How DEM Resolution and Sample Selection Variability Influence Deep Learning Susceptibility Models for Debris Flows
DOI:
https://doi.org/10.64862/ajeg.2025.2sp.137.299Keywords:
Debris flow, Deep Learning, Sampling StrategiesAbstract
Debris-flow susceptibility mapping requires precise topographic representation and balanced sampling strategies to ensure model robustness and generalizability. This study thoroughly quantifies the influence of digital elevation model (DEM) resolution (6.5, 12.5, 30, and 90 m) and sampling-strategy uncertainty on the predictive performance of advanced deep-learning (DL) architectures, including one and two-dimensional convolutional neural networks (CNN1D, CNN2D), recurrent neural networks (RNN), and long short-term memory (LSTM) models. A field-verified debris-flow inventory comprising 108 catchments and thirteen conditioning factors derived from multi-resolution DEMs and remote-sensing datasets was used for model construction. To evaluate sampling uncertainty, hundred symmetrical iterations of debris-flow and non-debris-flow samples were executed, resulting in a 6.7-10.5 % increase in mean accuracy with optimized sample selection. Factor importance derived from Random Forest and Frequency Ratio analyses identified rainfall as the dominant control on debris-flow occurrence. Model performance, assessed through multiple statistical metrics, revealed that the LSTM consistently outperformed other architectures, achieving a maximum accuracy of 0.929 and AUC of 0.973 at 12.5 m resolution. The proposed framework provides a reproducible and scalable approach for multi-resolution DFSM and quantification of sampling-related uncertainty in complex mountainous terrain.
References
Cama, M., Conoscenti, C., Lombardo, L., and Rotigliano, E. (2016). Exploring relationships between grid cell size and accuracy for debris-flow susceptibility models: a test in the Giampilieri catchment (Sicily, Italy). Environmental Earth Sciences, 75, 1–21. Available at https://link.springer.com/article/10.1007/s12665-015-5047-6
Cavazzi, S., Corstanje, R., Mayr, T., Hannam, J., and Fealy, R. (2013). Are fine resolution digital elevation models always the best choice in digital soil mapping? Geoderma, 195, 111–121. https://doi.org/10.1016/j.geoderma.2012.11.020
Daud, H., Dou, J., Khan, N. G., Xu, B., Dong, S., Dong, A., and Ma, H. (2025). Tree-Based Machine Learning and Flow Simulation for Debris Flow Susceptibility, Runout Propagation, and Dynamics in the Higher Himalayas. Mathematical Geosciences, 1–39. Available at https://link.springer.com/article/10.1007/s11004-025-10196-3
Daud, H., Tanoli, J. I., Asif, S. M., Qasim, M., Ali, M., Khan, J., Bhatti, Z. I., and Jadoon, I. A. K. (2024). Modelling of debris-flow susceptibility and propagation: a case study from Northwest Himalaya. Journal of Mountain Science, 21(1), 200–217. Available at https://link.springer.com/article/10.1007/s11629-023-7966-0
Dong, A., Dou, J., Li, C., Chen, Z., Ji, J., Xing, K., Zhang, J., and Daud, H. (2024). Accelerating Cross-scene Co-seismic Landslide Detection through Progressive Transfer Learning and Lightweight Deep Learning Strategies. IEEE Transactions on Geoscience and Remote Sensing. https://doi.org/10.1109/TGRS.2024.3424680
Huang, F., Teng, Z., Guo, Z., Catani, F., and Huang, J. (2023). Uncertainties of landslide susceptibility prediction: Influences of different spatial resolutions, machine learning models and proportions of training and testing dataset. Rock Mechanics Bulletin, 2(1), 100028. https://doi.org/10.1016/j.rockmb.2023.100028
Huang, F., Xiong, H., Jiang, S.-H., Yao, C., Fan, X., Catani, F., Chang, Z., Zhou, X., Huang, J., and Liu, K. (2024). Modelling landslide susceptibility prediction: A review and construction of semi-supervised imbalanced theory. Earth-Science Reviews, 250, 104700. https://doi.org/10.1016/j.earscirev.2024.104700
Khalid, M., Rahman, M. Z. bin A., Syed, J. H., Ali, N., Daud, H., Hussain, M. A., Ruslan, M. S., and Fauzi, O. F. (2025). Debris flow susceptibility and propagation modeling: a deep learning and flow-R framework. Bulletin of Engineering Geology and the Environment, 84(10), 455. https://doi.org/10.1007/s10064-025-04496-5
Ngo, P. T. T., Panahi, M., Khosravi, K., Ghorbanzadeh, O., Kariminejad, N., Cerda, A., and Lee, S. (2021). Evaluation of deep learning algorithms for national scale landslide susceptibility mapping of Iran. Geoscience Frontiers, 12(2), 505–519. https://doi.org/10.1016/j.gsf.2020.06.013
Shirzadi, A., Solaimani, K., Roshan, M. H., Kavian, A., Chapi, K., Shahabi, H., Keesstra, S., Ahmad, B. Bin, and Bui, D. T. (2019). Uncertainties of prediction accuracy in shallow landslide modeling: Sample size and raster resolution. Catena, 178, 172–188. https://doi.org/10.1016/j.catena.2019.03.017
Wei, R., Zeng, Q., Davies, T., Yuan, G., Wang, K., Xue, X., and Yin, Q. (2018). Geohazard cascade and mechanism of large debris flows in Tianmo gully, SE Tibetan Plateau and implications to hazard monitoring. Engineering Geology, 233, 172–182. Available at https://www.sciencedirect.com/science/article/abs/pii/S0013795217308438
Xu, Q., Zhang, S., Li, W. L., and Van Asch, T. W. J. (2012). The 13 August 2010 catastrophic debris flows after the 2008 Wenchuan earthquake, China. Natural Hazards and Earth System Sciences, 12(1), 201–216. https://doi.org/10.5194/nhess-12-201-2012
Zhang, R., Zhang, L., Fang, Z., Oguchi, T., Merghadi, A., Fu, Z., Dong, A., and Dou, J. (2024). Interferometric synthetic aperture radar (InSAR)-based absence sampling for machine-learning-based landslide susceptibility mapping: The Three Gorges Reservoir Area, China. Remote Sensing, 16(13), 2394. https://doi.org/10.3390/rs16132394
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Nepal Society of Engineering Geology
This work is licensed under a Creative Commons Attribution 4.0 International License.
