Capacity Estimation of Capillary Trapping of CO2 in Heterogeneous Saline Aquifers Through 3-D Invasion-Percolation Simulations at the cm-to-m Scale

Abstract:

During post-injection stage of geological carbon sequestration the primary forces governing CO₂ migration and entrapment are capillarity and buoyancy, raising doubts about the applicability of conventional Darcy-based flow simulators to predict plume saturation in such flow regimes. Moreover, evaluating the impact of geological heterogeneity at the cm-to-m scale is hindered by the computational cost of highly resolved geological models. This limitation is overcome by the ability of invasion-percolation models to rapidly simulate flow through complex geometries with strong contrasts in capillary threshold pressure. The current investigation originates from a previous study conducted on 2-D flow simulations of CO₂ invasion through stochastic and natural geologic models. This work addresses the influence of 3-D sedimentary structures and their variation in threshold capillary pressure on CO₂ plume distribution with the goal of developing a predictive method for volumetric storage capacity. A set of invasion-percolation simulations is performed in an attempt to identify emergent behavior related to capillary trapping capacity of CO₂ in brine-bearing sedimentary formations. Realistic depositional facies are generated with an existing Matlab code that relates the geometry of cross-bedding to the morphology of bedforms. Resulting 3-D binary models consist of matrix and lamina populated independently with probability density functions representative of sandstone facies with different grain size and sorting. Interpretation of numerical simulations reveals a correlation between percentage of model volume invaded by the plume and contrast in capillary threshold pressures of matrix and lamina, suggesting some predictive ability is achievable.