Patricia Montoya, Mark Hoefner, ExxonMobil Upstream Oil and Gas
The LaBarge field, located in SW Wyoming and operated by ExxonMobil, produces sour gas (66% CO₂, 5% H2S) in addition to hydrocarbons and helium from the Pennsylvanian-age carbonates of the Madison Formation.
Two Acid Gas Injection (AGI) wells located 35+ miles to the south of the producing field, have been successfully injecting H2S and CO₂ into the Madison below the field Gas Water Contact. Both disposal wells have demonstrated suitable injectivity and storage capacity of the Madison saline aquifer. ExxonMobil is planning to drill a third CO₂ disposal well ~9 miles away from the AGI wells at the Shute Creek Treatment Facility as part of a carbon capture expansion project. The expansion project will capture up to an additional 1 million metric tons of CO₂ annually at LaBarge. The new well will be primarily used for CO₂ disposal in the Madison but it will also include a potential secondary disposal target of Ordovician-age carbonates of the Bighorn formation. Both the existing AGI and future CO₂ disposal well are expected to continue to inject CO₂ and/or acid gas until end of field life.
A static earth model of the Madison and Bighorn geologic formations was built to help characterize the aquifer to the South of the producing LaBarge gas column. The model incorporates information from the two AGI wells, logs, cores, 2D and 3D seismic data. The static geologic model provided a framework for a dynamic model used for reservoir simulation with three objectives: (1) to history match AGI disposal wells, (2) to evaluate the impact of key parameters affecting CO₂ injection rates with time, and (3) to predict the size and location of CO₂ plumes through time in both reservoirs. The dynamic reservoir model incorporates historic production and injection data to account for field history to predict existing surface and subsurface conditions at the site of the new disposal well.
A history match of the AGI wells performance provides a calibration baseline for permeability thickness (kh), relative permeability, and skin parameters in the model, each a key source of uncertainty for injection of acid gas or CO₂ in an area of sparse well density. Scenario-based sensitivity analysis was used to demonstrate the impact of these 3 parameters to maintaining CO₂ injection rates over time. Once the new CO₂ disposal well is drilled and data are collected, the model will be calibrated and used to refine prediction of CO₂ injection rates, plume size, and potential for interference between wells over time.
Calculation of the volume-weighted average gas saturation at various time steps was used to determine the CO₂ gas plume expansion for the new CO₂ disposal well, with the plume boundary defined as the area with an average gas saturation of greater than 0.5%. The model was run out more than 80 years to increase confidence in maintaining a maximum injection rate for an extended period and to assess to what degree the injection from the three wells may begin to interfere over time.