Met office - wrong
Discussion
I know predicting the weather isn't an exact science - but why do the met office seem to get it consistently wrong.
For example - this is the current met office forecast for my area (i.e. no rain until ~4pm)
I just went outside however and it's raining.
I looked at both the met office live rainfall radar map and netweather live rainfall radar map and forecast and they show rain falling on my area right now. Netweather forecast rain all day - which looking at the rainfall radar I can well believe.
and
I can understand the met office not being able to predict isolated showers - but a huge slow moving swathe of rain like this must be easy to predict. The met office forecast doesn't even seem to be consistent with their own live rainfall data which shows it raining on my area - yet the forecast says no.
I looked at the live rainfall radar map before I went to bed last night and saw this huge band of rain sitting over Ireland. Just eyeballing the speed it was moving - I suspected it would hit us around 8am. Lo and behold.
I don't expect Back to the Future 2 levels of accuracy - but is 'broadly correct' and unreasonable expectation, especially since other sites seem to get it correct based off what is presumably the same data set.
For example - this is the current met office forecast for my area (i.e. no rain until ~4pm)
I just went outside however and it's raining.
I looked at both the met office live rainfall radar map and netweather live rainfall radar map and forecast and they show rain falling on my area right now. Netweather forecast rain all day - which looking at the rainfall radar I can well believe.
and
I can understand the met office not being able to predict isolated showers - but a huge slow moving swathe of rain like this must be easy to predict. The met office forecast doesn't even seem to be consistent with their own live rainfall data which shows it raining on my area - yet the forecast says no.
I looked at the live rainfall radar map before I went to bed last night and saw this huge band of rain sitting over Ireland. Just eyeballing the speed it was moving - I suspected it would hit us around 8am. Lo and behold.
I don't expect Back to the Future 2 levels of accuracy - but is 'broadly correct' and unreasonable expectation, especially since other sites seem to get it correct based off what is presumably the same data set.
Edited by Moonhawk on Sunday 3rd September 10:41
The weather is a chaotic system with many sources of influence.
It is a very fluid and constantly changing medium which is hard to predict.
The met office have a number of highly powerful computers that are constantly looking at current weather and making predictions about the future..... if I remember correctly it has something like 256 concurrently running timelines of the most likely future outcome that range over the course of a month.
As that month ahead comes to 2 weeks ahead, the possible futures are either eliminated or strengthened in probability, the computers then cut down to 64 likely possibilities.
Then as the future becomes on week ahead the computers update to 32 strong possibilities. And so on. Until the forecast comes down to 1 or 2 days in the future.
The computers by this time have compounded all the available data in to its best forecast, which the met office issues.
They also issue forecasts on many timescales but due to the number of possibilities that exist the further in to the future you go, their probability is less likely.
But if you compare how often the met office is correct to how often they are wrong, I think you'd find how good these computers are in their predictions.
The other thing to bear in mind is that forecasts are usually done as a general area coverage and not as a very local area. To predict what will exactly happen in a small local area raises the complexity of the prediction to a different level, and therefore the probability of being correct much lower.
(I'm no expert in this field, but I do remember a program not too long ago on the met office - which, being produced by the BBC, had the obligatory MMGW content thrown in to state how we were all going to die in the near future from floating icebergs with polar bear skeletons on them)
It is a very fluid and constantly changing medium which is hard to predict.
The met office have a number of highly powerful computers that are constantly looking at current weather and making predictions about the future..... if I remember correctly it has something like 256 concurrently running timelines of the most likely future outcome that range over the course of a month.
As that month ahead comes to 2 weeks ahead, the possible futures are either eliminated or strengthened in probability, the computers then cut down to 64 likely possibilities.
Then as the future becomes on week ahead the computers update to 32 strong possibilities. And so on. Until the forecast comes down to 1 or 2 days in the future.
The computers by this time have compounded all the available data in to its best forecast, which the met office issues.
They also issue forecasts on many timescales but due to the number of possibilities that exist the further in to the future you go, their probability is less likely.
But if you compare how often the met office is correct to how often they are wrong, I think you'd find how good these computers are in their predictions.
The other thing to bear in mind is that forecasts are usually done as a general area coverage and not as a very local area. To predict what will exactly happen in a small local area raises the complexity of the prediction to a different level, and therefore the probability of being correct much lower.
(I'm no expert in this field, but I do remember a program not too long ago on the met office - which, being produced by the BBC, had the obligatory MMGW content thrown in to state how we were all going to die in the near future from floating icebergs with polar bear skeletons on them)
In their defence, it is quite tough to predict the UK weather due to our position.
This might help explain it
http://www.bbc.co.uk/schools/gcsebitesize/geograph...
However I do agree that the Met are not the best at forecasting. I usually use other sources for long term forecasts.
This might help explain it
http://www.bbc.co.uk/schools/gcsebitesize/geograph...
However I do agree that the Met are not the best at forecasting. I usually use other sources for long term forecasts.
A forecast uses models to give probabilities of certain types of weather. Thresholds are applied to those probabilities to put the simple types of weather on a typical forecast. So for the rain instead of cloudy symbol some sites will use 40% chance and others 50%. They probably aren't going to update a forecast for current time as you can just look outside.
Atomic12C said:
The other thing to bear in mind is that forecasts are usually done as a general area coverage and not as a very local area. To predict what will exactly happen in a small local area raises the complexity of the prediction to a different level, and therefore the probability of being correct much lower.
But I wasn't complaining that they hadn't predicted a sharp localised shower - the rainfall they failed to predict was a blanket that covered most of the north west and most of wales as shown on the rainfall radar screenshot I posted in the OP. It rained almost constantly from around 8am and was still drizzling even at 11pm.Moonhawk said:
Atomic12C said:
The other thing to bear in mind is that forecasts are usually done as a general area coverage and not as a very local area. To predict what will exactly happen in a small local area raises the complexity of the prediction to a different level, and therefore the probability of being correct much lower.
But I wasn't complaining that they hadn't predicted a sharp localised shower - the rainfall they failed to predict was a blanket that covered most of the north west and most of wales as shown on the rainfall radar screenshot I posted in the OP. It rained almost constantly from around 8am and was still drizzling even at 11pm.
p.s.
I know what you mean though, when they get it wrong, sometimes its so obviously wrong we are left wondering what a complete bunch of plonkers they are.
You'd also think that given the amount of computing power they have, that getting it wrong should be few and far between. But maybe on the other side of the fence it just highlights how chaotic the weather system is.
Moonhawk said:
I know predicting the weather isn't an exact science - but why do the met office seem to get it consistently wrong.
For example - this is the current met office forecast for my area (i.e. no rain until ~4pm)
I just went outside however and it's raining.
...
Surely that suggests 10% chance of rain increasing to 40% chance at 16.00. For example - this is the current met office forecast for my area (i.e. no rain until ~4pm)
I just went outside however and it's raining.
...
Edited by Moonhawk on Sunday 3rd September 10:41
Nothing about no rain, then rain.
I watched an American news report on the path of hurricane Irma earlier and they showed a graphic of the predictions of its path from different forecast organisations - there were about 20 of them!
Interestingly the one the reporter seemed to put the most weight on was "the British" one - although he wasn't specific about who it was.
Interestingly the one the reporter seemed to put the most weight on was "the British" one - although he wasn't specific about who it was.
Sorry OP, in my opinion you're very wrong. I've worked outside for around 35 years and the accuracy of the forecasts from the MET is amazing compared with when I was young.
When I worked on a fishing boat in the early eighties, you'd get a vague idea whether there was likely to be a storm the following day, probably just as accurate as looking at the Western horizon the night before and they got it spectacularly wrong far too many times.
Nowadays, if you use their local 5 day forecast, it's accurate enough to the individual hour to help me to plan my working day if I need to avoid rain, wind and frost.
Of course it's not an exact science yet but it's pretty reliable.
When I worked on a fishing boat in the early eighties, you'd get a vague idea whether there was likely to be a storm the following day, probably just as accurate as looking at the Western horizon the night before and they got it spectacularly wrong far too many times.
Nowadays, if you use their local 5 day forecast, it's accurate enough to the individual hour to help me to plan my working day if I need to avoid rain, wind and frost.
Of course it's not an exact science yet but it's pretty reliable.
brrapp said:
Sorry OP, in my opinion you're very wrong. I've worked outside for around 35 years and the accuracy of the forecasts from the MET is amazing compared with when I was young.
When I worked on a fishing boat in the early eighties, you'd get a vague idea whether there was likely to be a storm the following day, probably just as accurate as looking at the Western horizon the night before and they got it spectacularly wrong far too many times.
Nowadays, if you use their local 5 day forecast, it's accurate enough to the individual hour to help me to plan my working day if I need to avoid rain, wind and frost.
Of course it's not an exact science yet but it's pretty reliable.
that differs from my experience . i spend every spare moment on the coast. in terms of wind strength ,direction and rain i would say it is 50/50 24 to 36 hours out(mostly north sea between the forth estuary and aberdeen plus a fair few days in the solway firth ). i have quite a few commercial skipper/owner mates, none of them use the met office for forecasts.When I worked on a fishing boat in the early eighties, you'd get a vague idea whether there was likely to be a storm the following day, probably just as accurate as looking at the Western horizon the night before and they got it spectacularly wrong far too many times.
Nowadays, if you use their local 5 day forecast, it's accurate enough to the individual hour to help me to plan my working day if I need to avoid rain, wind and frost.
Of course it's not an exact science yet but it's pretty reliable.
it is the computers that are the problem imo. the numbers going into them are virtually meaningless any more than a few hours out, similar issues of scale that the climate models see. short terms forecasting (days) seemed to be far more accurate back in the day from my perspective.
Unless you are having a wedding or judging on a life and death experience does it actually matter if right or wrong?
I went outside and got wet, I then went inside and cuddled up to my keyboard and PH. Panic over !
I spend a lot of time outside dog walking and check the rainfall radar and the daily forecast. I sometimes get wet, sometimes not. If I don't put on the right clothes then my fault.
My dog is naked apart from his collar, never complains about the Met office, ignorance is bliss. Something to learn from ?
I went outside and got wet, I then went inside and cuddled up to my keyboard and PH. Panic over !
I spend a lot of time outside dog walking and check the rainfall radar and the daily forecast. I sometimes get wet, sometimes not. If I don't put on the right clothes then my fault.
My dog is naked apart from his collar, never complains about the Met office, ignorance is bliss. Something to learn from ?
PS Modern day forecasts are very good, you just get used to them being fine and then spot them being wrong. Like driving a Morris Marina and then spotting them everywhere ( with a wave to them of course).
The other thing is that the UK is located with very different weather patterns over the year. If you go to the SE Asia tropics it might just be
Sunny in the morning, clouding over, showers late afternoon. Muggy. Repeat.
A FECKING CYCLONE
The other thing is that the UK is located with very different weather patterns over the year. If you go to the SE Asia tropics it might just be
Sunny in the morning, clouding over, showers late afternoon. Muggy. Repeat.
A FECKING CYCLONE
Edited by Gandahar on Wednesday 6th September 14:32
Gassing Station | Science! | Top of Page | What's New | My Stuff