A probabilistic assessment methodology for carbon dioxide enhanced oil recovery and associated carbon dioxide retention

The U.S. Energy Independence and Security Act of 2007 authorized the U.S. Geological Survey (USGS) to conduct a national assessment of the potential volume of hydrocarbons recoverable by injection of carbon dioxide (CO₂) into known oil reservoirs with historical production. The implementation of CO₂ enhanced oil recovery (CO₂-EOR) techniques could increase the U.S. recoverable hydrocarbon resource base. Use of anthropogenic CO₂ in the CO₂-EOR process could reduce the amount of CO₂ released to the atmosphere by allowing a percentage of the injected CO₂ to remain in reservoir pore space once occupied by produced oil and water or by CO₂ dissolution in oil and water in the reservoir.

The USGS has developed a new methodology for the national assessment of technically recoverable oil resources that may be produced by using current CO₂-EOR technologies. The methodology relies on a proprietary reservoir-level database, the comprehensive resource database (CRD). The CRD incorporates commercially available geologic and engineering data, and USGS-defined play averages or province averages of reservoir data were used to populate incomplete records. Values from the CRD are used to estimate the original oil in place (OOIP) for each reservoir. The inputs are reviewed by USGS geologists, particularly when play or province averages have been used. Monte Carlo simulation is used to produce a numerical probability distribution for the OOIP for each reservoir, with the mean defined as the value of the OOIP in the CRD. A reservoir model (CO₂ Prophet, developed for the U.S. Department of Energy by Texaco, Inc.) is used to determine the incremental recovery factors for oil during the CO₂-EOR process, on an individual reservoir basis. The model is also used to estimate the volume of CO₂ remaining in the reservoir after the CO₂-EOR process is complete. Empirical decline curve analysis and comparison with data from published papers and reports on CO₂-EOR projects are utilized to substantiate the simulation results. Numerical distributions of recovery factors are prepared for variations in the reservoir lithology (clastic or carbonate). The distribution of incremental oil is computed by multiplying the appropriate probability distribution of recovery factors by the individual reservoir distribution of the OOIP. A way to estimate the CO₂ remaining in the reservoir after the completion of the CO₂-EOR process is also included in the methodology.

Assessment results will be aggregated to play, petroleum province, regional, and national scales. This assessment methodology has been tested on the Horseshoe Atoll, Upper Pennsylvanian-Wolfcampian play in the Permian Basin Province in Texas; the play consists of 27 reservoirs having at least 2 billion barrels of OOIP that are amenable to the CO₂-EOR process. The play was selected as a test case because CO₂-EOR production data and published reports are available for several reservoirs within the play. Preliminary estimates of oil recoverable by implementation of miscible CO₂-EOR are comparable to those reported in the literature and obtained by reservoir decline curve analysis.