EVALUATING THE EFFECTS OF FRACTURE ROUGHNESS ON FLUID FLOW AND SOLUTE TRANSPORT: IS SCALING POSSIBLE?

Where connected open fractures are present, they dominate fluid flow and transport of solutes. A major challenge in predicting solute transport in fractured media is describing fracture physical characteristics at a scale that is appropriate for modeling. Assessing the impact of roughness on field-scale fluid flow through fractured media from samples of natural fractures on the order of 10 x 10cm assumes a relationship between fracture morphology and discharge is either scale invariant or smoothly transformable. Some research suggests that the length scale at which surface roughness significantly contributes to the discharge through a fracture falls within the size of a typical hand sample, but there are few data supporting an extension of small-scale relationships to larger scales and our previous studies on fracture roughness in granites showed a lack of stationarity. We analyze the results of flow tests on a single fracture through a large block of rhyolitic tuff, and the corresponding numeric modeling of the fracture surfaces imaged by computed tomography at The University of Texas HRXCT facility. These results are compared with relationships of smaller samples in similar tuffs. Fracture surface roughness of various tuffs are quantified by HRXCT and the data processed to yield regularly gridded surface elevations. Some of the fracture surfaces are self-affine and a corresponding fractal dimension can be ascribed to create representative models. Describing roughness as a ratio of surface area to footprint, variances of the roughnesses of equivalently sized square samples are plotted against sample size to determine if a representative surface exists. For fractures in tuffs measuring up to 25 x 10cm, a 3.2 x 3.2cm sample yields a reasonable expression of the roughness of the entire surface. We are currently comparing this to roughness at the 60 x 30cm scale. The results suggest that there may be a suitable scale of data for upscaling roughness, at least in welded tuffs. Upscaling of other factors, such as channeling, remains to be tested.