A Material-Balance Type Curve for CO2 Retention Evaluation from Production Data in CO2 Flooding

Summary Operating a carbon dioxide (CO2) flood requires a time-resolved estimate of the retained CO2 inventory, which follows from a cumulative injected-minus-produced component balance and is not measured directly. Existing estimates rely on compositional simulation or history matching, which are difficult to update from routine production data. This paper develops a material-balance type curve that maps the surface production gas/oil ratio (GOR) onto the CO2 retention degree. Matching a production trajectory to that curve returns the conditional retention degree, Eret, at an externally selected limiting GOR. The type-curve coefficients follow from the relative permeability and the fluid properties, and production data fix only the vertical position of the trajectory. Performance was assessed using 36 compositional simulations in paired 18-case designs above and below the minimum miscibility pressure (MMP). Four limiting ratios between 1,500 m3/m3 and 4,500 m3/m3 were compared, and 3,000 m3/m3 gave the largest set of complete evaluations. At that boundary, the 90th-percentile matching root mean squared error (RMSE) was 2.88 percentage points for the miscible cases and 2.14 for the immiscible cases. The 90th-percentile endpoint deviation did not exceed 2.01 percentage points. Representative matching intervals formed in 26 of the 36 trajectories, and both errors stayed within 3 percentage points for 23 of the 25 trajectories with independent endpoints. The 10 remaining trajectories spanned different tested levels of the four design factors, and no single factor provided an applicability cutoff. Oil viscosity and horizontal permeability produced the largest changes in Eret, whereas the effects of injection rate and porosity remained below 2 percentage points over the tested ranges. Field-scale use was illustrated with production records from the SACROC Northern Platform and Cranfield Field. The two field applications gave exploratory projections of Eret = 15.4% and 8.62%, with RMSE values of 0.94 and 0.79 percentage points, respectively. Applicability was assessed using six screening criteria developed from the 36 simulation cases; these criteria have not yet been validated under field conditions. Reliable application requires a representative matching interval and relative-permeability data that adequately characterize the displacement conditions under evaluation.