Large-scale CO2 Injection Analysis: understanding pressure variation in multiple compartments using analytical and computational analysis approaches

Injecting carbon dioxide into deep saline aquifers and then storing it by physically and chemically trapping, is one viable strategy for reducing greenhouse gas emissions and mitigating global warming. To ensure the security of storage, it is vital to comprehend the long-term variations in pressure after the injection of CO2 and the pressure limits during injection. The study area of this project is the Smeaheia area, located in the eastern margin of the Horda Platform, around 20~35 km offshore Western Norway. The Viking Group is the main storage group, where the Sognefjord Formation, the Fensfjord Formation and the Krossfjord Formation act as the major storage units and are the focus of this project. The problem is investigated using a basin fluid delta-pressure approach and the "box method", with a focus on parameter estimation. To combine with the real cases, the subsurface compartments are also divided into different scenarios based on the faults map to apply the "box method". The "box method" is a way of defining the pressure limit and the compressibility limit on basis of the initial estimation of parameters. Then the pressure variation in each compartment is obtained using an analytical method and the injection result for each compartment is calculated as well. After randomly sampling the variables, a more accurate range for each parameter is generated, on basis of this, the pressure situation and the mass of injected CO2 are estimated in different time periods. Machine learning approaches can also be applied to the results in future study. Furthermore, to figure out the deterministic factors, tornado plots are generated for the initial parameters and the PCA method is then applied to further analyse the results. The analysis indicates that both the geometric parameters and injectivity values can have a significant impact on the injection