Efficiency assessment of Modified Nilsson and Relative Effect models in landslide hazard zonation, case study: Parsian dam Watershed

From the viewpoint of landslide occurrence, Parsian dam Watershed is one of the high risk regions for landslide hazard based on its geological and climatological conditions. Landslide occurrence usually make huge damages and also high amount of sediments. Landslide control and management is planned according to Landslide Hazard Zonation (LHZ) that is obtainable through empirical models. The aim of this research was to assess the efficiency of two empirical models, Modified Nilsson and Relative Effect in LZH. First, prepared maps such as landslide distribution, slope, geology, land use, land cover, roads, drainages, faults, isohyet and hypsometric were imported into GIS environment and then the landslide hazard zonation was carried out by applying Modified Nilsson and Relative Effect models. Capability assessment of two models and determination of the more efficient one were evaluated using two statistical analysis methods namely quality summation (Qs) and correlation coefficient (R). Comparison of results from above two models with real map of happened landslides at Parsian dam Watershed showed that the relative effect model with quality summation of 1.00 and correlation coefficient of 0.941, is more efficient for landslide hazard zonation in Parsian dam Watershed.