Advancement, Assessment, and Application of Novel Landslide Monitoring Technologies
Slopes are often monitored by measuring deformation rates, or the factor of safety is estimated through the use of limit equilibrium stability models to evaluate the risk of failure. In this thesis, three novel landslide monitoring technologies are advanced, assessed, and applied using three strategically chosen field sites. Firstly, it was investigated if acoustic emissions (AE) could be used to measure the displacement of extremely slow-moving landslides. By installing a shallow and a deep AE sensor as well as a ShapeAccelArray (SAA) in a slope located in Peace River, Alberta, it was found that data from a shallow AE sensor allows noise to be filtered from the deep AE sensor data, and that smaller displacement rates than previously expected can be measured using an AE sensor. The second monitoring advance explored a method of measuring 3D slope displacement using digital image correlation (DIC) performed on hillshade images (image of shaded point cloud data). Through the use of synthetic movement tests and experimental data, it was found that 3D displacement can be measured using hillshade images at two different view angles and that small deformations leading to failure can be measured allowing the time to failure to be calculated. Thirdly, point cloud data produced from historical air photos was used to investigate whether historical slope profiles could be produced to serve as a quantitative baseline of historical landslide activity. This hypothesis was tested to investigate the impact of land-use change on retrogressive landslides occurring along a waterway. Natural revegetation of the area around the creek was observed to result in a significant decrease in the volume of landslide debris generated by geomorphic processes. Finally, back analyses were conducted to estimate the mobilised shear strength at failure through the use of monitoring data from two 3.5 metre high earth dams brought to failure. This unique dataset provided an assessment of the repeatability of the back analyses and showed that only a small component of apparent cohesion arises from a combination of dilation, unsaturated soil behaviour, or root reinforcement.