Scale-up of Pore Scale Spatiotemporal CO2 Dissolution Data

Abstract:

One of the potential risks associated with subsurface storage of CO₂ is potentially the seepage of CO₂ through existing faults and the studies devoted to this topic show that geochemistry plays an important role in rendering these faults as effective conduits for CO₂ movement while others show that mineralization due to CO₂ injection can result in seep migration and flow channeling. Therefore, understanding the changes in reservoir flow dynamics with time due to geochemical alteration of the porous media and accurately scaling up these changes for representation in field scale models is important to engineer the CO₂storage process.

After CO₂ is injected in a subsurface, dispersion results in mixing of CO₂ with aqueous species present in in-situ brine. The reactions of the dissolved CO₂ with rock minerals lead to dissolution and/or formation of precipitates which alter the pore structure by changing the porosity and the permeability. Scale-up of reservoir properties and flow response at a single snapshot of time has been described by other authors, however, scale-up for reactive processes cannot be correctly described just at a single snapshot of time. We use the concept of a Representative Elementary Volume (REV) to explore the scaling characteristics of the reactive transport process. We model the REV using a variance-based statistical approach using high-resolution pore-scale data. For comparison purposes, we also compute the REV for a conservative transport using 3-D pressure data. The REV for reactive process is modeled using three different types of data: CO₂ concentration, fluid/matrix pore-network data and dissolution data. For simplicity, we consider the spatial variations along a 2-D slice at various times, rendering this a 3D spatiotemporal dataset. The results indicate that the REV in reservoir with reactive flow changes with time and is greater when compared to the REV for conservative flow. The change in REV with time for reservoirs with reactive flow is due to changes in pore structure because of dissolution. Hence, these results suggest that changes in REV at pore-scale because of dissolution have important implications for CO₂ injection at field-scale. These results on pore-scale support the conclusions proposed by other authors about flow channeling and seep migration due to changes in pore structure.