Return to Website

Number Watch Web Forum

This forum is about wrong numbers in science, politics and the media. It respects good science and good English.

Number Watch Web Forum
Start a New Topic 
View Entire Thread
Re: What's the point of the Met Office

The issue raised in the Quentin Letts programme of the Met Office being pretty good at short-term weather forecasting but not so good at longer-term forecasting reminds me of an activity that Roger Harrabin tried to organise a few years ago called the "BBC Weather Test". This article by Harrabin from Sept 2010 describes the idea:

The idea was to compare the weather predictions for monthly and seasonal forecasts provided by various forecasters and determine who is the most accurate. Harrabin gives his reason for doing this as simple curiosity, but I would suspect that his main reason for wanting to carry out the exercise would be to see if he can use it to discredit long-range UK weather forecaster Piers Corbyn (brother of the current leader of the UK's Labour party, Jeremy Corbyn), who has for many years been a prominent AGW sceptic, and is also much more vocal in describing AGW as being a hoax than the average AGW sceptic. There is a tendency for long-range weather forecasters to be AGW sceptic, and they enjoy some credibility with the news media as at some point weather obviously turns into climate if the timescale is long enough.

The arrangement Harrabin appears to have set up looks rather overmanned - there is a 'steering group', and there seem to be about dozen academics involved rather than the one or two that you might think would be all that was necessary.

If this exercise was carried out objectively I think Piers Corbyn would actually come out as the top UK long-range weather forecaster. I had some familiarity with Corbyn's forecasts through following the Climate Realists blog (previously known as the CO2 Science blog) for several years, which heavily featured Corbyn's work in its blog posts (the blog is still online but has not been updated since 2013), and I came to the conclusion that while he was not as good as he claimed to be, he was moderately accurate and definitely better than the Met Office. However I suspect that the Met Office might come out better than the independent weather forecasters who provide the Daily Express newspaper with their notorious weather prediction headline stories every few weeks. But as Harrabin is running the exercise, there's bound to be some 'fiddle' where the Met Office somehow come out as the top UK forecaster.

Harrabin then provided an update of how the activity was progressing in Jan 2012:

In the update Harrabin announces the shortlist of organisations that will be invited to participate in the exercise. There has now been a major change in the types of weather forecast that are to be provided - there will be 1, 3 and 5 day forecasts and a seasonal forecast, but the Met Office don't want to provide a seasonal forecast (because they're not very good at long term forecasting). If the comparison exercise went ahead in this form the Met Office would come out of it much better, and might even be declared as the most accurate forecaster, because the exercise is now biased to short-term forecasting, and many of the independent forecasters specialise in longer-term forecasting. Harrabin also claims in the article that Piers Corbyn has already agreed to participate in the test.

A few weeks later the exercise seemed to fold up, as described in this blog post from the Climate Realists blog:

Most of the organisations in the shortlist didn't agree to take part, and Piers Corbyn (along with another participant) pulled out because he thought Harrabin was too strongly connected with the Green lobby.

Re: What's the point of the Met Office

Nice one, Dave

Re: What's the point of the Met Office

Perhaps they suspected that the results would be "homogenised", as Dave suggests.

"Homogenised." Such a nice euphemism for fiddled or fudged. I must see how it appears as such in a Thesaurus.

Still, whoever would have come out top, they win no prizes with me since the forecasts have been a problem for me this last week.
I have had to re-felt the garage roof and needed a couple of dry days. They were promised but I spent much of each day dragging the tarpaulin on and off. While they may hedge their bets with probabilities they certainly made no reference to the downpour and especially not the hail that fell. But as I understand it the Met office has been losing clients - allegedly since they modified their computer models to include the AGW model influences.
The BBC was one of the contracts they lost which adds something else to this Harrabin scheme.... any connection I wonder? Oh, and it is the BBC online forecasts I have been following so perhaps they have the most AGW of forecasts?

Re: What's the point of the Met Office

That's an interesting point J. In Harrabin's shortlist there are two weather forecasting organisations of substantial size, MeteoGroup and Metra, who could potentially act as replacements for the Met Office. The Met Office has lost its long running contract with the BBC, but the successor has not been announced as yet. It will be interesting to see if the new provider turns out to be MeteoGroup or Metra.

The other four weather forecasting organisations or individuals in Harrabin's shortlist - Piers Corbyn's Weather Action, Joe Bástardi, David King (not to be confused with the former UK Chief Scientist) and Positive Weather Solutions, are all small-scale forecasters who specialise in long-term forecasting and are probably all inclined towards AGW scepticism.

One of the four long range weather forecasters in the shortlist, Positive Weather Solutions (PWS), actually ceased trading only a few weeks after the shortlist was announced, as a result of adverse publicity inflicted upon it by another Green activist in the news media, the Guardian's George Monbiot.


Monbiot noticed that PWS was pretending to be a bigger organisation than it really was, in the form of inventing employees for its website who did not really exist. He is unlikely to be familiar with the business world, but it is not unknown for small businesses to do that. One of the humorous "Bluffer's Guide" series of books, "The Bluffer's Guide to Small Business", explains this issue as follows:

"Of course, the title 'Small Business' is something of a misnomer. No bluffer particularly wants to head a small business. He or she would probably prefer to own a big business. Alas, it is a great social injustice that few are able to go straight in at their preferred level. For most, the best way to own one's own big business is to start with a small one and make it bigger. And the best way of doing that is to convince everyone that the small business is already a big business.

Indeed, if a small business is to stand any chance at all, it is not merely desirable, but essential, that it appears to be what it hopes to become."

I believe the owner of PWS, Jonathan Powell, started up another company called "Vantage Weather Services" a few months later, but any 'reputation' that he might have been able to establish through PWS was ruined by Monbiot.