Development of Machine Learning for Debris Flow Event Prediction in a Volcanic Area
Downloads
The integration of machine learning (ML) into debris flow prediction in volcanic areas, exemplified by the Gendol River watershed of Mount Merapi, offers transformative potential for hazard mitigation. This study aimed to develop real-time, computationally efficient ML models capable of integrating multi-source data, rainfall intensity of 25 mm/hour linked to 300 cm Debris Flow heights, antecedent precipitation, and geomorphological variables to predict debris flows with actionable lead times. Key objectives included optimizing prediction accuracy, minimizing the false positive rate to 18.2% for "Debris Flow" events, and enhancing model interpretability for deployment in data-scarce volcanic regions. Results demonstrated that ensemble methods and deep learning architecture outperformed traditional models, with Efficient Logistic Regression and Linear SVM achieving an accuracy of 82.35%, and Cosine KNN attaining a prediction speed of 272 observations per second. Critical predictors included temporal rainfall patterns (contributing more than 50% to flow initiation) and ash deposit thickness (with a 70% influence on decision-making). However, challenges persisted: imbalanced datasets of nine training instances for "Debris Flow" events led to misclassification rates of 100% for hybrid events like "Rainfall and Debris Flow," while models like Naive Bayes exhibited instability (accuracy dropping to 50%). Research gaps highlighted data scarcity for high-magnitude events, limited geographic transferability, and the absence of standardized evaluation metrics. Technical limitations included reliance on low-resolution remote sensing data, high computational costs for ensemble models requiring 10 operational cost units, and the opacity of neural networks, which hindered stakeholder trust. Despite these constraints, ML models achieved 85% accuracy in non-event recognition and 76.47% precision in Bagged Trees, offering scalable frameworks for early warning systems. The study highlights the importance of enriched datasets, adaptive algorithms, and interdisciplinary collaboration in transforming volcanic risk management from a reactive approach, ultimately safeguarding vulnerable communities through data-driven, life-saving predictions.
Downloads
[1] Pastor, M., Tayyebi, S. M., Stickle, M. M., Yagüe, Á., Molinos, M., Navas, P., & Manzanal, D. (2021). A depth integrated, coupled, two-phase model for debris flow propagation. Acta Geotechnica, 16(8), 2409–2433. doi:10.1007/s11440-020-01114-4.
[2] Rizova, R., & Nikolova, V. (2021). Geomorphological and Sedimentological Characteristics of Debris Flows in the River Buyukdere Watershed (Eastern Rhodopes, Bulgaria). International Multidisciplinary Scientific GeoConference Surveying Geology and Mining Ecology Management, SGEM, 21(1.1), 43–50. doi:10.5593/sgem2021/1.1/s01.007.
[3] Iverson, R. M. (2013). Mechanics of debris flows and rock avalanches. Handbook of Environmental Fluid Dynamics, 1, 573-587, CRC Press, Boca Raton, United States.
[4] Pasculli, A., Zito, C., Sciarra, N., & Mangifesta, M. (2024). Back Analysis of a Real Debris Flow, the Morino-Rendinara Test Case (Italy), Using RAMMS Software. Land, 13(12), 2078. doi:10.3390/land13122078.
[5] Tian, S., Hu, G., Chen, N., Rahman, M., Han, Z., Somos-Valenzuela, M., & Maurice Habumugisha, J. (2022). Extreme climate and tectonic controls on the generation of a large-scale, low-frequency debris flow. CATENA, 212, 106086. doi:10.1016/j.catena.2022.106086.
[6] DeGraff, J.V., Ochiai, H., DeGraff, J., Ochiai, H. (2009). Rainfall, Debris Flows and Wildfires. Landslides – Disaster Risk Reduction. Springer, Berlin, Germany. doi:10.1007/978-3-540-69970-5_24.
[7] Yu, B., Wang, T., & Zhu, Y. (2016). Research on the topographical and rainfall factors of debris flows caused by shallow landslides. Shuikexue Jinzhan/Advances in Water Science, 27(4), 542–550. doi:10.14042/j.cnki.32.1309.2016.04.008.
[8] Schneuwly-Bollschweiler, M., & Stoffel, M. (2012). Hydrometeorological triggers of periglacial debris flows in the Zermatt valley (Switzerland) since 1864. Journal of Geophysical Research: Earth Surface, 117(2), 2011JF002262. doi:10.1029/2011JF002262.
[9] La Porta, G., Leonardi, A., Pirulli, M., Castelli, F., & Lentini, V. (2021). Rainfall-triggered debris flows: Triggering-propagation modelling and application to an event in Southern Italy. IOP Conference Series: Earth and Environmental Science, 833(1), 12106. doi:10.1088/1755-1315/833/1/012106.
[10] Dowling, C. A., & Santi, P. M. (2014). Debris flows and their toll on human life: A global analysis of debris-flow fatalities from 1950 to 2011. Natural Hazards, 71(1), 203–227. doi:10.1007/s11069-013-0907-4.
[11] Hou, J., Dou, M., Zhang, Y., Wang, J., & Li, G. (2021). An evaluation model for landslide and debris flow prediction using multiple hydrometeorological variables. Environmental Earth Sciences, 80(16), 515. doi:10.1007/s12665-021-09840-y.
[12] Li, R., Sun, P., Sang, K., Ke, C., & Zhang, S. (2024). Characteristics and initiation mechanism of the large mudstone Dongping landslide induced by heavy rainfall in Gansu Province, NW China. Geoenvironmental Disasters, 11(1), 39. doi:10.1186/s40677-024-00302-8.
[13] Yang, C. L., Chen, N. S., & Li, Z. L. (2011). Formation mode and mechanism for debris flow induced by Wenchuan earthquake. Journal of Natural Disasters, 20(3), 31–37.
[14] Sujatha, E. R., & Sridhar, V. (2017). Mapping debris flow susceptibility using analytical network process in Kodaikkanal Hills, Tamil Nadu (India). Journal of Earth System Science, 126(8), 116. doi:10.1007/s12040-017-0899-7.
[15] Hu, W., Xu, Q., Wang, G. H., van Asch, T. W. J., & Hicher, P. Y. (2015). Sensitivity of the initiation of debris flow to initial soil moisture. Landslides, 12(6), 1139–1145. doi:10.1007/s10346-014-0529-2.
[16] Moazzam, M. F. U., Vansarochana, A., Boonyanuphap, J., & Choosumrong, S. (2017). Landslide assessment using GIS-based frequency ratio method: a case study of mae-phun sub-district, Laplae district, Uttaradit, Thailand. 23-27 October, 2017, 38th Asian conference on remote sensing, New Delhi, India.
[17] Mohammadi, N., & Sasanpour, F. (2021). Landslide and debris flow risk analysis in Haraz and Lavasanat roads. Water and Soil Management and Modeling, 1(4), 14–29. doi:10.22098/mmws.2021.9138.1023.
[18] Yu, G., Zhang, M., Cong, K., & Pei, L. (2015). Critical rainfall thresholds for debris flows in Sanyanyu, Zhouqu County, Gansu Province, China. Quarterly Journal of Engineering Geology and Hydrogeology, 48(3–4), 224–233. doi:10.1144/qjegh2014-078.
[19] Zhang, S., Guo, X., Cheng, J., & Li, Y. (2025). Rainfall thresholds of debris flows built with assistance of artificial intelligence in a small catchment. Journal of Hydrology, 660, 133440. doi:10.1016/j.jhydrol.2025.133440.
[20] Nikolopoulos, E. I., Borga, M., Creutin, J. D., & Marra, F. (2015). Estimation of debris flow triggering rainfall: Influence of rain gauge density and interpolation methods. Geomorphology, 243, 40–50. doi:10.1016/j.geomorph.2015.04.028.
[21] Huang, L. (2023). On the assessment of debris flow hazard based on partially ordered set. Journal of Safety and Environment, 23(6), 2009–2016. doi:10.13637/j.issn.1009-6094.2022.0139.
[22] Mei, Y., Hong, F., Jia, Z., & Kang, Z. (2018). Debris Flow Forecasting of Northwest of Yunnan Province Based on LR, SVM, and RF Statistical Models. 2018 26th International Conference on Geoinformatics, 1–7. doi:10.1109/geoinformatics.2018.8557115.
[23] Li, Y., Shen, J., Huang, M., & Peng, Z. (2023). Debris Flow Classification and Risk Assessment Based on Combination Weighting Method and Cluster Analysis: A Case Study of Debris Flow Clusters in Longmenshan Town, Pengzhou, China. Applied Sciences (Switzerland), 13(13), 7551. doi:10.3390/app13137551.
[24] Du, Y., Liu, H., Li, H., Xie, M., Chicas, S. D., Wu, J., Lv, F., & Wu, Y. (2024). Exploring the initiating mechanism, monitoring equipment and warning indicators of gully-type debris flow for disaster reduction: a review. Natural Hazards, 120(15), 13667–13692. doi:10.1007/s11069-024-06742-7.
[25] Chen, N., Tian, S., Zhang, Y., & Wang, Z. (2021). Soil mass domination in debris-flow disasters and strategy for hazard mitigation. Earth Science Frontiers, 28(4), 337–348. doi:10.13745/j.esf.sf.2020.6.39.
[26] Jiang, Y., Wei, F., Zhang, J., Gu, L., Deng, B., & Liu, H. (2007). Regional prediction of impending debris flow based on Doppler weather radar. Wuhan University Journal of Natural Sciences, 12(4), 627–632. doi:10.1007/s11859-006-0325-5.
[27] Zhao, Y., Li, Y., Zheng, J., Wang, Y., Meng, X., Yue, D., Guo, F., Chen, G., Qi, T., & Zhang, Y. (2025). A new rainfall Intensity−Duration threshold curve for debris flows using comprehensive rainfall intensity. Engineering Geology, 347. doi:10.1016/j.enggeo.2025.107949.
[28] Li, Y., Zhang, J., Jarsve, K. T., Ma, F., Huang, Y., Zhao, W., Zhao, Y., Meng, X., Wang, M., Li, G., Chen, G., Qi, T., Guo, F., & Yue, D. (2025). Debris flow forecasting: disastrous rainfall threshold matters. Landslides, 22(7), 2461–2474. doi:10.1007/s10346-025-02492-0.
[29] Yan, Y., Zhang, Y., Hu, W., Guo, X. Jun, Ma, C., Wang, Z. Ang, & Zhang, Q. (2020). A multiobjective evolutionary optimization method based critical rainfall thresholds for debris flows initiation. Journal of Mountain Science, 17(8), 1860–1873. doi:10.1007/s11629-019-5812-1.
[30] Chang, F. J., Hsu, K., & Chang, L. C. (Eds.). (2019). Flood forecasting using machine learning methods. MDPI Books, Basel, Switzerland. doi:10.3390/books978-3-03897-549-6.
[31] Takahashi, T. (2007). Debris Flow. Taylor & Francis, London, United Kingdom. doi:10.1201/9780203946282.
[32] Duhita, A. D. P., Rahardjo, A. P., & Hairani, A. (2020). Effect of Slope on Infiltration Capacity and Erosion of Mount Merapi Slope Materials. Journal of the Civil Engineering Forum, 1000, 71–84. doi:10.22146/jcef.58350.
[33] Dash, R. K., Gupta, N., Falae, P. O., Pati, R., & Kanungo, D. P. (2024). A comparative evaluation of statistical and machine learning approaches for debris flow susceptibility zonation mapping in the Indian Himalayas. Environment, Development and Sustainability, 1-34. doi:10.1007/s10668-024-05398-4.
[34] Roten, D., Block, J., Crawl, D., Lee, J., & Altintas, I. (2022). Machine Learning for Improved Post-fire Debris Flow Likelihood Prediction. 2022 IEEE International Conference on Big Data (Big Data), 1681–1690. doi:10.1109/BigData55660.2022.10020574.
[35] Wang, X., Tian, M., Qin, Q., & Liang, J. (2023). Hybridization of Machine Learning Algorithms and an Empirical Regression Model for Predicting Debris-Flow-Endangered Areas. Advances in Civil Engineering, 9465811. doi:10.1155/2023/9465811.
[36] Jiang, H., Zou, Q., Zhu, Y., Li, Y., Zhou, B., Zhou, W., Yao, S., Dai, X., Yao, H., & Chen, S. (2024). Deep learning prediction of rainfall-driven debris flows considering the similar critical thresholds within comparable background conditions. Environmental Modelling and Software, 179. doi:10.1016/j.envsoft.2024.106130.
[37] Chmiel, M., Walter, F., Wenner, M., Zhang, Z., McArdell, B. W., & Hibert, C. (2021). Machine Learning Improves Debris Flow Warning. Geophysical Research Letters, 48(3), 2020GL090874. doi:10.1029/2020GL090874.
[38] Zhao, Y., Meng, X., Qi, T., Li, Y., Chen, G., Yue, D., & Qing, F. (2022). AI-based rainfall prediction model for debris flows. Engineering Geology, 296. doi:10.1016/j.enggeo.2021.106456.
[39] Chen, Z., Quan, H., Jin, R., Lin, Z., & Jin, G. (2024). Debris flow susceptibility assessment based on boosting ensemble learning techniques: a case study in the Tumen River basin, China. Stochastic Environmental Research and Risk Assessment, 38(6), 2359–2382. doi:10.1007/s00477-024-02683-6.
[40] Wang, J., Tie, Y., Bai, Y., Gao, Y., Wang, D., & Zhang, M. (2025). Application and prospects of machine learning for rockfalls, landslides and debris flows. Hydrogeology and Engineering Geology, 52(4), 228–244. doi:10.16030/j.cnki.issn.1000-3665.202402011.
[41] McGrane, S. J. (2016). Impacts of urbanisation on hydrological and water quality dynamics, and urban water management: a review. Hydrological Sciences Journal, 61(13), 2295–2311. doi:10.1080/02626667.2015.1128084.
[42] Manfreda, S., McCabe, M. F., Miller, P. E., Lucas, R., Madrigal, V. P., Mallinis, G., Dor, E. Ben, Helman, D., Estes, L., Ciraolo, G., Müllerová, J., Tauro, F., de Lima, M. I., de Lima, J. L. M. P., Maltese, A., Frances, F., Caylor, K., Kohv, M., Perks, M., … Toth, B. (2018). On the use of unmanned aerial systems for environmental monitoring. Remote Sensing, 10(4), 641. doi:10.3390/rs10040641.
[43] Nellyana, R., Listriani, S., Mahfud, M., Rivaldi, A., Chaliluddin, M. A., & Bustami, B. (2023). Law Enforcement in Plastic Waste Management in Aceh Waters: Transboundary Marine Pollution Concern. IOP Conference Series: Earth and Environmental Science, 1221(1), 012074. doi:10.1088/1755-1315/1221/1/012074.
[44] Liu, D., Zhou, J., Sang, X., Tang, D., Zhang, S., & Chen, Q. (2025). Machine learning-based identification of potential debris flow catchments in the Wenchuan earthquake region. Earth Science Informatics, 18(4), 515. doi:10.1007/s12145-025-01990-y.
[45] Chen, Y., Li, N., Xing, F., Xiang, H., & Chen, Z. (2025). Study on debris flow vulnerability of ensemble learning model based on spy technology A case study of upper Minjiang river basin. Scientific Reports, 15(1), 22480. doi:10.1038/s41598-025-03479-6.
[46] Cai, S., Zhang, Z., Yang, X., Lv, Q., Liu, X., Lai, R., Yu, X., & Hu, Y. (2025). The modified theoretical model for debris flows predication with multiple rainfall characteristic parameters. Scientific Reports, 15(1), 12402. doi:10.1038/s41598-024-84199-1.
[47] Lebedeva, E. V., Baldina, E. A., Chernomorets, S. S., Kharchenko, S. V., Kotenkov, A. V., Kuramagomedov, B. M., & Medvedev, A. A. (2025). Slope mass movements and debris flow activity in the Geysernaya River valley (Kamchatka, Russia). Geomorphology, 491. doi:10.1016/j.geomorph.2025.110029.
[48] Guo, S., Gu, N., Zhou, C., Liu, L., Wan, R., Wang, Y., & Liu, Z. (2025). Hydrodynamic characterization and efficient prediction of tapered trash-blocking nets for coastal nuclear intakes: Integration of experiment, simulation, and machine learning. Ocean Engineering, 341. doi:10.1016/j.oceaneng.2025.122442.
[49] De Haas, T., Braat, L., Leuven, J. R. F. W., Lokhorst, I. R., & Kleinhans, M. G. (2015). Effects of debris flow composition on runout, depositional mechanisms, and deposit morphology in laboratory experiments. Journal of Geophysical Research: Earth Surface, 120(9), 1949–1972. doi:10.1002/2015JF003525.
[50] Shen, P., Zhang, L., Wong, H. F., Peng, D., Zhou, S., Zhang, S., & Chen, C. (2020). Debris flow enlargement from entrainment: A case study for comparison of three entrainment models. Engineering Geology, 270, 105581. doi:10.1016/j.enggeo.2020.105581.
[51] Shimokawa, E., & Jitousono, T. (1997). Field survey for debris flow in volcanic area. Recent Developments on Debris Flows, Lecture Notes in Earth Sciences, Springer, Berlin, Germany. doi:10.1007/BFb0117761.
[52] Julien, P. Y., & O'Brien, J. S. (1997). Selected notes on debris flow dynamics. Recent Developments on Debris Flows. Lecture Notes in Earth Sciences. Springer, Berlin, Germany. doi:10.1007/BFb0117766.
[53] Kim, S. (2025). Sluice Gate Operation and Managed Water Levels Improve Predicted Estuarine Lake Water Quality. Civil Engineering Journal, 11(1), 244–278. doi:10.28991/CEJ-2025-011-01-015.
[54] Lavigne, F., Thouret, J. C., Voight, B., Suwa, H., & Sumaryono, A. (2000). Lahars at Merapi volcano, Central Java: An overview. Journal of Volcanology and Geothermal Research, 100(1–4), 423–456. doi:10.1016/S0377-0273(00)00150-5.
[55] Yulinsa, N. (2015). Snake Line Analysis for Lahar Flow Warning System (Case Study in Putih River, Mount Merapi). Journal of the Civil Engineering Forum, 1(1), 37-42. doi:10.22146/jcef.22729.
[56] Rizal, N. S., Umarie, I., Munandar, K., & Wardoyo, A. E. (2023). Calibration and Validation of CN Values for Watershed Hydrological Response. Civil Engineering Journal (Iran), 9(1), 72–85. doi:10.28991/CEJ-2023-09-01-06.
[57] Zhang, X., Li, H., Fan, Y., Zhang, L., Peng, S., Huang, J., Zhang, J., & Meng, Z. (2025). Predicting the Dynamic of Debris Flow Based on Viscoplastic Theory and Support Vector Regression. Water (Switzerland), 17(1), 120. doi:10.3390/w17010120.
[58] Qiu, C., & Geng, X. (2024). Travel distance estimation of landslide-induced debris flows by machine learning method in Nepal Himalaya after the Gorkha earthquake. Bulletin of Engineering Geology and the Environment, 83(10), 395. doi:10.1007/s10064-024-03883-8.
[59] Ming, Z., Zhang, J., He, H., Zhang, L., Chen, R., & Jia, Y. (2025). Addressing accuracy challenges in machine learning for debris flow susceptibility: Insights from the Yalong River basin. Journal of Mountain Science, 22(6), 2034-2052. doi:10.1007/s11629-024-9316-2.
[60] Deng, A. A. N., Nursetiawan, Ikhsan, J., Riyadi, S., & Zaki, A. (2024). Intelligent Forecasting of Flooding Intensity Using Machine Learning. Civil Engineering Journal (Iran), 10(10), 3269–3291. doi:10.28991/CEJ-2024-010-10-010.
[61] Le, X. H., Ho, H. V., Lee, G., & Jung, S. (2019). Application of Long Short-Term Memory (LSTM) neural network for flood forecasting. Water (Switzerland), 11(7), 1387. doi:10.3390/w11071387.
[62] MLIT. (2024). Guidelines for Construction Technology Transfer: Development of Warning and Evacuation System against Sediment Disasters in Developing Countries. Ministry of Land, Infrastructure and Transport Infrastructure Development Institute, Tsukuba, Japan. Available online: https://www.mlit.go.jp/sogoseisaku/inter/keizai/gijyutu/pdf/sediment_e.pdf (accessed on December 2025).
[63] Tercini, J. R. B., & Mello Júnior, A. V. (2023). Impact of Hydroclimatic Changes on Water Security in the Cantareira Water Production System, Brazil. Atmosphere, 14(12), 1836. doi:10.3390/atmos14121836.
[64] Egbelakin, T., Omotayo, T., Ogunmakinde, O. E., & Ekundayo, D. (2025). Eliciting social themes of flood mitigation and community engagement studies through text mining. International Journal of Building Pathology and Adaptation, 43(1), 29–49. doi:10.1108/IJBPA-02-2023-0022.
[65] Faluyi, M. O., & Irmak, S. (2023). Northeastern American Forests: Natural Disturbances, Climate Change Impact, and the Utilization of Increasingly Damaged Forest Trees for Biofuel Production. Forests, 14(12), 2409. doi:10.3390/f14122409.
[66] Wang, J., Lin, W., Chen, Z., Nikolaeva, V. O., Alimi, L. O., & Khashab, N. M. (2024). Smart touchless human–machine interaction based on crystalline porous cages. Nature Communications, 15(1), 1575. doi:10.1038/s41467-024-46071-8.
[67] Xue, X., & Yang, X. (2016). Seismic liquefaction potential assessed by support vector machines approaches. Bulletin of Engineering Geology and the Environment, 75(1), 153–162. doi:10.1007/s10064-015-0741-x.
[68] Hayder, I. M., Al-Amiedy, T. A., Ghaban, W., Saeed, F., Nasser, M., Al-Ali, G. A., & Younis, H. A. (2023). An Intelligent Early Flood Forecasting and Prediction Leveraging Machine and Deep Learning Algorithms with Advanced Alert System. Processes, 11(2), 481. doi:10.3390/pr11020481.
[69] Yang, L., Ge, Y., Chen, B., Wu, Y., & Fu, R. (2024). Machine-Learning-Based Prediction Modeling for Debris Flow Occurrence: A Meta-Analysis. Water (Switzerland), 16(7), 923. doi:10.3390/w16070923.
[70] Zhou, Q., Tang, H., Hibert, C., Chmiel, M., Walter, F., Dietze, M., & Turowski, J. M. (2025). Enhancing Debris Flow Warning via Machine Learning Feature Reduction and Model Selection. Journal of Geophysical Research: Earth Surface, 130(4), e2024JF008094. doi:10.1029/2024JF008094.
[71] Wang, T., Ge, Q., Ma, T., Chen, H., Luo, R., Wang, X., Zhang, K., Chu, Z., Ni, X., & Sun, H. (2025). A novel method for predicting debris flow hazard: a multi-strategy fusion approach based on the light gradient boosting machine framework. Stochastic Environmental Research and Risk Assessment, 39(10), 4867–4890. doi:10.1007/s00477-025-02955-9.
[72] Rey-Devesa, P., Carthy, J., Titos, M., Prudencio, J., Ibáñez, J. M., & Benítez, C. (2024). Universal machine learning approach to volcanic eruption forecasting using seismic features. Frontiers in Earth Science, 12. doi:10.3389/feart.2024.1342468.
[73] Li, K., Zhao, J., Chen, G., & Li, Y. (2025). Debris-flow susceptibility assessment using deep learning algorithms with GeoDetector for factor optimization. Bulletin of Engineering Geology and the Environment, 84(6), 278. doi:10.1007/s10064-025-04343-7.
- Authors retain all copyrights. It is noticeable that authors will not be forced to sign any copyright transfer agreements.
- This work (including HTML and PDF Files) is licensed under a Creative Commons Attribution 4.0 International License.
