Assessment of landslide hazards in Miocene Basins based on Modified Frequency Ratio and Limit Equilibrium models: A case study in Aomar Miocene basin, Bouira, Algeria

Abstract

The Neogene basins are well known for their high susceptibility to landslides. This feature is mainly favoured by the lithological nature of the filling materials and slope morphology. This study proposes a combined frequency ratio and limits equilibrium methodology to perform a regional analysis of slope stability in the Neogene basin of Aomar Ben Haroun (Bouira, Algeria). For this purpose, different work steps have been conducted. Initially, the field site selection was made according to the surface geology underlying ground motion and the extent of landslide events. Subsequently, the landslide survey was built on existing studies and on high-resolution aerial photography and satellite imagery analysis, as well as on detailed fieldwork involving mapping and compilation of triggering factors. Besides, the evaluation of the landslide susceptibility around the study area was performed using the frequency ratio (FR) in the GIS environment. Note that the geotechnical characterization of each formation was determined from the physical and mechanical parameters of recent and old geotechnical studies carried out in the region. Moreover, the determination of the​​ landslide-prone areas with a characteristic stability factor for each slope was then computed using a two-dimensional limit equilibrium model. Thanks to the provided results, the classes of slopes that are characterized by parting surfaces susceptible to landslides are clearly defined. The result maps of the Aomar Miocene basin were categorized into low, moderate and high susceptibility classes. We have then been able to validate the obtained results by each used method. In the case of the inventory and susceptibility maps, we have also compared the two results based on the results of the area under the curve (AUC). For the frequency ratio (FR) and Limit Equilibrium (LE) models, the AUC are (83.23%) and (89.25%) respectively, which are indicating good and very good accuracy in the identification of landslide susceptibility zones of a basin and can be used in land use planning. In conclusion, this methodology highlights the fact that the landslides susceptibility map, which is obtained using the limit equilibrium model, was significantly more reliable and consistent than the susceptibility map completed by the frequency ratio model after field verification.