Using time-lapse seismic monitoring to identify trapping mechanisms during CO2 sequestration

We show that it may be possible to distinguish between structurally and capillary trapped CO2 using time-lapse seismic monitoring of geological CO2 sequestration. Commercial reservoir simulation software is used to predict CO2 saturation in a saline aquifer over time. The output is combined with a rock physics model to calculate the elastic and seismic properties of the aquifer. As the seismic response depends on both fluid content and distribution, appropriate end-member fluid-distribution models are used to predict the possible range of seismic responses. We propose that different fluid-distribution models should be used for capillary and structurally trapped CO2 in a reservoir; the Hill average should be used for high, mobile CO2 saturations found during structural trapping whilst the Gassmann-Reuss average should be used for low, immobile CO2 saturations resulting from residual trapping. Far-offset seismic data is used to interpret reflections produced by the different trapping phases and cross-plotting is used to determine the trapping phase. Results indicate that structurally trapped CO2 can be seismically imaged irrespective of fluid distribution and, importantly, a reflection may be generated off the interface between residually and structurally trapped CO2. (c) 2012 Elsevier Ltd. All rights reserved.