تحلیل فضایی مخاطرات محیطی

تحلیل فضایی مخاطرات محیطی

پهنه‌بندی خطر سیلاب در شهرستان ساری با استفاده از الگوریتم‌های یادگیری ماشین و رویکرد تلفیقی

مقالات آماده انتشار
نویسندگان
کمیل عبدی
همت اله رورده
دانشگاه مازندران
چکیده
هدف: سیلاب یکی از مهم‌ترین بلایای طبیعی در استان مازندران و به‌ویژه شهرستان ساری به شمار می‌رود که هر ساله خسارات گسترده‌ای در ابعاد اقتصادی، اجتماعی و زیست‌محیطی به‌همراه دارد. هدف این پژوهش، شناسایی و پهنه‌بندی خطر سیلاب با بهره‌گیری از الگوریتم‌های یادگیری ماشین جنگل تصادفی (RF) و ماشین بردار پشتیبان (SVM) و همچنین استفاده از رویکرد تلفیقی برای افزایش دقت پیش‌بینی‌ها و کاهش عدم‌قطعیت مدل‌ها است.

روش پژوهش: در این مطالعه، مجموعه‌ای از داده‌های مکانی شامل مدل رقومی ارتفاع (DEM)، کاربری اراضی حاصل از تصاویر ماهواره‌ای، شاخص‌های ژئومورفولوژیکی (شیب، جهت شیب و تراکم زهکشی)، داده‌های زمین‌شناسی، فاصله از جاده‌ها و آبراهه‌ها، شاخص پوشش گیاهی (NDVI) و متغیرهای اقلیمی (بارش و دما) گردآوری شد. داده‌ها با استفاده از ابزارهای GIS و RS پردازش و برای آموزش و اعتبارسنجی مدل‌ها آماده گردیدند. عملکرد مدل‌ها با معیارهای ارزیابی شامل دقت، F1، AUC و منحنی ROC سنجیده شد.

یافته‌ها: نتایج نشان داد که هر دو مدل RF و SVM عملکرد بالایی در پهنه‌بندی خطر سیلاب دارند، به‌طوری‌که مقادیر شاخص‌های ارزیابی بیانگر دقت قابل قبول آن‌ها است. همچنین، رویکرد تلفیقی منجر به بهبود نتایج و کاهش خطاهای ناشی از پیش‌بینی منفرد شد. بر اساس نقشه‌های تولیدشده، بخش قابل توجهی از شهرستان ساری در طبقات خطر زیاد و خیلی زیاد قرار دارد که با مناطق دارای بارش‌های شدید، تراکم زهکشی بالا و شیب تند همپوشانی دارد.

نتیجه‌گیری: پژوهش حاضر تأکید می‌کند که الگوریتم‌های یادگیری ماشین، به‌ویژه در قالب رویکرد تلفیقی، ابزار مؤثری در شناسایی مناطق مستعد سیلاب هستند. نتایج این تحقیق می‌تواند به‌عنوان مبنای علمی در برنامه‌ریزی شهری، مدیریت بحران و کاهش خطرپذیری سیلاب در شهرستان ساری و سایر مناطق مشابه مورد استفاده قرار
کلیدواژه‌ها
پهنه‌بندی خطر
جنگل تصادفی (RF)
رویکرد تلفیقی
سیلاب
شهرستان ساری
ماشین بردار پشتیبان (SVM)
یادگیری ماشین

عنوان مقاله English

Flood Hazard Zoning in Sari County Using Machine Learning Algorithms and an Ensemble Approach

نویسندگان English

Komei Abdi
hematolah Roradeh
University of Mazandaran
چکیده English

Objective: Floods are among the most significant natural disasters in Mazandaran Province, particularly in Sari County, where they cause widespread economic, social, and environmental damages each year. The main objective of this research is to identify and map flood hazard zones using machine learning algorithms, namely Random Forest (RF) and Support Vector Machine (SVM), and to apply an ensemble approach in order to enhance prediction accuracy and reduce model uncertainty.

Method: In this study, a set of spatial datasets including a Digital Elevation Model (DEM), land use/land cover derived from satellite imagery, geomorphological indices (slope, aspect, and drainage density), geological data, distance from roads and streams, vegetation index (NDVI), and climatic variables (precipitation and temperature) were collected. These datasets were processed using GIS and RS techniques and prepared for model training and validation. The models’ performance was assessed using evaluation metrics such as Accuracy, F1-score, AUC, and ROC curve analysis.

Findings: The results indicated that both RF and SVM demonstrated high performance in flood hazard mapping, as reflected by strong evaluation metrics. Moreover, the ensemble approach improved prediction reliability and reduced errors compared to single-model predictions. The generated maps revealed that a significant portion of Sari County falls within high and very high hazard zones, which overlap with are::as char::acterized by intense rainfall, high drainage density, and steep slopes.

Conclusion: This research highlights that machine learning algorithms, particularly when applied in an ensemble framework, are powerful tools for identifying flood-prone areas. The findings can serve as a scientific basis for urban planning, disaster management, and flood risk reduction strategies in Sari County and other comparable regions.

کلیدواژه‌ها English

: Ensemble Approach
Flood
Hazard zoning
Machine Learnig
Random Forest (RF)
Sari County Support Vector Machine (SVM)
1. Abedini, M., & Fathi, M. H. (2015). Flood hazard zoning using the Analytic Network Process (Case study: Khiyav Chay watershed). Hydrogeomorphology, 2(3), 99–120. (in Persian)
2. Ahmad, I., et al. (2025). Improving flood hazard susceptibility assessment by ensemble machine learning in Hunza–Nagar region. Natural Hazards. https://doi.org/10.1007/s11069-025-07109-2
3. Ahmad, I., Farooq, R., Ashraf, M. et al. Improving flood hazard susceptibility assessment by integrating hydrodynamic modeling with remote sensing and ensemble machine learning. Nat Hazards 121, 7839–7868 (2025). https://doi.org/10.1007/s11069-025-07109-2
4. Avand M, Janizadeh S, Jafari F. Evaluating the Efficiency of Machine Learning Models in Preparing Flood Probability Mapping. Degrad Rehabil Nat Land 2020; 1 (1) :19-32 URL: http://drnl.sanru.ac.ir/article-1-141-fa.html (in Persian)
5. Azadi, F., Sadough, S. H., Ghahroudi, M., & Shahabi, H. (2020). Flood susceptibility mapping in Kashkan watershed using WOE and EBF models. Geography and Environmental Hazards, 9(33), 45–60. Retrieved from https://sid.ir/paper/526271/fa (in Persian)
6. Beven, K., & Binley, A. (1992). The future of distributed models: Model calibration and uncertainty prediction. Hydrological Processes, 6(3), 279–298.
7. Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324
8. Choubin, B., et al. (2019). An ensemble prediction of flood susceptibility using meta-heuristic and machine learning. Journal of Hydrology. https://pubmed.ncbi.nlm.nih.gov/30321730/
9. Cortes, C., & Vapnik, V. (1995). Support-vector networks. Machine Learning, 20(3), 273–297. https://doi.org/10.1007/BF00994018
10. Cutler, D. R., Edwards, T. C., Beard, K. H., Cutler, A., Hess, K. T., Gibson, J., & Lawler, J. J. (2007). Random forests for classification in ecology. Ecology, 88(11), 2783–2792. https://doi.org/10.1890/07-0539.1
11. Dietterich, T. G. (2000). Ensemble methods in machine learning. In Multiple classifier systems (pp. 1–15). Springer.
12. Esfandiari, M., Abdi, G., Jabari, S., McGrath, H., & Coleman, D. (2020). Flood hazard risk mapping using a pseudo supervised random forest. Remote Sensing, 12(19), 3206.
13. Karami, P., Eslaminezhad, S. A., Eftekhari, M., Akbari, M., & Rastgoo, M. (2023). Flood susceptibility mapping using machine learning methods improved by genetic algorithm. Journal of Natural Environment, 76(1), 43–60. https://doi.org/10.22059/jne.2022.350170.2485 (in Persian)
14. Kazemi Ghehi, H., Mansouri, N., & Jouzi, S. A. (2021). Flood hazard zonation in Nowshahr city using machine learning models. Housing and Rural Environment, 40(176), 71–86. https://doi.org/10.22034/40.176.71 (in Persian)
15. Karami, P., Eslaminezhad, S. A., Eftekhari, M., Akbari, M., & Rastgoo, M. (2023). Flood susceptibility mapping using machine learning methods improved by genetic algorithm. Journal of Natural Environment, 76(1), 43–60. https://doi.org/10.22059/jne.2022.350170.2485 (in persian)
16. Opitz, D., & Maclin, R. (1999). Popular ensemble methods: An empirical study. Journal of Artificial Intelligence Research, 11, 169–198.
17. Pham, B. T., et al. (2021). Flood risk assessment using SVM and Random Forest models: A case study. Journal of Hydrology, 592, 125–140.
18. Pham, B. T., et al. (2021). Improved flood susceptibility mapping using best-first ensemble models. Geomatics, Natural Hazards and Risk, 12(1), 2380–2406. https://doi.org/10.1016/j.jogr.2020.10.013
19. Pham, B. T., Jaafari, A., Van Phong, T., Yen, H. P. H., Tuyen, T. T., Van Luong, V., ... & Foong, L. K. (2021). Improved flood susceptibility mapping using a best first decision tree integrated with ensemble learning techniques. Geoscience Frontiers, 12(3), 101105. in Hunza–Nagar region. Natural Hazards. https://doi.org/10.1007/s11069-025-07109-2
20. Rahmati, O., Pourghasemi, H. R., & Zeinivand, H. (2016). Flood susceptibility mapping using frequency ratio and weights-of-evidence models in Golestan Province, Iran. Geocarto International, 31(1), 42–70.
21. Rahmati, O., Yousefi, S., Kalantari, Z., Uuemaa, E., Teimurian, T., Keesstra, S., ... & Tien Bui, D. (2019). Multi-hazard exposure mapping using machine learning techniques: A case study from Iran. Remote Sensing, 11(16), 1943.
22. Razavi-Termeh, S. V., Safari Bazargani, J., Sadeghi-Niaraki, A., Song, H., & Choi, S. M. (2025). Virtual reality-assisted visualization of flood susceptibility using optimized machine learning models. Applied Water Science, 15(10), 257.SpringerLin
23. Shahabi H, Shirzadi A, Ghaderi K, Omidvar E, Al-Ansari N, Clague JJ, Geertsema M, Khosravi K, Amini A, Bahrami S, et al. Flood Detection and Susceptibility Mapping Using Sentinel-1 Remote Sensing Data and a Machine Learning Approach: Hybrid Intelligence of Bagging Ensemble Based on K-Nearest Neighbor Classifier. Remote Sensing. 2020; 12(2):266. https://doi.org/10.3390/rs12020266
24. Shannon, C. E. (1948). A mathematical theory of communication. Bell System Technical Journal, 27(3), 379–423.
25. Tahmasebi, M. R., Shabanlou, S., Rajabi, A., & Yousefvand, F. (2021). Flood occurrence probability mapping using Random Forest and Support Vector Machine in northern Iran. Journal of Water Management and Irrigation, 11(2), 223–235. https://doi.org/10.22059/jwim.2021.317527.856 (in Persian)
26. Tehrany, M. S., Kumar, L., & Shabani, F. (2019). A novel GIS-based ensemble technique for flood susceptibility mapping using evidential belief function and support vector machine: Brisbane, Australia. PeerJ, 7, e7653.
27. Tehrany, M. S., Pradhan, B., & Jebur, M. N. (2015). Flood susceptibility assessment using GIS-based SVM. Catena, 125, 91–101. https://doi.org/10.1016/j.catena.2014.10.017
28. Tehrany, M. S., Pradhan, B., & Jebur, M. N. (2015). Flood susceptibility assessment using GIS-based support vector machine model. Catena, 125, 91–101. https://doi.org/10.1016/j.catena.2014.10.017
29. Tehrany, M. S., Pradhan, B., Mansor, S., & Ahmad, N. (2015). Flood susceptibility assessment using GIS-based support vector machine model with different kernel types. Catena, 125, 91-101.
30. Valentini, G., & Dietterich, T. G. (2004). Bias-variance analysis of support vector machines for the development of SVM-based ensemble methods. Journal of Machine Learning Research, 5, 725–775.
31. Wahba, M., Essam, R., El-Rawy, M., Al-Arifi, N., Abdalla, F., & Elsadek, W. M. (2024). Forecasting of flash flood susceptibility mapping using random forest regression model and geographic information systems. Heliyon, 10(13).
32. Yousefi,H. , Yonesi,H. A. , Davoudimoghadam,D. , Arshia,A. and Shamsi,Z. (2022). Determination of Flood potential Using CART, GLM and GAM Machine learning Models. Irrigation and Water Engineering, 12(4), 84-105. doi: 10.22125/iwe.2022.150684 (in persian)
33. Naemitabar,M. , zanganeh asadi,M. A. and Boroughani,M. (2025). Flood susceptibility mapping in the semi-arid region of Tabas using machine learning algorithms. (e220164). Journal of Arid Regions Geographic Studies, (), e220164 doi: 10.22034/jargs.2025.509539.1182 (in persian)
34. Hanifinia, A. and Abghari, H. (2025). Predicting flood-prone areas using generalized linear and maximum entropy machine learning models. Journal of Natural Environmental Hazards, 14(43), 19-34. doi: 10.22111/jneh.2024.47730.2021 (in persian)
35. Sadri, M. A. (2024). Flood risk assessment in flood prone areas using probabilistic models and machine learning case study: Gharesu and Gorganroud Watersheds, Golestan Province. , 18.1(55), (in persian)
36. gholami,F. , Entezari,M. , zakerinezhad,R. and karimi,H. (2025). Flood susceptibility prediction and zoning using maximum entropy model in Ilam province. Quantitative Geomorphological Research, 13(4), 116-138. doi: 10.22034/gmpj.2025.475889.1520
تحلیل فضایی مخاطرات محیطی

مقالات آماده انتشار، اصلاح شده برای چاپ
انتشار آنلاین از 01 بهمن 1399