Integration of Geoscience and Engineering Concepts to Account for Natural Fractures in Fluid Flow within Shale Reservoirs

Abstract

Unconventional shale reservoirs are characterized by low porosity and ultra-low permeability. Natural fractures are known to be present and considered a critical factor for the enhanced post-stimulation productivity. Accounting for natural fractures with existing techniques has not been widely adopted owing to their complexity or lack of validation. Ongoing research efforts are striving to understand how natural fractures can be accounted for and accurately modeled in fluid flow of the subject reservoirs. This study utilized Eagle Ford well data comprising reservoir properties, stimulation metrics, production, microseismicity and permeability measurements from a core plug. The methodology comprised use of production data to extract a linear flow regime parameter. This was coupled with fracture geometry, predicted from hydraulic fracture modeling and microseismicity, to estimate the system permeability. From interpreting microseismic events as slips on critically stressed natural fractures, explicit modeling incorporating a discrete fracture network (DFN) assumed activated natural fractures supplement conductive reservoir contact area. Thus, allowed the estimation of matrix permeability. For validation, the aforementioned was compared with core plug permeability measurements. Results from modeling of planar hydraulic fractures, with microseismicity as validation, predicted planar fracture geometry which when coupled with the linear flow parameter resulted in a system permeability. Incorporation of DFNs to account for activated natural fractures yielded matrix permeability in picodarcy range. A review of laboratory permeability measurements exhibited stress dependence with the value at the maximum experimental confining pressure of 4000 psi in the same range as the computed system permeability. However, the confining pressures used in the experiments were less than the in situ effective stress. Correction for representative stress yielded an ultra-low matrix permeability in the same range as the DFN-based picodarcy matrix permeability. Thus, supporting the adopted drainage architecture and often suggested role of natural fractures in shale reservoir fluid flow. This study presents a multi-discipline workflow to account for natural fractures, and contributes to understanding that will improve laboratory petrophysics and the overall reservoir characterization of the subject reservoirs. Given that the Eagle Ford is an analogue of emerging shales elsewhere, results from this study can be widely adopted.