Reconstruction of rough rock joints: 2D profiles and 3D surfaces

Abstract

The 3D reconstruction of rock joint surfaces is crucial for quantifying rock joints. Through the uniform discretization of 2D and 3D rock joints, it was determined that both the climbing angles along the shear direction and the projection lengths of the consecutive ascending/descending segments follow normal distributions . This statistical distribution of climbing angles and projection lengths provides an important reference for rock joint reconstruction, although hypothesis testing indicated that some joint profiles/surfaces do not follow normal distributions. A normal distribution-based method and a bubbling method were used to reconstruct the 2D joint profiles. The bubbling method demonstrated good applicability for the generation of 3D rock joint surfaces, especially because it included surfaces with different roughness in two orthogonal directions. Therefore, ten typical 3D rock joint surfaces were reconstructed using the bubbling method, where both the size of asperities (waviness or unevenness) and the joint roughness coefficient values were quantified. To reconstruct a joint surface with a certain roughness coefficient, scaling the elevation of a generated surface was demonstrated as an effective supplementary approach. • Rock joints were statistically analyzed via normal distributions owing to a uniform discretization operation. • The normal distribution-based method and the bubbling method were both suitable for reconstructing 2D joint profiles. • The bubbling method could generate 3D joint surfaces with different roughness in two orthogonal directions. • Rock joints with different joint roughness coefficients were reconstructed by scaling their elevation coordinates.