Application and comparison of shallow landslide susceptibility models in weathered granite soil under extreme rainfall events

Landslide susceptibility maps are important for risk management and land-use planning in mountainous countries such as Korea. Many of the slopes of the mountainous areas of Korea are composed of residual soils that originate from weathering of the granite bedrock. The purpose of this study was to compare the application of the SHALSTAB and SINMAP models to slope stability evaluation in Deokjeok-ri Creek. The SINMAP and SHALSTAB models are terrain stability models that are based on a steady-state hydrologic model coupled with an infinite-slope stability equation. A digital elevation model based on LIDAR data was used to calculate the slope and wetness indices. In situ and laboratory tests were performed to find the geotechnical parameters. A geographic information system-based landslide inventory map of 748 landslide locations was prepared using data from previous reports, aerial photographs, and extensive field work. This inventory of landslide scars was used to document sites of instability and to provide an evaluation of model performance by comparing the observed landslide locations with predictions from the models. Both SHALSTAB and SINMAP models identified the northern part of the catchment as the most unstable area. Receiver operator curve plots were used to show that the susceptibility maps produced using SHALSTAB had better prediction accuracy (82.4 %) than the SINMAP model (62.58 %) had. It is concluded that SHALSTAB was relatively successful in delineating slopes where shallow landslides have been observed previously. SHALSTAB is therefore more applicable to weathered granite soil.