A systematic multi-isotope approach to unravel methane origin in groundwater: Example of an aquifer above a gas field in southern New Brunswick (Canada)

Abstract

Following the large increase in unconventional hydrocarbon production in North America and elsewhere in the last 15 years, many jurisdictions have implemented groundwater monitoring programs to verify whether these subsurface industrial activities impact shallow groundwater quality. The interpretation of groundwater monitoring results typically relies mostly on dissolved alkane chemical and isotopic composition to infer the potential presence of thermogenic hydrocarbons presumed to originate from a deep source, which may indicate contamination. However, ambiguous results are frequently obtained, and post-genetic processes are often suspected to have modified the original gas composition. Here, we present a systematic approach to identify alkane origin with greater certainty, by thoroughly investigating four processes that may affect dissolved hydrocarbon gas: 1) late-stage methanogenesis, 2) oxidation (of methane or higher alkanes), 3) mixing between different gas sources, and 4) secondary methanogenesis. This is achieved by using empirical equations and fractionation factors available in the literature, combined to site-specific isotopic tracers (δ13CCH4, δ2HCH4, δ2HH2O, δ13CDIC) in groundwater samples. This approach is being tested and applied to a study area located in southern New Brunswick, Canada. The area overlies the McCully gas field from which unconventional natural gas is produced since 2003, and the nearby Elgin area, a prospective area for condensates. Results demonstrate that the presence of methane in groundwater is not related to the proximity of gas wells. In a few shallow wells located very close to producing gas wells in the McCully gas field, methane and ethane were detected, and the compositional and isotopic data sometimes seemed to point towards a thermogenic origin. However, consideration of the four processes led to the conclusion that the gas was of microbial origin, and that it had been oxidized to various levels. In contrast, thermogenic gas was detected in groundwater in the Elgin area, where no commercial production has yet taken place. In this area the natural hydrocarbon gas context is more complex, and the gas from some of the wells was affected by mixing, oxidation, and late-stage methanogenesis. Finally, the approach used in this paper has proven capable of disentangling the original isotopic signature from post-genetic modifications and, despite initial ambiguity, has shown no evidence that past hydraulic fracturing in the McCully gas field has affected shallow groundwater quality.