Answers are grounded in this paper and cite their sources.
Time-lapse seismic monitoring is essential for verifying CO2 containment in geological storage, yet quantitative interpretation remains challenging due to the complex relationship between fluid distribution and seismic response. Patchy saturation models are often utilized with a saturation-independent patch size, failing to capture the dynamic evolution of fluid patches during the injection process. To address this limitation, we develop a model that integrates percolation theory within the continuous random medium framework. This model dynamically links the spatial correlation length of fluid patches to saturation while accounting for mesoscopic patch stiffening due to capillarity. We apply this model to high-frequency continuous active-source seismic monitoring data from the Frio-II CO2 injection project. Our model reproduces the observed spatiotemporal evolution of P-wave velocity and attenuation. The results reveal a distinct “inverse percolation” mechanism: as non-wetting CO2 invades the sandstone reservoir, the initially continuous brine phase undergoes fragmentation, transitioning from meter-scale connectivity to centimeter-scale isolated patches. Furthermore, we demonstrate that a capillary stiffening parameter is essential to match the observed velocity trends, confirming the significant influence of capillary forces even under high-rate injection conditions. By establishing a physics-based link between mesoscopic multiphase flow and seismic properties, our model may enhance the accuracy of CO2 saturation estimates in 4D monitoring projects.
No takes yet. Share an insight, caveat, or question.
Liu et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: