TUNCAY, KAGAN¹, ANTHONY J. PARK¹, DOROTHY F. PAYNE¹, SPRING ROMER¹, KENNETH R. SUNDBERG², and THOMAS E. HOAK^3
1Laboratory for Computational Geodynamics, Indiana University, Bloomington, IN
²Phillips Petroleum Company, Bartlesville, OK
³SAIC and Kestrel Geoscience, LLC, Littleton, CO

Abstract: Three-Dimensional Forward Modeling of Naturally Fractured Reservoir Formation

The prediction of locations and characteristics of fracture zones is difficult because of the complex networks of strongly coupled, reaction-transport-mechanical (RTM) processes. In order to address this problem, we have developed a three-dimensional forward modeling approach which uses seismic, core and surface data as input. This model accounts for the coupling of important processes, for example the following: fracturing causes an increase in permeability, which reduces fluid pressures, allowing fractures to close. Our incremental stress rheology integrates poroelasticity, nonlinear viscosity, pressure solution, faulting and fracturing.

Simulation results show good agreement with observed reservoir characteristics in the naturally fractured Permian Basin in Texas and the Piceance Basin in Colorado. General results suggest: 1) the propensity for fracturing is a function of the rheological properties; 2) fractures are induced by flexure, extension, salt migration, and in the absence of extensional stresses, overpressure and hydrocarbon generation; and 3) the dynamic and often episodic nature of fracturing play a key role in the timing and geometry of petroleum expulsion, migration and trapping. This model has important implications for the prediction of naturally fractured reservoirs as it predicts the evolution and distribution of fracture length, aperture, intensity, orientation, and anisotropic permeability.

AAPG Search and Discovery Article #90928©1999 AAPG Annual Convention, San Antonio, Texas