Using Inverse Watershed Analysis to Predict CO2 Plume Migration in Saline Aquifer Systems: A Numerical Approach

The effectiveness of underground geological storage for CO2 relies heavily on the underlying geological structures and trapping mechanisms, due to the buoyant nature of CO2 in relation to natural groundwater in aquifers. As a result, it is crucial to establish a clear understanding of plume tracking and timing when developing a modeling strategy for a CO2 storage site. Comprehensive modeling is necessary to accurately predict the behavior of migrating CO2 plumes, particularly in unconfined aquifer systems where trapping is influenced by the geological formation itself. The primary goal of this approach is to assess the regional suitability for carbon storage, which is especially critical in open aquifer storage systems where future storage capacity may be directly impacted by plume migration patterns. This numerical approach simulates the natural water drive in an aquifer by modeling inverse groundwater flow in porous media. The method involves creating a Darcy velocity vector field within a porous sandstone, which generates a groundwater flow system with defined velocity magnitude and direction. Since the flow is gravity-driven, the only influencing factor is the surface slope, which can be derived from depth-converted seismic surfaces. According to Darcy's Law, the Darcy velocity (q) in a porous medium is calculated as the product of the hydraulic conductivity (K) and the head gradient. The hydraulic conductivity (K) can be determined from the transmissivity (T) and thickness (b) of the aquifer, using the formula K = T/b. By applying Darcy's Law, the flow field can be calculated, representing a vector field of groundwater seepage flow velocities. A particle track can be induced within this flow field, providing velocity and magnitude at each time step. Although the particle track is unidirectional, it can be used to simulate groundwater flow in one direction or pressure-driven water drive in the opposite direction. This makes it useful for modeling CO2 plume migration, as the vector field can be calibrated to match the buoyant CO2 plume velocity from modeling. To visualize the plume, a porous-puff analysis can be introduced, generating a statistical Gaussian plume at each time step. This analysis simulates solute transport in a porous medium using advection and hydrodynamic dispersion, allowing for a more accurate representation of the CO2 plume migration. The outcomes of this tool demonstrate its capability to accurately predict three key aspects of CO2 plume migration: The total area where a CO2 plume can potentially migrate, defined as the inverse watershed. The optimal injection placements, as well as the subsequent CO2 plume paths. The maximum length of the CO2 plume track for any given point within the inverse watershed. This method has been successfully applied in a regional analysis of a potential open aquifer CO2 storage site in Saudi Arabia, assessing its capacity and viability for subsurface geological storage. The results have proven the tool to be an essential component in selecting areas for further detailed dynamic modeling. Notably, the workflow was developed and patented by Aramco in 2025 and has since been integrated into Saudi Aramco's workflow for screening areas and aquifers suitable for CO2 storage, highlighting its significance and effectiveness in the field.