Distributed Acoustic Sensing of debris-flow activity in the Öschibach torrent (Swiss Alps)

The slope instability of Spitze Stei supplies large sediment volumes that accumulate at the slope toe and are subsequently remobilized as debris flows and debris floods in the adjacent Öschibach torrent thus threatening the nearby village of Kandersteg, Switzerland. Since early 2020, continuous monitoring and preventive measures have been implemented in the area. While long-term monitoring has documented frequent torrential activity, the dynamic linkage between sediment supply from the rock slope and debris-flow activity in the torrent remains poorly constrained due to the spatial limitations of point sensors.

In summer 2025, we deployed a dense seismic array on the rock slope and interrogated an existing dark optical fiber running along the ~4 km-long Öschibach torrent using Distributed Acoustic Sensing (DAS). The DAS setup enabled spatially continuous strain-rate measurements at meter-scale resolution with a sampling frequency of ~600 Hz. For the three-month acquisition period, our aim is to detect and characterize debris-flow and debris-flood activity using DAS methods, supported by relative water-level time series and data from nearby seismic stations.

A catalog of possible debris flows and debris floods is generated leveraging an established pre-warning water-level increase threshold (set at 0.6 m), using moving average windowing and duration filtering. This discharge inventory was characterized using the DAS array, whose ~850 channels have been geolocalized with tap test, based on strain rate amplitudes visualized in logarithmic waterfall plots. Analysis of Power Spectral Density (PSD) for the corresponding DAS recordings reveals an increase in seismic energy at high frequencies (~20-40 Hz) concentrated on channels closest to the stream. Vertically offset waveform comparison plots demonstrate high coherence between DAS channels and wavefields recorded at the seismic stations, from which the apparent speed of seismic sources can be estimated. We also observe other coherent signals along the fiber, including mass movements from the Spitze Stei rock slope (e.g., rockfalls and granular flows), as well as local and tele-seismic earthquakes.

Our assessment of signal quality and coherence provides a basis for subsequent event detection, source location, and characterization using array-based methods, particularly during the event initiation phase. Our multisensor approach highlights the potential of DAS to provide spatially dense observations of torrential processes in steep Alpine catchments.