Landslide susceptibility assessment using frequency ratio, statistical index and certainty factor models for the Gangu County, China

Abstract

The purpose of this paper is to produce a reliable susceptibility mapping using frequency ratio (FR), statistical index (SI), and certainty factor (CF) models with the aid of geographic information system (GIS) for the Gangu County, Gansu Province, China. First, a total of 328 landslide locations were detected by literatures, aerial photographs and field surveys; meanwhile, a landslide inventory map was constructed mainly based on landslide locations. Then, 230 (70 %) landslides were randomly selected for modeling, and the remaining 98 (30 %) landslides were used for the model validation. In order to produce a susceptibility map, 12 landslide influencing factors were selected from the database: slope angle, slope aspect, plan curvature, profile curvature, altitude, distance to faults, distance to rivers, distance to roads, NDVI, land use, rainfall, and lithology. Whereafter, the landslide susceptibility maps were mapped using landslide influencing factors based on the FR, SI, and CF models. Finally, the accuracy of the landslide susceptibility maps developed from the three models was validated using area under the curve (AUC) analysis. Through the analysis, it is seen that the prediction accuracy of the three models was 75.62 % for FR model, 75.71 % for SI model, and 75.56 % for CF model, respectively. According to the results, three models show almost similar results, while SI model performs slightly better than other models and the map produced by SI model represents the most appropriate properties. In addition, the study area was classified into five classes, such as very low, low, moderate, high, and very high. The landslide susceptibility maps can be helpful to select site and mitigate landslide hazards in the study area.