Landslide hazard zoning using frequency ratio, entropy methods and TOPSIS decision-making methods (Case study: Fahliyan basin, Fars)

Distinguishing the susceptible areas to landslide using appropriate experimental models of landslide susceptibility mapping is one of the primitive and basic works to reduce probable damages and reduce risk in country's watersheds. In this research, thirteen factors, including altitude, slope, aspect, rainfall, plan curvature, profile curvature, Topographic Wetness Index (TWI), Stream power index (SPI), distance from the river, distance from road, distance from the fault, lithology, and land use are identified as effective parameters in landslide occurrence in the studied area. The landslide landslide distribution map was prepared using Landsat 2017 satellite imagery and Field studies (2017). Then, the landslide susceptibility mapping was performed with three methods, including frequency ratio, entropy and TOPSIS. The results were evaluated using receiver operating characteristic (ROC) curve. The area under the curve (AUC) of applied models shows the accuracy of 84, 87.7, and 91 percent for entropy frequency ratio and TOPSIS models, respectively. The results are indicative statistical models have a better accuracy in comparison with MCDM models and expert approaches. The results of the present study could be useful for better managing the flood affected areas and to reduce its damage.