Pre-drilling water-quality data of groundwater prior to shale gas drilling in the Appalachian Basin: Analysis of the Chesapeake Energy Corporation dataset

Unconventional natural gas production in the Marcellus and Utica formations of the Northeastern United States raises concerns about potential impacts to shallow groundwater. We examined and interpreted 13,040 analyses from pre-drilling groundwater samples from domestic water wells in northeastern (NE) Pennsylvania and 8004 samples from water wells in the Western Area which includes southwest Pennsylvania, eastern Ohio, and north-central West Virginia. These samples were acquired on behalf of Chesapeake Energy Corporation as part of its local pre-drilling water supply monitoring program. We evaluated concentrations of major ions and metals relative to federal drinking-water-quality standards upon which regulatory decisions are often based. Chesapeake's dataset, the most comprehensive for these areas, shows that exceedance of at least one water-quality standard occurs in 63% of water well samples in NE Pennsylvania and 87% in the Western Area. In NE Pennsylvania, 10% of the samples exceeded one or more of the United States Environmental Protection Agency's (USEPA) primary maximum contaminant levels (MCLs) for drinking-water supplies, 46.1% of the samples exceeded one or more of USEPA secondary maximum contaminant levels (SMCLs), and another 7% exceeded one or more of USEPA health advisory or regional screening levels for tap water. In the Western Area 8% of samples exceeded one or more MCLs, 65% exceeded one or more SMCLs, and 15% exceeded one or more health advisory or regional screening levels for tap water. Chesapeake's dataset, orders of magnitude larger than any in previously published literature, shows that water-quality exceedances relate to factors such: as where the sample occurs within the groundwater flow system, the natural groundwater chemical type (hydrochemical facies), the geologic unit producing the water, and/or the topographic position (valley versus upland). Our comparison of these results to historical groundwater data from NE Pennsylvania, which pre-dates most unconventional shale gas development, shows that the recent pre-drilling geochemical data is similar to historical data. We see no broad changes in variability of chemical quality in this large dataset to suggest any unusual salinization caused by possible release of produced waters from oil and gas operations, even after thousands of gas wells have been drilled among tens of thousands of domestic wells within the two areas studied. Our evaluation also agrees with early researchers such as Piper (1933) and Lohman (1939, 1937) who found that the saline waters in both areas underlie fresher groundwater. The saline water is naturally-occurring connate brine or salt water which has not been flushed by circulating meteoric water; rather than vertical migration of salt water from deep strata such as the Marcellus shale as suggested by Warner et al. (2012). Elevated metals concentrations, particularly iron and manganese, partly relate to sample turbidity; dissolved metals would provide a more accurate measurement of metals in shallow groundwater than does the total metals analysis typically required by regulations. (C) 2015 Elsevier Ltd. All rights reserved.