Abstract:
Rainfall-induced landslides pose a common problem in areas with slopes steeper than the friction angle of the soil. A series of such landslides in North Switzerland inspired a detailed geophysical and geotechnical site investigation prior to a monitoring experiment. High-resolution 2D and 3D electrical resistivity tomography (ERT) was used to derive a detailed subsurface image, which was verified by direct penetration tests, boreholes and laboratory analysis of soil samples with respect to grain size distribution and plasticity. Resolution analysis of ERT configurations proved a combination of Wenner-, Schlumberger- and Dipole-Dipole data to be a reasonable compromise between measurement time and model accuracy. Furthermore, a statistical approach to reducing subjectivity in the interpretation of 3D resistivity models is suggested. Applying this classification scheme to field data yields a model in very good agreement with the geotechnical model. The 3D resistivity model is then interpreted quantitatively using laboratory data and a constitutive relation accounting for clay and silt contents. The dominant influence of saturation on resistivity predicted by this model is confirmed and exemplified during repeated surveys in a dry and a wet period. In wet summer 2004, a silty sand layer of high water saturation is confined between two less permeable layers, the sandstone bedrock below and a clayey sand layer on top. This layer may locally form an aquifer, which becomes rapidly saturated during heavy rainfalls and contributes to the risk of failure. The combined ERT and geotechnical survey helped to optimize the design of the forthcoming monitoring experiment and may be used as a guideline for the investigation of similar slope conditions. © 2006 Elsevier B.V. All rights reserved.
