We present 3D microseismic depletion delineation (MDD) numerical experiments for stimulated rock volume (SRV) estimation using a novel finite difference method for simulating fast fully coupled 3D quasistatic poroelasticity in fractured rock. We demonstrate MDD using parameters from Duverney oil field in Canada. The results of the simulation provide an explanation for the mechanics of the MDD process and illustrate the conditions for MDD's success.
Understanding how the depletion process effects the reservoir is important in effective development of unconventional plays, particularly for quantifying the stimulated reservoir volume (SRV). MDD is a promising tool for monitoring SRV. It has been demonstrated in the field (Dohmen et al., 2013, 2014), theoretically (Dohmen et al., 2017, Wang et al. 2019) and numerically in 2D (Norbeck and Horne, 2016, 2018; Jin and Zoback, 2018, 2019). However, to our knowledge, no fully coupled poroelastic MDD numerical study has been conducted in 3D. The reason for this stems from the fact that most geomechanical modeling tools coupled with flow are based on implicit implementation of finite difference (finite volume or finite element). Thus, it requires solving of large 3D linear systems (e.g., Jha and Juanes, 2014; Castelletto et al., 2019; Frigo et al., 2019) with a significant computational effort. This effort is directed towards the construction and storage of the matrix, and the practical methods on how to solve the linear system. Use of high-performance computing (HPC) becomes an integral part of this effort. This limitation sets bounds on maximum dimensions that can be used in the simulation. In this study, we present a novel method for fast 3D fully coupled quasistatic Biot's poroelastic finite difference modeling for investigating 3D MDD using parameters from Duverney oilfield in Canada. Our study is divided into two parts. In the first part, we provide the mathematical derivation of the new method which is based on a rescaling of the solid rock and the fluid flow density parameters. In the second part, we compute MDD in the medium with stimulated and natural fractures using the developed fast fully coupled 3D poroelastic modeling and Mohr-Coulomb (MC) failure criteria (e.g., Juvinall, 2007). We conduct two MDD experiments. One experiment is in the presence of the liquid water and the second in the presence of a gas-condensate. The simulations of the MDD support the physical observations conducted in the field and provide insight on the MDD's success.