Towards empirically validated models of soft-rock landslides' occurrence, activity, and sediment delivery : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Earth Science at Massey University, Manawatū, New Zealand

Thumbnail Image
Open Access Location
Journal Title
Journal ISSN
Volume Title
Massey University
The Author
Within New Zealand, soft-rock landslides present a severe hazard to infrastructure and contribute to the degradation of river systems by delivering large amounts of sediment to waterways. Updates to New Zealand’s national policy statement for freshwater management necessitate accurate accounting of freshwater sediment sources, but current sediment budget models do not account for the sediment inputs from soft-rock, and other large slow-moving landslides. To understand which factors lead to the occurrence and continued activity of these landslides and the role they play in New Zealand’s river sediment dynamics, I have completed the following objectives. (i) I have mapped large landslides within the Whanganui-Rangitikei soft-rock hill country in the North Island of New Zealand and conducted a geostatistical analysis to determine which factors control their occurrence. (ii) I have developed a novel remote sensing framework for monitoring large, slow-moving landslides that is based upon time-series Interferometric Synthetic Aperture Radar (InSAR) and time-series sub-Pixel Offset Tracking (sPOT) analyses. Furthermore, I have shown that this framework can identify large landslide activity with an accuracy of 91% and measure the movement of landslides moving with an average velocity of 2.05 m/yr with a mean absolute error of 0.74 m/yr. (iii) I have applied this framework to the landslides of the Whanganui-Rangitikei soft-rock hill country and used its results to perform a geostatistical analysis to determine which factors control a landslide’s current activity state and to estimate the total sediment mass delivered by soft-rock landslides to the rivers of this region. In total, I mapped 1057 large landslides in this region and identified 66 of them as currently active. I find that low slopes, river incision, alignment between bedding planes and slopes, and forest cover are predictive of landslide occurrence, but that low slopes and high annual precipitation rates best predict the current activity states of these landslides. I also find that soft-rock landslides contribute a 10±2% of the total sediment mass delivered to the river systems of this region. Overall, this thesis advances our understanding of why soft-rock landslides occur and provides a framework that will allow future studies to monitor these landslides at region to country-wide scales.
Figure 2.2 (=Crozier & Glade, 2005 Fig 2.1) was removed for copyright reasons. Figure 2.3 is re-used under a Creative Commons Attribution 4.0 International (CC BY 4.0) license.
Landslides, Soil erosion, Landslide hazard analysis, Landslides, Remote sensing, New Zealand, Manawatu-Wanganui