How millions of equations, physical data create weather models
Did you know meteorologists forecast the weather using a bunch of different physics and mathematical equations?
They use these equations to create what's known as numerical weather models to predict what's going to happen in the atmosphere.
Meteorologists normally just say 'models' on air, but what exactly is a model?
But we have to start with a quick history lesson. Back in the early 1900s, Lewis Fry Richardson is credited with creating the first numerical weather models using only physics and math equations to predict the weather for six hours in the future.
The caveat is that it took him about six weeks to complete these calculations by hand.
Fast forward to today, we have supercomputers that can do these calculations in a matter of microseconds. These are where modern weather models come from.
Now, how do they work? You have to think of the Earth as a grid. Models take these hundreds of thousands of equations and input all the different parts of the atmosphere, like heat, moisture, wind, just to name a few, for one section of the grid. Then, the process is repeated on millions of grid points horizontally and vertically in the atmosphere.
The models then take these millions of calculations and combine them with physical data from weather balloons, satellites, radar, etc. The combination of the millions of equations with the physical data of what's currently happening in the atmosphere is what ultimately creates the weather models.
What's the difference between the American model and the European model?
Now, diving a little bit deeper, you may have heard us mention the American vs the European weather model – one run by the U.S. and the other run by nations in Europe. But other than that, what's the big difference?
Going back to thinking of the Earth as a grid, the smaller the grid box, the higher the resolution. Just like a phone with more pixels creates a picture with better clarity, higher resolution in a model can show more detail with weather processes that may have been missed or poorly sampled.
The European model uses smaller grid boxes and has a higher resolution than the American model. Therefore, the European model is normally regarded as more accurate because it has a higher quality and is more detailed. But, the GFS is still useful and can pick up on phenomena the EURO cannot – like Hurricane Dorian.
And even though the EURO and GFS are two of the big models, there are so many other ones to look at as well.
If you're into statistics, you may have heard the phrase "All models are wrong, but some are useful," and that is exactly the case in meteorology as well.
Models do not represent the true atmosphere: it's up to meteorologists to identify when, where, and which models to use in specific situations to help us most accurately forecast the weather.
