<![CDATA[The study area (Enfraz to Addis Zemen) is located in northwestern Ethiopia, which frequently experiences landslides, causing damage to farmland, engineering structures, infrastructures, and villages, as well as animal and human fatalities. To manage this catastrophic hazard, a comprehensive GIS-based frequency ratio model (FR) was applied to produce a landslide susceptibility map. In this study, 134 landslides were identified from detailed fieldwork and Google Earth imagery analysis, split into 70% to develop the model and 30% for model validation. The relationship between landslide probability with landslide factor classes of lithology, annual mean rainfall, slope, aspect, curvature, elevation, distance to the river, and land use-land cover was analyzed in a GIS environment. FR model assigns weights to each factor class based on observed frequencies. These weighted factors were summed using a raster calculator to produce landslide susceptibility indexes (LSIs), which were classified into very low, low, moderate, high, and very high susceptibility classes using the natural break classification method. The model’s accuracy and performance were validated using the area under the curve of the receiver operating characteristics curve (ROC), which showed an AUC success rate of 92.2% and a predictive rate of 86.05%. These results confirm that the FR model is effective in landslide susceptibility modeling. The generated map can support decision-makers, urban planners, and researchers in land use planning, landslide mitigation strategies, and future research.]]>
Copyrights © 2025
