Articles | Volume 25, issue 7
https://doi.org/10.5194/nhess-25-2371-2025
© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/nhess-25-2371-2025
© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Brief communication
| 
15 Jul 2025
Brief communication |
| 15 Jul 2025
| 15 Jul 2025
Brief communication: AI-driven rapid landslide mapping following the 2024 Hualien earthquake in Taiwan
Department of Earth Sciences, University of Cambridge, Cambridge, UK
Department of Geography, University of Cambridge, Cambridge, UK
Machine Intelligence and Slope Stability Laboratory, Department of Geosciences, University of Padova, 35129 Padua, Italy
Alessandro Novellino
British Geological Survey, Nicker Hill, Keyworth, NG12 5GG, UK
Chengyong Fang
State Key Laboratory of Geohazard Prevention and Geoenvironment Protection, Chengdu University of Technology, 610059, Chengdu, China
Kushanav Bhuyan
State Key Laboratory of Geohazard Prevention and Geoenvironment Protection, Chengdu University of Technology, 610059, Chengdu, China
Machine Intelligence and Slope Stability Laboratory, Department of Geosciences, University of Padova, 35129 Padua, Italy
Kathryn Leeming
British Geological Survey, Nicker Hill, Keyworth, NG12 5GG, UK
Itahisa Gonzalez Alvarez
British Geological Survey, Nicker Hill, Keyworth, NG12 5GG, UK
Claire Dashwood
British Geological Survey, Nicker Hill, Keyworth, NG12 5GG, UK
Sophie Doward
British Geological Survey, Nicker Hill, Keyworth, NG12 5GG, UK
Rahul Chahel
British Geological Survey, Nicker Hill, Keyworth, NG12 5GG, UK
Emma McAllister
British Geological Survey, Nicker Hill, Keyworth, NG12 5GG, UK
Sansar Raj Meena
Machine Intelligence and Slope Stability Laboratory, Department of Geosciences, University of Padova, 35129 Padua, Italy
Filippo Catani
Machine Intelligence and Slope Stability Laboratory, Department of Geosciences, University of Padova, 35129 Padua, Italy
Related authors
Lorenzo Nava, Maximilian Van Wyk de Vries, and Louie Elliot Bell
EGUsphere, https://doi.org/10.5194/egusphere-2025-2795, https://doi.org/10.5194/egusphere-2025-2795, 2025
This preprint is open for discussion and under review for Natural Hazards and Earth System Sciences (NHESS).
Short summary
Short summary
We introduce TerraTrack, an open-source tool for detecting and monitoring slow-moving landslides using Sentinel-2 data. It automates image acquisition, landslide identification, and time-series generation in an accessible and cloud-based workflow. TerraTrack supports early warning, complements InSAR, and offers a scalable solution for landslide hazard identification and monitoring.
Lorenzo Nava, Alessandro Mondini, Kushanav Bhuyan, Chengyong Fang, Oriol Monserrat, Alessandro Novellino, and Filippo Catani
Geosci. Model Dev. Discuss., https://doi.org/10.5194/gmd-2024-230, https://doi.org/10.5194/gmd-2024-230, 2025
Preprint under review for GMD
Short summary
Short summary
This paper presents a new framework for landslide detection using radar and deep learning, informed by data from 73000 landslides across diverse regions in the world. The method showed high accuracy and rapid response potential regardless of weather and illumination conditions. By overcoming the limits of optical satellite imagery, it offers a powerful tool for global landslide detection, benefiting disaster management and advancing methods for monitoring hazardous terrains.
Chengyong Fang, Xuanmei Fan, Xin Wang, Lorenzo Nava, Hao Zhong, Xiujun Dong, Jixiao Qi, and Filippo Catani
Earth Syst. Sci. Data, 16, 4817–4842, https://doi.org/10.5194/essd-16-4817-2024, https://doi.org/10.5194/essd-16-4817-2024, 2024
Short summary
Short summary
In this study, we present the largest publicly available landslide dataset, Globally Distributed Coseismic Landslide Dataset (GDCLD), which includes multi-sensor high-resolution images from various locations around the world. We test GDCLD with seven advanced algorithms and show that it is effective in achieving reliable landslide mapping across different triggers and environments, with great potential in enhancing emergency response and disaster management.
Sansar Raj Meena, Lorenzo Nava, Kushanav Bhuyan, Silvia Puliero, Lucas Pedrosa Soares, Helen Cristina Dias, Mario Floris, and Filippo Catani
Earth Syst. Sci. Data, 15, 3283–3298, https://doi.org/10.5194/essd-15-3283-2023, https://doi.org/10.5194/essd-15-3283-2023, 2023
Short summary
Short summary
Landslides occur often across the world, with the potential to cause significant damage. Although a substantial amount of research has been conducted on the mapping of landslides using remote-sensing data, gaps and uncertainties remain when developing models to be operational at the global scale. To address this issue, we present the High-Resolution Global landslide Detector Database (HR-GLDD) for landslide mapping with landslide instances from 10 different physiographical regions globally.
Lorenzo Nava, Maximilian Van Wyk de Vries, and Louie Elliot Bell
EGUsphere, https://doi.org/10.5194/egusphere-2025-2795, https://doi.org/10.5194/egusphere-2025-2795, 2025
This preprint is open for discussion and under review for Natural Hazards and Earth System Sciences (NHESS).
Short summary
Short summary
We introduce TerraTrack, an open-source tool for detecting and monitoring slow-moving landslides using Sentinel-2 data. It automates image acquisition, landslide identification, and time-series generation in an accessible and cloud-based workflow. TerraTrack supports early warning, complements InSAR, and offers a scalable solution for landslide hazard identification and monitoring.
Lorenzo Nava, Alessandro Mondini, Kushanav Bhuyan, Chengyong Fang, Oriol Monserrat, Alessandro Novellino, and Filippo Catani
Geosci. Model Dev. Discuss., https://doi.org/10.5194/gmd-2024-230, https://doi.org/10.5194/gmd-2024-230, 2025
Preprint under review for GMD
Short summary
Short summary
This paper presents a new framework for landslide detection using radar and deep learning, informed by data from 73000 landslides across diverse regions in the world. The method showed high accuracy and rapid response potential regardless of weather and illumination conditions. By overcoming the limits of optical satellite imagery, it offers a powerful tool for global landslide detection, benefiting disaster management and advancing methods for monitoring hazardous terrains.
Alejandro Builes-Jaramillo, Annie E. G. Winson, Emma Bee, Nancy Quirós, Dairo Urán, James Rúa, Luis Alejandro Rivera-Flórez, Camilo Restrepo-Estrada, Ingry Natalia Gómez-Miranda, Claire Dashwood, and João Porto de Albuquerque
Nat. Hazards Earth Syst. Sci. Discuss., https://doi.org/10.5194/nhess-2024-221, https://doi.org/10.5194/nhess-2024-221, 2025
Preprint under review for NHESS
Short summary
Short summary
This study addresses data gaps in urban neighborhoods often excluded from official hazard maps by co-developing a vulnerability assessment framework with community researchers in El Pacífico, Medellín. Through participatory methods, we created detailed building-scale hazard maps, improving community planning and advocacy for public interventions. This approach highlights how local knowledge can enhance risk assessments and support disaster risk reduction in marginalized areas.
Chengyong Fang, Xuanmei Fan, Xin Wang, Lorenzo Nava, Hao Zhong, Xiujun Dong, Jixiao Qi, and Filippo Catani
Earth Syst. Sci. Data, 16, 4817–4842, https://doi.org/10.5194/essd-16-4817-2024, https://doi.org/10.5194/essd-16-4817-2024, 2024
Short summary
Short summary
In this study, we present the largest publicly available landslide dataset, Globally Distributed Coseismic Landslide Dataset (GDCLD), which includes multi-sensor high-resolution images from various locations around the world. We test GDCLD with seven advanced algorithms and show that it is effective in achieving reliable landslide mapping across different triggers and environments, with great potential in enhancing emergency response and disaster management.
Kathryn A. Leeming, John P. Bloomfield, Gemma Coxon, and Yanchen Zheng
Hydrol. Earth Syst. Sci. Discuss., https://doi.org/10.5194/hess-2023-202, https://doi.org/10.5194/hess-2023-202, 2023
Preprint withdrawn
Short summary
Short summary
In this work we characterise annual patterns in baseflow, the component of streamflow that comes from subsurface storage. Our research identified early-, mid-, and late-seasonality of baseflow across catchments in Great Britain over two time blocks: 1976–1995 and 1996–2015, and found that many catchments have earlier seasonal patterns of baseflow in the second time period. These changes are linked to changes in climate signals: snow-melt in highland catchments and effective rainfall changes.
Sansar Raj Meena, Lorenzo Nava, Kushanav Bhuyan, Silvia Puliero, Lucas Pedrosa Soares, Helen Cristina Dias, Mario Floris, and Filippo Catani
Earth Syst. Sci. Data, 15, 3283–3298, https://doi.org/10.5194/essd-15-3283-2023, https://doi.org/10.5194/essd-15-3283-2023, 2023
Short summary
Short summary
Landslides occur often across the world, with the potential to cause significant damage. Although a substantial amount of research has been conducted on the mapping of landslides using remote-sensing data, gaps and uncertainties remain when developing models to be operational at the global scale. To address this issue, we present the High-Resolution Global landslide Detector Database (HR-GLDD) for landslide mapping with landslide instances from 10 different physiographical regions globally.
Sansar Raj Meena, Silvia Puliero, Kushanav Bhuyan, Mario Floris, and Filippo Catani
Nat. Hazards Earth Syst. Sci., 22, 1395–1417, https://doi.org/10.5194/nhess-22-1395-2022, https://doi.org/10.5194/nhess-22-1395-2022, 2022
Short summary
Short summary
The study investigated the importance of the conditioning factors in predicting landslide occurrences using the mentioned models. In this paper, we evaluated the importance of the conditioning factors (features) in the overall prediction capabilities of the statistical and machine learning algorithms.
Sansar Raj Meena, Florian Albrecht, Daniel Hölbling, Omid Ghorbanzadeh, and Thomas Blaschke
Nat. Hazards Earth Syst. Sci., 21, 301–316, https://doi.org/10.5194/nhess-21-301-2021, https://doi.org/10.5194/nhess-21-301-2021, 2021
Short summary
Short summary
Comprehensive and sustainable landslide management, including identification of landslide-susceptible areas, requires a lot of organisations and people to collaborate efficiently. In this study, we propose a concept for a system that provides users with a platform to share the location of landslide events for further collaboration in Nepal. The system can be beneficial for specifying potentially risky regions and consequently, the development of risk mitigation strategies at the local level.
Cited articles
Amatya, P., Kirschbaum, D. B., Stanley, T., and Tanyaş, H.: Landslide mapping using object-based image analysis and open-source tools, Eng. Geol., 282, 106000, https://doi.org/10.1016/j.enggeo.2021.106000, 2021. a
Amatya, P., Scheip, C., Déprez, A., Malet, J.-P., Slaughter, S. L., Handwerger, A. L., Emberson, R., Kirschbaum, D. B., Jean-Baptiste, J., Huang, M.-H., Clark, M. K., Zekkos, D., Huang, J.-R., Pacini, F., and Boissier, E.: Learnings from rapid response efforts to remotely detect landslides triggered by the August 2021 Nippes earthquake and Tropical Storm Grace in Haiti, Nat. Hazards, 118, 2337–2375, https://doi.org/10.1007/s11069-023-06096-6, 2023. a, b, c
Bhuyan, K., Tanyaş, H., Nava, L., Puliero, S., Meena, S. R., Floris, M., Van Westen, C. J., and Catani, F.: Generating multi-temporal landslide inventories through a general deep transfer learning strategy using HR EO data, Sci. Rep., 13, 162, https://doi.org/10.1038/s41598-022-27352-y, 2023. a, b, c, d
Catani, F.: Landslide detection by deep learning of non-nadiral and crowdsourced optical images, Landslides, 18, https://doi.org/10.1007/s10346-020-01513-4, 2021. a
Chang, J. M., Chao, W. A., Yang, C. M., and Huang, M. W.: Coseismic and subsequent landslides of the 2024 Hualien earthquake (M7.2) on April 3 in Taiwan, Landslides, 21, 2591–2595, https://doi.org/10.1007/s10346-024-02312-x, 2024.
Chen, Y., Song, C., Li, Z., Chen, B., Chen, Y.-C., Hu, J.-C., Cai, X., Zhu, S., Wang, Q., Ma, Y., Gu, C., Li, S., and Peng, J.: Preliminary analysis of landslides induced by the 3 April 2024 Mw 7.4 Hualien, Taiwan earthquake, Landslides, 22, https://doi.org/10.1007/s10346-025-02465-3, 2025. a, b
Chuang, R. Y., Wu, B. S., Liu, H. C., Huang, H. H., and Lu, C. H.: Development of a statistics-based nowcasting model for earthquake-triggered landslides in Taiwan, Eng. Geol., 289, 106177, https://doi.org/10.1016/j.enggeo.2021.106177, 2021. a
Dai, L., Fan, X., Wang, X., Fang, C., Zou, C., Tang, X., Wei, Z., Xia, M., Wang, D., and Xu, Q.: Coseismic landslides triggered by the 2022 Luding Ms 6.8 earthquake, China, Landslides, 20, 1277–1292, https://doi.org/10.1007/s10346-023-02061-3, 2023. a
Fan, X., Juang, C. H., Wasowski, J., Huang, R., Xu, Q., Scaringi, G., and Havenith, H. B.: What we have learned from the 2008 Wenchuan Earthquake and its aftermath: A decade of research and challenges, Eng. Geol., 241, 25–32, https://doi.org/10.1016/j.enggeo.2018.05.004, 2018. a
Fang, C., Fan, X., Zhong, H., Lombardo, L., Tanyaş, H., and Wang, X.: A novel historical landslide detection approach based on LiDAR and lightweight attention U-Net, Remote Sens., 14, 4357, https://doi.org/10.3390/rs14174357, 2022. a
Fang, C., Fan, X., Wang, X., Nava, L., Zhong, H., Dong, X., Qi, J., and Catani, F.: A globally distributed dataset of coseismic landslide mapping via multi-source high-resolution remote sensing images, Earth Syst. Sci. Data, 16, 4817–4842, https://doi.org/10.5194/essd-16-4817-2024, 2024a. a, b, c, d
Fang, C., Fan, X., and Wang, X.: GDCLD, Zenodo [data set], https://doi.org/10.5281/zenodo.11369484, 2024b. a
Hung, J. J.: Chi-Chi earthquake induced landslides in Taiwan, Earthquake Eng. Eng. Seismol., 2, 25–33, 2000. a
Lee, C. T., and Fei, L. Y.: Nationwide landslide hazard analysis and mapping in Taiwan, Eng. Geol. Soc. Territ. 2: Landslide Processes, 2, 971–974, 2015. a
Lombardo, L., Opitz, T., Ardizzone, F., Guzzetti, F., and Huser, R.: Space–time landslide predictive modelling, Earth-Sci. Rev., 209, 1–32, https://doi.org/10.1016/j.earscirev.2020.103318, 2020. a
Meena, S. R., Nava, L., Bhuyan, K., Puliero, S., Soares, L. P., Dias, H. C., Floris, M., and Catani, F.: HR-GLDD: a globally distributed dataset using generalized deep learning (DL) for rapid landslide mapping on high-resolution (HR) satellite imagery, Earth Syst. Sci. Data, 15, 3283–3298, https://doi.org/10.5194/essd-15-3283-2023, 2023. a
Nava, L., Bhuyan, K., Meena, S. R., Monserrat, O., and Catani, F.: Rapid mapping of landslides on SAR data by attention U-Net, Remote Sens., 14, 1449, https://doi.org/10.3390/rs14061449, 2022. a
Nava, L., Mondini, A., Bhuyan, K., Fang, C., Monserrat, O., Novellino, A., and Catani, F.: Sentinel-1 SAR-based Globally Distributed Co-Seismic Landslide Detection by Deep Neural Networks, Geosci. Model Dev. Discuss. [preprint], https://doi.org/10.5194/gmd-2024-230, in review, 2025. a
Nava, L., Mondini, A. C., Fang, C., Bhuyan, K., Monserrat, O., Novellino, A., and Catani, F.: SAR-LRA: A Synthetic Aperture Radar-Based Landslide Rapid Assessment Tool, Zenodo, https://doi.org/10.5281/zenodo.14898556, 2025a. a
Nava, L., Mondini, A. C., Fang, C., Bhuyan, K., Monserrat, O., Novellino, A., and Catani, F.: SAR-LRA: A Synthetic Aperture Radar-Based Landslide Rapid Assessment Tool, Zenodo [data set], https://doi.org/10.5281/zenodo.15159492, 2025b. a
Novellino, A., Pennington, C., Leeming, K., Taylor, S., Alvarez, I. G., McAllister, E., Arnhardt, C., and Winson, A.: Mapping landslides from space: A review, Landslides, 21, 1041–1052, https://doi.org/10.1007/s10346-024-02215-x, 2024a. a, b
Novellino, A., Nava, L., Fang, C., Leeming, K., bhuyan, K., Gonzalez Alvarez, I., Dashwood, C., Doward, S., Chahel, R., McAllister, E., Meena, S. R., Fan, X., Tang, X., and Catani, F.: AI-driven rapid landslides mapping following the 2024 Hualien City Earthquake in Taiwan (1.0), Zenodo [data set], https://doi.org/10.5281/zenodo.11519683, 2024b. a
Novellino, A., Nava, L., Fang, C., Leeming, K., bhuyan, K., Gonzalez Alvarez, I., Dashwood, C., Doward, S., Chahel, R., McAllister, E., Meena, S. R., Fan, X., Tang, X., and Catani, F.: AI-driven rapid landslides mapping following the 2024 Hualien City Earthquake in Taiwan, Zenodo, https://doi.org/10.5281/zenodo.15027165, 2025. a
Pennington, C. V., Bossu, R., Ofli, F., Imran, M., Qazi, U., Roch, J., and Banks, V. J.: A near-real-time global landslide incident reporting tool demonstrator using social media and artificial intelligence, Int. J. Disast. Risk Re., 77, 103089, https://doi.org/10.1016/j.ijdrr.2022.103089, 2022. a
Prakash, N., Manconi, A., and Loew, S.: A new strategy to map landslides with a generalized convolutional neural network, Sci. Rep., 11, 9722, https://doi.org/10.1038/s41598-021-89015-8, 2021. a
Roback, K., Clark, M. K., West, A. J., Zekkos, D., Li, G., Gallen, S. F., Chamlagain, D., and Godt, J. W.: The size, distribution, and mobility of landslides caused by the 2015 Mw7.8 Gorkha earthquake, Nepal, Geomorphology, 301, 121–138, https://doi.org/10.1016/j.geomorph.2017.01.030, 2018. a
Shou, K.-J., and Chen, J.: On the rainfall-induced deep-seated and shallow landslide hazard in Taiwan, Eng. Geol., 288, 106156, https://doi.org/10.1016/j.enggeo.2021.106156, 2021. a
Tang, X., Tu, Z., Wang, Y., Liu, M., Li, D., and Fan, X.: Automatic detection of coseismic landslides using a new transformer method, Remote Sens., 14, 2884, https://doi.org/10.3390/rs14122884, 2022. a
Tanyaş, H., Görüm, T., Fadel, I., Yıldırım, C., and Lombardo, L.: An open dataset for landslides triggered by the 2016 Mw 7.8 Kaikōura earthquake, New Zealand, Landslides, 19, 1405–1420, https://doi.org/10.1007/s10346-022-01869-9, 2022. a
Xu, Y., Ouyang, C., Xu, Q., Wang, D., Zhao, B., and Luo, Y.: CAS Landslide Dataset: A large-scale and multisensor dataset for deep learning-based landslide detection, Sci. Data, 11, 12, https://doi.org/10.1038/s41597-023-02847-z, 2024. a
Short summary
On 2 April 2024, a Mw 7.4 earthquake hit Taiwan's eastern coast, causing extensive landslides and damage. We used automated methods combining Earth observation (EO) data with AI to quickly inventory the landslides. This approach identified 7090 landslides over 75 km2 within 3 h of acquiring the EO imagery. The study highlights AI's role in improving landslide detection efforts in disaster response.
On 2 April 2024, a Mw 7.4 earthquake hit Taiwan's eastern coast, causing extensive landslides...
Altmetrics
Final-revised paper
Preprint