Numerical modelling of long-term CO2 storage mechanisms in saline aquifers using the Sleipner benchmark dataset

Abstract

We present numerical modelling of long-term CO2 storage in saline aquifers using field dataset in the Sleipner carbon capture and storage project in offshore Norway. A compositional reservoir simulation model was constructed for an entire section of the reservoir whose thickness is approximately 200 m. We used a fine grid system with a thickness of down to 0.2 m to capture thin intraformational mudrock barriers which can play an important role in the upward migration of injected CO2 as observed in 4D seismic datasets. The threshold capillary pressure and vertical permeability of these mudrock barriers were tuned to have nine separated CO2 plumes below the mudrock layers as observed in the corresponding seismic data. The mismatches in the CO2 plume between our simulation result and 4D seismic observation suggested further model tuning. The simulation was extended without further CO2 injection. During CO2 injection, the dominant storage mechanism was structural trapping, which was influenced by the properties of intraformational mudrock barriers. After the end of CO2 injection, the amount of residual gas trapping started to increase. The significant change in the storage mechanisms occurred during a period of 20 years after shut-in. At 100 years and onward, solubility trapping took effect with the decrease of residually trapped CO2. Mass diffusion of CO2 in formation brine controlled the degree of the invasion of CO2 into the cap rock over 1000 years of our simulation limit.