Seismic Monitoring Feasibility Study of CO2 Injection into Saline Aquifer – A Case Study from Otway Basin, Australia

Abstract

CO2 sequestration into underground formations requires the application of well panned monitoring methodology. One of the key elements in the design of the monitoring strategy is the prediction of rock property changes resulting form injection of CO2. In this study we investigate whether seismic methods can be used to detect small quantities of CO2 injected into saline aquifer. To do this we generated accurate, noise polluted time lapse synthetic models. We pushed the noise level to an extreme case to arrive at the limiting conditions for the application of time-lapse survey. Despite the fact that synthetic noise models can never quite simulate the in situ conditions, results of the modelling shows that we can expect to detect time-lapse seismic response. Moreover this is true even for the worst possible data quality we experienced during repeated seismic test surveys in Otway basin. Thus as much as qualitative analysis is concerned it is clear that even sparse time lapse 3D seismic surveys are sufficient for monitoring of CO2 injection at Naylor site. However, to map distribution of CO2 plume in space or to attempt quantitative analysis we need the highest possible S/N ratio and high resolution seismic data.

Application of Diffracted Wave Analysis to Time-lapse Seismic Data for CO2 Leakage Detection

“…The synthetic datasets were generated using 50-Hz Ricker wavelet. The velocities are based on the interval velocities calculated from zero-offset VSP data acquired at Naylor-1 monitor well (Pevzner et al 2009). The fluid substitution modeling was performed to calculate the physical properties of the host formation in order to predict the change of the acoustic/elastic properties of the rock after injection, i.e.…”