GIS-based statistical and limit equilibrium models in the assessment of slope stability and landslide susceptibility: the case study of the Aomar Miocene basin, Bouira, Algeria

Abstract

Landslides are the main gravitational phenomena in the Neogene basins of Algeria. This feature is mainly favored by the lithological nature of the fill materials, the climate, and the slope morphology. They are known for their high residential concentration and urban development. For better management and planning of future projects in landslide-prone areas, the use of Landslide Susceptibility Maps (LSMs) is indispensable. This study proposes a combined statistical and limit equilibrium method for the analysis of slope stability and for the production of LSMs using GIS in the Aomar Neogene basin (Bouira, Algeria). For this purpose, different work steps were carried out. Initially, a landslide inventory map was prepared based on existing studies, high-resolution aerial photography and satellite imagery analysis, as well as on detailed fieldwork. Besides, the geotechnical characterization of each geological formation was determined from the physical and mechanical parameters of recent and old geotechnical studies conducted in the region. A total, 169 landslides were compiled and randomly divided into training (70%) and validation (30%) datasets. Furthermore, eight factors influencing landslide occurrence (including slope angle, lithology, slope aspect, elevation, drainage density, distance to the rivers, distance to the roads, and rainfall) were selected and applied for the analysis. Moreover, the inventoried landslide and their associated factors were taken into account when mapping landslide susceptibility in the Aomar Neogene basin using Frequency Ratio (FR) models in GIS. On the other hand, a two-dimensional limit equilibrium (LE) model was used to determine the landslide-prone zones with a distinctive stability factor for each slope and the failure surface depth. Then, we have been able to verify and validate the obtained results for the statistical method using the Area Under the Curve (AUC) value of the Receiver Operating Characteristics Curves (ROC) method. For the frequency ratio models, the AUC is 83.23%, and for the limit equilibrium method, the safety factor is less than one for a slope greater than 16°. Based on these findings, we draw the conclusion that the maps may be exploited as practical instruments for land use planning and risk reduction in the Aomar Neogene basin.