Akademik Veri Yönetim Sistemi
Jeomorfolojik Araştırmalar Dergisi (Online), sa.15, ss.170-188, 2025 (TRDizin)
Floods, whose frequency and severity have increased due to both climate change and anthropogenic effects such as urbanization, deforestation, and land use changes, continue to pose serious risks to human life, infrastructure, and ecosystems worldwide. In regions like southern Türkiye, where complex topography, orographic precipitation, and rapid urban growth intersect, understanding flood dynamics is particularly critical. This study evaluates the flood susceptibility of 24 river basins that drain into the Gulf of İskenderun, focusing on the districts of Erzin, Dörtyol, İskenderun, Arsuz and Belen in Hatay Province. In this study, we developed a comprehensive framework for assessing spatial flood risk by integrating morphometric analysis with statistical classification methods. Fourteen morphometric parameters derived from 10-meter resolution digital elevation models were processed using GIS-based analyses. The proposed methodology involves two complementary analytical techniques: the Normalized Morphometric Flood Index (NMFI) and Principal Component Analysis (PCA). The Normalized Morphometric Flood Index (NMFI) plays a significant role in understanding and identifying flood-prone basins. This method allows the morphometric-based evaluation results of flood-prone basins to be normalized, enabling the obtained values to range between 0 and 1, and classifying flood susceptibility into four distinct categories. The Principal Component Analysis (PCA), on the other hand, considers the dynamic parameters influencing the occurrence of flood events and highlights the most dominant and effective parameters contributing to flooding. As a result of evaluating 24 river basins draining from the Amanos Mountains into the İskenderun Gulf, it was found that, although some differences exist between the two methods, both approaches identified several basins with high floodgeneration potential and exhibited many similarities. Moreover, a portion of these 24 basins was classified within the moderate and high flood susceptibility categories. Furthermore, the results derived from the PCA method demonstrated superior performance compared to the NMFI method in terms of classification accuracy, recall rate, and overall reliability. According to the analysis, the drainage density (Dd), bifurcation ratio (Rb), time of concentration (Tc), circularity ratio (Rc), and basin relief (Bh) were identified as the most influential factors affecting flood potential across the 24 basins. The findings from both methods reveal that these approaches are critically important for understanding flood potential and identifying flood-prone basins. Moreover, they can be effectively applied to small-, medium-, and large-scale basins. These results are particularly valuable for conducting rapid and probabilistic assessments in watersheds and support hydraulic modelingbased flood hazard and risk analyses in areas with high flood potential, thereby contributing to a more efficient decision-support process in flood management.