Real-Time Wellbore Integrity Monitoring in CO2 Injection Wells Using Distributed Temperature Sensing Data

Wellbore integrity is a critical factor for safe and efficient well operations, as well as for the secure storage of reservoir fluids. It plays a key role in minimizing environmental risks during both hydrocarbon production and carbon dioxide (CO2) injection. The present study is situated within this framework and investigates the application of continuous Fiber Optic (FO) monitoring technologies, with a particular focus on Distributed Temperature Sensing (DTS). By enabling real-time acquisition of temperature profiles along the wellbore, DTS offers a promising approach for the early detection of thermal anomalies potentially associated with leakage events. The objective of this research is to evaluate the sensitivity and reliability of DTS in identifying such anomalies, thereby contributing to the development of advanced monitoring strategies for wellbore integrity assurance. The approach utilizes continuous FO measurements data collected along the well profile to map, in real-time, the thermal behavior during full stable operation. To evaluate leak detection capabilities of DTS systems, a series of synthetic simulations were performed using a dynamic multiphase flow simulator. Pressure and temperature profiles along the wellbore were modelled, also mimicking DTS data acquisition during leak events in case of a CO2 injection well. Different scenarios were considered in terms of leak location and injection regimes. At the same time, surface and bottomhole temperature and pressure profiles were also generated to complement the analysis. Finally, the simulated DTS profiles were perturbed with gaussian noise to replicate the uncertainty typical of the measurements. The resulting datasets served as inputs to different algorithms with the aim of ranking their effectiveness in correctly identifying and locating the leak depth. Among the tested methods, the so-called Z-Score approach proved to be robust and effective in distinguishing all the simulated leak signals from temperature background. Moreover, the mathematics behind the method allows for real-time analysis of DTS data through an automatic alarm setting. This was demonstrated bymeansofa case studyapplicationwhere the DTS profiles from a real CO2 injection well were continuously processed during injection operations, showing no critical issues to be highlighted, as expected. The developed approach provides new insights into the thermal and flow dynamics of representative leak scenarios during CO2 injection, specifically in the context of DTS‐based real‐time monitoring. In addition, the study demonstrates the potential of DTS as real-time solution for thermal anomaly detection, integrated with pressure and temperature measured from standard gauges. While the initial results are promising, further analyses acrossotherwell configurationsandoperational conditions are important to enhance the generalizability and robustness of the approach. However, this work lays the groundwork for advanced and automated permanent FO monitoring systems aimed at ensuring wellbore integrity and environmental safety in subsurface CO2 storage operations.