Answers are grounded in this paper and cite their sources.
We perform time-lapse elastic full waveform inversion to estimate velocity changes associated with CO2 injection at the Nagaoka carbon capture and storage site in Japan. A series of repeating cross-well surveys were conducted to delineate the subsurface changes within a few hundred meters from the injection well. As the acquisition is optimized for traveltime tomography, the data sets pose substantial challenges in handling amplitude variations, non- pressure source radiation patterns and previously unknown anisotropy of the reservoir. We validate the anisotropy and deviatoric source excitation by waveform simulation and develop an anisotropy model. We preprocess the data set by utilizing strong tube waves and estimate source mechanisms from waveform information as the combined pressure and deviatoric sources. With these scrutinizing, we successfully estimate P-wave velocity models for baseline and monitoring data sets. The elastic full waveform inversion velocity models show much detailed structures than previous traveltime based models and do not contain artefacts evident in acoustic full waveform inversion results. Elastic full waveform inversion retrieves the CO2 plume as velocity decreases along a thin layer and suggests that the plume does not migrate upward but is trapped by a high-velocity layer.
No takes yet. Share an insight, caveat, or question.
Nakata et al. (2022) studied this question.