Abstract: To further improve accuracy and reliability of describing the roughness feature of the structural plane, and more comprehensively reflect influences of the morphological differences, the fluctuation of the profile was decomposed into three principal characteristics: amplitude, inclination, and curvature, which represent the change characteristics of height, angle, and bending of the profile. The data of the three characteristics were obtained by 10 standard profiles established by Barton. And the distribution curves of the three characteristics were analyzed. Then, the power spectral density (PSD), which can well describe the microscopic morphology, was used to analyze the spectral characteristics of amplitude, inclination and curvature. The fractal parameters (corresponding to the slope of the PSD curve) and roughness parameters (corresponding to the intercept of the PSD curve) were got from the logarithmic PSD curves. The relationship among , and the joint roughness coefficient (JRC) was studied, and the expressive forms of JRC with the new index were established. Meanwhile, the new index was extended to reveal the characterization of three-dimensional morphology, which was also used to calculate surface roughness of a marble surface obtained by the shear test. The results show that the new index has an extremely slight evaluation error, and can reflect the anisotropic property and effectively character morphology features of the structural plane, which provides a reference for characterizing the roughness of the structural plane and it is of great significance to study shear mechanical properties of the structural plane.

Key words: structural plane, roughness characterization, power spectral density, curvature