HR: 1340h
AN: H33F-0529    [Abstracts]
TI: Transport Property Modeling in Partially-saturated Rocks Using Pore-scale Simulations
AU: * Keehm, Y
EM: keehm@stanford.edu
AF: Stanford University, 397 Panama Mall, Geophysics, Stanford, CA 94305-2215 United States
AU: Nur, A
EM: anur@stanford.edu
AF: Stanford University, 397 Panama Mall, Geophysics, Stanford, CA 94305-2215 United States
AB: The earth sciences are undergoing a gradual but massive shift from description of the earth and earth systems, toward process modeling and simulation. This shift is very challenging because the underlying physical and chemical processes are often nonlinear and coupled. In addition, we are especially challenged when the processes take place in strongly heterogeneous systems. One example is multiphase fluid flow in rocks, which is a nonlinear, coupled and time-dependent problem and occurs in complex porous systems. To understand and simulate these complex processes, the knowledge of underlying pore-scale processes is essential. To this end, we have initiated computational rock physics to rigorously simulate rock/reservoir properties. The computational rock physics framework is based on digital representations of rocks, which consist of minerals and fluids, and may evolve with time. It also contains modular physical property simulators, with which we directly simulate physical properties of rocks. This computational environment significantly complements the physical laboratory: 1) rigorous prediction of the physical properties, 2) interrelations among the different rock properties using shared digital porous media, and 3) simulation of dynamic problems with multiple physical responses. As a continuing effort of this framework, we investigated transport properties in partially-saturated rocks. Two digital rock structures were used in this study - X-ray tomographic Fontainebleau sandstone and random dense pack of spheres. Partially-saturated rock samples were obtained through two-phase flow simulations. We used both steady-state and unsteady-state simulations to compare static and dynamic properties at different partial saturations. We then performed single-phase and electrical flow simulations to calculate permeability and electrical conductivity. We found that water phase percolates around Sw=20% and air phase loses connectivity around Sw=60-70%. We also observed electrical conductivity hysteresis between drainage and imbibition simulations. The amount of hysteresis shows strong anisotropy, which shows strong dependence on the orientation of fluid flow. However, angular averaged electrical conductivity did not show any significant hysteresis. The relation between peremeability and electrical conductivity varies with different rock geometry. It then is possible to characterize the relation between electrical conductivity and permeability in a given rock formation, and to provide rigorous links to understanding and modeling geological processes.
DE: 5109 Magnetic and electrical properties
DE: 5112 Microstructure
DE: 5114 Permeability and porosity
DE: 5139 Transport properties
SC: Hydrology [H]
MN: 2004 AGU Fall Meeting