year.

Tad Fox, Battelle's principal investigator, noted that the statistical approach gave government personnel confidence that both best-case and worst-case scenarios had been identified with respect to the rate of change in contamination levels. This statistical approach was first presented in a paper at Battelle's Fourth In Situ and On-Site Bioremediation Symposium in 1997.

About the author: 
Overlay Init