2002 Denver Annual Meeting (October 27-30, 2002)

Paper No. 4
Presentation Time: 2:15 PM

INTEGRATED SUBSURFACE IMAGING TECHNIQUES FOR DETECTING CAVITIES IN OKLAHOMA EVAPORITE KARST


TARHULE, Aondover1, HALIHAN, Todd2, DEWERS, Thomas3, YOUNG, Roger3 and WITTEN, Alan3, (1)Department of Geography, Univ of Oklahoma, 100 East Boyd Street, Suite 684, Norman, OK 73019, (2)School of Geology, Oklahoma State Univ, 105 Noble Research Center, Stillwater, OK 74078, (3)School of Geology and Geophysics, Univ of Oklahoma, 100 East Boyd Street, Suite 810, Norman, OK 73019, atarhule@ou.edu

Rapid dynamics of gypsum karst development constitutes a serious collapse hazard for surface structures, including highways, on a human rather than geological time scale. Gypsum and anhydrite deposits occur in the shallow subsurface (< 30 m below ground surface; m.b.g.s.) in about 4% of Oklahoma, USA. This study presents preliminary results of an integrated study utilizing subsurface imaging techniques to non-intrusively detect and map solutional cavities in the Permian Blaine gypsum karst of western Oklahoma. The study employed electrical resistivity tomography (ERT), broadband electromagnetic induction (EMI), seismic reflection, and seismic refraction to detect, delineate and map karst geohazards within the Nescatunga cave system of Northwestern Oklahoma. The combined use of several subsurface imaging techniques provides opportunities for evaluating the relative effectiveness of the methods in detecting cavities and minimizing uncertainty in the interpretation of subsurface features, especially where reasonable agreement was achieved among the methods. Subsurface cross sectional plots from the various methods were superimposed in ARCVIEW GIS and the Spatial Analyst tool was used to establish the magnitude of error in detected features from their true location. Interpreted cavern locations are compared to three dimensional digital maps of caverns obtained by a laser positioning method (discussed elsewhere in this session). The results provide a measure of relative accuracy of modeled cross sections. The approach adopted is superior to conventional site characterization methods, which typically involve intensive field drilling programs. The ultimate goal is to produce a methodology for locating sinkholes and cavities that constitute geohazards in the karst environments of Oklahoma, and perhaps elsewhere in the United States.