MO/BBC TMax Record Error - Was It Today?

[shedhead]

Forecast 20c for London today after the BBC News at 22.35hrs last evening, looks to have made 10c. I can't recall a max temp clanger of this magnitude inside 24hrs, can anyone else?

[Summer Sun]

Forecast 20c for London today after the BBC News at 22.35hrs last evening, looks to have made 10c. I can't recall a max temp clanger of this magnitude inside 24hrs, can anyone else?

The BBC / met office have failed for the whole week IMO John Hammond admitted the other day there forecast didn't go as planned and that just tops it off, I think cloud / fog amounts have been higher than they expected though I may be wrong

[Costa Del Fal]

I have a feeling the Meto/BBC like to ramp the temperatures too much in cold and warm spells. For example, in Summer when 28c seems more likely, they will say 30c probably because it sounds and looks better. Yesterday is a prime example. 20c was forecast in London yet only 17/18c was reached. Whilst 19c was the overall Max, im sure they said 20c just so it sounds better and gives an even better impression.

[AderynCoch]

It makes a massive difference to temperatures whether or not the sun comes out at the moment, and because it's been hard forecasting cloud amounts it's also been hard to nail the temperatures.

Today was the first sunny day here all week and the difference in feel was very noticeable. A contrast to yesterday's chilly gloom.

[Chris K]

I recall them saying earlier in the week in these scenarios it can be notoriously difficult to predict where the sun is going to poke through the cloud, as well as saying IF and WHERE it does come out, temperatures will respond the best. They may have been wrong on various occasions this week for some areas but in my opinion its justified and they have not 'failed'. They have been spot on for my area this week anyway.

[johnholmes]

I have regularly tried to explain the difficulty less than 24 hours out to predict cloud amounts in the situation that has been with us for the past week. Get the cloud breaking in the forecast when it does not in reality and your maximum temperature will also be wrong.

That said it is not too impressive that one of the 2 world class centres, Met's own suggestion(!), fail for the capital city in such a big way as reported above.

[J10]

I think the Met Office have done better this week generally than the GFS, which has been suggesting wall to wall sunshine this week.

In terms of temps clangers, I would say another bad one was at the end of Jan 2006, GFS was forecast temps up to 10c, when in fact due to misty conditions temps were around freezing for most areas.

http://www.wetterzentrale.de/pics/archive/ra/2006/Rrea00120060131.gif

http://www.wetterzentrale.de/pics/archive/ra/2006/Rrea00220060131.gif

I have regularly tried to explain the difficulty less than 24 hours out to predict cloud amounts in the situation that has been with us for the past week. Get the cloud breaking in the forecast when it does not in reality and your maximum temperature will also be wrong.

That said it is not too impressive that one of the 2 world class centres, Met's own suggestion(!), fail for the capital city in such a big way as reported above.

Why is it so difficult to determine cloud amounts?

[Chris K]

I dont expect a forecast to be more accurate just because its the capital city (stating the obvious I know). Plus the mentioned forecast was yesterday night. Im sure they updated as the next day progressed? If not then yes I agree that is a bit bad.

Edited March 16, 2012 by Bugganuts

[johnholmes]

Its a matter of very fine mesh outputs or watching sat piccs, the very latest skew-T ascents and very close watching of surface data.

Both require much effort and time to get it anywhere near correct. 24 hours ahead and model outputs of, temperature, depoint trajectories, humidy forecasts in the lowest 2000ft data is crucial.

Hence my comment suggesting some surprise that the premier forecast centre was, assuming the comment at the top is correct, that they got it so far out. I would have thought/hoped that at least on the morning if it was going wrong an amendment would have been issued.

[shedhead]

Not withstanding the current tricky forecast situation regarding next maxima, I'm pretty sure the MO have what they consider to be an acceptable degree of error/tolerance, irrespective of the synoptics. I'd be very surprised if that was +/- 4 or 5c to be honest, let alone +/- 9 or 10c, indeed the additional heat coming off some faces in Exeter this afternoon probably far outstriped values in London....

[Terminal Moraine]

The difficulties in predicting exactly where the cloud will break, and the implications for the weather at any one locality, are well illustrated by the following.

Buxton is at a similar altitude to my own site and is 20 miles north west, slightly more exposed to the north west than here and about equally exposed to the south west.

On Sunday the 11th we had a W/NW air flow across the Peak District. Buxton remained cloudy virtually all day with less than 1 hour of sunshine and a maximum of 7.8c; here the maximum was 14.3c with almost 10 hours of sunshine.

Yesterday with a westerly wind I recorded a maximum of 7.8c with just under 2 hours of sunshine early in the afternoon whereas Buxton had 6.6 hours of sunshine and a maximum of 13.3c. On the 14th there was no sunshine at all here and a max' of 6.0c but at Buxton the max' was 11.7c with 4.4 hrs of sunshine.

With such seemingly random cloud breaks giving widely varying conditions over such a short distance it's next to impossible to forecast within a few hours let alone a day ahead.

Edited March 16, 2012 by Terminal Moraine

[Geordiesnow]

I think the Met Office have done better this week generally than the GFS, which has been suggesting wall to wall sunshine this week.

Agreed with that, I also think the UKMO model handled this situation much better than our American friends, the UKMO predicted there will be alot of cloud and for alot of areas it has turned out this way. I did say in another thread that most of us, myself included would think when you look at the set up on paper, its going to be a very warm sunny week with perhaps some low cloud in western areas but even here as the high centres over the UK more, sunshine should become more widespread, in reality there was alot more cloud than what the GFS model and many of us on here thought there would be.

It is one of those set ups where the low cloud will either break up quickly and sunshine could be widespread or the cloud is reluctant to break up and it stays dull and chilly all day. Personally I think the BBC forecasts has been pretty sound but obviously there has been some errors regarding overestimating temperatures at times.

[knocker]

Isn't the formation and dissipation of Sc one of the more interesting and difficult areas of forecasting? JH will obviously know more about this than me. I'm not convinced that models are any use in this.

Edited March 16, 2012 by weather ship

[johnholmes]

yes I forgot to add that topography plays a large part in some situations.

One day in particular, I cannot remember which but I did comment about it on one thread, was quite definitiely overturning of the lower St underneath the inversion and at about Peak District height when cloud broke first over that area.

weather ship mentiones Sc cloud-that is a somewhat different situation to St but yes it is probably the least most progressive improvement in anything to do with meteorology. Its no better now than it was 50 years or so ago. Hugely complex even more so than some of the features I have mentioned regarding St.

Not withstanding the current tricky forecast situation regarding next maxima, I'm pretty sure the MO have what they consider to be an acceptable degree of error/tolerance, irrespective of the synoptics. I'd be very surprised if that was +/- 4 or 5c to be honest, let alone +/- 9 or 10c, indeed the additional heat coming off some faces in Exeter this afternoon probably far outstriped values in London....

its +/- 2C to my knowledge, 3C and over is considered an incorrect forecast

I would be very interested to have a copy of the late evening forecast shed if you could get one please?

Also if anyone had a forecast copy of what was forecast on their 03-0400 issue the next morning. I simply cannot believe that it would still be so far out.

Edited March 16, 2012 by johnholmes

[Interitus]

Don't know about the 20c forecast but the archived Met Office data output gave a forecast of 12c for Heathrow from the previous day.

You can download archived forecast data and actual readings from this site - http://data.gov.uk/m...ce-data-archive

Edited March 19, 2012 by Interitus

[johnholmes]

thanks for that link-a useful site to have