- Full (HTML) - Related EHP Articles - Purchase This Issue

Timothy L. Jacobs,¹ John M. Warmerdam,¹ Miguel A. Medina,¹ and Warren T. Piver²

¹Department of Civil and Environmental Engineering, Duke University, Durham, NC 27708-0287; ²National Institute of Environmental Health Sciences, Research Triangle Park, NC 27709 USA

Abstract
Pollutants in groundwater aquifers may constitute a significant human health risk. A large variation in response may result among human populations experiencing the same level and duration of exposure to pollutants. Variability in response, as a result of exposure to a carcinogenic contaminant such as trichloroethylene (TCE) , can be represented by a distribution function of safe doses. Spatial variability in aquifer characteristics and contaminant transport parameters requires the use of stochastic transport models to quantify variability in exposure concentrations. A second moment method is used to evaluate the probability of exceeding safe dose levels for a contaminated aquifer. The name of this method stems from the fact that the formulation is based on the first and second moments of the random variables. With this method, the probability is a function of the variability of contaminant concentration (which incorporates variability in hydrogeologic parameters such as hydraulic conductivity) and the variability in response in the human population. In this manner, the severity of the health risk posed by a contaminated aquifer and the evaluation of appropriate strategies and technologies for aquifer remediation are a function of contaminant concentrations and human health risks. The applicability and limitations of this method are demonstrated with data on groundwater contaminated by TCE at Hill Air Force Base, Utah. Key words: contaminant transport, groundwater flow, health risk, second moment method, stochastic modeling, trichloroethylene. Environ Health Perspect 104:866-870 (1996)

Address correspondence to T.L. Jacobs, Department of Civil and Environmental Engineering, Duke University, Box 90287, Durham, NC 27708-0287 USA.

Field data used in this article were provided by the U.S. Air Force under contract F08635-92-C-0009 to M. A. Medina and T. L. Jacobs, Duke University, to develop an advisory system for modeling and decision-making under conditions of uncertainty. Several site visits were conducted at Hill Air Force Base, Utah, coordinated by project officers Thomas B. Stauffer and Thomas P. de Venoge, Armstrong Laboratory, Tyndall AFB, Florida.

Recieved 1 August 1995 ; accepted 24 April 1996.