ABSTRACT:
Outcrop-scale directional roughness profiles can be extracted from photogrammetric or laser scanner point clouds using a straightforward five step procedure consisting of: 1) calculation of the dip direction and magnitude for use as a geologically significant reference vector; 2) rotation of the coordinate system into alignment with the dip-line and strike-line; 3) interpolation of scattered points onto a grid in a plane defined by the dip-line and strike-line; 4) extraction of linear profiles in desired directions relative to the dip-line; and 5) estimation of quantities such as Patton asperity angles (i) and Barton joint roughness coefficients (JRC) using published empirical relationships. Because rotation of the coordinate system effectively reduces the dimensionality of the problem, rock surface roughness can be conveniently visualized using contour maps, shaded relief images, or three-dimensional surface plots. Profile data extracted from point clouds can also be used as the basis for more advanced methods such as spectral or fractal analysis of rock surface roughness. Extraction of profiles and estimation of i and JRC values is illustrated using data from a large joint surface encountered in a highway rock slope stabilization project along Interstate 90 near Snoqualmie Pass, Washington, USA.
1 INTRODUCTION
Practical measurements of rock surface roughness have traditionally been made using methods that require direct contact with the rock (e.g. Wyllie & Norrish 1996). Although direct methods can produce accurate and useful results, they have some significant limitations. Traditional field-based measurements can be made only on rock surfaces that are accessible, which can require the use of technical rock climbing equipment or mechanical lifting devices that are expensive, time consuming, and may expose field personnel to hazardous conditions. Some practitioners use pocket-sized mechanical contour gauges that, while being simple and easy to carry, can measure profiles only 15 to 20 cm long.
