Assessment of LNRF model in landslide Hazard mapping using GIS at Ziarat watershed, Gorgan

Recognition of regions susceptible to Landslide event is one of the basic measures taken to decrease the possible risk, and hazard management. The main purpose of this study was to produce landslide hazard mapping by LNRF model based on geographic information system (GIS) for the Ziarat watershed, Golestan province, Iran. At first, a landslide inventory map was prepared by using earlier reports and aerial photographs, and a total of 50 landslides were mapped and out of which 35 (70%) were randomly selected for building landslide Hazard model, while the remaining 15 (30%) were used for validating the model. Thirteen input data layers were employed as landslide conditioning factors in the landslide Hazard modeling. These factors are slope percent, slope aspect, altitude, plan curvature, precipitation amount, lithology, land use, distance from faults, distance from rivers, distance from roads, topographic wetness index (TWI) and stream power, CTI (Sediment Transport Index)and stream power index (SPI). Afterward, landslide susceptibility map was produced using LNRF model. For verification, receiver operating characteristics curve (ROC) and area under the curve (AUC) approaches were used. Verification results showed 82.15% accuracy. According to the results of the AUC (area under curve) evaluation, the map produced exhibits satisfactory properties.