Identification of Landslide-prone Zones Using a GIS-based Multi-Criteria Decision Analysis and Region Growing Algorithm in Uncertain Conditions

Abstract

Landslides are considered to be one of the most significant natural hazards. Detection of landslide-prone zones is an important phase in landslide hazard assessment and mitigation of landslide-related losses. AHP as one of the most effective methods for GIS-based multi-criteria decision analysis is increasingly being used in susceptibility mapping. However, its weights have some degree of uncertainty that interval comparison matrix (ICM) method can be used to deal with this problem. The importance of this study is to propose an interval number distance-based region growing (IDRG) method based on ICM for the identification of landslide-prone zones in the Urmia lake basin, Iran. To assess the capability of the proposed IDRG method, a landslide susceptibility map was produced using common AHP, too. To generate the maps, the weights of nine conditioning factors were determined using both traditional pairwise comparison matrices (PCM) of the AHP method and ICM. The accuracy of the produced maps was assessed through ROC (receiver operating curve) and using a dataset of known landslide occurrences. The results indicate an improvement in accuracy of about 11% by identifying the landslide-prone zones using the IDRG method. This improvement was achieved by minimizing the uncertainty associated with criteria ranking/weighting in a traditional AHP and identifying the prone zones as areas instead of pixels.