Original URL: http://www.theregister.co.uk/2011/01/07/met_office_forecast_not_forecast/

Met confirms secret Gov forecast of Brass Monkey winter

Didn't share it with us, though

By Andrew Orlowski

Posted in Government, 7th January 2011 11:18 GMT

The Met Office gave Government confidential weather advice that contradicts their published public statements, according to the BBC's Roger Harrabin. Harrabin says the Cabinet Office was told to expect "an exceptionally cold winter", and he sounds pretty certain about it. But back in October, the Met published maps showing a high probability of a warmer-than-average winter.

The Met vigorously denies the map, reproduced below, is a "forecast". However it has confirmed that it "provided a long-range forecast to the Cabinet Office at the end of October highlighting the risk of a cold start to the winter".

According to separate internal documents released under the FOI Act, saying two things to two audiences is advisable. The BBC News website published excerpts from several internal executive reports that bemoan the ability of the public to understand probabilistic forecasts. This resulted in "less 'intelligent' (and potentially hostile) sections of the press, competitors and politicos" conspiring to damage the Met Office "brand".

By contrast, people who the Met deems "interested customers" should be told the three-month outlook will be available on the research pages of the website. "'Intelligent' customers (such as the Cabinet Office) find probabilistic forecasts helpful in planning their resource deployment."

The Met recommended that three-month forecasts should be available on the research pages of its website but "this message should not be used with our mainstream audiences" [our emphasis].

Critics have called for a Parliamentary enquiry by the Science and Technology Select Committee. Why the fuss?

In October 2008, the Met predicted "a milder than average" winter, only for Britain to experience its coldest in 10 years. The following year, the Met predicted that "the trend to milder and wetter winters is expected to continue" – only for Britain to experience the coldest winter in 30 years. This winter has proved the coldest since records begin. Transport Minister Phil Hammond told Parliament he'll ask the Met if three cold winters in a row indicates a "steep change" in British climate.

In October, the Met responded to newspaper reports based on a probability map:

Most definitely not a forecast

It's not our fault, we need bigger computers

The map showed an above average likelihood of milder weather for the next three months. The Met says that this data requires "expert interpretation" and "the need to be combined with a range of other information before you can make a seasonal forecast" (Oct 28). After criticism by London Mayor Boris Johnson, the Met again denied making a forecast of a "mild winter" (Dec 20).

The Met announced that it would stop making long range public forecasts in March this year – so the controversy centres on what exactly on what was said to whom, and when. The latter should be easy to clear up – once we find out exactly what communications passed between the Met Office and The Cabinet Office. But the Met vigorously denies "forecasting" anything.

A probabilistic map suggesting the likelihood of warmer-than-average temperatures prompted a flurry of news stories in October – followed by a denial from the Met that it the map constituted a prediction or forecast.

Chief Scientist at the Met Office Julia Slingo used the opportunity to campaign for a supercomputer upgrade.

"It's quite clear that if we could run our models at a higher resolution we could do a much better job – tomorrow – in terms of our seasonal and decadal predictions," she told Nature blog last week. "It's so frustrating. We keep saying we need four times the computing power."

Last year Slingo said "hundreds of millions of pounds" of new computer kit was needed.

It must be frustrating, indeed. The Met Office only took delivery of its £33m IBM supercomputer in February, following a deal signed in 2008.

The same models are used to predict both weather and climate, Slingo told Parliament in March last year. ®