Is weather forecasting a pseudoscience?

[elmuzzy66.]

I live in Hereford, which has it's own micro-climate due to all the hills/mountains that surround us. So I don't really listen to the weather. Although Sunday/Monday is supposed to be around 30c. Niiiiiice!!

Same as over here in gibraltar. We got the sierra nevada mountains to the north, atlas reef mountains to the south. This plays havoc with the model.
Some people think that bad weather sends us a fax or gives us a call when its gona strike

 

[Valentino]

It's pretty damn good at predicting hurricanes and their paths! You can see patterns such as moving high and low points, but properly accurate weather like 'It will rain at 10 past 5 for 13 minutes on tuesday in a week" won't ever happen i don't think. On a large scale weather is very predictable, but when you zoom in it all changes rather quickly.

 

[me227]

The Saudi weather reporters are usually pretty accurate.

 

[[TW]Sponge]

Same as over here in gibraltar. We got the sierra nevada mountains to the north, atlas reef mountains to the south. This plays havoc with the model.
Some people think that bad weather sends us a fax or gives us a call when its gona strike

I just look out the window every now and again and decide what it's going to do before I go anywhere. It's the most reliable way!

 

[Castiel]

Predicting the weather is about as scientific as it gets. It's absolutely all about predictive modelling, which is pure science.

The trouble is there's a shedload of variables and the modelling is very complicated, so it's not nearly as accurate as some other sciences.

My brother (who used to design climate mainframes) says the computing power needed just for simple models is enormous and we do not have suffient capability as yet to accurately model long term climate let alone local weather which is more difficult apparantly.

There are hopes for quantum computing providing the raw processing power required.

 

[Judgeneo]

Hes not always right, it is the weather after all.

This is his summary of predictions his predictions in April and may, and an outlook to the summer. If you want his forecasts ahead of schedule, you have to pay for them

Also, warning in advance, hes a gigantic global warming sceptic

Note how he predicted the heavy tornadoes in the states, and although he got the dates wrong, the heavy rain in England recently (the met-office has been saying drought conditions).


If only he was a better talker.

 

[Morbius]

The trouble with weather forecasting is that you have a chaotic system. Any error in your initial variables blows up over time until your solution is completely different to reality.

Also, you're numerically solving the system over a discretised space, so you need vastly powerful computers to run simulations over large areas.

 

[kylew]

It IS a fake science. For all the ponce and data, they still can't get it right. There is just too much data, too many variables, too much unknown.

It's not fake science at all, it's just a prediction to the best of their ability with current technology, of how they think the weather will be over a given time period.

If the forecast is wrong, it just means things didn't happen how they predicted they would, based on the current weather paterns.

For example, it's like getting upset about a train arriving late, and calling liars and fakes. Times trains are due are a prediction based on a few simple facts that they can rely on. The distance between stations is fixed, so it's it's 100 miles from Station A to Station B and the train can average 100 miles in an hour, then you'd predict it'd take an hour to arrive.

 

[Brian8bit]

Any way the wind blows doesn't really matter to me.