Fibre-Optic Monitoring of Seismic Events from an Alpine Slope Instability: Insights into Spatial and Temporal Dynamics

Slope instabilities represent a significant hazard to communities and infrastructure across various regions worldwide. Climate change and resultant increasing severe precipitation events potentially raise the risk of failing mass movements. Therefore, a fundamental understanding of slope failure processes is vital for reducing risks. Established remote-sensing and synthetic aperture radar technologies provide valuable data on the surface movement of landslides, but only provide limited information on the instability’s internal state. In contrast, seismic imaging and monitoring techniques can provide critical complementary information on the subsurface structure, physical properties, and time-dependent processes linked to the slope instability dynamics.

The ‘Cuolm da Vi’ slope instability near Sedrun (central Switzerland) represents one of the Alps’ largest active landslides, with an estimated volume of around 150 million m³ and maximum displacement rates of up to 20 cm per year. While the instability currently does not pose an imminent danger, the slope's surface displacement is under constant observation. However, little is known about the Cuolm da Vi internal structure and dynamics at depth. The primary objective of our project is to advance our understanding of the subsurface structures and processes over time, with potential implications for deepening our fundamental knowledge of toppling instabilities in general.

In the summer of 2022, we established an extensive seismic observation network at Cuolm da Vi. This seismic sensor setup included over 1’000 autonomous seismic nodes and a 6-kilometer-long trenched fibre-optic cable. The fibre-optic sensing system was designed for long-term Distributed Acoustic Sensing (DAS) and Distributed Strain Sensing (DSS) observations. This multi-sensor geophysical network provides a unique spatial and temporal resolution for studying the Cuolm da Vi instability, allowing us to observe time-dependent changes across a wide range of spatial and temporal scales. Between summer 2022 and 2024, we gathered a comprehensive data set, including long-term continuous recordings from the nodal, DAS, and DSS systems.

Using a DAS dataset continuously collected from February to July 2023, we developed a wavefield coherence-based workflow to detect and cluster over 7’000 events recorded along the fibre-optic cable. These event clusters of highly similar seismic signals were manually classified into categories such as regional earthquakes, anthropogenic noise, rockfalls, and local seismic events, based on their time- and frequency domain characteristics. The spatial and temporal distribution of several local seismic event clusters exhibits distinct patterns that correlate closely, for example, with the surface displacement measurements. We are currently analysing these clusters of local events and investigating whether spatial links to known tectonic structures can be established, and whether the observed seismic signals allow refining the hazard scenarios and associated early warning strategies.