Rock, unlike most engineered materials, possesses a large degree of inherent variability in physical characteristics. This creates a number of unique challenges when creating engineering works involving rock. When characterizing a rock, engineers must collect sufficient data so that the epistemic uncertainty is small relative to the aleatory variability. The aleatory variability is an inherent characteristic of rockmasses, and refers to the natural variation within the rock, which occurs in even the most homogeneous rocks with uniform size, shape, and orientation of grains. Thus, while epistemic uncertainty, which relates to a lack of knowledge can be minimized, variability in the uniaxial compressive strength values contributes to the overall engineering property variability of the material, and needs to be understood in order to characterize a rock. To quantify the extent of spatial correlation, covariograms and semivariograms from core specimen images are calculated from a 2D covariogram and semivariogram image, and the quantitative metrics of spatial heterogeneity can be used to represent aleatory variability. This then can be applied to multiple rocks to qualify and quantify their relative geomechanical variability.
Modeling Geologic and UCS Variability Through Covariogram Analysis of Rock Core Images
- Share Icon Share
- Search Site
Boyd, D. Lane, Walton, Gabriel, and Whitney J. Trainor-Guitton. "Modeling Geologic and UCS Variability Through Covariogram Analysis of Rock Core Images." Paper presented at the 51st U.S. Rock Mechanics/Geomechanics Symposium, San Francisco, California, USA, June 2017.
Download citation file: