# Why are there Two Hurricane Models, the European Model and the American Model when physics and statistics are the same everywhere?

278 views

Why are there Two Hurricane Models, the European Model and the American Model when physics and statistics are the same everywhere?

In: 293

The models take different things into account in different degrees. Weather is driven by a lot of randomness.

It’s more like predictions about a sports season. We all know the Yankees are going to do well, but one forecast will have them winning the division while another forecast will have them barely getting the wildcard spot.

Models as complicated as the ones used for predicting the behavior of weather systems are *extremely* complex, and for a “perfect” model would require an immense amount of processing power, time, and an unrealistic amount of data to feed into the model.

Thus, the only way to make the models functional is to make assumptions about the physics and use the models to provide a “best-guess” of a weather system’s behavior. If you have two different models with different core assumptions, then you can end up with different results.

The European model is generally a “stronger” model in that it makes less assumptions/more valid assumptions, but the cost of this is that it also requires a much more powerful computer to run than the American model.

If they jammed all of the European and American data together what would happen then?

Because one of them is based on the freezing and boiling point of water and the other one is based on a random mixture of ice, water and ammonium chloride **/s**

Because we are not able to make a complete model encompassing all of physics.

So we simplify and leave stuff out. And two different groups can do that and arrive at different models.

Physics is the same but we don’t have a full understanding of atmospheric physics nor the ability to get all the right measurements

We know that strong upper level winds hurt storm formation, but how much? How strongly does that interact with the other 24 variables like mid and low level winds? Once you’re beyond projectile motion in a vacuum and need to start factoring in other non ideal variables everything is an approximation

Within the US and European models there are multiple models. It’s not that each group made one model, various US groups made models and various European groups made slightly different ones so we bundle them up. Some models are really good at predicting the next 72 hours but less accurate after while others are great at predicting the 1 week track but less accurate for tomorrow. The actual forecast you see takes input from a half dozen different models and takes a best guess at the track from the inputs

Aren’t there many models? When I watch the news it shows many models at once. I looks like 10 models overlaying at once.

If I remember correctly!! The difference is the time between “data points”.
The US on average only take data every 6 hours and use these in their models.
The UK (Met) and EU takes data points every hour, so you get a slightly different outcome to models as you have a more fine grain set of data to put into your models.