Dynamic Landslide Susceptibility Mapping Using InSAR and Machine Learning Fusion
DOI:
https://doi.org/10.64862/ajeg.2025.2sp.120.274Keywords:
Landslide susceptibility, InSAR, Machine learning, Threshold method, Potential landslide areasAbstract
Reservoir-induced landslides pose serious hazards in mountainous regions. Landslide susceptibility assessment (LSA) and predicting potential landslide areas (PLA) in reservoir regions play a vital role in disaster prevention and mitigation. This study proposes an integrated framework combining ascending and descending orbit InSAR and machine learning (ML) to generate dynamic LSA. we fuse the susceptibility map with InSAR-derived deformation using threshold intersection method to delineate potential landslide zones. Results show that the Random forest (RF) model with InSAR-sampled negatives achieves the performance (AUC = 0.814). The threshold intersection method predicted a PLA of 7.91 km² (including 44 known cataloged areas). This integrated approach provides a robust methodology for LSA in reservoir-affected areas.
References
Agliardi, F., Scuderi, M. M., Fusi, N., and Collettini, C. (2020). Slow-to-fast transition of giant creeping rockslides modulated by undrained loading in basal shear zones. Nature Communications, 11(1), 1–11. https://doi.org/10.1038/s41467-020-15093-3
Dou, J., Yunus, A. P., Bui, D. T., Merghadi, A., Sahana, M., Zhu, Z., Chen, C. W., Han, Z., and Pham, B. T. (2020). Improved landslide assessment using support vector machine with bagging, boosting, and stacking ensemble machine learning framework in a mountainous watershed, Japan. Landslides, 17, 641–658. https://doi.org/10.1007/s10346-019-01286-5
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Nepal Society of Engineering Geology (NSEG)
This work is licensed under a Creative Commons Attribution 4.0 International License.
