Probabilistic stability evaluation and seismic triggering scenarios of submerged slopes in Lake Zurich (Switzerland)

Abstract

Subaqueous landslides and their consequences, such as tsunamis, can cause serious damage to offshore infrastructure and coastal communities. Stability analyses of submerged slopes are therefore crucial, yet complex steps for hazard assessment, as many geotechnical and morphological factors need to be considered. Typically, deterministic models with data from a few sampling locations are used for the evaluation of slope stabilities, as high efforts are required to ensure high spatial data coverage. This study presents a simple but flexible approach for the probabilistic stability assessment of subaqueous slopes that takes into account the spatial variability of geotechnical data. The study area (~2 km2) in Lake Zurich (northern Switzerland) shows three distinct subaquatic landslides with well-defined headscarps, translation areas (i.e. the zone where translational sliding occurred) and mass transport deposits. The ages of the landslides are known (~2,210 and ~640 cal. yr BP, and 1918 AD), and their triggers have been assigned to different mechanisms by previous studies. A combination of geophysical, geotechnical, and sedimentological methods served to analyse the subaquatic slope in great spatial detail: 3.5 kHz pinger seismic reflection data and a 300 kHz multibeam bathymetric dataset (1 m grid) were used for the detection of landslide features and for the layout of a coring and an in situ cone penetration testing campaign. The assignment of geotechnical data to lithological units enabled the construction of a sediment-mechanical stratigraphy that consists of four units, each with characteristic profiles of bulk density and shear strength. The thickness of each mechanical unit can be flexibly adapted to the local lithological unit thicknesses identified from sediment cores and seismic reflection profiles correlated to sediment cores. The sediment-mechanical stratigraphy was used as input for a Monte Carlo simulated limit-equilibrium model on an infinite slope for the assessment of the present slope stability and for a back analysis of past landslides in the study area, both for static and earthquake-triggered scenarios. The results show that the location of failure initiation in the model is consistent with stratigraphic analysis and failure-plane identification from sediment cores. Furthermore, today’s sediment-charged slopes are failure-prone, even for a static case. This approach of including an adaptable sediment-mechanical stratigraphy into a limit-equilibrium slope stability analysis may be applied as well to the marine realm.