Landslide suspectibility mapping of Karaburun Peninsula with geographic information systems

Landslides are one of the most damaging types of natural disasters. Geographical Information Systems (GIS) provide to facilitate both for mapping the current landslides and for predicting the landslides that may be experienced in the future. Within the scope of the study, landslide susceptibility mapping study was carried out in order to reduce the damages of possible landslides in Karaburun Peninsula of İzmir province. In accordance with this purpose; firstly, landslide inventory map was produced from related databases and then 10 parameters (altitude, slope, aspect, curvature, lithology, distance to roads, distance to faults, distance of stream, land use and vegetation) map which are thought to affect landslide formation in the region were created. Landslide susceptibility maps were produced by logistic regression and frequency ratio methods. The maps made with these methods were compared with ROC analysis and it was found that the map produced by logistic regression was more accurate than the map produced by frequency method. As a result, it was determined that the suspectibility maps created by both methods are compatible with the existing landslide inventory map and can be used as an important data in the future.