EVALUATION OF BACKGROUND MERCURY CONCENTRATIONS IN THE SRS GROUNDWATER SYSTEM.

JACKSON, DENNIS G., BRIAN B. LOONEY, and MILES E. DENHAM, SRTC-Environmental Sciences & Technology, Westinghouse Savannah River Site, Aiken SC

Methods of order statistics have been combined with probability plotting and partitioning techniques to estimate the background concentrations of constituents at the Savannah River Site. Methods of order statistics are commonly used in the field of exploration geochemistry to extract information about ore distributions and to identify mineral reserves. These methods have been applied at the SRS to estimate the background groundwater concentration of mercury in the site's groundwater system. The estimated concentration obtained from this analysis is below the nominal detection limit of traditional EPA approved analysis methods. The data used in developing this estimate includes values reported as estimated concentrations, values below nominal detection limits, and actual reported concentrations. The analysis provides a reasonable upper bound on the ambient, background concentration for mercury in the SRS groundwater system. The methods utilized are not limited exclusively to mercury analysis. The robust techniques are valid for analysis and partitioning of large data sets that are postulated to comprise a combination of normal or log normal populations. Through the application process the hypothesis is validated in a simple graphical manner and quantitative results are obtained. The analysis illustrates one possible technique of incorporating data that is traditionally unused and data that is considered unusable with validated data.