Inferring migration of CO₂ plume using injection data and a probabilistic history matching approach

Large-scale implementation of geologic carbon storage (GCS) will require reliable techniques for monitoring the movement of the CO₂ plume in the subsurface. The movement of CO₂ plumes beyond the region permitted for storage will be of particular interest both to regulators and to operators. However, the cost of many monitoring technologies, such as time-lapse seismic, limits their application. Given that injection data (pressures, rates) from wells are readily available and inexpensive, we examine whether they can be used as a viable alternative for monitoring and predicting plume migration. In this paper, we have implemented a probabilistic history matching approach to creating models of the aquifer for predicting the movement of the CO₂ plume. The geologic property of interest for example hydraulic conductivity is updated conditioned to geological information and injection pressures. The resultant aquifer model which is geologically consistent can be used to reliably predict the movement of the CO₂ plume in the subsurface. We tailor the method to CO₂ sequestration by considering only injection data in the matching process. We also introduce a two-step approach to stochastically simulate high-permeability features such as oriented sets of natural fractures that occupy only a small fraction of the storage formation. We illustrate the approach by applying it to data from the In Salah Gas project. The final history-matched models contain high permeability features consistent with the field observation of rapid arrival of injected CO₂ at a suspended well and with surface deformation data obtained by remote sensing. We conclude that the approach can provide a probabilistic assessment of plume migration at the field scale.