Why Is America’s Weather Model Falling Short of Europe’s?
The uncertainty of Hurricane Joaquin stirs up underlying problems
When comparing the United States’s weather models with Europe’s, the U.S. tends to come up short. Take Hurricane Joaquin. The U.S. models initially predicted the storm was headed for various locations along the East Coast, while the European model predicted a sharp turn eastward. And as we roll into the weekend, it looks as though the Europeans were right.
It’s not the first time. If you google “European vs American weather model” you’ll see what I mean. Back in 2013, a lack of snow when there was supposed to be a big storm led National Geographic to ask “Why Are Europeans Better at Predicting Weather?“
And a 2015 non-blizzard had The New York Times explaining “Blizzard Questions, Including Why a European Weather Model (Usually) Excels at U.S. Forecasts..” And of course, there was Hurricane Sandy, in which the European model correctly predicted a turn toward the East Coast, as Scientific American notes in their post asking “Are Europeans Better than Americans at Forecasting Storms?“
So in recent years at least, the answer to that last question is generally, “yes.” The European weather model has beat out the United States’s Global Forecasting System (GFS) in numerous situations. And now we may know why…
According to The New York Times, even though NOAA upgraded the GFS modeling system in January 2015, its higher-resolution still lacked the ability to compete with the European model.
That’s because just gaining more computing power won’t solve the underlying data problems the GFS faces. “The problems run deeper, all the way down to the description and modeling of the basic physics of radiation, clouds, precipitation and turbulence,” The New York Times reports. Here’s hoping it will get better with time.
Correction: this article originally incorrectly stated that NOAA upgraded the GFS in January 2016, which was temporally impossible and false. We have since updated the piece with the correct date and regret the error.