SHEAR SLIP FAULT REACTIVATION AND INDUCED SEISMICITY ASSOCIATED WITH HYDRAULIC FRACTURING IN THE DUVERNAY FORMATION, FOX CREEK AREA, ALBERTA: INSIGHTS FROM 3D RESERVOIR GEOMECHANICAL MODELING
The analysis of ant-tracking attribute of 3D seismic data revealed a presence of a 1.4-km long linear discontinuity that may represent a fault zone aligned with the linear zone of seismicity. The inferred fault extends vertically for ~680 m from the Precambrian basement to the top of the Ireton Formation. Discrete Fracture Network (DFN) was introduced to 3D structural model in the Duvernay interval. Hydraulic fracturing (HF) and formation pressure buildup were simulated in the shale reservoir along two horizontal wells closest to the linear zone of induced seismicity, respecting field management history data. The initial Pp in the Duvernay Formation is ~52-59 MPa (gradient 16-17 kPa/m). Maximum bottomhole pressure during HF reached 79-80 MPa (~23 kPa/m) during wells treatment. The simulation results show that HFs grow parallel to the orientation of present-day maximum horizontal stress (N43°E), interacting with DFN. The pressure increase by ~20 MPa propagates along the HFs and reaches the fault zone.
The 3D one-way coupled reservoir geomechanical modeling was conducted to analyze changes in horizontal stresses and fault shear instability as a result of Pp increase by 20 MPa in the fault zone. Plastic shear strain in fault elements indicates that shear slip occurs at the depth interval of the Duvernay and Lower Ireton Formations. It is likely that Pp increase associated with fracturing stimulation was the main cause of the induced earthquakes. This study helps to identify the root causes of fault reactivation during HF, assisting in mitigation of future risks of induced seismicity associated with HF of the Duvernay Formation, one of the major producing gas shale plays in the Alberta Basin.
We thank Pulse Seismic for donation of seismic data, Schlumberger for consolidated license donation to the University of Alberta. This work was funded by NSERC through grants RGPIN 2019 04397 and DGECR 2019 00186.