Sampling and Distribution Parameter Analysis for Estimating Three-dimensional Fracture Orientation Distributions from a Circular Sampling Window

Abstract

Precise evaluation of three-dimensional (3D) fracture orientation distributions is vital to a reliable discrete fracture network (DFN) model for rock mass. This study analyzed the accuracy and dispersion of the evaluated orientation distributions of fracture sets with different fracture geometric parameters. The geometric probability method by two-dimensional (2D) trace data is introduced and compared with the three methods of the Terzaghi family by one-dimensional (1D) orientation data. The Monte Carlo method generated many groups of 3D fracture networks to analyze the influence of uniform orientation distribution, average fracture diameter, and volume density on the evaluated orientation distributions. The results indicated that the piecewise distribution increases the computational complexity of the geometric probability method. This method can be used to effectively evaluate the orientation distributions of fracture sets with an average diameter greater than 0.25 m and volume density greater than 0.2 m−3.