Environmental Factors Associated With Natural Methane Occurrence in the Appalachian Basin |
| |
| Authors: | Lisa J. Molofsky John A. Connor Thomas E. McHugh Stephen D. Richardson Casper Woroszylo Pedro J. Alvarez |
| |
| Affiliation: | 1. GSI Environmental Inc., Houston, TX;2. GSI Environmental Inc., Austin, TX;3. Department of Civil and Environmental Engineering and Department of Statistics, Rice University, Houston, TX |
| |
| Abstract: | The recent boom in shale gas development in the Marcellus Shale has increased interest in the methods to distinguish between naturally occurring methane in groundwater and stray methane associated with drilling and production operations. This study evaluates the relationship between natural methane occurrence and three principal environmental factors (groundwater redox state, water type, and topography) using two pre‐drill datasets of 132 samples from western Pennsylvania, Ohio, and West Virginia and 1417 samples from northeastern Pennsylvania. Higher natural methane concentrations in residential wells are strongly associated with reducing conditions characterized by low nitrate and low sulfate ([NO3?] < 0.5 mg/L; [SO42?] < 2.5 mg/L). However, no significant relationship exists between methane and iron [Fe(II)], which is traditionally considered an indicator of conditions that have progressed through iron reduction. As shown in previous studies, water type is significantly correlated with natural methane concentrations, where sodium (Na) ‐rich waters exhibit significantly higher (p<0.001) natural methane concentrations than calcium (Ca)‐rich waters. For water wells exhibiting Na‐rich waters and/or low nitrate and low sulfate conditions, valley locations are associated with higher methane concentrations than upland topography. Consequently, we identify three factors (“Low NO3? & SO42?” redox condition, Na‐rich water type, and valley location), which, in combination, offer strong predictive power regarding the natural occurrence of high methane concentrations. Samples exhibiting these three factors have a median methane concentration of 10,000 µg/L. These heuristic relationships may facilitate the design of pre‐drill monitoring programs and the subsequent evaluation of post‐drill monitoring results to help distinguish between naturally occurring methane and methane originating from anthropogenic sources or migration pathways. |
| |
| Keywords: | |
|
|
