UKC

Yesterday on the BBC weather website:

Nottingham's forecast for Wednesday: 7°c, light snow
Nottingham Airport's forecast for Wednesday: 6°c, sunny intervals

Today on the BBC weather website:

Nottingham's forecast for Wednesday: 6°c, heavy rain (colder than predicted yesterday, so naturally the snow will now be rain), wind 21mph from the West.
Nottingham Airport's forecast for Wednesday: 9°c, unbroken blue skies, wind 10mph from the West.

Nottingham Airport is about 2 miles from Nottingham.

So who is to blame for such incompetence?

The Met Office who provide the raw data to the Beeb, or the Beeb, who then turn this data into their forecast?

In reply to Chris Harris:
non linear mathematics.
It's God's fault.

In reply to Chris Harris:

Nottingham Airport is the other side of the Trent though, it's a whole different weather system down there! Often when I drive over the bridge at Gunthorpe the temperature jumps a couple of degrees.

In the recent snowy weather there was a huge difference between Nottingham weather (Beeston area) and Hucknall weather.

Anyway, I always look at both and choose which one I like best

In reply to Chris Harris:

There's only a 70% chance of actually predicting the right weather for anywhere in the UK if I remember rightly.

In reply to The Green Giant:
Some graphs here. Seems to be something strange about 4 day predictions.

“BBC Weather made a perfect prediction within 5 days 57% of the time, which made it superior to both the Met Office (47%) and Weather.co.uk (34%).”

Looks like the BBC have the deluxe crystal ball.

http://www.whatprice.co.uk/weather.html

In reply to The Green Giant: What non-one seems to do is to take uncertainty into account (which all good scientists should when there is a significant level of uncertainty). Thus, the Nottingham eather forecast would more usefully read :

6-8 degs broken cloud cover (0/8-6/8), rain/snow showers

The wider the range, the more you know that anything could happen. Trying to make one uncertain prediction fit the whole day is just asking for trouble. It amazes me that the weather forecasters don't bound the best/worst limits

In reply to Chris Harris: I don't really see what this issue is here.

It is a forecast, based on opinion, estimation, and averages, not fact. It is then condensed into a few phrases, leaving room for interpretation and variation.

Personally, I don't think the differences are that huge between these two forecasts - there is no such thing as never or always in forecasts. Why would there be? They are, after all, only predictions.

In reply to nickinscottishmountains:

Scorchio!!!

In reply to nickinscottishmountains:

So you think heavy rain, 6°c, 21 mph wind is similar to sunny, 9°c, 10mph wind?

In reply to nickinscottishmountains:

I never questioned whether or not the forecasts will turn out to be accurate. What I questioned was two such different forecasts for 2 places a couple of miles apart.
So you're saying "room for interpretation" is the reason why one place has been forecast for heavy rain and the other sunny?

In reply to Chris Harris: No I am not saying that. A few miles can make a difference though, do you not think?

In your OP, Nott yesterday was predicted light snow, now predicted heavy rain. Not a huge change. In your OP, Nott airport was yesterday predicted sunny intervals, today predicted unbroken blue skies. Not a huge change.

I don't think there are huge changes there, especially as with 24 more hours of info, the forecast comed a bit more precise.

In reply to Chris Harris: Heavy rain and sunny were the predictions on different days for different places. why do you expect two predictions on different days for different places to be so similar?

In reply to Chris Harris:

How often does it snow at 7 degrees?

In reply to nickinscottishmountains:

I have no problem with the predictions for each place changing from day to day. The point is that 2 places so close together have such different forecasts for the same day.

In reply to nickinscottishmountains:


Very definitely, especially in mountain or coastal environments.

However, in a mildly undulating area of the East Midlands, I cannot see how anyone can defend a forecast that is 4 days in advance with such different weather at 2 spots about 2 miles apart!

In reply to nickinscottishmountains:

I'll try one last time. The issue is the difference between the forecasts for the 2 locations, not the minor change in forecast from day to day.

In reply to Chris Harris:

Well you may be getting forecasts for what to you are places close together but in reality the observations are miles apart. For example, Harrogate gets some of its forecasts from LBA but the cold weather payments are worked out from Church Fenton, about 40 miles apart.
Llanberis gets its forecast from Valley in Angelesy as well as its cold weather payments so only got one last winter but Capel Curig weather station which would give a more accurate forecast is ignored, and they got four paymnets.

Nottingham, put there into here and see the different places they get readings from;
http://www.netweather.tv/index.cgi?action=uk7dayx7;type=;sess=

Do the same for Sheffield and it gives you choices there as well.

In Harrogate I can have one sort of weather, Pateley and Brimham another, Skipton and Malham another, York another, and they are all in a 20 mile radius of me.

In reply to Chris Harris: But your OP goes on about incompetence. Where is the incompetence? Weather changes in a matter of miles. What is the big deal with two reports that reflect the change in a few miles?

In reply to ericoides:

Bueno Estente!!

that's the weather.. for now->
youtube.com/watch?v=nXr2lwYMXhc&

In reply to nickinscottishmountains:

I'll try to make it simple.

"Weather changes in a matter of miles". Correct. That is why accurate forecasting is so difficult. That is why they struggle to get the next day right so often. That is why a forecast for 4-5 days ahead should never be more than regional. That is why the forecast for 2 places so close together should be the same.

To have a system that throws up two such different forecasts for 2 places 2 miles apart is incompetent.

Or are you suggesting that they have done their homework properly, and have predicted that on Wednesday there will be a front sat somewhere around the Gamston/West Bridgford area of Nottingham, with rain to the west and sun to the east?

In reply to Chris Harris:
agreed- the local forecasts should be roughly consistent (why not smooth them out?)
a short descriptive summary eg wintry showers would be more informative, or even Dan's tablecloths and swiss cheese

In reply to subalpine:

How do you smooth out sunny and heavy rain?

In reply to Chris Harris:
Mate, I fundamentally disagree with the second statement of yours here; in the first (edited) one, you have admitted weather can change a lot in the space of a few miles so I am at a loss to see why you see this as incompetent.

I'm out. All the best buddy.

In reply to subalpine:

More information is available at the BBC weather pages and from the Met Office so you can see for yourself what the situation is. It helps to look at the larger scale weather maps. I think through tuesday wednesday and thursday a series of fronts are forecast to pass over the UK. It's not possible to say exactly when they'll arrive in Nottingham, but it certainly gives you and idea of what weather to expect over the next few days.

In reply to nickinscottishmountains:

even on a 4 day forecast?

In reply to Chris Harris:
What incompetence?

IIRC the raw data (from the met office) is fed into the customers own supercomputers and run through their model(s) ten's if not hundreds of times. Small variations in the starting conditions can lead to massive changes in the result (see choas maths).
Highly qualified people then look at the simulations and try to see what are the likely patterns from the varying results. There is no right or wrong answer. It isn't that simple.

“Nonlinear problems are of interest to engineers, physicists and mathematicians because most physical systems are inherently nonlinear in nature. Nonlinear equations are difficult to solve and give rise to interesting phenomena such as chaos. The weather is famously nonlinear, where simple changes in one part of the system produce complex effects throughout.”

http://en.wikipedia.org/wiki/Nonlinear_system

In reply to nickinscottishmountains:
The incompetance kicks in when the met office give the 'worst case' forecast as it regularly does for the west coast here... Often gale forecasts amount to nothing... I was sailing once when they came up with a F11 imminent forecast. Naturally I ran for shelter, checked with the coastguard where this wind was... for 24hours... and it never appeared.
My work depends on an accurate forecast so I look for accurate forecasts... I did meteorology as part of an environmental science subsidiary course at uni... so not particularly qualified in the subject, but I can read synoptic charts well enough to see when the met office are talking total b***ocks!
As for the 'worst case' output, one of the senior forecasters admitted on telly that they did this after the Michael Fish 'hurricane'

In reply to nickinscottishmountains:

The point I'm making is that because it's so hard to predict what will happen, when, and where, it's daft to make such specific predictions.

If they said "Warm and sunny early in the week, rain spreading from west to east mid week, probably on Wednesday, followed by showery and breezy weather", I'd be perfectly happy.

But they aren't saying that. They're saying "On Wednesday our multi million pound computer modelling system, is saying heavy rain in Nottingham and bright sunshine 2 miles away", which is patently absurd.

Yes, it can be quite easily raining in one place and sunny 2 miles away. Predicting that this will happen at a specified location, several days in advance is just daft.

In reply to Chris Harris:
Why do you still use the forecasts?

In reply to anonymouse: did postmanpat get ever get a response from the beeb about his weathermap colour query?

In reply to subalpine:
The one with the black and white cat, or the grumpy one who posts here?

In reply to Chris Harris:
Against all expectations, I can report that the weather at Nottingham Airport was pretty much the same as that in Nottingham today.

In reply to Chris Harris:

Yes but even if it's only regional (like, chop the country into 5 or 6 areas and give a forecast for that), then you're always going to get places sat just either side of the line. I grant you they could "smooth it out", so they dont just give 5 or 6 temperature predictions, but a range over the border areas. It's a bit harder to do with "sunny" and "broken cloud" though, isn't it?

Tim

By Nottingham Airport, do you actually mean East Midlands airport in Leicestershire?

Do you not think (presumably not) that the forecasts the BBC delivers are regional (rather than exact to a spot) and that the different post code for East Midlands airport as opposed to Nottingham town centre.

Frankly, you argument (or lack of) stems from the fact that two locations in different counties have different average weather forecast despite being relatively close (11 miles according to google maps).

This happens everywhere. On Radio 4 in the morning they may say snow showers in Scotland but generally bright and dry in England and Wales. Does this mean that those people 1 mile from the border in England should be surprised that their weather is the same as those 1 mile North of the border?

My mistake, it's actually in Derbyshire not Leicestershire. Everything else remains true, as it is still in a different county.

In reply to Philip:

Nottingham aiport is in Nottinghamshire

http://www.google.co.uk/maps?f=q&source=s_q&hl=en&geocode=&...

East Midlands Airport is somewhere different

In reply to Philip: No, I mean Nottingham Airport, in Nottinghamshire, about 3 miles from Nottingham city centre.

You're thinking of East Midlands Airport at Castle Donington.

In reply to Chris Harris:
Beaten to it by Ebygomm.

Nottingham Airport is a small local airport, mostly light airctaft/local flying club type stuff. Through the summer they get a gentle trickle of families with kids/suitcases etc rolling up for their flight to Benidorm.....

In reply to ebygomm:
I could have my own weather site. It wouldn't have very much in the way of detailed forecasts though, simply a statement saying "The weather tomorrow will be very similar than that of today" - It'd score quite well against the proper sites too!

In reply to Chris Harris:
WTF has the weather forecast got to do with getting out and enjoying yourself!

In reply to Philip:

Fail on all counts I'm afraid.

You're talking about the wrong airport, and the one you're talking about is in Leicestershire, not Derbyshire.

In reply to Philip:


That bit I agree with. A general forecast of regional weather - just what we need. Anyone with half a brain will realise that the chosen border is a bit arbitrary and that the snow/no snow boundary to the weather will probably be somewhere in the Borders area.
The example I quoted in my OP is equivalent to issuing a forecast saying sunny in Cornhill on Tweed & blizzards in Coldstream, which is patent nonsense.

In reply to Chris Harris:
Forecast for tomorrow in the Nottingham area (not exactly what you'd call long range in a challenging mountain/coastal area):

Nottingham will be sunny, and -1°c Thursday night
Nottingham Airport will heavy rain, and +4°c Thursday night