Northeastern Section (39th Annual) and Southeastern Section (53rd Annual) Joint Meeting (March 25–27, 2004)

Paper No. 2
Presentation Time: 8:20 AM

FORECASTING DRINKING-WATER RISKS IN KARSTIC TERRANES


FIELD, Malcolm S., Office of Research and Development, U.S. Environmental Protection Agency, National Center for Environmental Assessment (8623D), 1200 Pennsylvania Ave., NW, Washington, DC 20837-2242, field.malcolm@epa.gov

Assessing risks from exposure to contaminated ground water requires detailed knowledge of the ground-water system itself as well as the contaminants in the ground water Exposure assessment is often the most difficult part of a ground water risk assessment because it requires quantification of several facets: pollutant release rates from the pollutant sources, environmental fate-and-transport through the ground water system, concentrations at exposure points as a function of time, the number and types of people exposed, and exposure modeling for different pathways (drinking, bathing, irrigation, recreation). If releases have already occurred, then exposures and risks may be computed from measured concentrations at receptor locations abrogating the need for additional studies and modeling. However, it is desirable that risks be forecast prior to any pollutant releases so that management plans may be set in place and human health adequately protected.

Forecasting environmental and human health risks to drinking-water supplies in karstic terranes are best accomplished by conducting numerous quantitative-tracer tests from each potential source location to each exposure point so that acceptable solute-transport parameters may be estimated from analyses of the breakthrough curves. Breakthrough curves developed from quantitative tracer tests are readily amenable to numerical analyses and modeling efforts which yield basic solute fate-and-transport parameters such as time-of-travel, velocity, dispersion, and decay. Alternatively, pollutant fate-and-transport parameters may be estimated using basic measured field parameters combined in functional relationships and applied in the advection-dispersion model. Once these parameters are established, the models can be run to generate a time history of concentrations at exposure points which feed directly into the exposure assessments.

The estimated exposures may then be used to forecast risks. Calculated exposures and subsequent risks will still depend on the type of release (discrete or continuous) and whether the substance is a chemical or biological agent. If a biological agent is released, then it may also be necessary to assess the capability of the biological agent to be transmitted as a disease through a population via person-to-person contact.