Pore-to-Core-Scale Network Modelling of CO₂ Migration in Porous Media

Pore network modelling offers a versatile and efficient means for examining the complex interplay of a variety of microscopic processes affecting subsurface migration of injected for storage. We present a dynamic pore-to-core network model capable of simulating the full range of migration processes under the influence of capillary and gravity forces, including dissolution in brine. A parametric sensitivity study investigating four variables that define the microscopic Bond number, viz: mean pore radius, -brine interfacial tension, brine- density difference, and network height, was performed. Two broad classes of behaviours were identified-one quasi-stable and the other unstable (migratory)-and critical gas saturation was found to change in a non-monotonic way with transition from quasi-stable to migratory regime. The model predicts strong effects of gravity at the scale typical of continuum-type simulator gridblocks, and pore size distribution variance and pore connectivity were found to have a major impact on which cannot be predicted a priori through the use of Bond number scaling. For temperatures and pressures above the critical point, and flow regimes in brine displayed generally similar characteristics, suggesting that flow coefficients (e.g. relative permeability) of and in brine could be used interchangeably in continuum-type simulators with effectively the same results.