CHARACTERIZING FAULTS IN A KARST AQUIFER USING BOREHOLE LOG AND AQUIFER TEST ANALYSES

In Appalachian karst terrains, faults are zones of high permeability along which increased fracture density serves to promote dissolution and conduit development. Commonly, structural geologists rely upon the location of springs, sinkholes, or other expressions of karst geomorphology to infer the mapped location of displacement. As a result, groundwater hydrologists may regard mapped fault locations as proximate, particularly in mature karst regions where outcrop data are limited. In the Great Valley of West Virginia, borehole geophysical logs show that the inferred locations of displacement are consistent with zones of increased porosity that, in cases, are ideal for targeting water-supply wells. However, drilling for large supplies in karst aquifers is complicated by turbidity issues that may preclude a well from use. Further, due to the complexity of karst development, variability in fracturing, and difficulty in predicting subsurface heterogeneities in carbonate aquifers, characterizing fault zones prior to exploratory drilling is challenging. Site specific data are used to discern trends from one location that may be useful in predicting the karstification of an area prior to drilling. Using fractal analysis, systematic variation in karst features is depicted from borehole logs collected in diffuse, conduit, and faulted regions of the aquifer. Similarly, field measurements of fracture and calcite vein exposures conform to a power-law distribution that can be used to describe the karst continuum of the Great Valley and fractures attributed to displacement along faults. Additional data from constant discharge tests show fault-induced heterogeneity with periods of linear and radial flow that reflect variation of fracture density and karstification in specific areas. Although no general conclusions of flow in karst systems can be drawn, information from detailed examination of individual faults may be useful in resource management decisions.