CO2 Storage Characterization Driven by Images of a Prior Injection: CO2CRC's Otway Project

Abstract

To characterise geological features that control the fluid flow in the subsurface from seismic data, I develop a multi-attribute analysis using an artificial neural network. The network is trained on the plume of CO2 injected into a saline aquifer as part of the CO2CRC Otway Project, using the plume’s time-lapse seismic image as ground truth. The results highlight geological features controlling CO2 flow and guide static and dynamic modelling for future injection.