Leakage pathway estimation using iTOUGH2 in a multiphase flow system for geologic CO2 storage

The objective of this study is to apply an inverse analysis using the iTOUGH2 model to estimate the location of a leakage pathway in multiple brine reservoirs when CO2 is injected. If a reservoir exhibits leakage, brine or CO2 is able to migrate into a permeable reservoir overlying the storage reservoir. Fluid pressure anomalies induced by leaks in the overlying reservoir can be distributed differently depending on the leakage locations and rates. Thus, the application of an inverse model utilizes specific pressure anomalies for leakage pathway detection. Prior to applying the inverse analysis, a forward simulation and a sensitivity analysis are conducted. The result of forward simulation demonstrates the interrelation between migrations of brine or CO2 through the leakage pathway and pressure anomalies in the leakage pathway and reservoirs. The sensitivity analysis is performed to evaluate/identify the most influential model inputs on the observed pressure signals and the most appropriate monitoring wells for leakage pathway estimation. The inverse modeling examines the impact of the input parameter's uncertainties, the number of monitoring wells, observed periods of leakage signal, and noises in the measurements on the leakage pathway estimation through thirteen simulation scenarios. Residual (between the measured pressure and the calculated pressure) analysis illustrates that pressure anomalies in the overlying reservoir induced by leaks are critical information for leakage pathway estimation. The accuracy of the leakage detection using inverse analysis can significantly depend on the number of monitoring wells and the magnitude of the pressure anomalies.