A determination of effective factors on shallow landslide occurrence in Khiov-Chai watershed

Shallow landslide susceptibility assessment by using appropriate methods and determine of effective factors in reduce of its hazards is so effective. The aim of this study is to determine the effective factors on shallow landslide occurrence and investigation of Statistical Index Method (SIM) efficiency for landslide susceptibility mapping. So, determination of each class of factors’ weights was accomplished by using SIM. That was done by adoption of inventory landslide map and ten initial factors including: slope, aspect, rainfall, altitude, drainage density, plan curvature, land use, geology, geomorphological faces and rock unit sensitivity to erosion in ArcGIS 9.3. Then model efficiency was evaluated by using percentage of area under ROC curve and the results showed high accuracy (0.95) of SIM. Principal Component Analysis (PCA) was used for determination of primary causative factors of shallow landslide occurrence. Results showed that five variables of rainfall, slope, aspect, rock unit sensitivity to erosion and plain curvature are the most effective factors on landslide occurrence, respectively.