Modeling of geohydrological processes in geological CO2 storage – with focus on residual trapping

Sammanfattning: Geological storage of carbon dioxide (CO2) in deep saline aquifers is one approach to mitigate release from large point sources to the atmosphere. Understanding of in-situ processes providing trapping is important to the development of realistic models and the planning of future storage projects. This thesis covers both field- and pore-scale numerical modeling studies of such geohydrological processes, with focus on residual trapping. The setting is a CO2-injection experiment at the Heletz test site, conducted within the frame of the EU FP7 MUSTANG and TRUST projects.The objectives of the thesis are to develop and analyze alternative experimental characterization test sequences for determining in-situ residual CO2 saturation (Sgr), as well as to analyze the impact of the injection strategy on trapping, the effect of model assumptions (coupled wellbore-reservoir flow, geological heterogeneity, trapping model) on the predicted trapping, and to develop a pore-network model (PNM) for simulating and analyzing pore-scale mechanisms.The results include a comparison of alternative characterization test sequences for estimating Sgr. The estimates were retrieved through parameter estimation. The effect on the estimate of including various data sets was determined. A new method, using withdrawal and an indicator-tracer, for obtaining a residual zone in-situ was also introduced.Simulations were made of the CO2 partitioning between layers in a multi-layered formation, and parameters influencing this were identified. The results showed the importance of accounting for coupled wellbore-reservoir flow in simulations of such scenarios.Simulations also showed that adding chase-fluid stages after a conventional CO2 injection enhances the (residual and dissolution) trapping. Including geological heterogeneity generally decreased the estimated trapping. The choice of trapping model may largely effect the quantity of the predicted residual trapping (although most of them produced similar results). The use of an appropriate trapping model and description of geological heterogeneity for a site when simulating CO2 sequestration is vital, as different assumptions may give significant discrepancies in predicted trapping.The result also includes a PNM code, for multiphase quasi-static flow and trapping in porous materials. It was used to investigate trapping and obtain an estimated trapping (IR) curve for Heletz sandstone.