Coupled Flow and Geomechanical Modeling of Reservoir Seismicity: Effects of Hydraulic Communication and Well Type

Abstract Oilfields in California are heavily faulted, stress sensitive and need to be monitored for geomechanical effects such as fracturing, subsidence and seismicity induced by production or injection in the field. The objective of this paper is to investigate well production and injection-induced activation of a fault based on two factors: 1) hydraulic communication between the fault and the well i.e. intersecting vs. non-intersecting reservoir fault, and 2) sign of pressure change in the reservoir i.e. positive during injection vs. negative during production. Using a novel coupled flow and geomechanical simulation framework, we model the physical processes involved in activation of a fault and the effects of hydraulic communication and pressure change factors. We use the finite element method for mechanics sub-problem, the finite volume method for the flow sub-problem and the fixed-stress sequential solution strategy to solve the coupled sub-problems. We use the Mohr Coulomb failure criterion to determine mechanical stability of the fault against shear failure. Our study provides key observations for long-term oilfield management in stress-sensitive and faulted reservoirs. We find that during injection pore pressure changes dominate the stability of a fault when it is in hydraulic communication with the well, whereas total stress changes dominate the stability of the fault when it is hydraulically isolated from the well. Injection-induced seismicity is a canonical example of the former case. In terms of the sign of pressure change, production from one side of a sealing fault always destabilizes the fault regardless of hydraulic communication between the fault and well whereas injection can stabilize a fault that is not in hydraulic communication with the well. In conclusion, we propose a novel approach to classify induced seismicity events as a function of reservoir geology (fault-to-well hydraulic communication) and reservoir operation (injection or production). The framework can be applied to monitor reservoir performance in faulted and stress-sensitive fields such as the ones in California.