Turning ground movement into actionable infrastructure evidence
Geohazard monitoring must distinguish between recording an event, identifying its impact on infrastructure and predicting its occurrence. These are different objectives. Distributed acoustic sensing has demonstrated value in recording vibrations associated with rockslides and avalanches, but successful event recording should not be presented as proof that the same system can forecast when a slope will fail. Research in Switzerland and Norway shows both the potential of the technology and the importance of this distinction.
Existing fiber can reveal events across an infrastructure corridor.
During the June 2023 Brienz rockslide in Switzerland, researchers recorded the event using a DAS system connected to a 10-kilometre telecommunications fiber buried alongside a road and railway. The installation provided a dense record of ground-strain fluctuations across the valley. Its significance was the ability to observe a large geological event through infrastructure that had originally been installed for communication. It was not evidence that DAS had predicted the landslide or initiated the earlier evacuation.
An operationally focused example comes from NORSAR’s avalanche-monitoring work in Norway. A pilot installed at Holmbuktura during the 2022–2023 winter was developed to identify whether an avalanche had reached the road and whether vehicles were present in the affected area. That information can support road closure and emergency response, preventing additional vehicles from entering an already hazardous section.
The most valuable question may be: “Has the asset been affected?”
For a road, railway or pipeline operator, rapid confirmation of an affected section can be more actionable than a general indication of seismic activity. Monitoring requirements should therefore distinguish an event occurring somewhere nearby from an event intersecting a protected route.
This is also where the limitations become important. A 2025 study of avalanche signals in Grasdalen found that road traffic could interfere with automatic detection. Large vehicles and avalanches produced overlapping signal characteristics, and the researchers identified improved traffic-noise discrimination as necessary for reliable automation.
Development should focus on combining evidence.
A sensible development path is to combine distributed acoustic observations with the measurements needed for the specific hazard: for example, dedicated deformation monitoring, weather observations, cameras or established geotechnical instrumentation. The purpose would be to build a better-supported picture of asset exposure, not to imply that one sensor measures every failure mechanism.
A pilot should start with a defined decision: identifying a road impact, prioritising a railway inspection or locating a disturbed pipeline section. Evaluation should include known events, ordinary traffic and periods without hazardous activity. The resulting evidence can then support a proportionate expansion of the monitoring scope.
