Horizontal drilling with hydraulic fracturing at shallow depths (less than 2,200 ft) in the Devonian Berea Sandstone oil and gas play, along with the potential for high-volume hydraulic fracturing in the nascent Cambrian Rogersville Shale gas play, have generated a renewed interest in protecting groundwater quality in eastern Kentucky. A critical component of protection is an accurate understanding of the distribution of fresh water in the subsurface. The “Fresh-Saline Water Interface Map of Kentucky” by H.T. Hopkins, published by the U.S. Geological Survey and Kentucky Geological Survey in 1966, has been a critical reference for assessing the maximum depth of fresh groundwater and is an important guidance document for well operators and regulatory agencies. To create the map, Hopkins assumed that total depth of domestic water wells equaled the base of fresh groundwater (total dissolved solids less than 1,000 ppm). Most domestic wells fail to penetrate the deepest fresh groundwater, however, and consequently, Hopkins’s map likely underestimates the depth of the fresh-saline water interface.
Our study also used total depths of wells to map the base of fresh groundwater, but increased the data density by adding data from domestic water wells drilled after 1966. In the 14-county study area, the number of wells increased from 50 used by Hopkins to 4,824 in this study. Total well depths were contour mapped using Petra software. Despite the increased data density, the inclusion of a greater number of shallow wells produced contour patterns that impeded resolution of deep fresh groundwater distribution (i.e., noise). To limit the influence of shallow wells, we eliminated wells with total depths above the elevations of watershed pour points in each watershed defined by 14- and 11-digit hydrologic unit codes. This excluded wells that did not penetrate the deepest fresh groundwater in low-order watersheds. We then created maps based on all wells with total depths below the elevations of their respective pour points in 14- and 11-digit hydrologic units (n = 3,203 and 854, respectively), as well as maps based on the single deepest well in the 14- and 11-digit hydrologic units (n = 1,420 and 74, respectively). The pour-point method improved the resolution of deep fresh groundwater distribution, and the map using the single deepest well depth in each 11-digit hydrologic unit provided the clearest illustration of deep fresh groundwater distribution.
Throughout most of the study area, the estimated depth of fresh groundwater derived from the 11-digit hydrologic unit deepest-well map is, on average, 147 ft deeper than the interface shown on the Hopkins map; in eastern Lawrence County, the difference exceeds 500 ft. Even though our study resulted in an improved estimate of maximum fresh groundwater depth, uncertainties remain in the data and methods. To reflect this uncertainty, the term “deepest observed fresh water” should be used as an alternative to “fresh-saline water interface.”
Report of Investigations 7
Digital Object Identifier (DOI)
Davis, E.S.L., Parris, T.M., and Grider, J., 2021, Deepest observed fresh groundwater contour lines for the Cumberland Plateau of eastern Kentucky: Kentucky Geological Survey Research Data, https://doi.org/10.13023/kgs.data.2021.01.
Well data were retrieved and supplied by Bart Davidson and Carrie Pulliam.
Davis, Ethan S. L.; Parris, Thomas M.; and Grider, Jerrad, "Using Watershed Pour-Point Elevations to Evaluate the Base of Fresh Groundwater in the Cumberland Plateau of Eastern Kentucky" (2019). Kentucky Geological Survey Report of Investigations. 46.