Slope stability analysis and landslide susceptibility assessment on the SE foot slopes of Mt Cameroon

Abstract

Landslides are well known processes that hamper sustainable development efforts, particularly in developing countries where extreme poverty is already a household to societal scale problem. Though a worldwide issue, landslides remain enigmatic especially with regards to their geometry, causative factors, triggering mechanisms, relationship between triggering factors and landslide occurrence, and the susceptibility of any area to mass wasting phenomena. This thesis presents an inventory and thorough analysis of small–scale devastating landslides that occurred in recent years around the city of Limbe at the base of Mt Cameroon. A multidisciplinary approach was adopted in order to understand the complex and multi-phase nature of landslide processes. This involved integrating results of traditional field mapping, rainfall measurements, analyses of geotechnical properties, soil mineralogy and geochemistry to get an idea of the processes and mechanisms involved in landslide initiation and mobilisation. From these results, two conceptual models are proposed to explain the occurrence of landslides affecting areas populated by vulnerable people. The thesis thus provides first-hand field-measured geometric characteristics of landslide scars and systematically documents and quantifies for the first time the overall features of volume-limited devastating slides around Limbe. Hence, it provides constraints for modelling, monitoring and remediation efforts that might be applied in other areas affected by similar small-scale failures. Field observations, geotechnical and mineralogical characterisation suggests that slope failure within the study area is enhanced by human intervention through anarchical construction and slope undercutting whereas natural slope failure is generated as a result of fracture enhanced permeability. This implies that in the absence of external factors the slopes within the study area would generally remain stable for long periods. It also demonstrates that understanding landslide triggering processes and relationships between triggers and slide occurrence could be improved through more accurate recording of landslide events and the corresponding triggering process (intensity, magnitude and duration). This knowlegde can be used in landslide susceptibility assessment which was one of the most significant achievements of this thesis, and in the development of early-warning systems that might enable a reduction in the number of casualties and economic loss from future landslide phenomena. The findings from the present study have generic value and can be used to tackle small-scale slope instability problems elsewhere in the subtropics.