AI provides debris flow warnings
Together with his team, D-BAUG Professor Fabian Walter has developed an innovative alarm system that can provide early warning of debris flows at Illgraben in Valais. For this, the scientists used seismic sensors and machine learning. The new AI detector increases the warning time by at least 20 minutes compared with existing systems. The study is the first milestone in a research collaboration conducted by WSL and Swisscom Broadcast in the early detection of natural hazards in Alpine regions.
Workshop discussion: New IT solution for faster warning of natural hazards
Video in german
TV broadcast from 26.03.2021 in the Swiss news program 10vor10
Video in german
Find ETH News article by Michael Keller
Original publication:
Chmiel M, Walter F, Wenner M, Zhang Z, McArdell BW, Hibert C: Machine Learning Improves Debris Flow Warning. Geophysical Research Letters (2021), external pagedoi: 10.1029/2020GL090874call_made
- Workshop discussion: external pageNatural hazards - innovative monitoring system successfully testedcall_made (Video in german)
- Fabian Walter is also a lecturer in an on-the-job training course of the D-BAUG for natural hazard specialists: CAS ETH in Natural Hazard Risk Management