Abstract:
This thesis describes the application of airborne and surface geophysical remote sensing to the study of a prehistoric landslide on the tephra-mantled coastline, south of Port Waikato at Ohuka, North Island, New Zealand. The landslide morphology includes an arcuate headscarp, lateral scarps, back-tilted blocks, and pull-apart zones, incised by post-failure gullying. The toe of the landslide has been removed by coastal erosion, so the original down-slope extent of the landslide is unknown. The project also involved an initial desk study, field surveys, and laboratory analyses in order to provide detailed information about the nature and extent of the landslide in the area. The airborne remote sensing utilised an unmanned aerial vehicle (UAV) mounted camera to provide a high resolution image dataset. The imagery was then processed and modelled using the structure-from-motion (SfM) technique, to provide a digital elevation model (DEM) of the landslide, comparable with LiDAR in quality. The subsurface of the landslide was investigated using electrical resistivity tomography (ERT), to provide insights into the stratigraphy and present hydrogeological conditions. In situ logging of exposures and strength testing of materials allowed ground-truthing of geophysical data, and samples were further investigated in the laboratory using strength index properties. Sensitive clay materials were identified and characterised using X-Ray Diffraction (XRD) and Scanning Electron Microscopy (SEM). Finally, numerical modelling provided insights into the initial prehistoric failure mode, and the calculated factors of safety (FoS) suggest current stability at the site. It is clear from the results of this study that in order to determine the process-mechanics of prehistoric landslides in these types of clay-dominated terrains, a combination of techniques should be used. Such coastal landslides, on the Waikato and Bay of Plenty coasts, do not have existing toe deposits, due to Holocene transgression, but they do exhibit distinct features that allow failure mechanics to be determined and reconstructed. In particular, the method of UAV-SfM remote sensing coupled with electrical resistivity tomography described in this thesis represents a rapid method of terrain evaluation and landslide hazard assessment, which can be undertaken prior to more costly borehole investigations.