Matching Simulated Fracture Data with Field Measurements Using Joint Intensity
Abstract
Three-dimensional models of fracture patterns were generated using different measures of joint intensity (lambda), a measure of joint surface area per unit volume. The simulated data were compared to field measurements and the models were regenerated until the two matched as close as possible. Mean vertical joint spacing was used for the comparison. Because only a sample of trace lengths was measured in each outcrop, lambda was initially defined for each joint set as nl/hw, where n = number of joints, and l = mean trace length, h = sampling area height, and w = sampling area width. Simulated mean vertical joint spacings using this definition were significantly lower than the means for the field measurements. Mean simulated vertical spacing is 0.12 m, whereas mean measured vertical spacing is 0.23 m, indicating that lambda at 12.10 is too high. Consequently, the definition of lambda was reevaluated. Sampling area widths varied for each joint set in each outcrop, so joint set intensities were weighted proportionally with respect to sampling area width. Joint intensities for the horizontal sets were recalculated using the length of the longest joint in that set as the sampling area height. In addition, the second set of models was generated in increasing termination percent order which better represents geologic conditions. Mean simulated intensity for these models is 7.34, and mean simulated vertical spacing is 0.20 m. These results are realistic with respect to the field data.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 1996
- Accession Number
- ADA354984
Entities
People
- Judy Ehlen