Landslide susceptibility mapping in Lesalso (Laelay Maichew), Northern Ethiopia: a GIS approach using frequency ratio and analytical hierarchy process methods
摘要
Landslides in hilly regions lead to significant damage to natural resources, property loss, and numerous fatalities. This research aims to produce landslide susceptibility maps of ‘Lesalso’’ area, employing field investigations and GIS-based Frequency Ratio (FR) and Analytic Hierarchy Process (AHP) methods for the first time. A total of 235 landslide inventories covering 6.729 km2 were identified, and nine causative factors—slope steepness, slope aspect, elevation, lithology, rainfall, land use/land cover (LULC), and proximity to discontinuities, streams, and roads were analyzed for their contribution to landslide initiation. The FR method identified high correlations with lithology (basalt, laterite sandstone, and phyllite), proximity to streams and discontinuities, rainfall (over 800 mm), elevation (above 2000 m), and slopes greater than 15°. The AHP analysis highlighted similar critical factors, including lithology (basalt and laterite sandstone), LULC (bare land and agriculture), rainfall (over 900 mm), elevation (1950–2000 m), and slope steepness (20°-25°). The FR-derived LSM indicated very high (8.32%), high (46.47%), medium (41.01%), and low (4.2%) susceptibility zones, whereas the AHP approach yielded 5.51%, 44.92%, 48.89%, and 0.68% susceptibility zones, respectively. Both methods produced comparable results, with AUC values demonstrating excellent model performance (FR: 81.3%, AHP: 75.16%). These maps are critical for informing future land use and land cover planning, ultimately contributing to reducing landslide risks.