CSIRO In-Situ Lab: A multi-pronged approach to surface gas and groundwater monitoring at geological CO2 storage sites

Abstract In February 2019, at the CSIRO In-Situ Laboratory CCS project, a test was conducted where 38 t of gaseous CO2 were injected over 5 days into a fault zone at a depth of approximately 340 m. As a release test, this project enabled the testing and validation of surface and shallow well monitoring strategies at intermediate depths (i.e. depths much deeper than previous release projects and shallower than reservoirs used for CO2 storage). One of the aims of this project is to understand how CO2 would behave at intermediate depths if it did migrate from deeper depths (i.e. from a storage reservoir); the CO2 was not intended to migrate to the shallow subsurface or to surface/atmosphere. To verify that the injected CO2 remained in the subsurface, and to comply with environmental performance requirements on site, a comprehensive surface gas and groundwater monitoring program was conducted. The monitoring strategy was designed such that any leakage(s) to the surface of injected CO2 would be detected, mapped and, ultimately, quantified. The surface air monitoring program was comprised of three different but complementary approaches allowing data to be efficiently collected over different spatial and temporal scales. These approaches included continuous soil-gas chamber measurements at fixed locations, periodic soil-gas chamber measurements on gridded locations and near-surface atmospheric measurements on a mobile platform. The surface air monitoring approaches gave self-consistent results and reduced the risk of “false negative” test results. The only anomalous CO2 detected at the surface flowed from the observation well and could be directly attributed to a breach in the well casing at the injection depth providing a conduit for CO2/water to rise to the surface. Groundwater monitoring program revealed no impact on the groundwater resources attributable to the carbon injection project. Based on this work, we demonstrate that this multi-pronged monitoring strategy can be utilized to minimize the overall resources devoted to monitoring by increasing the number of monitoring approaches and diminishing the resources devoted to each technique. By maximizing the effectiveness of each element of the monitoring program, a cost-efficient and robust monitoring strategy capable of early leak detection and attribution of any leaking CO2 can be achieved.