Geophysical imaging of shallow subsurface topography and its implication for shallow landslide susceptibility in the Urseren Valley, Switzerland

Abstract

Abstract Landslides and soil erosion are an ever present threat to water management, building construction, vegetation formation and biodiversity in the Swiss Alps. Improved understanding of the mechanics and causative factors of soil erosion is a key factor in mitigation of damage to Alpine natural resources. Recently, much progress has been achieved in the forecasting of landslides on Alpine slopes with a new generation of shallow landslide models. These models perform well in spatial predictions, but temporal control on the occurrence of shallow landslides is less successful. Realistic soil composition and geometry of interfaces are necessary input for better predictions. Geophysical methods have so far not been widely considered to obtain these parameters, in spite of their ability to cover much ground with high-resolution. In this study we successfully use such methods to derive adequate subsurface topography as input to dynamic spatially distributed hydrological and soil mechanical models. Trench, GPS, electrical resistivity tomography and ground penetrating radar data were collected, resulting in revealing images of the composition and geometry of past and future landslides. A conceptual model for the occurrence of local shallow landslides is derived, spanning from block-wise steady creep of detaching soil units to rapid sliding and downslope deposition of soil units via varying sliding planes. Significant topography was observed in the soil interfaces acting as sliding planes, leading to a more complex role of groundwater flow in the initiation of shallow landslides. Hydrogeologic models should be revised accordingly.