Flood Susceptibility Mapping in Erbil City Using Machine Learning Techniques
Journal of Studies in Civil Engineering · 20.06.2026 · 3 (1) · 93-111
DOI: 10.53898/jsce2026316 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Studies in Civil Engineering, online erschienen am 20.06.2026. Verfasst von Kaifi Chomani. Die Arbeit nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The objective of this research was to evaluate the performance of various machine learning approaches for identifying flood-risk areas in Erbil City, the Kurdistan Region of Iraq. In this study, three machine learning methods, Maximum Likelihood Classifier (MLC), Support Vector Machine (SVM), and Random Forest (RF), were compared to identify flood risk areas and assess their performance using historical flood data. Multiple factors such as Normalized Difference Vegetation Index (NDVI), aspect, Topographic Wetness Index (TWI), distance to roads, lithology, soil data, Normalized Difference Built-Up Index (NDBI), Land Use Land Cover (LULC), flow accumulation, rainfall, slope, Stream Power Index (SPI), drainage density, and elevation have been used to create susceptible flood areas. The results of different approaches for very high flood susceptibility were as follows: 22.7% (206.4 km²) for SVM, 26.5% (241.7 km²) for RF, and 9.1% (83.2 km²) for MLC. The Receiver Operating Characteristic – Area Under the Curve (ROC–AUC) accuracy assessments reported the following AUC values for flood susceptibility mapping: 87.3% for the MLC, 88.8% for RF, and 94.1% for SVM. The comparison analysis showed that SVM effectively maps flood susceptibility. This study offers insights into assessing flood hazards across different scales and data conditions, aiding better risk understanding and mitigation planning.
Zitieren
APA
Chomani, K. (2026). Flood Susceptibility Mapping in Erbil City Using Machine Learning Techniques. Journal of Studies in Civil Engineering, 3(1), 93-111. https://doi.org/10.53898/jsce2026316
Vancouver
Chomani K. Flood Susceptibility Mapping in Erbil City Using Machine Learning Techniques. Journal of Studies in Civil Engineering. 2026;3(1):93-111. doi:10.53898/jsce2026316
BibTeX
@article{Chomani2026,
author = {Chomani, Kaifi},
title = {Flood Susceptibility Mapping in Erbil City Using Machine Learning Techniques},
journal = {Journal of Studies in Civil Engineering},
year = {2026},
volume = {3},
number = {1},
pages = {93--111},
doi = {10.53898/jsce2026316}
}
EUVIMED