Weather forecasts - why so wrong?
Discussion
OK, so i admit I know very little about the complexities of weather and forecasts and things, but how come they sometimes get it so wrong?
Throughout yesterday there were weather flash bulletins on TV saying that they can be as certain as possible there would be lots of snow today, including in my area. But when i wake up this morning, no snow at all, just lots of rain, and milder weather.
Throughout yesterday there were weather flash bulletins on TV saying that they can be as certain as possible there would be lots of snow today, including in my area. But when i wake up this morning, no snow at all, just lots of rain, and milder weather.
Weather forecasts from the met office are paid for by businesses all around the country - and not cheap either. That giant powerful supercomputer wasn't just for fancy graphics on the BBC, it was for commercial reasons.
I can't imagine a business paying for something thats unreliable and poor quality, so tbh I reckon the weathermen get it right most of the time.
I can't imagine a business paying for something thats unreliable and poor quality, so tbh I reckon the weathermen get it right most of the time.
Compare it to a horse race... You can study the form, the going, etc and make a prediction based on the most likely outcome. Thats what the weather forecast is. There are so many variables that that can have a dramatic effect on the outcome that to be honest i dont think the met office does too badly 
I'm still waiting for the heatwave We were suppossed to get several years ago. it was all over the news so I booked a weeks holliday. Pissed it down every day.
not seen -14 here yet. in fact its quite mild.
wits
I have no faith in the weather forcasts so I only look at then when I am going somewhere etc etc. when looking at them in this way they are overwhelmingly wrong. I cannot remember a time when they have been right for that matter.
not seen -14 here yet. in fact its quite mild.
wits I have no faith in the weather forcasts so I only look at then when I am going somewhere etc etc. when looking at them in this way they are overwhelmingly wrong. I cannot remember a time when they have been right for that matter.
Given to me on good faith by a friend who deals with this sort of thing, so if it turns out im wrong blame him not me 
All forecasting is done by a supercomputer over in Japan that is the most powerful in commercial use, but still not powerful enough. Supposedly at its current speed level it can analyse all the parameters given to it from all the monitoring stations, and give an almost perfect forecast - however it takes 2 weeks to do this. In other words, it can "predict" the weather but by the time its done so the day it was predicting passed a fortnight ago. Therefore it cuts right down on the parameters in order to be able to give a forecast thats actually in the future, and thats why they are always wrong
All forecasting is done by a supercomputer over in Japan that is the most powerful in commercial use, but still not powerful enough. Supposedly at its current speed level it can analyse all the parameters given to it from all the monitoring stations, and give an almost perfect forecast - however it takes 2 weeks to do this. In other words, it can "predict" the weather but by the time its done so the day it was predicting passed a fortnight ago. Therefore it cuts right down on the parameters in order to be able to give a forecast thats actually in the future, and thats why they are always wrong
paulhol said:
Given to me on good faith by a friend who deals with this sort of thing, so if it turns out im wrong blame him not me
All forecasting is done by a supercomputer over in Japan that is the most powerful in commercial use, but still not powerful enough. Supposedly at its current speed level it can analyse all the parameters given to it from all the monitoring stations, and give an almost perfect forecast - however it takes 2 weeks to do this. In other words, it can "predict" the weather but by the time its done so the day it was predicting passed a fortnight ago. Therefore it cuts right down on the parameters in order to be able to give a forecast thats actually in the future, and thats why they are always wrong
I think Deep Blue (the chess computer that beat Kaspirov) does this, a quick search on Google reveals lots of the supercomputers round the World are employed to predict the weather.
The Met office now uses 'first production NEC SX-8 supercomputer in the world' which is currently at number 357 in the World supercomputer list. Oddly the Korean Met office is at number 16.
paulhol said:
Given to me on good faith by a friend who deals with this sort of thing, so if it turns out im wrong blame him not me
All forecasting is done by a supercomputer over in Japan that is the most powerful in commercial use, but still not powerful enough. Supposedly at its current speed level it can analyse all the parameters given to it from all the monitoring stations, and give an almost perfect forecast - however it takes 2 weeks to do this. In other words, it can "predict" the weather but by the time its done so the day it was predicting passed a fortnight ago. Therefore it cuts right down on the parameters in order to be able to give a forecast thats actually in the future, and thats why they are always wrong
In the UK at least, its dealt with by people who have seen lord of the rings 231 times and can speak klingon fluently, they meet around a table with a big map and predict completely the wrong weather.
Personally, I think it's a demise of the good old pressure chart, seeing what is where when it is and making your own predications. People need to be spoon fed these suns and drips from idiotic bints where as the good old understanding of a chart, your own microclimate and experience tells you what is actually going to happen. Having my life governed by pressure charts has led me to believe the reliablilty of my own forecasting skills, tomorrow, the surf at Whitsand bay will be chest to head 'n' half high on sets, it will be moderately clean with a bit of a W breeze, making it not so clean, it will be drizzly, a bit colder than you like but winter-warm front warmth, about 11 deg C, so not so bad and I'll be spanking it whilst you lot lie in.
The problem with weather forecasts is that the public expect to be spoonfed the exact information which is dependent on the microclimate. So, if they say you're going to get "a drip" or "a sun" they could both be right, it's up to you to understand your local area and know what's going to happen.
If you are unhappy with the dross punted out on BBC (for chrissakes) go to the met website and get a proper chart, know what it means and know what is LIKELY to happen. It ain't a science but the information has it's probabilities.
Stu
The problem with weather forecasts is that the public expect to be spoonfed the exact information which is dependent on the microclimate. So, if they say you're going to get "a drip" or "a sun" they could both be right, it's up to you to understand your local area and know what's going to happen.
If you are unhappy with the dross punted out on BBC (for chrissakes) go to the met website and get a proper chart, know what it means and know what is LIKELY to happen. It ain't a science but the information has it's probabilities.
Stu
We have a friend who (with a PHD) is a meterologist at the Met. The reality is that they are actually very accurate - but the weather systems are so complex that 100% accuracy can never be achieved.
The difference between a few inches of snow and a heavy downpour can be as little as half a degree - and that is small enough to be impossible to predict with any certainty.
The difference between a few inches of snow and a heavy downpour can be as little as half a degree - and that is small enough to be impossible to predict with any certainty.
Gassing Station | The Pie & Piston Archive | Top of Page | What's New | My Stuff