Kerekes

LANDSLIDE SUSCEPTIBILITY EVALUATION USING GIS. CASE STUDY: SILVANIA HILLS (ROMANIA)
KEREKES ANNA‐HAJNALKA1, HORVÁTH CS.1
1 Babeș‐Bolyai University, Faculty of Geography, 5‐7 Clinicilor Street, Cluj‐Napoca, Romania, annakrks@yahoo.com, hcsaba@gmail.com
ABSTRACT. – Landslide Susceptibility Evaluation Using GIS. Case Study: Silvania Hills (Romania). Landslides are destructive natural or human‐induced hazards, therefore an assessment of landslide susceptibility becomes essential in an area that is prone to landslide through its geographical features. Landslide susceptibility maps provide valuable information for disaster mitigation works and land planning strategies. Silvania Hills are highly prone to landslide due their lithological and geomorphological structure. The purpose of this paper is to prepare a reliable landslide susceptibility map, which was obtained using the maximum entropy model. The model was run considering twelve environmental factors: lithology, slope, aspect, land use, land cover type, precipitation, temperature, terrain roughness, depth of fragmentation, drainage density, profile and plan curvature. The resulting map was grouped into five landslide susceptibility classes: very low, low, moderate, high and very high class. The results indicated that the land use and ‐cover type, slope and depth of fragmentation are the three most influential landslide predisposing factors. The accuracy of the resulted map was verified by generating a receiver operating characteristic (ROC) curve. The area under the curve (AUC) showed a good performance (0.847) of the analysis.
Keywords: natural hazard, landslide, susceptibility analysis, mapping susceptibility, database, MaxEnt model, Silvania Hills, GIS, ROC curve.
FULL TEXT