A Tool for Performing Automatic Kinematic Analysis on Rock Outcrops

Abstract

The assessment of rock outcrops’ predisposition to the main possible kinematisms represents the preliminary step of stability analysis: Markland’s tests for sliding and toppling constitute a milestone due to the ease of use and interpretation of results. Orientation and friction angles of the main discontinuity sets and orientation of rock faces are required as input to perform the test on a stereonet graphically. However, for natural outcrops, the orientation of rock faces could vary significantly, and the test should be performed assuming all the representative ones. To speed up this process, the authors set up an automatic procedure based on the GIS environment working principles and developed it in Matlab language. Main discontinuity sets orientation and relative friction angles, along with slope and aspect data representing the rockface orientation of the considered outcrop, are the input data. The slope and aspect data are in GeoTIFF format, the most common format for georeferenced raster files employed in a GIS environment. The Matlab code performs Markland’s tests for planar and wedge sliding and flexural toppling, considering all the possible sets or intersections of sets, and provides the output with the same extent and georeferencing of the input data. The outputs are a series of GeoTIFF raster files describing the result for each kinematism separately and globally, which can be imported directly into GIS. The global results can also be used to map source areas for 3D rockfall numerical simulations. The code was validated through a case study by comparing its results with those obtained by performing the conventional tests singularly on a number of significant rock faces. The results obtained in the case study show that the algorithm produces reliable results consistent with those provided by traditional methods.

A Design Scenario Approach for Choosing Protection Works against Rockfall Phenomena

“…In this study, we follow the approach to block size description proposed by Lu & Latham (1999) [20], expanded in later works such as the one presented in Umili et al (2023) [18]. In these works, the approach to the quantification of block size relies on probability distributions describing the spacing of the discontinuity sets identified in the rock mass: on their basis, the block volume distribution can be computed, employing the well-established Palmstrom's formula [37]; this equation, which considers the block generated by the intersection of three discontinuity planes, is not geometrically proven and, actually, it tends to either underestimate or overestimate the block size, depending on the reciprocal orientation of the sets involved. A solution to this problem was recently found, and a new formulation was introduced [18].…”

“…The original version of the tests relies on stereographic projections and allows only for one rockface orientation to be tested at a time. A much more practical solution was proposed by Taboni et al (2022) [37], who introduced an algorithm capable of performing Markland's test requiring a DSM of the rockface and the geometric and mechanical features of the discontinuities. The advantage of this tool (named Automatic Markland's Test Tool, AMTT) is clear: the complexity of the rockface geometry, which originally required the stereoplot tests to be replicated as many times as different orientations could be identified, is an overcome problem; the use of a DSM allows for inferring the orientation of the rockface in each of its points, the only limit being the spatial resolution and accuracy of the DSM itself.…”

“…To identify the source area, the AMTT tool [37] was used: this algorithm requires the reference orientation values for each of the three discontinuity sets, the slope angle and Having selected the reference volume of 6.5 m 3 and knowing the shape distribution, the 3D numerical simulations were performed. Rockyfor3d requires a set of ten grid files containing all the features of the slope, of the source area and of the blocks to be simulated.…”

“…To identify the source area, the AMTT tool [37] was used: this algorithm requires the reference orientation values for each of the three discontinuity sets, the slope angle and…”

“…The ten classes of soils and surfaces identified on the slope of the chase study area, their soil type and the range of normal restitution coefficient (R n ) values, as presented in the software's manual [27]. To identify the source area, the AMTT tool [37] was used: this algorithm requires the reference orientation values for each of the three discontinuity sets, the slope angle and the aspect angle of the considered rock face. These two values were extracted from the DSM obtained in Phase 1.…”