Advances in Attribution Monitoring Techniques for CO2 Anomalies: Discriminating Injection Leakage from Natural Variability

Summary Large-scale CCUS deployment requires solid evidence demonstrating secure containment of CO2 throughout the entire industrial chain with no leakage occurring. Atmospheric CO2 levels vary due to multiple sources. Effective CO2 attribution monitoring techniques help reduce monitoring costs, ease public concerns, and provide reliable evidence for carbon accounting. This study aims to resolve the challenge of anomaly source identification in geological CO2 storage monitoring. The study follows a three-step methodology based on published literature, including books, agency reports, peer-reviewed journals, and conference publications. First, we review the principles and methods of different CO2 attribution monitoring technologies. Next, we perform comparative assessments of their technical feasibility, costs, and field application results. Finally, an optimized MRV procedure is proposed. CO2 attribution monitoring technologies can be classified into five major categories: eddy covariance, accumulation chamber monitoring, tracer-based techniques, process-based analysis, and deep learning algorithms. Eddy covariance, accumulation chamber monitoring, and deep learning techniques require extensive baseline data for comparison. Tracer-based techniques and process-based analysis demonstrate higher sensitivity and operational efficiency and therefore are recommended for prioritized application. Building upon conventional MRV frameworks, an optimized workflow is proposed by adding CO2 attribution monitoring plan between base-case plan and contingency plan.