Risk Evaluation and Management of CO2 Storage in the Bahrain Oil Field

Abstract This paper presents a risk-focused assessment of CO2 storage in the Mauddud reservoir of the Bahrain Oil Field, emphasizing subsurface uncertainty and technical risk management. A detailed geological model was constructed using core analysis, well log interpretation, and formation micro-imaging (FMI), with the secondary gas cap (248,000 acre-ft) identified as the primary containment unit for CO2 injection. A volumetric-based storage estimation was coupled with a Monte Carlo simulation using Oracle Crystal Ball, resulting in a mean CO2 storage potential of 37.57 million metric tons (MMt), with P10 = 26.88 MMt and P90 = 49.47 MMt. A sensitivity analysis, conducted using both rank correlation and a machine learning-based Random Forest model, revealed that storage efficiency, porosity, and formation volume factor were the dominant parameters influencing storage capacity, while irreducible water saturation exhibited minimal impact. A Bowtie risk assessment framework was used to identify and evaluate technical risks, including caprock integrity under pressure buildup, fault reactivation near injection zones, potential wellbore leakage due to aging completions, and uncontrolled plume migration. Targeted mitigation strategies such as zonal isolation, corrosion-resistant completions, and real-time pressure monitoring were proposed to address these risks. This integrated approach, combining probabilistic modeling, machine learning insights, and a structured risk evaluation framework, enhances confidence in safe, long-term CO2 storage and offers a replicable model for future carbon capture and storage (CCS) projects in similar carbonate reservoirs.