Updates on Pseudo-3D FWI: Understanding the upside and limitation in preparation of a Snowflake data application

Anton Ziegon, Jinji Li, Kristopher A. Innanen

Reliable and cost-effective monitoring of carbon capture and storage (CCS) projects is essential for ensuring containment and conformance of injected CO₂. Seismic full waveform inversion (FWI) offers high-resolution subsurface imaging but is often limited by its computational demands, specifically in 3D implementations. This report presents aPseudo-3D FWI approach, which leverages coupled 2D FWI frameworks in a cooperative inversion scheme to improve model consistency along intersecting 2D sections, enabling effective 3D interpretation in a multi-offset, multi-azimuth VSP setting. Two coupling formulations are evaluated, with the normalized approach providing superior stability and consistency. Synthetic studies based on the Snowflake VSP dataset at the Field Research Station (FRS) demonstrate that Pseudo-3D FWI achieves comparable imaging quality to benchmark 3D FWI, accurately recovering plume location and preferential flow direction,even in the presence of noise and sparse acquisition. The method delivers a major reduction in computational cost (over 90% runtime savings) without significant loss in imaging fidelity for gentle, layered geologies, that are commonly targeted for carbon sequestration projects. Limitations arise in settings with strong 3D effects, highlighting the suitability for relatively benign subsurface conditions. These findings suggest Pseudo-3D FWI as a promising, efficient alternative for rapid and reliable CCS monitoring.