Weather information. Will it be really (very) cold next week?
After the relative mildness of the past few days, temperatures will drop next week. In this context, excerpts from the North American weather model have been widely distributed on social networks.
They indicate a severe cold season in France, with temperatures dropping below -10 or even -15°C, with no daytime thaw. In the question? A depression in the Norwegian Sea resulting in a cold drop that will plunge Western Europe into these freezing temperatures.
An isolated “model”, not a prediction
But as Météo-villes.com reminds us, “a cold snap is a phenomenon that is difficult to predict by models” – especially a week before the deadline. First of all, this American model seems isolated compared to all other meteorological models, especially European models.
Generally, they are looking at seasonal temperatures, if not slightly below normal.
It will be colder next week
According to Météonews meteorologist Frédéric Decker, “we’re heading for a fairly significant drop in temperatures this weekend. From Monday we’ll be back to normal, even a little below,” he explains.
Talk: An anticyclone in the British Isles, then in the North Sea, which will cause dry cold. “Obviously we’re not going to be in the numbers predicted by the American model, but we’re going to have frost all over the area over the next week, Mr Decker predicts. Due to clear weather at night, it sometimes reaches -7 or -8°C locally. »
However, according to the meteorologist, “there will be no melting zone during the day”. Therefore, the weather will be cold and dry: with few exceptions, there will be no precipitation, either rain or snow.
A badly “regulated” American model?
“American models often go to extremes,” says Frédérick Decker, noting that “certain parameters are not well-tuned” for reliable predictions about our continent.
Professional meteorologists, who are increasingly used to seeing circulating outputs from models presented as reliable forecasts, warn: they “do not constitute a weather forecast”, as Météo France reminds us.
“We see extreme cards like this very regularly”
How can one model differ from others?
Christophe Mertz, meteorologist at MeteoNews: “The models differ over a long period of time because their initial conditions, geographic areas, and parameters are not exactly the same,” explains Christophe Mertz, meteorologist at MeteoNews. Such a difference occurs very regularly after seven days. gives, so it can be predicted quite low.
As the days pass, as the deadline approaches, most models converge in one direction or another, or toward a “median” scenario between different scenarios, and then reliability increases. »
What are these predictions worth?
CM: “It’s a pity that such maps generated from models are distributed in a ‘raw state’, because it is not a prediction and the networks catch fire very quickly. Meteorologists see such extreme maps very regularly, and if we include them in communications or forecasts, they no longer make sense. »
Do we still have to wait for a cold frost?
CM: “The risks of cold waves modeled by the GFS are very low. We’re more likely to head for slightly cooler weather than these days, with clearer skies from the northeast and a continental flow that will bring the return of night/morning frosts, but by no means a real cold snap. . »
The Weather Channel, for its part, also heralds colder weather next week, while noting the “differences” between weather models.
It presents the Météo Consult model as “more reasonable” in terms of the expected cold intensity.
