GSA Annual Meeting, November 5-8, 2001

Paper No. 0
Presentation Time: 8:45 AM

COMBINED INTERPRETATION OF RADAR DIFFERENCE-ATTENUATION TOMOGRAMS AND SALINE-TRACER TESTS IN FRACTURED ROCK


DAY-LEWIS, Frederick D., Department of Geology, Bucknell Univ, Carnegie Building, Lewisburg, PA 17837, LANE Jr, John W., Branch of Geophysics, U.S. Geol Survey, 11 Sherman Place, U-5015, Storrs, CT 06269, HARRIS, Jerry M., Department of Geophysics, Stanford Univ, Mitchell Building, Stanford, CA 94305 and GORELICK, Steven M., Department of Geological and Environmental Sciences, Stanford Univ, Bldg 320, Stanford, CA 94305, fdaylewi@bucknell.edu

The combination of difference-attenuation radar tomography, saline-tracer doublet tests, and inverse flow and transport models can aid in aquifer characterization and delineation of preferential flow paths. We demonstrate a combined interpretation of tracer and cross-hole difference-attenuation data from an experiment at the U.S. Geological Survey Fractured Rock Hydrology Research Site in Grafton County, New Hampshire. A novel sequential inversion approach is used to generate time-lapse difference-attenuation tomograms in three planes at ten-minute intervals throughout the experiment. The method inverts the tomographic radar data on a time series of 3-D nodal meshes and accounts for staggered data acquisition times and space-time correlation of model parameters. The tomograms are then used to infer tracer arrival times and relative tracer concentrations in the imaged planes. Based on the geophysical data and the tracer concentration history at the pumping well, a suite of three-dimensional groundwater-flow and solute-transport models were constructed and using non-linear regression were calibrated to (1) tracer concentrations, (2) arrival time of peak tracer concentration at the outlet, and (3) steady-state hydraulic head at the outlet. We considered conceptual hydrologic models involving different combinations of zonal aquifer heterogeneity and rate-limited mass transfer as possible explanations for the observed tracer and geophysical data. The results of the hydrologic inversion procedure indicate that tracer is likely transported through two pathways and that much of the tracer left the tomographic imaged planes. Fast transport occurs primarily through a preferential channel within a previously identified highly transmissive zone. The quantitative combination of high-resolution geophysical, tracer, and hydraulic data yields a more detailed and robust interpretation of heterogeneity at the site and demonstrates an effective method for characterizing flowpaths between boreholes in fractured rock.