A day in the life of the Met Office’s London forecast

How the Met Office’s general forecast for London has changed over a 24-hour period.

24 hours ago

12 hours ago


If this is the amount of change the Met Office makes to a forecast in one day how can we possibly rely on longer range forecasts?  Perhaps they would do better to spend our money trying to get the basics right, you know, just a day or two ahead so we know what weather to expect before it arrives.

15 Responses to “A day in the life of the Met Office’s London forecast”

  1. 1 scud 15/01/2013 at 12:34 am

    kinda brings into question does it not the whole ‘reality’ of satellite weather mapping. One would have thought that if we were indeed able to look down from above upon our weather systems and able to photograph, track, estimate temperature, humidity, velocity etc, etc then the forecasts should be pretty bloody accurate…yet they are not.

    Where are most ‘satellites’ reckoned to reside? Within the Thermosphere…where temperatures regularly reach 2,500 degrees celsius not to mention the full monty of ionising extreme UV, X and Gamma waves…in other words, a place where no man made artefact could possibly function.

    There is more to this AM…much more.

  2. 2 Steve 15/01/2013 at 7:50 am

    There is a couple of degrees change in the 2-3 day outlook, and a difference in opinion on the amount of cloud. The Monday difference (between second and last) is because the last image was from part way through Monday after the rain had passed. So there isn’t that much change.

    It would be easier to find bigger differences, A rain band that is due to pass through the region may go through quicker or slower than expected leading to a change from sun to rain in one symbol and from rain to sun in the next. That’s the British weather for you.

    Given that the Met Office is one of the two best operational centres in the world and exports its model to the US, Australia, S Korea, India, S Africa…, you may struggle to get better from elsewhere. By paying their staff 2-3 times more than the Met Office (using EU money) and buying more computers, the ECMWF model is arguably better, but they do not fully test their model in a short term forecast system.

  3. 3 Flyinthesky 15/01/2013 at 7:00 pm

    It’s as obvious as the nose on my face, It’s the rise of Co2 that’s giving rise to this phenomenon: the inability to accurately predict the weather a couple of days in advance, however it does continue to postulate what’s going to happen in a hundred years with some inferred certainty.

  4. 4 dearieme 15/01/2013 at 7:29 pm

    The Science Is Settled, but the weather isn’t.

  5. 5 TomO 15/01/2013 at 7:55 pm

    The problem as I see it – and I do see it on a professional basis working on weather sensitive projects is that the Met Office is a bureaucratic sausage factory with no quality control.

    A bunch of people shoveling ingredients into a system they only see parts of – that homogenises the constituent parts and produces an utterly lacklustre product that’s snazzily branded.

    Seeing what’s coming is pretty straightforward most of the time, with confidence decreasing with time and the complexity of local weather. UKMO never seem to put confidence numbers on anything even vehement climate scaremongering.

    As far as weather forecasting is concerned problem is – the Met Office doesn’t seem to want to increase the density of physical measurements – something that in this day and age should be cheap and easy. They are a bureaucracy, and evidently going outside is frowned on as is looking out the bluidy window.

  6. 6 Steve 15/01/2013 at 9:43 pm

    “UKMO never seem to put confidence numbers on anything even vehement climate scaremongering.”

    The Met Office have put a lot of research into how to explain uncertainty to the general user of their forecasters. As a small example:


    And, for example, the seasonal forecast includes a lot of probability info:

    Click to access A3-layout-precip-AMJ.pdf

  7. 7 james higham 15/01/2013 at 10:33 pm

    Think it’s just meant for entertainment.

  8. 8 TomO 16/01/2013 at 12:39 am

    infotainment … please

    some time back I was involved in sanctioning significant payments on UKMO detailed offshore forecasts – I now shop elsewhere.. There’s an old documentary about the development of the Britannia field offshore Scotland where the forecasting only got accurate when significant pressure was applied and a forecaster was actually deployed to the construction site where he had to endure some of the consequences of inaccurate forecasting..
    Your links are to output from UKMO PR dept – I suspect strongly that even the existence of a Met Office PR department is an indication that all is not well in Michael Fish land. ..

  9. 9 Ian Hills 16/01/2013 at 4:03 am

    Bit of IPCC sauce –

    “The long-term prediction of future climate states is not possible”


  10. 10 Steve 16/01/2013 at 6:56 am

    TomO the second link is to the detailed output from the 3-month outlook including all the identified uncertainties and probabilities that you claimed the Met O do not do. The first link is the “PR output” to a simple but useful game identifying how people view different ways of publicising probabilities. But without the PR they wouldn’t have got thousands of people playing the game (it’s not a very exciting game).

    I suspect things have moved on since the Britannia oilfield was developed. Then, you would have got some poor Scientific Officer being paid 37p a day extra to risk his life on a dodgy oil platform. These days I suspect you’ll get a glossy brochure from all the weather companies and a price list for automatic forecasts, or someone on the end of the phone, or daily meetings with a forecaster or someone on-site 24/7.

    Michael Fish left the Met Office many years ago!

    PS. I noticed last night the Met Office are predicting a low possibility of snow in the SW on Friday. Accepting the fact that predicting the amount of snow 3 days ahead in this case is difficult, should they tell people, given it might turn out to be rain. Or should they keep quiet till they are sure. Posting the daily change in graphics could be entertaining, but for the few of us who have to plan carefully to avoid grim weather, understanding the forecast in more depth is important.

  11. 11 Anoneumouse 16/01/2013 at 10:07 am

    it would seem there are traces of horse DNA in the MET ORIFICE forecasts

  12. 12 umbongo 16/01/2013 at 2:10 pm

    That Piers Corbyn of Weather Action


    makes a decent living says all (or most of what) you need to know about the competence or otherwise of the Met Office.

  13. 13 TomO 17/01/2013 at 8:53 pm

    Britannia weather guy was onshore in portacabin on the construction site. .

    What nettled the oil company consortium beyond the inaccuracy was the offhand attitude to providing forecasts on which both lives and a considerable investment depended.

    Needless to say – the accuracy of forecasts improved out of all recognition when the “weather guy” was in the next office.

    As I’ve commented before one of the things that a non meteorologist weather watcher can pick up quite quickly – is that where there is sparse field data (i.e. observations) the software simulations can diverge dramatically from actual weather experienced.

    In terms of confidence numbers – it would be a useful exercise to map out computed vs. observed – a standard metric in a number of scientific and engineering disciplines……

  14. 14 Steve 18/01/2013 at 9:22 am

    TomO, it’s frightening but social media/internet efforts such as WoW and Twitter bring home the accuracy/inaccuracy of forecasts much more directly than it used to. The Met Office is judged by auditors on its day-to-day forecast performance at a wide range of locations. On average the errors should be spotted…

    For the record it snowed a bit in the SW. People in the hills got a bit snowed in, and in Wales and the Midlands where the previous warning was focussed it is worse. This would seem to be one example of a cracking forecast where the central prediction 2.5 days in advance pretty much hit the nail on the head.

  15. 15 Wood Bench Plans 23/03/2013 at 9:42 am

    Hi there would you mind stating which blog platform you’re working with? I’m
    going to start my own blog in the near future but I’m having a hard time deciding between BlogEngine/Wordpress/B2evolution and Drupal. The reason I ask is because your design seems different then most blogs and I’m looking for something
    completely unique. P.S Sorry for being off-topic but I had to ask!

Comments are currently closed.

Enter your email address below

The Harrogate Agenda Explained

Email AM

Bloggers for an Independent UK

AM on Twitter

Error: Please make sure the Twitter account is public.

STOR Scandal

Autonomous Mind Archive

%d bloggers like this: