Evaluation of the efficiency of Haeri-Audio experimental model and bivariate statistical in landslide risk zoning (Case study: Qazvin province)

Landslides are one of the most destructive types of erosion on slopes, which causes a lot of financial and human losses. The purpose of this study is to evaluate the efficiency of landslide potential zoning models in Qazvin province and to provide management strategies to reduce landslide risk. The proposed relationships of this model were used for the Haeri-Samiei experimental model. For the statistical model, after preparing the layers of effective factors (land use, slope, slope direction, height, etc.) and cutting them with a landslide distribution map, the area and percentage of landslide in each floor were determined from the effective factors map. Then, using the frequency ratio, the importance of each class of effective factors was calculated and a map was prepared based on the importance of effective factors in the GIS environment. Finally, to prepare a landslide hazard map, the maps were grouped together based on the rate of factors and classified according to the method of intervals equal to seven floors. Density ratio and total utility indices were used to evaluate the efficiency of the models and select the superior model. The results showed that the bivariate statistical model with Qs equal to 0.938 has a higher accuracy than the Haeri-Samiei model with Qs equal to 0.622. To reduce damage and manage landslide risk, three non-action management policies for low risk classes, prevention for medium risk classes and treatment for high risk classes were considered and some necessary measures were provided for each management policy.