Linking Experimental Capillary Pressure-Saturation ...

Object Details

View

XVI International Conference on Computational Methods in Water Resources (CMWR-XVI) Ingeniørhuset

Linking Experimental Capillary Pressure-Saturation Data with Lattice Boltzmann Simulations.
Paper
Author:Marcel Schaap <mschaap@ucr.edu> (University of California, Riverside and GEBJ Salinity Lab)
Britt Christensen <brc@er.dtu.dk> (nstitute of Environment & Resources, Technical University of Denmark, DK-2800 Kongens Lyngby, Denmark)
Mark Porter <porterm@geo.oregonstate.edu> (Department of Geosciences, Oregon State University, Corvallis, OR 97331, USA)
Dorthe Wildenschild <wildend@geo.oregonstate.edu> (Department of Geosciences, Oregon State University, Corvallis, OR 97331, USA)
Presenter:Marcel Schaap <mschaap@ucr.edu> (University of California, Riverside and GEBJ Salinity Lab)
Date: 2006-06-18     Track: Special Sessions     Session: Boltzmann Methods in Water Resources
DOI:10.4122/1.1000000273
DOI:10.4122/1.1000000274

Recent advances in observational and computational techniques have facilitated the study of fluid dynamics and interfacial geometry in porous media. Within some experimental limitations, computed tomography X-ray (CMT) and magnetic resonance imaging (MRI) are now able to accurately map the 3D structure of porous geometries. Computational advances largely concern Lattice Boltzmann (LB) method that has been shown to be useful in simulating microscale flow in porous media. With some phenomenological or thermodynamic extensions, the LB method is also able to deal with microscale interfacial phenomena in single or multiphase systems. The goal of this presentation is to provide insight into what is needed to make a link between 3D experimental observations of interfacial geometry and LB simulations. The experimental data consist of CMT observations several Sotrol-water displacements inside a glass bead system with a resolution of 17 microns. Also available are capillary pressure-saturation curves between 0 and 1kPa. The LB model is that of Shan-Chen as modified by Martys and Chen (1996). We present the most parsimonious way to calibrate the surface tension and contact angle in the model, define space, pressure and time scaling. We will also identify potential problems relating to pore-size and digitization effects that are present in the simulations, but not in the original observations. The analyses are partly performed on idealized systems and finally applied to large scale (107 voxel) simulations of the real physical systems. Observations are simulations are compared in terms of pressure-saturation curves, and where possible, in terms of fluid distribution and interfacial curvatures.