A framework for full waveform modeling and imaging for CO 2 injection at the FRS project

Davood Nowroozi, Donald C. Lawton


The Field Research Station (FRS) is a project developed by CMC Research Institutes, Inc. (CMC) and the University of Calgary. It is a CO 2 injection and test site in the south east of Alberta, near Brooks. A well has been drilled to a depth of 550 m and a full set of well log data has been acquired. It is ready for the small volume of CO 2 injection in the shallow targets to be monitored using seismic and other survey types.

During the injection CO 2 in the target layer (300 m depth), dynamic parameters of the reservoir as pressure and phases saturation will change and they can be derived of fluid simulation result. For the project, strategy is five years’ injection with constant mass of CO 2 equal to 1000 t/yr. In this case, the CO 2 saturation increases to a maximum of 70% in the injection zone adjacent to the well but is generally between 10 to 50 percent; the CO 2 plume shape is an ellipsoid with radius of 120 m radius and a thickness of 12 m. Based on well log data and dynamic reservoir parameters (CO 2 and brine saturation, and reservoir pressure) the P-wave velocity and density were determined through fluid substitution methods. The bulk modulus of dry rock, fluids, minerals and density after injection was calculated and saturated bulk modulus extracted using Gassmann’s equation. Fluid substitution causes a change in acoustic impedance value in injection zone of reservoir.

Time-lapse seismic analysis of reservoir was assessed by seismic finite difference time domain (FDTD) modeling based on an acoustic velocity-stress staggered leapfrog scheme. The FDTD is 2nd order in time and 4th order in space on Central Finite Difference (CFD). The boundary conditions are set on all edges except surface, based on a perfectly matched layers (PML) approach. The effect of CO 2 substitution is a time delay in time domain seismic data under the reservoir because of velocity reduction and also a change in amplitude of reservoir reflections. Bases on synthetic models, the difference between base model and time-lapse model after 5 years of CO 2 injection reveals a significant seismic result, because it is a near-surface reservoir. Given that the seismic resolution is high because of the shallow target depth and acquisition parameters, it is expected to improve that seismic monitoring will be an effective method to monitor the CO 2 injection.

View full article as PDF (4.08 Mb)