PNNL

Abstract

Geologic carbon storage (GCS) is a critical element of carbon management strategies focused on reducing anthropogenic carbon dioxide (CO2) emissions. Substantial investment by the international research, development, and deployment (RD&D) community has developed the science base, field-experience, and industry best-practices needed to give confidence in the safe and effective implementation of GCS projects. This is evidenced by the successful storage of more than 30 million tonnes of CO2 in deep saline reservoirs through a series of pilot- and commercial-scale projects. Prior to initiating CO2 injection, each of these projects performed a site risk assessment (RA) to identify potential hazards and to provide confidence that the injected CO2 could be stored with minimal risk to human health and the environment. The RA methodologies implemented ranged from qualitative to fully quantitative, with the chosen RA approach driven by the questions being addressed and the availability of site-specific data and information. While these projects demonstrate that multiple approaches have merit and can be used to secure regulatory approval for commercial-scale GCS projects, there are efficiencies to be gained with a unified approach. Specifically, qualitative RAs broadly examine project risks while quantitative RAs constrain technical risks and improve the efficiency and effectiveness of operations (e.g., monitoring, measurement, and verification). This paper documents how the integrated risk assessment tool (NRAP-Open-IAM) developed by the National Risk Assessment Partnership complements the Bowtie RA developed by Shell at the Quest Carbon Capture and Storage Facility near Edmonton, Alberta through development of a risk-based Area of Review (AoR) for the site. The results support a further reduction to the established AoR for the Quest site due to the low groundwater impact risk if CO2 and/or brine were to leak from the storage reservoir. Integrating both approaches early in the site characterization period could significantly benefit commercial-scale projects by drastically reducing the region over which leakage risks (e.g., legacy wells) need to be evaluated, i.e., the AoR. Further, basing monitoring network design on regions where impacts are most likely to occur has the potential to result in more efficient deployment of monitoring technologies while also improving the likelihood of detecting a leak should one occur.