document preview

Abstract
The effects of heterogeneity in Carbon Capture and Storage (CCS) in saline aquifers have been investigated extensively and are known to have important bearings on the storage capacity of the aquifer. In CCS projects, the time-lapse seismic survey has been proposed as a valuable tool for monitoring of CO2 movement. However, the potential of the time-lapse seismic data for heterogeneity characterization and geologic model updating has not been fully explored. One of the biggest challenges in the quantitative use of time-lapse seismic data during CCS is the complex movement of the CO2 influenced by compositional effects, geochemical reactions, phase changes and gravity segregation.

In this paper, we first introduce compositional streamlines to understand and visualize the flow and transport of CO2 in the presence of mineral precipitation/dissolution, residual trapping and buoyancy effects. To start with, individual component fluxes are generated by a finite difference fully implicit compositional simulator incorporating all the relevant physics of CO2 sequestration. The fluxes are then utilized in novel streamline tracing algorithms to generate phase and component streamlines depicting the movement and the trapping of CO2 in the aquifer. Next, we utilize the compositional streamlines to determine the sensitivity of the time-lapse seismic attributes specifically, interpreted saturation differences, to changes in reservoir properties such as permeability and porosity. The sensitivities are then used in an inverse modeling algorithm to calibrate the geologic model to time-lapse seismic data. The outcome is an improved description of permeability heterogeneity that is consistent with the 4-D seismic response and improved predictions of the CO2 storage capacity.

We have investigated the benefits of time-lapse seismic data integration in improving the performance assessment of CO2 sequestration using examples involving CO2 injection under realistic conditions. The first example examines the value of the 4-D seismic data integration in the estimation of storage capacity. The second example systematically studies the impact of viscous to gravity ratio on the performance of time-lapse seismic monitoring and heterogeneity characterization during CCS.

Introduction
Carbon sequestration in brine aquifers faces many different challenges in both engineering and economical aspects. There are several sources of uncertainties associated with the injection of CO2 in deep saline aquifers. Engineering problems such as the leakage of CO2 can compromise the integrity of fresh waters, ecosystems and the health of populations exposed to high concentration of CO2 (Ha-Duong, 2003; Gasda et al., 2004). There are also economic threats associated with legal disputes and fines imposed by regulatory agencies.

Monitoring, verification and accounting (MVA) are the activities directed to determine the location of the injected CO2 and the presence of possible leaks in order to provide public assurance. Many techniques have been developed for monitoring the performance of CO2 injection projects and the migration of CO2 in geologic formations. Time lapse seismic surveillance data is one of the most mature and effective techniques for monitoring changes in the fluid saturation and pressure and has been extensively used by the oil industry.

The viability of time lapse seismic data as a monitoring tool has been investigated in laboratory and field experiments (Wang and Nur 1989; Wang et al 1998; Korneev et al 2004; Hovorka et al 2006; Daley et al 2007). The characteristics of seismic responses under the complex compositional and geochemical interactions of the injected CO2 in geologic formations have been modeled by Kumar et al. (2008). Also, the effectiveness of traditional rock physics models to invert the seismic responses into changes in the acoustic impedance of the rock during CO2 sequestration has been investigated by Vanorio et al. (2010).