Integrating Calibrated Mud Gas Analysis into Formation Evaluation Workflows - Experience from a Norwegian Oil Field

Summary This work presents case studies demonstrating how continuous fluid logs derived from mud gas analysis provide quantitative insight into reservoir fluid variability along the wellbore. We integrate advanced gas chromatography with real-time geosteering visualization to enhance well placement decisions during drilling. Machine learning algorithms are applied to multi-component gas data to infer predictive fluid properties, transforming mud gas from a qualitative show indicator into a quantitative reservoir characterization tool. The workflow operates entirely at surface, without interrupting drilling, making it a low-risk, cost-effective enhancement to subsurface evaluation. Across multiple field deployments, the approach has improved decision-making, reduced non-productive time, and enhanced reservoir understanding. The methodology demonstrates how continuous, AI-enhanced mud gas analysis can close key data gaps in geosteering and reservoir navigation—particularly in complex and laterally heterogeneous plays—supporting more effective reservoir management and production optimization.