Mapping landslide susceptibility in the Debretabor-Alember road sector, Northwestern Ethiopia through geospatial tools and statistical approaches

• DOI: 10.15243/jdmlm.2024.112.5169
• Publication Date: 2024-01-01T03:31:35Z
• Authors: Betelhem Tesfaye; Muralitharan Joth...; Zerihun Dawit
• Keywords: Elevation (ballistics); Lithology; Land Cover

This study aimed to locate areas along the Debretabor-Alember route segment in northern Ethiopia that are susceptible landslides. Geospatial tools, specifically frequency ratios (FR) and information values (IV), were used develop landslide susceptibility maps (LSMs). A comprehensive on-site investigation analysis of Google Earth imagery conducted, resulting detection 89 landslides, including current historical events. The dataset for validation comprised 78% previously documented whereas remaining 22% was training. Several factors considered this determine susceptibility, "slope, aspect, curvature, elevation, lithology, distance from streams, land use cover, precipitation, normalized difference vegetation index (NDVI)", FR IV models. Based on results obtained using approach, specific exhibited different levels ranging very low moderately high, medium, high. These covered a total 18.4 km2 (19.9%), 18.9 (20.5%), 19.7 (20.3%), 17.7 (20%), (19%), respectively. LSMs generated by model indicated multiple classes area, varying revealed (19.8%), 18.8 (19.5%), 18.3 (19.8%) area fell into these classes. density indicator method employed validate LSMs. models demonstrated significant proportion confirmed past records (72.16% 73.86%, respectively) occurred regions with high or Overall, model, which utilized latent variable structural modeling (LSM) independent outperformed fixed effects regression (FR).