Search for a command to run...
Summary This work describes the analysis of wellbore storage (WBS) to monitor induced fracture dimensions over time in a waterflooded field transitioned to fractured injection (FI) conditions. A pivotal risk to manage in FI is the loss of containment (LOC), due to induced fractures growing and eventually connecting to offsetting faults, potentially creating a leak path into the overburden and beyond. Hence, fracture size relative to distance to faults requires strict monitoring to sustain safe injection operations. The required surveillance program, methodology, tools, and workflow implementation are illustrated in this article, with field examples. A fit-for-purpose surveillance program was designed to monitor fracture size via zonal step-rate-tests (SRTs) and short-duration pressure fall offs (PFOs), targeting early-time data analysis and focusing on WBS coefficient ("C") estimations, used as a proxy for induced fracture size. To assess fracture size, a modular Python-based tool was adopted. The tool comprises modules to i) extract raw injection performance data; ii) quality check data; and iii) match PFO to estimate "C" and the equivalent fracture half-length (Xf). The well used to illustrate the workflow (Well-X) is a multi-zone injector, with zonal PDHGs and ICVs, operated in FI conditions since late 2023. Well-X is 100m away for the nearest fault and 120m away from the closest offset well, imposing inherent safety limits on the allowable induced fracture size. Well-X zonal SRTs were used to identify the fractured zone. Then, using the WBS tool, raw injection data was extracted, and shut-in events were identified and inspected to detect data disturbances/noise to determine their suitability for reliable analysis. Early PFO data was matched to estimate "C", followed by parameters’ trends visualization and zonal injection rate corrections to generate base-case estimations. Zonal trend plots showed a low value of "C" (close to wellbore volume) up to June-2024, followed by a significant increase attributed to the presence of a propagating fracture in alignment with SRT results. The estimated WBS was translated into Xf, by assuming fracture height scenarios based on well logs and well completion configuration, and by considering elastic rock properties. The maximum estimated Xf was 65m, with a range depending on height assumptions (the smaller the height, the longer Xf). To tackle reduce this uncertainty and constrain Xf estimations, DTS/DAS (Distributed Temperature Sensing and Distributed Acoustic Sensing) surveillance was deployed. The incorporation of DTS/DAS technology allowed direct measurement of near-wellbore fracture height to constrain Xf and rule out disturbances across or close to the caprock generally present in the case of uncontrolled fracture height overgrowth. In contrast to the use of conventional PTA tools/methods, the implemented WBS methodology allows for a custom-tailored interpretation of the very early pressure fall-off data, minimizing required shut-in times and associated water injection deferment. It also allows to take advantage of unplanned short shut-ins to monitor FI wells and offers an alternative to wells not meant to inject under fractured conditions, requiring checks for the presence of undesired induced fractures.