Flood Hazard Potential Zoning Using Machine Learning and Multi-Criteria Decision-Making Methods in the Izeh Basin, Northeast of Khuzestan Province

Nazarpour, Ahad; Joveili, Arman

doi:10.22111/jneh.2026.54299.2151

Flood Hazard Potential Zoning Using Machine Learning and Multi-Criteria Decision-Making Methods in the Izeh Basin, Northeast of Khuzestan Province

Articles in Press

Document Type : Research Article

Authors

Ahad Nazarpour ¹
Arman Joveili ²

¹ Associate Professor, Department of Geology, Ahvaz Branch, Islamic Azad University, Ahvaz, Iran

² MSc Student, Department of Remote Sensing and GIS, Ahvaz Branch, Islamic Azad University, Ahvaz, Iran

10.22111/jneh.2026.54299.2151

Abstract

Flooding is one of the most significant natural hazards, causing extensive human and economic losses every year. In this study, flood susceptibility mapping of Izeh County was carried out using nine influential factors, including distance to waterways, rainfall intensity, digital elevation model (DEM), land use/land cover, slope, vegetation cover, soil texture, erosion, and soil moisture. The weighting of these criteria was performed using the Analytic Hierarchy Process (AHP). The results indicated that distance to waterways plays the most critical role in flood occurrence, followed by factors such as rainfall intensity, elevation, and land use. For a more detailed assessment, two artificial intelligence methods, Random Forest (RF) and Artificial Neural Network (ANN), were employed for flood hazard zoning. A comparison of the results demonstrated that the Random Forest model exhibited a higher capability in identifying high-risk areas and overall provided more accurate performance than the Artificial Neural Network model. The generated maps revealed that the western and central parts of Izeh, particularly areas adjacent to rivers, have the highest flood susceptibility, whereas regions such as Pian and Dehdaz fall within the low-risk category. Furthermore, multi-criteria decision-making analysis using the TOPSIS method indicated that the Eastern Susan and Morgha areas show higher vulnerability to flooding due to the high density of waterways and specific environmental conditions. Finally, by integrating the results obtained from the AHP, Random Forest, and Artificial Neural Network methods, a final flood susceptibility map of the region was produced. The integration of this map with the vulnerability map derived from the TOPSIS method resulted in a comprehensive flood hazard and vulnerability map. By utilizing remote sensing data and integrating multiple modeling approaches, this study enhanced the accuracy of flood hazard assessment and emphasized the importance of spatial planning, construction of flood control structures, dredging of waterways, implementation of early warning systems, and regulation of construction activities in high-risk areas.

Keywords

Main Subjects

Environmental Hazards

References

References (in Persian)

Abdollahzadeh, S., & Sharifzadeh, M. S. (1402). Identifying the determinants of participation in flood risk mitigation measures: A study of rural areas exposed to flood in Golestan province. Scientific Quarterly of Crisis Prevention and Management Knowledge, 13(3), 318-335. https://doi: 10.32598/DMKP.13.3.762.1. [In Persian]

Hejazi, A., Andaryani, S., Almaspour, F., & Mokhtari Asl, A. (1394). Using multi-criteria decision making and remote sensing techniques in GIS environment to investigate areas sensitive to flooding in Liqvan Chai basin. Hydrogeomorphology, 2(3), 61-80. https://doi: 10.32598/DMKP.13.3.762.1. [In Persian]

Jensi, Z., Mahmoudi Chenari, H., Rahimipour Sheikhaninejad, M. A., Maskani, H., & Abedi, T. (1402). Layered analysis of the causes of instability of Iran's wetlands. Majles and Rahbord, 30(114), 5-39. https://doi: 10.22034/mr.2022.5133.4932. [In Persian]

Mahmoudi, P., Fattahi, E., Heydari, M., Rigi, A., Ansari, A., Ghaemi, A., & Rezaei, J. (1404). Analysis of precipitation extremes trends in the Persian Gulf and Oman Sea Basin in Southern Iran. Journal of Natural Environmental Hazards, 14(44), 63-88. https://doi: 10.22034/mr.2022.5133.4932. [In Persian]

Negahban, S., Ganja'iyan, H., Ebrahimi, A., & Gheysarian, S. S. (1404). Analysis of the roles of environmental factors in the occurrence of floods using the Google Earth Engine system (Case Study: West of Golestan Province). Geography and Environmental Planning, 35(4), 1-18. https://doi: 10.22108/gep.2024.142342.1659. [In Persian]

References (in English)

Jonkman, S.N., Curran, A. & Bouwer, L.M. Floods have become less deadly: an analysis of global flood fatalities 1975–2022. Nat Hazards 120, 6327–6342 (2024). https://doi.org/10.1007/s11069-024-06444-0