Skip to content
Watchberg
Articles

8 October 2026/5 min read

Triage, not alarm: where satellites fit in the Swiss hazard workflow

Switzerland already has a layered system for slope hazards. Satellite screening belongs at the front of it, as decision support for experts, not as a siren.

By Julian Walder

Switzerland does not lack a system for mountain hazards. It has one of the most developed in the world, and any satellite service that wants to be useful has to understand where it fits.

How the system works

Responsibility is layered. The cantons produce the hazard maps, intensity maps and hazard indication maps, co-financed by the Confederation, and publish them on their geoportals; a harmonised national dataset is being built. Cantonal natural-hazard offices run measurement stations, early-warning services and event registers, and advise the communes. Communes often own the local monitoring systems, with intervention planning coordinated by the canton; at Brienz and Kandersteg the municipality is the operator or client of the instruments on the slope. Infrastructure operators, from railways to cable cars and dam owners, run their own hazard management for their own assets. Engineering and geology consultancies do much of the specialist work on contract, including the design and operation of monitoring on individual slopes.

Below all of this sits the hardware. Ground-based interferometric radar measures a slope face every few minutes. Deformation cameras, GNSS receivers, total stations, inclinometers and seismic sensors add independent lines of evidence. Periodic laser scans and drone surveys capture volumes. And in Valais a network of about 90 local observers, people who know their mountain, is a formal part of the preventive layer; the alert that led residents of Blatten to leave their homes on 19 May 2025 came from one of them.

This is a system built to produce hours of warning on a slope it is already watching. Its weakness is the slope it is not watching.

The honest limitation

A 2026 review of collapsing mountains in a warming climate reaches a sober conclusion: even with long-term monitoring, the exact timing and size of a collapse remain hard to predict. The record supports it. At Blatten the ground radar arrived nine days before the end and did its job. At Kandersteg, where the Spitze Stei mass above the village has shown GNSS peaks of 70 centimetres per day, the warning target is 48 hours, from instruments on site.

Satellites play a different game. They revisit every slope on a fixed schedule, every 6 days for Sentinel-1 and every 12 days for NISAR, and they see millimetre-to-metre motion over whole regions. What they offer is early detection, measured in years to weeks: a slope has started to accelerate and deserves a closer look. What they do not offer is the final countdown. Satellite revisit is days; the final acceleration of a rock-ice mass plays out in hours. That phase belongs to ground systems, and no satellite service should pretend otherwise.

Satellites tell an expert where to look first. Ground systems tell the expert when to act.

What this looks like in practice

The Canton of Valais showed the workflow in 2026, without any dedicated product. Satellite analyses and field observations found anomalies and isolated collapses at the Oigschtchummun Glacier, a few kilometres from Blatten. The response was a laser scan on 9 May 2026, a fixed camera from mid-May, and the closure of a road and a trail. On 26 June the road reopened under an interferometric radar, a new Doppler radar and a traffic-light system. Break-offs of up to 10,000 cubic metres had been recorded since December 2025.

Satellite anomaly, expert review, ground instruments, operational decision. That chain is what Watchberg is built to feed, on every slope in a region rather than the few that happen to be on someone's desk.

What a Watchberg briefing contains

Each flagged slope produces a short briefing addressed to the responsible geologist or operator. It has five parts.

  • What changed: the slope moved from stable to accelerating in a stated period, with the rate before and after.
  • The evidence: which satellites, which dates, which processing, with the time-series plots, so that every number traces back to a dataset.
  • Confidence and gaps: how strong the signal is, whether it appears in more than one sensor and geometry, where winter gaps or layover limit the view.
  • What is exposed: the settlements, roads, rail lines, cable cars and dams inside a simple runout envelope below the slope.
  • The recommended next step: an expert review of the data, a drone or laser survey, or the deployment of ground radar, with the option of a hand-off to a monitoring provider.

The briefing is written for an expert and ends with a recommendation to that expert. It does not end with an instruction.

What it never does

Three lines are drawn deliberately.

Watchberg never issues a public warning. In Switzerland the decision to close a road, stop a train or ask residents to leave belongs to the canton, the commune and the operator, under their legal responsibility. Nothing Watchberg produces is phrased as an instruction to the public, and nothing is published to the public.

Watchberg never predicts the hour of a collapse. It reports a change in behaviour and a trajectory, with confidence. Timing in hours is the domain of ground instruments and the experts who read them.

Watchberg never replaces ground monitoring. It is the layer before it. The best outcome of a briefing is that a radar, a camera or a geologist goes to a slope that would otherwise have been seen too late.

All of this is projected. A pilot programme is planned for 2027, in which pilot experts receive weekly briefings for a region and a corridor and judge for themselves whether the triage is right, early enough and worth their time. The system will be measured by their verdict.

Sources

  • Triage
  • Switzerland
  • Workflow

Related

Join the 2027 pilot programme.

Watchberg is preparing pilots with hazard offices and operators in three Swiss cantons. If you carry responsibility for alpine slopes, let us show you what a weekly briefing would look like for your region.

Request a pilot briefingRead the articles

Pilot programme 2027. In discussion with hazard offices in three cantons.