2009 Portland GSA Annual Meeting (18-21 October 2009)

Paper No. 3
Presentation Time: 9:00 AM-6:00 PM

HOW CHANGING APERTURES AFFECT THE TRANSMISSIVITY OF ROUGH FRACTURES


CRANDALL, Dustin, NETL, Morgantown, WV 26507-0880 and BROMHAL, Grant S., National Energy Technology Lab, Department of Energy, 3610 Collins Ferry Road P.O. Box 880, Morgantown, WV 26507, grant.bromhal@netl.doe.gov

Fractures have the potential to enable fast fluid transport pathways through otherwise low permeability geologic media; thus, properly characterizing how fluid transport within fractures occurs is crucial to estimating migration of sequestered CO2 or other fluids through naturally fractured systems. Fracture properties within field settings are known to be dynamic, varying with changing stress fields, mineralization upon walls, and dissolution of matrix material. For decades researchers have examined the effects of fracture wall-roughness on fluid transport by separating fractures, measuring the fracture surfaces, reconstructing analogue models, and performing simulations and experiments within the reconstructed geometries. We have obtained computed tomography scans of a fracture created within sandstone and performed a series of fine-grid computational fluid dynamic simulations within the reconstructions of this rough-walled geometry. By changing the ‘physical’ aperture within our model, we have shown that the transmissivity of rough-walled fractures is well described by the cubic-law when the average fracture aperture is much larger than the surface deviations. Conversely, when the ‘physical’ fracture aperture is similar to the surface variations along the fracture walls the cubic law over-estimates flow through the fractures. The ability to control the fracture dimensions within our fine-grid computational models enables us to determine relationships between the fracture geometry and the observed flow through these naturally complex potential escape pathways. By linking these relationships to discrete fracture reservoir-scale models we are working towards more accurate estimation procedures to plan leakage monitoring activities.