Time-lapse data matching, strategy and FWI of Snowflake DAS VSP data at CaMI.FRS

He Liu, Xin Fu, Xiaohui Cai, Daniel O. Trad, Kristopher A. Innanen

Full-waveform inversion (FWI) using DAS vertical seismic profile (VSP) data has been promising for long-term and low-cost monitoring CO₂ injection and sequestration. In 2018 and 2022, two 3D DAS VSP datasets were acquired at CaMI before and after an amount of CO₂ injection, providing feasible field data for the feasibility study of monitoring the time-lapse physical parameter change introduced by the CO₂ injection. In this work, the biggest challenge to detecting the time-lapse change by applying FWI to DAS VSP data is that we observe no obvious data difference resulting from the CO₂ injection in the target reservoir-Basal Belly River Sandstone (BBRS) layer (a 7 m-thick sandstone formation). To pursue successful imaging of time-lapse change using DAS data, we start by developing a data-matching approach in the first place. Then we design a practical time-lapse inversion strategy. At last, we perform time-lapse FWI using the matched datasets based on 2D acoustic FWI to image the Vp change in the BBRS formation.