Lawrence Berkeley National Laboratory

In carbon capture and sequestration (CCS), developing rapid and effective imaging techniques is crucial for real-time monitoring of the spatial and temporal dynamics of CO2 propagation during/after injection. We propose an efficient “hybrid” time-lapse workflow that combines physics-based full-waveform inversion (FWI) and data-driven machine-learning (ML) inversion. The developed hybrid methodology simultaneously predicts the variations in velocity and saturation and achieves a high spatial resolution in the presence of realistic noise in the data. The method is validated on a synthetic CO2-sequestration model based on the Kimberlina storage reservoir in California. The scarcity of the training data with the ML algorithm is addressed by developing a new data-generation technique with physics constraints. The proposed approach synthesizes a large volume of high-quality, physically realistic training data, which is critically important in accurately characterizing the CO2 movement in the reservoir.