Integrated analysis of landslide susceptibility: geotechnical insights, frequency ratio method, and hazard mitigation strategies in a volcanic terrain

Abstract

This study investigates landslide susceptibility in the Baingo and Njinekejem areas with elevations ranging from 504 to 2863 m along the Cameroon Volcanic Line (CVL) in Central Africa. In this work, multifaceted approaches integrating geotechnical characterization, remote sensing data, and Geographic Information System (GIS) techniques were used to create a comprehensive landslide susceptibility map. The research methodology combines multiple factors contributing to landslide susceptibility, including topography, geology, land use, and soil properties, which are analyzed and integrated into a frequency ratio (FR) model. Detailed geotechnical analysis was conducted on soil samples from three locations (Ba1, Ba2, and NJ) to understand the mechanical behaviour of volcanic soils under saturated conditions. The soil analysis revealed significant physical and mechanical properties, including high porosity (55–61%), substantial void indices (1.228–1.561), and over-consolidation ratios around 21, indicating compressible but non-swelling characteristics. Additional parameters examined included unit weight and Atterberg limits to provide comprehensive insights into soil behaviour. The model’s accuracy was validated using the area under curve (AUC) method, achieving a value of 0.84, which demonstrates strong predictive capability. The resulting landslide susceptibility map identifies areas with varying degrees of susceptibility, from very low to very high, enabling stakeholders to develop targeted risk mitigation strategies for the region.