Abstract

Although JRC profiles are commonly used for shear strength assessment of rock joints, they can also serve as a base model for weathering assessment. In this study, JRC profiles were characterized using eight promising roughness algorithms from diverse fields such as geomorphology, rock mechanics and signal processing. Roughness indices of digitized JRC profiles were assessed at three different sampling intervals to observe how scale impacts their correlation with JRC. Plots of JRC versus roughness indices exhibit one of two general shapes: either a generally (but sometimes monotonically increasing) curve with occasional small peaks at JRC 6-8 and JRC 12-14 or a bimodal shape that peaks consistently at JRC 10-12 and JRC 14-16 followed by a sudden drop in roughness. Z2 and Mean Absolute Angle show the highest linearity of the eight algorithms. Z2 has been used numerous times to assess JRC profiles but Mean Absolute Angle is a new and promising roughness algorithm. Standard Deviation (s) and Signal Energy (Es) demonstrate remarkable consistency between sampling intervals but their relative non-linearity with JRC suggests they are insufficient in quantifying roughness. A new JRC relationship, based on Z2 and intercept of the log-log plot of Z2 versus sampling interval, is presented

1. INTRODUCTION

Roughness and associated shear strength of discontinuities has long been an important aspect of rock engineering. One of the first studies relating normal force and joint roughness to shear behavior was conducted by Patton [1]. As this research progressed, joint roughness became an input into rock mass classification systems. All students, practitioners, and researchers in the field of rock mechanics are familiar with the joint roughness coefficient profiles (JRC) developed by Barton and Choubey [2]. This seminal tool has been instrumental in the development of roughness and shear strength assessment of discontinuities.

Quantifying roughness using various algorithms to compute roughness indices (RI) is far from a novel idea and has been explored very soon after the publication of the JRC profiles. However, agreement is seldom found in regards to the proper roughness index [3, 4, 5, and 6], discretization methods such as sampling interval selection [4, 7, 8, and 9], and even the statistical nature of the JRC profiles and other rough surfaces [10 and 11], proving that the task can be deceptively complex. As such, the particulars of proper JRC roughness assessment are still being openly discussed.

This content is only available via PDF.
You can access this article if you purchase or spend a download.