A GIS-based comparative study of frequency ratio, statistical index and weights-of-evidence models in landslide susceptibility mapping

The aim of this study is to generate reliable susceptibility maps using frequency ratio (FR), statistical index (SI), and weights-of-evidence (WoE) models based on geographic information system (GIS) for the Qianyang County of Baoji City, China. At first, landslide locations were identified by earlier reports, aerial photographs, and field surveys, and a total of 81 landslides were mapped from various sources. Then, the landslide inventory was randomly split into a training dataset 70 % (56 landslides) for training the models, and the remaining 30 % (25 landslides) was used for validation purpose. In this case study, 13 landslide-conditioning factors were exploited to detect the most susceptible areas. These factors are slope angle, slope aspect, curvature, plan curvature, profile curvature, altitude, distance to faults, distance to rivers, distance to roads, Sediment Transport Index (STI), Stream Power Index (SPI), Topographic Wetness Index (TWI), and lithology. Subsequently, landslide-susceptible areas were mapped using the FR, SI, and WoE models based on landslide-conditioning factors. Finally, the accuracy of the landslide susceptibility maps produced from the three models was verified by using areas under the curve (AUC). The AUC plot estimation results showed that the susceptibility map using FR model has the highest training accuracy of 83.62 %, followed by the SI model (83.45 %), and the WoE model (82.51 %). Similarly, the AUC plot showed that the prediction accuracy of the three models was 79.40 % for FR model, 79.35 % for SI model, and 78.53 % for WoE model, respectively. According to the validation results of the AUC evaluation, the map produced by FR model exhibits the most satisfactory properties.