CHARACTERIZING GROUNDWATER ALTERATION BY HYDAULIC FRACTURING USING RADIONUCLIDES AND STABLE ISOTOPES

Hydraulic fracturing used for shale gas extraction has garnered a great deal of attention and criticism regarding its potential to pollute shallow groundwater and surface water. Over the past decade, several studies have shown that hydraulic fracturing may cause contamination of groundwater and/or surface water. However, the broader scientific literature on this subject is still limited in terms of methodological scope and geographic context, and has been focused primarily on conventional, water-based hydraulic fracturing techniques implemented in the Marcellus Shale region of western Pennsylvania, USA. As a result, the applicability of these findings in alternative shale gas plays – e.g., the shallower Chattanooga Shale, which is fractured using nitrogen foam instead of water – remains largely untested.

This study characterizes groundwater in Letcher County, Kentucky, USA using chemical and isotopic tracers to identify indicators of water pollution from hydraulic fracturing in the Chattanooga Shale. Groundwater samples collected from private drinking water wells have been analyzed for: 1) concentrations of major ions, metals, methane gas, and radon gas; 2) δ¹³C and δ²H composition of CH₄; and 3) δ³⁴S and δ¹⁸O composition of SO₄. Results from these analyses have been analyzed for each sample site using multiple regression to determine correlations between drinking well water composition and proximity to hydraulically fractured shale gas wells. The methods and analyses implemented in this study are meant to largely mimic previous research in the Marcellus Shale region to verify its applicability in shallower drilling contexts that use less hydraulically-intensive methods to fracture shale deposits.

The chemical composition of the studied groundwater from the Chattanooga Shale has shown different variations compared to the Marcellus Shale. Concentrations of methane gas have shown to marginally increase with proximity to the hydraulically fractured shale gas wells. In contrast, concentrations of radon gas have not exhibited a pattern as a function of proximity to hydraulically fractured wells. The wide variation of δ³⁴S in dissolved sulfate (-1.2 to +34.3 ‰) suggests mixing processes between waters of different origin and/or subsequent alteration by microbial sulfate reduction.