Item

Abstract

Within the ERA-NET co-funded ACT project Pre-ACT (Pressure control and conformance management for safe and efficient CO2 storage – Accelerating CCS Technologies), a monitoring concept was established to distinguish between CO2 induced saturation and pore pressure effects. As part of this monitoring concept, geoelectrical cross-hole surveys have been designed and conducted at the Svelvik CO2 Field Lab, located on the Svelvik ridge at the outlet of the Drammensfjord in Norway. The Svelvik CO2 Field Lab has been established in summer 2019, and comprises four newly drilled, 100 m deep monitoring wells, surrounding an existing well used for water and CO2 injection. Each monitoring well was equipped with modern sensing systems including five types of fiber-optic cables, conventional- and capillary pressure monitoring systems, as well as electrode arrays for Electrical Resistivity Tomography (ERT) surveys. With a total of 64 electrodes (16 each per monitoring well), a large number of measurement configurations for the ERT imaging is possible, requiring the performance of the tomography to be investigated beforehand by numerical studies. We combine the free and open-source geophysical modeling library pyGIMLi with Eclipse reservoir modeling to simulate the expected behavior of all cross-well electrode configurations during the CO2 injection experiment. Simulated CO2 saturations are converted to changes in electrical resistivity using Archie's Law. Using a finely meshed resistivity model, we simulate the response of all possible measurement configurations, where always two electrodes are located in two corresponding wells. We select suitable sets of configurations based on different criteria, i.e. the ratio between the measured change in apparent resistivity in relation to the geometric factor and the maximum sensitivity in the target area. The individually selected measurement configurations are tested by inverting the synthetic ERT data on a second coarser mesh. The pre-experimental, numerical results show adequate resolution of the CO2 plume. Since less CO2 was injected during the field experiment than originally modeled, we perform post-experimental tests of the selected configurations for their potential to image the CO2 plume using revised reservoir models and injection volumes. These tests show that detecting the small amount of injected CO2 will likely not be feasible.