2004 Denver Annual Meeting (November 7–10, 2004)

Paper No. 10
Presentation Time: 4:10 PM


JONES, S.W., 761 Emory Valley Road, Oak Ridge, TN 37830, WHEAT, J.D., Oak Ridge, TN 37830, DAVIES, G.J., 109 Dixie Lane, Oak Ridge, TN 37830 and BENFIELD, Robert C., Department of Energy, Division of Oversight, State of Tennessee, 761 Emory Valley Rd, Oak Ridge, TN 37830, sid.jones@state.tn.us

The Oak Ridge Reservation (ORR) was placed on the U.S. EPA National Priorities List in 1989. Paleozoic carbonate rocks underlie about 60 percent of this U.S. Department of Energy (DOE) site in East Tennessee. Karst on the ORR was acknowledged since carbonates were the host to significant contamination. However, ground-water tracing with fluorescent dyes, a standard tool for determining ground-water flow paths and travel times in any carbonate setting, was rarely attempted prior to 1995. A site conceptual model suggested that rapid ground-water flow was generally limited to interflow after storms, and that some carbonate units behaved as aquitards. Numerical models of ground-water flow in areas with carbonate rocks were constructed ignoring rapid flow through karst pathways.

Tennessee Department of Environment and Conservation (TDEC) staff was not convinced that the carbonates within ORR boundaries were significantly different than carbonates elsewhere. TDEC staff, in conjunction with Cambrian Ground Water Company (CGWC) of Oak Ridge, demonstrated the significance of rapid flow along karst pathways with ground-water tracing tests. Beginning in 1995, ten ground-water tracing tests using fluorescent dyes were completed by TDEC and CGWC. The results showed that ORR ground-water flow paths could be kilometers long and that velocities were rapid and similar to ground-water velocities in carbonates in the rest of the world. In some cases, so-called scaling effects (differences between hydrogeologic data from well tests and tracer data) were striking, with traced velocities being many orders of magnitude faster than those inferred from numerical models.

A comparison of tracing data with pump, slug, and packer test data from numerous DOE investigations, confirmed a significant scaling effect on hydrogeologic parameters. This effect clearly demonstrates the limitation of using potentiometric data and information from small-scale hydrogeologic tests to deduce groundwater velocities or water budgets at the watershed or basin scale in this carbonate setting. To date, however, DOE has done little to modify the original site conceptual model, which held that rapid groundwater velocities in bedrock were the exception on the ORR.