A versatile monitoring framework for detecting out-of-zone leakages across active and abandoned wells using pressure transient analysis
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Leakage out of the target formation is a major risk associated with subsurface injection activities, including those in hydrocarbon reservoirs, geothermal systems, and geological CO₂ storage. While previous studies often focused on individual leakage features in isolation, this paper proposes a holistic monitoring approach that simultaneously considers multiple potential pathways, including induced fractures and both active and legacy wells. These features may become conductive due to injection-induced pressure buildup and fluid migration within the reservoir. This simulation study is structured in two main parts. The first addresses the mechanisms of induced fracturing and fracture propagation through caprock, potentially leading to out-of-zone leakage. Then, it presents a modeling approach and a Pressure Transient Analysis (PTA) method for early detection of leakage into shallow formations. The second part examines casing-cement-rock integrity failure mechanisms that can result in behind-casing leakage or crossflow. This is followed by a dedicated modeling approach and a PTA-based monitoring method for identifying leakage behind the casing. The proposed monitoring technology utilizes real-time pressure data from Permanent Downhole Gauges (PDGs) installed in wells. A mechanistic well-reservoir model assembled based on a representative geological setup of a sandstone formation offshore Norway was used to test the methodology. The results have demonstrated the capability of the PTA-based methods to detect leakage and estimate leakage rates, with high accuracy for high-rate events and decreasing precision for smaller rates. Compared to costly methods like time-lapse seismic, these methods provide a cost-effective solution for enhancing reservoir containment monitoring workflows in both large-scale injection projects and smaller ones with limited data and resources.