Simulating discontinuous gas flow and methane reactive transport in the saturated zone

The release of natural gas to shallow groundwater systems from energy wells suffering integrity issues (termed gas migration, GM) can lead to adverse impacts on groundwater quality, safety concerns associated with explosion risks, and the release of greenhouse gases to the atmosphere. The physiochemical processes occurring during GM are difficult to characterize and model. However, modeling is important to develop strategies for detecting and monitoring GM and formulate conceptual models. In this study: (i) A previously developed numerical model based on macroscopic invasion percolation (macro-IP) and multicomponent mass transfer was enhanced to simulate gas releases in the shallow subsurface. Model simulations were compared to previously conducted bench-scale gas injection experiments and results show that gas flow is highly sensitive to the entry pressure field distribution assigned within the model domain and the critical gas saturation used to model gas-water flow based on macro-IP. However, mass transfer and the resulting domain-scale dissolved-gas transport was relatively insensitive to these parameters. (ii) Hypothetical natural gas releases were simulated in a two-dimensional shallow confined aquifer to identify and evaluate GM indicators. Free-phase gas movement was modeled using macroscopic invasion percolation. The resulting free-phase gas distribution was then inputted to the multicomponent reactive transport model MIN3P. A variety of scenarios were tested to understand the impact on dissolved gas concentrations downgradient of the release. The scenarios included the presence of multiple sources of free-phase gas and different biogeochemical conditions in the aquifer. Simulations show that exsolution of dissolved background gases during simultaneous dissolution of free-phase hydrocarbons cause delayed and variable breakthrough of dissolved gas concentrations at monitoring points across the domain. The results suggest that monitoring of background dissolved gases with total dissolved gas pressure can help enhance the monitoring network at sites impacted by GM. It is clear from the modeling results that multicomponent mass transfer and transport need to be considered when interpreting GM indicators. The findings from this study will aid in taking necessary steps towards up-scaling and implementing numerical models at larger scales to simulate GM scenarios.Science, Faculty ofEarth, Ocean and Atmospheric Sciences, Department ofGraduat