The models used until today to estimate this risk include flaws, estimates a team of Franco-Québécoise researchers.
by
In December 1999, the Blayais region in Gironde bordered on the disaster during the passage of the Martin storm. Part of the nuclear power plant, located on the edge of the Gironde estuary, is flooded. Its emergency cooling system is affected. Two reactors are arrested urgently. Although the worst has been avoided, the event questions the protection of the five nuclear power plants located on the French coast. Several measures are then taken to prevent such a situation from happening again. Calculations of the risk of flooding are in particular studied again.
Now, the models used until today to estimate this risk include flaws, judge researchers from the National Institute of Scientific Research (INRS), the Institute of Radiation Protection and Nuclear Safety (IRSN ) and the Gustave-Eiffel University, in a study published in March in the Water Resources Research review.
The sea level is determined by two factors: astronomical tide and overout. The astronomical tide depends on the rotation of the earth and its gravitational attraction with the moon and the sun. Equations make it possible to determine its future or past value. However, the overcrowding is more difficult to anticipate. This is the difference between the water level observed and the astronomical tide. It is due to weather conditions such as wind or atmospheric pressure. To predict its magnitude, the idea is to rely on the values it has taken previously.
increased risk of flood due to climate change
In the case of French coastal nuclear power plants, protective measures are notably established according to the millennial surge, that is to say an exceeded value once every thousand years. However, the data recorded by the swamps generally offer too little perspective to make this type of calculation, estimates Yasser Hamdi, researcher at the IRSN and co -author of the study: “We use, for example, old -year -old data of thirty years To extrapolate over a thousand years. It makes no sense in terms of representativeness. “
Aware of the problem, the IRSN has undertaken for ten years a vast collection of information on the French coast. Helped by historians, the researchers traced, among other things, via the local press, the episodes of extreme sea levels of the last centuries. “If we have significant sea levels that cause floods and human and environmental damage, we are almost sure that they were listed in the press of the time,” notes Laurie Saint-Criq, doctoral student in coatural France-Québec between IRSN and INRS and first author of the study.
You have 28.93% of this article to read. The continuation is reserved for subscribers.