Groundwater Quality and Simulation of Sources of Water to Wells in the Marsh Creek Valley at the U.S. Geological Survey Northern Appalachian Research Laboratory, Tioga County, Pennsylvania: Usgs Scientific Investigations Report 2012-5042

This report provides a November 2010 snapshot of groundwater quality and an analysis of the sources of water to wells at the U.S. Geological Survey (USGS) Northern Appalachian Research Laboratory (NARL) near Wellsboro, Pennsylvania. The laboratory, which conducts fisheries research, currently (2011) withdraws 1,000 gallons per minute of high-quality groundwater from three wells completed in the glacial sand and gravel aquifer beneath the Marsh Creek valley; a fourth well that taps the same aquifer provides the potable supply for the facility. The study was conducted to document the source areas and quality of the water supply for this Department of Interior facility, which is surrounded by the ongoing development of natural gas from the Marcellus Shale. Groundwater samples were collected from the four wells used by the NARL and from two nearby domestic-supply wells. The domestic-supply wells withdraw groundwater from bedrock of the Catskill Formation. Samples were analyzed for major ions, nutrients, trace metals, radiochemicals, dissolved gases, and stable isotopes of oxygen and hydrogen in water and carbon in dissolved carbonate to document groundwater quality. Organic constituents (other than hydrocarbon gases) associated with hydraulic fracturing and other human activities were not analyzed as part of this assessment. Results show low concentrations of all constituents. Only radon, which ranged from 980 to 1,310 picocuries per liter, was somewhat elevated. These findings are consistent with the pristine nature of the aquifer in the Marsh Creek valley, which is the reason the laboratory was sited at this location. The sources of water and areas contributing recharge to wells were identified by the use of a previously documented MODFLOW groundwater-flow model for the following conditions: (1) withdrawals of 1,000 to 3,000 gallons per minute from the NARL wells, (2) average or dry hydrologic conditions, and (3) withdrawals of 1,000 gallons per minute from a new well 3,500 feet to the southwest that was drilled to provide water for Marcellus gas-well operations. Results of simulations indicate that during average hydrologic conditions, infiltration from Straight Run, a tributary to Marsh Creek, provides nearly all the water to the NARL wells. During dry conditions, the areas contributing recharge expand such that Asaph Run contributes about half of the water to the NARL wells when withdrawals are 1,000 or 2,000 gallons per minute. The addition of a simulated withdrawal of 1,000 gallons per minute from the nearby new well does not substantially affect the sources of water captured by the NARL wells. These results are subject to some limitations. The water-quality samples represent a snapshot of groundwater chemistry for only one hydrologic condition; the concentrations of some constituents may change temporally. In addition, samples were collected and analyzed for hydrocarbon gases, but not organic constituents associated with hydraulic fracturing; additional sampling for these constituents would provide a more complete water-quality baseline. The sources contributing water to the NARL wells and the new well were simulated by use of a simplified one-layer model of the glacial sand and gravel aquifer for steady-state conditions that in reality are never achieved. Steady-state simulations of dry hydrologic conditions show that it is possible for the NARL wells to capture water from Asaph Run; however, maps of simulated groundwater time-of-travel indicate that a dry period of unusually long duration would be required. A better analysis could be done by recalibrating the groundwater-flow model with a finite-difference grid having multiple layers, cells smaller than the 200-foot by 200-foot cells used in this study, and transient stress periods.