Lee Hunt1, Eric Street1, Graham Hack1, Jason Schweigert2 and Matthew Allen2



This is part 2 of a 3-part paper series describing a new monitoring, measurement, and verification (MMV) method for onshore carbon capture and storage (CCS). The method is called Theseus 24D, and it focuses on reducing unnecessary seismic effort in MMV. In part 1 of the series, we demonstrated that by altering the inside and outside areas of repeat monitoring 3D surveys and integrating the 2D and 3D schedule, capital savings on the order of 26% can be achieved, and similar reductions can be made in the surface—or environmental—disturbance at the time of project closure. This part shows how the method can be extended to its limit, where repeat 3D is either minimized or eliminated altogether. By doing so, we show that the capital savings in a model storage project could be as high as 57%, with a reduction in surface disturbance at closure of 96%.

These large reductions in effort and disturbance can only be achieved if 2D and 3D can be integrated and used effectively in the 24D repeat method. Integration in CCS is not wholly new; VSP and surface 3D seismic data has some history in the literature of being shot in an integrated fashion for CCS MMV. There has, however, been no clear method for efficiently integrating 2D and 3D surface seismic until now. We use an example program to show the value of 24D. The question of how to achieve these ends while achieving the goals of monitoring is then explored in a 3D design exercise. To do this, we define two end-member approaches for this data integration, focusing on how to best manage the interpretation of the expanding carbon dioxide (CO2) plume over time. The two methods that we investigate are: a controlled 2D extraction from a 3D geometry, and an uncontrolled method. In the controlled case, the 3D is designed to include explicit source receiver lines matching those of subsequent repeat 2D surveys. We demonstrate that while this method can produce the 2D experiment within a 3D survey, it fails to achieve other key goals. The second method is a less controlled experiment where a 3D survey is designed to deliver the best geometry for wavefield reconstruction. Such a survey serves well in its capacity as 3D seismic and can also produce any baseline 2D geometry that is desired after reconstruction. The critical point with this method is its dependency on the wavefield reconstruction to accurately represent the 2D seismic lines which will be extracted. We show that the uncontrolled reconstruction method is likely to work best, provided the baseline 3D design is adequate.

The third part in this series involves a test of the reconstruction concept that we describe in this paper.