Assessment and Spatial Prediction of Landslide Hazard using Logistic Regression and Certainty Factor Model (Case Study: Along the Khalkhal-Sarcham Road)

  The occurrence of landslides is the result of the interaction of complex and diverse environmental factors. These factors are divided into the trigger and the primary cause. Landslide occurrence triggers include weathering, earthquakes, rainfall and snow melting. Human activity like construction of roads and buildings on steep slopes and dispersal of water from supply systems and sewers could also trigger the occurrence of the phenomena (Cubito et al., 2005). Landslide susceptibility mapping involves handling, processing and interpreting a large amount of territorial data. Thus, Geographical Information Systems (GIS) have proved to be very useful in susceptibility evaluation (Aleotti and Chowdhury, 1999; Ayalew et al., 2005), as it allows frequent updating of the database related to spatial distribution of the landslide events and their predisposing factors, as well as the susceptibility assessment procedures (Aleotti and Chowdhury, 1999). Landslide susceptibility mapping relies on a rather complex knowledge of slope movements and their controlling factors. The reliability of landslide susceptibility maps depends mostly on the amount and quality of available data, the working scale and the selection of the appropriate methodology of analysis and modeling. In this research, several quantitative approaches in order to landslide hazard zonation were used for the area along the Khalkhal-Sarcham Road. The study area with geographic coordinates between 37º 19′ to 37º 36′ N latitudes and 48º 50′ to 21º 01′ E longitudes is located in Ardabil province.

In this research, for assessment and landslide zonation ten factors affecting landslide occurrence, which include: elevation, slope, aspect, streams, faults, topographic features, lithology, land use, vegetation and communication road was considered. Required data were obtained from the topographic maps with scale of 1:25000, geological maps with scale of 1:100000, digital elevation model (DEM) 12.5 meter from ALOS – PALSAR satellite, Sentinel (Spatial resolution of 10 meter), Google Earth satellite images and field studies. In landslide hazard zonation using the GIS, the most important part of the study is preparation of the landslide distribution map or landslide inventory. Therefore, field works was done in order to identification of landslides and preparation of landslide inventory. In this study for assessment and landslide zonation were used certainty factor and logistic regression quantitative approaches. In this research, landslide events risk use certainty factor and logistic regression quantitative approaches was evaluated. Certainty factor model is a probabilistic bivariate model and suitable for landslide hazard zonation. This model provides reasonable results for the study area. Also for zonation by using multivariate statistical methods, was used logistic regression, which is one of the most suitable methods for landslide hazard zonation. The results from this method were detected as the most favorable model among the investigated models. For the studied area, 98 large and small landslides were identified and delineated through satellite images of high resolution satellite image and field observations. Most of these landslides have occurred from the north-east, south-west trend, from the Kahran and Esmarud villages to the Gheshlag and Gorjagh villages. A significant number of landslides have occurred around the Khalkhal-Sarcham communication road between of Kabodchi to Gheshlagh villages. The results show that total landslide area of the region is about 650 hectares. In terms of percentage of high and very high risk class area, these methods represent fairly similar results, so that can say that approximately 23 percentage of the study area is located in the high and very high risk class.

This research has been carried out to identify areas of potential Landslide in the Khalkhal-Sarcham communication road. In this regard, logistic regression and certainty factor model were used. This communication road is very important for landslide occurrence. According to the results that between the factors affecting the occurrence of landslides in the region, lithological conditions play an important role. Construction of a communication road on these landslide sensitive units has increased the risk of landslide events. The slope and elevation variables also have a large effect on the landslide occurrence in the area. The impact of other conditions can be considered locally. Therefore, slope instabilities in all of the spatial planning should be considered in this area