2009 Portland GSA Annual Meeting (18-21 October 2009)

Paper No. 13
Presentation Time: 4:55 PM

UTILIZING GROUND PENETRATING RADAR (GPR) TO VISUALIZE AND CHARACTERIZE HYDRAULICALLY CONDUCTIVE BEDROCK FRACTURES


BAKER, Gregroy S., Department of Earth and Planetary Sciences, University of Tennessee, 306 Earth and Planetary Sciences Building, Knoxville, TN 37996-1410, gbaker@tennessee.edu

Fluid flow in the subsurface is an important topic related to both “clean” and “dirty” water-related issues. While fluid flow in unconsolidated media is fairly well defined by numerical approximations and somewhat predictable, fluid flow through fractured media is typically poorly constrained and the current governing field-scale hydrologic equations are difficult to test. One method for simulating groundwater flow in fractured media is to model the target environment as an equivalent porous media. Inherent in this model is the cubic law that illuminates the significance of reliable fracture characterization, e.g., doubling the fracture aperture size will result in an 8-fold increase in discharge. In spite of the importance of fracture aperture, however, there are significant challenges when determining fracture aperture distribution in situ. One method that had been used successfully to characterize bedrock fractures in situ is ground penetrating radar (GPR) and several case studies will be presented. Various GPR techniques have been used to generate both static and time-lapse images of fractures and fluid migration in fractures. To date, however, these images have only yielded quantitative or loosely qualitative fracture estimates due to a poor understanding of the in situ relationship between fracture aperture and GPR reflection amplitudes. Data will be presented indicating that “accepted” equations defining GPR reflection amplitude as a function of fracture aperture fail when applied under controlled conditions. In developing GPR for future use in hydrogeology to aid in determining the critical parameter of fracture aperture, relevant relationships must be identified and more appropriately defined.