Migration of CO2 through storage reservoirs can be monitored using time lapse seismic reflection surveys. At the Sleipner Field, injected CO2 is distributed throughout nine layers within the reservoir. These layers are too thin to be seismically resolvable by direct measurement of the separation between reflections from the top and bottom of each layer. Here we develop and apply an inverse method for measuring thickness changes of the shallowest layer. Our approach combines differences in traveltime down to a specific reflection together with amplitude measurements to determine layer thicknesses from time lapse surveys. A series of synthetic forward models were used to test the robustness of our inverse approach and to quantify uncertainties. In the absence of ambient noise, this approach can unambiguously resolve layer thickness. If a realistic ambient noise distribution is included, layer thicknesses of 1–6 m are accurately retrieved with an uncertainty of ±0.5 m. We used this approach to generate a thickness map of the shallowest layer. The fidelity of this result was tested using measurements of layer thickness determined from the 2010 broadband seismic survey. The calculated volume of CO2 within the shallowest layer increases at a rate that is quadratic in time, despite an approximately constant injection rate into the base of the reservoir. This result is consistent with a diminished growth rate of the areal extent of underlying layers. Finally, the relationship between caprock topography and layer thickness is explored and potential migration pathways that charge this layer are identified.

Numerical modeling plays an essential role in both identifying and assessing sub-surface reservoirs that might be suitable for future carbon capture and storage projects. Accuracy of flow simulations is tested by benchmarking against historic observations from ongoing CO 2 injection sites. At the Sleipner project located in the North Sea, a suite of time-lapse seismic reflection surveys enables the three-dimensional distribution of CO 2 at the top of the reservoir to be determined as a function of time. Previous attempts have used Darcy flow simulators to model CO 2 migration throughout this layer, given the volume of injection with time and the location of the injection point. Due primarily to computational limitations preventing adequate exploration of model parameter space, these simulations usually fail to match the observed distribution of CO 2 as a function of space and time. To circumvent these limitations, we develop a vertically-integrated fluid flow simulator that is based upon the theory of topographically controlled, porous gravity currents. This computationally efficient scheme can be used to invert for the spatial distribution of reservoir permeability required to minimize differ

a b s t r a c t CO 2 migration in a saline aquifer is governed by viscous, capillary and gravitational fluid forces at an early stage of injection, where the dominant flow regime is site specific and controls the fluid migration in the pore space. This study combines the CO 2 saturation inverted from time-lapse seismic methods with an analytical expression to define the CO 2 flow regime, saturation distribution and layer thickness in the Tubåen Fm. following CO 2 injection. Quantitative estimates of the CO 2 saturation from time-lapse seismic amplitude versus offset (AVO) and spectral decomposition are compared to a viscous dominated analytical expression of CO 2 injection into a saline aquifer. The spatial extent of the CO 2 plume obtained from time-lapse spectral decomposition and inverted from time-lapse AVO analysis display good agreement with the analytical expression. The CO 2 is limited to an area close to the injection well, with an elongated shape in the channel direction. Comparison between the time-lapse seismic and analytical expression shows that the fluid flow is dominated by viscous forces. CO 2 saturation within the plume is constant and close to the residual brine saturation. The influence of gravity is ignorable on the reservoir CO 2 flow. CO 2 fills the entire sandstone unit up to approximately 50 m away from the injection before the CO 2 layer thickness is reduced to a thin wedge that propagates below the overlying shale unit. Reduction in CO 2 saturation away from the injection well is a reduction in effective CO 2 saturation relative to the thickness of the horizon. The maximum radius of the CO 2 layer from the analytic expression is 750 m, of which 400 m is above the time-lapse noise level. Time-lapse seismic analysis reveals the CO 2 layer radius is 405 m in the direction of the local fluvial channel and 273 m in the perpendicular direction.

No abstract