Comprehensive characterization and clustering of orientation data: A case study from the Songta dam site, China

The quantitative description of orientation data in rock masses is fundamental for understanding the engineering geological properties of rock masses. However, the features of joint orientation distributions are still not easy to describe impractical projects. Orientation data are randomly distributed in a rock mass, and its poles are also randomly distributed on a stereonet. The delineation of orientation data on the stereonet is difficult, but the density of poles on the stereonet indicates some characteristics of joint geometric patterns, which will influence the mechanical and hydraulic properties of rock masses. Considering the nonlinear distribution of poles density, fractal geometry is introduced to describe the orientation data in this study. Accordingly, a framework combining the idea of fractal geometry, graphic display and the Schmidt upper-hemisphere equal-area projection plot is proposed for comprehensive characterization of orientation data. The monofractal and multifractal descriptions have been used to delineate the orientation data, and a fractal indicator is proposed for the dispersion of orientation data. In addition, the identification and delineation of joint sets with similar orientations are discussed, and the optimal method is recommended. Finally, the methods proposed in this study were further applied to a real data set collected from a survey at the dam site of the Songta hydropower station, China. (C) 2017 Elsevier B.V. All rights reserved.