No abstract

No abstract

Numerical models of geologic carbon sequestration (GCS) in saline aquifers use multiphase fluid flow-characteristic curves (relative permeability and capillary pressure) to represent the interactions of the non-wetting CO2 and the wetting brine. Relative permeability data for many sedimentary formations is very scarce, resulting in the utilisation of mathematical correlations to generate the fluid flow characteristics in these formations. The flow models are essential for the prediction of CO2 storage capacity and trapping mechanisms in the geological media. The observation of pressure dissipation across the storage and sealing formations is relevant for storage capacity and geomechanical analysis during CO2 injection. This paper evaluates the relevance of representing relative permeability variations in the sealing formation when modelling geological CO2 sequestration processes. Here we concentrate on gradational changes in the lower part of the caprock, particularly how they affect pressure evolution within the entire sealing formation when duly represented by relative permeability functions. The results demonstrate the importance of accounting for pore size variations in the mathematical model adopted to generate the characteristic curves for GCS analysis. Gradational changes at the base of the caprock influence the magnitude of pressure that propagates vertically into the caprock from the aquifer, especially at the critical zone (i.e. the region overlying the CO2 plume accumulating at the reservoir-seal interface). A higher degree of overpressure and CO2 storage capacity was observed at the base of caprocks that showed gradation. These results illustrate the need to obtain reliable relative permeability functions for GCS, beyond just permeability and porosity data. The study provides a formative principle for geomechanical simulations that study the possibility of pressure-induced caprock failure during CO2 sequestration.

National techno-economic pathways to reduce carbon emissions are required for the United Kingdom to meet its decarbonisation obligations as mandated by the Paris Agreement. Analysis using energy systems models indicate that carbon capture and storage is a key technology for the UK to achieve its mitigation targets at lowest cost. There is potential to significantly improve upon the representation of the CO2 storage systems used in these models, but sensitivities of a given reservoir system to future development pathways must be evaluated. To investigate this we generate a range of numerical simulations of CO2 injection into the Bunter Sandstone of the UK Southern North Sea, considered to be one of the most important regional aquifers for CO2 storage. The scenarios investigate the sensitivity of CO2 storage to characteristics of regional development including number of injection sites and target rates of CO2 injection. This enables an evaluation of the impact of a range of deployment possibilities reflecting the range of scenarios that may be explored in an energy system analysis. The results show that limitations in achieving target injection rates are encountered at rates greater than 2 MtCO2/year-site due to local pressure buildup. The areal location of injection sites has minimal impact on the results because the Bunter Sandstone model has good regional connectivity. Rather, the depth of the site is the most important factor controlling limits on CO2 injection due to the relationship between the limiting pressure and the lithostatic pressure gradient. The potential for model simplification is explored by comparison of reservoir simulation with analytical models of average reservoir pressure and near-site pressure. The numerical simulations match average pressure buildup estimated with the “closed-box” analytical model of Zhou et al. (2008) over a 50 year injection period. The pressure buildup at individual sites is estimated using the Mathias et al. (2011) formulation and compared to the simulation response. Discrepancies in the match are mostly due to the interaction of signals from multiple injection sites and heterogeneous permeability in the numerical simulations. These issues should be the focus of further development of simplified models for CO2 storage in an energy systems analysis framework

No abstract