Acoustic Emissions for the Prediction of Avalanches

Pierre de Guibert dguibert@cict.fr and Vincent Gibiat gibiat@cict.fr
Laboratoire PHASE (Université Paul Sabatier- Toulouse – France)

Popular version of paper 2pPAa6, "Acoustic emissions for the prediction of avalanches"
Presented: Tuesday Afternoon, July 1, 2008
Acoustics’08 Paris, France

Nowadays we know quite well the main zones where occur snow avalanches. The challenge is to control and prevent dangerous zones, not only for amateurs of winter sports in resorts, in or out tracks but also for inhabitants and users of the road infrastructures in the mountainous zones. Recent natural disasters such as cyclones and earthquakes remind us how nature can be cruel and above all unpredictable, or at least difficult to control.

Concerning avalanches, numerous studies were led not only to understand better the snow phenomenon, its evolution, transformation and its cape structure deposit according to weather conditions, but also to study the avalanche way while operating. We now know that a snow coat is made of numerous snow slopes and that the embrittlement of a slope causes the most dramatic avalanches (slab avalanches). Difficulty of prevention is due to the fact that the snow coat is constantly evolving and that it is often impossible to walk through dangerous areas in order to study and observe the potential snow coat failure.

Professionals and mountain specialists of natural risks have the knowledge of the ground and event history. As a consequence, they could give their opinion and recommendations about the potential risk. In agreement with the meteorologist forecasters, they act in the interests of security (artificial purges with explosives, road or ski resort closure, alert bulletins). The object of our study is to propose a more reliable method to prevent from disasters caused by avalanches.

In order to help mountain forecasters, PHASE laboratory has settled a system to study the snow slope instability (on the basis of Sommerfeld work [1,2] initiated in 1977). Instability creates stress, fissure and failure states generating acoustic emissions whose intensity, frequency and recurrence can lead to forecast avalanches. The aim is to detect in situ these precursors, that initiate a low frequency signal propagating in snow. We have settled an experiment in avalanche corridors to monitor and register this low frequency signal. It includes two geophones to detect the acoustic signals, a numerical acquisition system, one aerial of temperature sensors and a wireless connection to a PC deported. The whole system is controlled by ADSL from our office and real-time treatment can be achieved. By comparing the acoustic signal to temperature, evolution of the snow pack, we could define an alarm degree allowing them to secure the area. Generally specialists of nivology (science of snow) are studying the state of snow layers far from dangerous areas. They study the repartition of different snow flake states. On the contrary our system (including temperature and vibrating tools ) is located inside the avalanche channel on the ground and in the layers.

By studying the evolution of temperature variation and temperature profile in the snow-covered coat in the same sector, we can make a local snow study and follow the evolution of the snow in order to correlate it with audio signals, allowing us to give alerts that have confirmed the “intuition” of professionals of snow