Ph.D. Thesis Defense: Kamal Rana
Ph.D. Thesis Defense
The Geometry and Topology of Landslides
Kamal Rana
Imaging Science PhD Candidate
Chester F. Carlson Center for Imaging Science, RIT
Abstract: Landslides are often catastrophic, causing loss of life and destruction of property and infrastructure. Landslide susceptibility and hazard modeling help mitigate these losses by locating regions prone to landslides and providing probabilistic forecasting of landslide occurrence. However, these landslide susceptibility and hazard models' efficacy depends on the quality of existing databases that often lack crucial information, like the underlying trigger and failure mechanism of a landslide. In this Ph.D. project, we developed techniques to identify landslide triggering and failure mechanisms using the geometric and topological properties of the debris field, improving the efficacy of existing databases. Our methods employ several mathematical and machine-learning-based state-of-the-art approaches such as Topological Data Analysis (TDA), Convolutional Neural Networks (CNN), and decision tree-based algorithms. For identifying the triggering mechanism, our methods achieve above 80% classification performance in the landslide databases spread over the Japanese archipelago. Similarly, our techniques for determining the failure mechanisms of landslides achieve over 90% per class performance in the datasets from the US's Pacific Northwest and Italy. As part of this Ph.D., we also developed and made publicly available a python library named Landsifier, which provides functionalities for estimating likely triggers of mapped landslides. Our methods and python library can potentially mitigate the impacts of landslides by enhancing the performance of the predictive models in accurately locating landslide-prone regions and predicting landslides.
Intended Audience: Undergraduates, graduates, and experts. Those with interest in the topic.
To request an interpreter, please visit https://myaccess.rit.edu
Event Snapshot
When and Where
Who
Open to the Public
Interpreter Requested?
No