F Kamranzad; E Mohasel Afshar; M Mojarab; H Memarian
Abstract
Landslide is one of the natural phenomena which can cause catastrophic losses or damages in life and property each year. Hence, it is very important to recognize landslide-prone areas and apply methods to prevent or reduce slope instabilities and landslide hazard and risk. For this purpose, landslide ...
Read More
Landslide is one of the natural phenomena which can cause catastrophic losses or damages in life and property each year. Hence, it is very important to recognize landslide-prone areas and apply methods to prevent or reduce slope instabilities and landslide hazard and risk. For this purpose, landslide hazard zonation is one of the indirect and efficient methods. This study aims to apply data-driven and AHP methods to provide a zonation map of landslide hazard potential in the Tehranprovince of Iran. First, six essential and available factors including slope, slope direction, geologic background, distance from faults, earthquake acceleration and rainfall were selected to be classified in GIS based on engineering judgment. By superposing data layers over landslide distribution map in data-driven method and expert judgment in AHP method, layers and sub-layers were weighted and combined. The landslide-hazard zonation map was then produced for each of the methods in GIS. Results showed that in data-driven method 92.9% of landslides fall into the perilous zone (i.e. hazardous and very hazardous zones) having an area of 7135.15 km2, which is 37.2% of total area of Tehran province. For the AHP method, 96.47% of the landslides were in perilous zone with an area of 10344.7 km2, which is 53.9% of the total area of the province. Finally, the ratio of percentage of landslides in the perilous zone to the percentage of total area of the zone was calculated. The ratio is 2.5 for the data-driven and 1.79 for the AHP method. The larger ratio in the data-driven method indicates its better consistency than the AHP method, implying more coverage of landslides in a smaller perilous area by the data-driven method. This result represents better accuracy of the data-driven method than the AHP method in landslide hazard zonation.
