Focused sparse seismic data planning for CO2 sequestration monitoring

Abstract

Abstract Carbon Capture and Storage (CCS) monitoring relies on geophysical methods to track CO2 plume migration and ensure long-term storage security. While conventional 4D or repeated 3D seismic surveys provide comprehensive subsurface imaging, their high costs limit acquisition frequency, reducing the ability to detect unexpected migration events in a timely manner. Focused sparse seismic monitoring, which targets specific subsurface regions using strategically selected source-receiver pairs, offers a cost-effective, high-frequency complement to conventional methods. However, current approaches for selecting monitoring locations are often heuristic and do not explicitly evaluate their effectiveness in reducing CO2 plume uncertainty. To address this limitation, a systematic framework is developed to optimize monitoring locations in focused seismic surveys. The methodology integrates geostatistical modeling of subsurface heterogeneity, coupled flow and seismic forward simulations to relate reservoir properties to seismic responses, Bayesian belief updating to refine model ensembles, and Kullback-Leibler (KL) divergence to quantify expected information gain. Numerical experiments compare the uncertainty reduction achieved by full-scale 3D seismic data and localized sparse observations. While sparse data provide limited global constraints, they effectively reduce uncertainty near observation points. Monte Carlo evaluation of candidate monitoring configurations based on expected information value offers a quantitative foundation for optimizing resource allocation in future focused seismic acquisitions.