An example of damage sustained during the Aquila earthquake on 6 April 2009 in the Abruzzo region in Italy, here to a private home in the Roio Piano neighbourhood (Italy, 2009). © BRGM - Julien Rey

SURICATE-Nat platform on natural risks: data collection, information and prevention via social networks

The BRGM and the Troyes University of Technology, with support from the MAIF Foundation, have launched SURICATE-Nat, a platform using social networks as a new source of citizens' information to support natural risks vigilance.

The situation when a natural disaster strikes is typically very difficult to describe. There is always a critical phase when the major problem is to make use of as much information as possible from the ground to build up a realistic picture of what is happening.

The speed of people's reactions on social networks when they witness an event means they can almost act as "sensors".

A web platform to collect and make use of real-time reports on social networks

The SURICATE-Nat platform, with Version 1 launched in 2017 by the BRGM and UTT (Troyes University of Technology) with support from the MAIF Foundation, collects tweets on natural disasters from Twitter users and processes the data to provide information on risks and risk prevention. 

In practice, every "tweet" in French containing terms that refer to earthquakes is collected and automatically processed by AI algorithms. The messages are analysed to build up indicators of the intensity of phenomena as they unfold.  The messages are then sent out to internet users whose contributions to the analyses will complete and improve the predictive models produced.

Version 1 of the platform covers earthquakes. Version 2 will extend its coverage to floods, then to other natural hazards.

Developing risk awareness among citizens

The main goal is to involve "connected citizens" actively in natural disaster prevention and safety. The project therefore makes a signficant contribution to the development of risk awareness within the population.

Earth tremor detected within one minute thanks to social networks

On 28 April 2016, an earth tremor occurred at 8:46 local time at La Rochelle in France. By 8:47 it had already been detected from the first Twitter posts.

By 8:50, the flow of tweets had provided an initial estimation of the zone affected, with the first official notification from seismological networks issued at 9:08.

BRGM - 3 avenue Claude-Guillemin - BP 36009 45060 Orléans Cedex 2 - France Tel.: +33 (0)2 38 64 34 34