Seismic ambient noise auto-correlation imaging in a CO2 storage area

Abstract

Carbon capture and storage is an effective technique for achieving 'carbon neutrality'. Seismology plays an important role in detecting and monitoring whether CO2 has been safely sealed. However, active seismic surveys are usually incapable of continuous observations. We can extract the body wave and image the subsurface interface directly and continuously via single sensor ambient noise auto-correlation. However, because of the weak body wave in the ambient noise and the multiples in the imaging results, it is challenging to improve the imaging resolution. This study provides a reference for the development of the ambient noise (phase) auto-correlation imaging method for future use in carbon storage areas. In this study, we performed numerical experiments to show that the phase auto-correlation method is better than the auto-correlation in terms of weakening artefacts and detecting subsurface thin CO2 layers. Then, we applied the phase auto-correlation method in the Ketzin (Germany) CO2 storage area. Using 223 body-wave dominated noise panels automatically selected from totally 4680 panels, we imaged two primary interfaces, which are consistent with the common midpoint stacking results obtained using active seismic surveys. We applied a multiple removal approach to the imaging results and obtained high precision results. Then, we applied the proposed workflow to three passive seismic survey lines north of the CO2 injection well. The results of the three passive surveys demonstrate the workflow is robust.