The North American weather model GFS projects in successive starts temperatures of up to 47 ° C in the Porto Alegre region and highs close to 50 ° C in Rio Grande do Sul, central Argentina and the Uruguay at the end of next week.
It would be the greatest heat of all time, the “mother of all heat waves” with values that would decimate the maximum temperature records that have lasted for more than a century and with a recurrence of perhaps thousands of years. ‘years.
These extraordinary values, bordering on the unimaginable, have appeared on several sites and applications that offer computer-generated weather forecasts automatically from one or more models, without the intervention of a meteorologist.
First of all, it is necessary to understand what a weather model is. These are projections made by supercomputers that process billions of calculations from mathematical equations that are performed every determined number of hours from weather observations around the world from satellites, surface stations and the atmosphere by soundings carried out by meteorological balloons.
All this data is initialized in the computer which processes it and makes it into future simulations. The model is therefore a weather forecasting tool, raw data, and not the final weather forecast made by a meteorologist. Several times, meteorologists here and abroad have observed such simulations indicating unrealistic scenarios days before, such as multiple simultaneous hurricanes in the Atlantic that never materialized. Or blizzards in the United States with huge accumulations that never happened.
Numerical modeling technology has come a long way and is a revolution in weather forecasting, but imperfect.
There are many different weather models and they show different results. At the subscriber, for example, we offer data from models such as North American, Canadian, German, WRF and others.
The GFS model, which projected an unusually high temperature to an absurd level for the middle of this month, is an acronym for the Global Forecasting System and operated by NOAA, the National Oceanic and Atmospheric Administration, the government’s weather and climate forecast. American. United States agency.
So when we do a weather forecast, we are not just looking at a particular model, but a whole set of data. Around the world, meteorologists pay special attention to simulations of the European model ECMWF, which has a higher accuracy rate than the North American model.
This is what, based on processing data from polar satellites, was able to accurately predict the positioning of Storm Sandy that caused the disaster in the New York area in 2012, while other simulations failed. could not anticipate what was going to happen.
In the case of Rio Grande do Sul, the technical literature notes that the models exaggerate the high temperatures in the subtropics during the hot season, summer, while negative biases are observed during the cold months throughout the region (Menéndez , CG, and co-authors, 2010: Downscaling extreme month-long anomalies in south South America. Climatic Change, 98, 379-403, doi: 10.1007 / s10584-009-9739-3).
The model’s projections for seven to ten days in summer are less reliable than in winter, because the warmer and more unstable atmosphere favors more frequent and drastic changes in forecast, while in the cold season, trends of seven to ten days tend to be more regular and have fewer changes.
So what the model says today, January 7, for January 15, may be drastically different from what the same model says for the same January 15 on January 13 or January 14. What the GFS Model Has Designed
Round after round indicated the heat in levels that are absurdly high and beyond what might be conceivable for areas such as Rio Grande do Sul. Highs as high as nearly 50 ° C and up to 47 ° C for Porto Alegre have been reported.
Using a single point on the grid, in this case the capital of Rio Grande do Sul, the GFS model indicated during its departure from 12Z yesterday in Porto Alegre maximums of 41 ° C on 12.39 ° C on 13, 44 ° C on 14th and 46 ° C on day 15. Starting yesterday at 18Z, the GFS projected 42 ° C for the city on 12th, 37 ° C on 13th, 39 ° C on 14th and 42 ° C on 15th. In today’s 0Z round, the same pattern reported Porto Alegre 39 ° C on day 12, 42 ° C on day 13, 45 ° C on day 14 and 46 ° C on day 15. In On today’s 6Z cycle, the GFS projected 38 ° C on day 12, 37 ° C on day 13, 40 ° C on day 14, and 47 ° C on day 15.
That one of the world’s major existing and most widely used weather models – due to free access – projects such absurdly high values obviously lights a warning light for us meteorologists.
“Oops, it’s extreme, you have to watch it closely.” Even because, even in the territory of the absurd, the last few months around the world have seen events conceived as nearly impossible, such as the heatwave in the western United States and Canada in June 2021 which left 800 people dead. and with estimates of up to 1,400 deaths.
Seattle (USA) recorded 42 ° C, while the maximum average in June is 22 ° C, which would be equivalent, for comparison and illustration, to 50 ° C for Porto Alegre, since the maximum average in January is 30 ° C.
Normally, we wouldn’t even make this data public and comment almost ten days in advance because, as explained, that can change a lot of things.
Not to mention the risk that some media and part of the public misunderstand that MetSul is predicting a heat of nearly 50 ° C, which it does not, in an era of chronic misinformation and hyperbole across social media. such as Whatsapp and Internet portals.
There are times when questions from the public multiply, who see such values in various popular websites and mobile apps that use GFS, and people need to understand what they are seeing and have a professional word.
When we see the model indicate a maximum of 47 ° C for Porto Alegre and a temperature at the pressure level of 850 hPa (1,500 meters above sea level) of 31 ° C, we are obviously obliged to consider such projections with a mixture of skepticism, caution and attention. Skepticism because a maximum of 47 ° C would decimate not only the Porto Alegre record of 40.8 ° C from 1943, but also the Rio Grande do Sul record of 42.6 ° C of January 19, 1917 (Alegrete) and of January 1, 1943 (Jaguarão).
And don’t stop there. He would break Brazil’s record of 44.8 ° C on November 4-5, 2020 in Nova Maringá, Mato Grosso. Worse, it would be close to the official South American heat record recognized by the World Meteorological Organization (WMO / OMM) of 48.9 ° C in Ridavia, Argentina, on December 11, 1905. climatology, which could not be achieved. expect that in the worst case scenarios. (now unlikely) of a global warming of 4 ° C to 5 ° C according to projections for the year 2100.
Caution and caution because, even if the model exaggerates, there is a signal of extreme event risk. In this sense, as I explained previously, weather forecasting by meteorologists is not done using a single model. In my case and that of my colleagues at MetSul, we use several and the European model is generally the “true one of the scales” due to its high success rate.
By METSUL
“Amateur web enthusiast. Award-winning creator. Extreme music expert. Wannabe analyst. Organizer. Hipster-friendly tv scholar. Twitter guru.”