Implications of reservoir heterogeneity for CO2 storage monitoring: A combined experimental and theoretical rock physics study 

Abstract

<p>Carbon Capture Utilization and Storage (CCUS) is an essential technology to meet net-zero carbon emission targets. Due to CO<sub>2</sub> injection, original reservoir properties are altered. Early warning of potential CO<sub>2</sub> injection-induced reservoir instability depends on our correct interpretation of the geophysical remote sensing data, particularly seismic and electromagnetic datasets. Using joint elastic-electrical datasets is a proven effective approach to simultaneously characterize mineral skeleton properties and pore fluid distribution, and therefore a powerful reservoir monitoring tool for CCUS.</p><p>To interpret large-scale elastic-electrical datasets, original rock properties and fundamental CO<sub>2</sub>-fluid-rock interactions are  preliminarily investigated by combining available well-logging data, lab-controlled experiments using rock samples, and rock physics theories; the latter two are inevitably dependent on one another. Despite we can mimic changing reservoir conditions in the lab and generate datasets that provide essential information to understand specific processes at the micro- and meso-scales (and serve as inputs for large-scale reservoir simulations), every experiment carries limitations inherent to the particular lab capabilities, together with the obvious time- and space-scale related uncertainties (i.e., core-scale experiments only partially describe the events occurring in the field). Then, we need theoretical rock physics to make experimental assumptions, and reciprocally we use the experimental data to validate models.</p><p>Physical and petrographic properties of reservoir rocks condition the degree of heterogeneity and anisotropy of the CO<sub>2</sub> storage unit that, in turn, influence the total storage capacity and fluid migration. Original clay content, grain size distribution, mineralogy, porosity and permeability are among the most influencing parameters, particularly for low reactive siliciclastic formations (i.e., desired CO<sub>2</sub> storage reservoirs). But these properties randomly change to some extent within any reservoir formation.</p><p>Here, we investigate how reservoir heterogeneity influences our geophysical interpretation of the potential CO<sub>2</sub> storage site Aurora, offshore Norway. Recent studies suggest high clay content and porosity variability within the Johansen Formation sandstone, Aurora’s primary reservoir. Due to lack of Johansen Fm. samples, we selected three sandstone samples from the Central Graben, Offsore UK (Forties Formation), formed in a similar depositional environment, with similar mineralogical composition, and porosity (20 to 28%), clay content (10 to 26%) and permeability (1 to 8 mD) ranges. The elastic (from P- and S-wave velocities) and transport (from permeability and resistivity) properties of the tested samples were used to assess the influence of their intrinsic properties on the pore fluid distribution during CO<sub>2</sub> injection and the permanent CO<sub>2</sub>-induced changes in the Aurora reservoir complex. We apply well-known rock physics theories, including effective stress law, Archie’s relationship, and the Biot-Stoll and White and Dutta-Ode models, for both to impose the most similar reservoir conditions according to our lab limitations and to assess the experimental results. We observe (i) elastic and transport properties variations (up to 15% and 30%, respectively) between samples, mainly related to porosity differences, and (ii) more significant permanent alterations post-CO<sub>2</sub> injection in those with higher porosity and clay content. Our results show the importance of accounting for heterogeneity-related changes in sandstone reservoirs during/after subsurface CO<sub>2</sub> storage activities for enhanced geophysical interpretation.  </p>