Interpreting | 2010.01.26
Fair weather friends | Michael Fish
Long term forecasting is a science in its infancy. The BBC should stick with Met Office expertise
It\'s open season, it seems, on the Met Office . The BBC is reported to be weighing up its contract with the state-owned forecaster, and considering a new deal with an alternative company, which would sever a 90-year link with the broadcaster. In my many years working for the Met Office and appearing on BBC programmes, I and my colleagues saw little of the wrangling behind the scenes, but renegotiation of contracts happened on at least two occasions. The BBC would be foolish not to assess the way its money is being spent. They are not alone in needing to find ways to save pennies. But the weather is slightly different to any old contract – this is a vital service, upon which millions of pounds, and human lives, depend.
To my mind the Met Office is the most experienced and thorough forecaster in the world, home to the best and brightest brains in the business. Twenty four hours a day, 365 days a year, the Met Office delivers its information to Britain through the national broadcaster.
Forecasters for the BBC need to have two distinct but crucial parts. First, they must be fully qualified meteorologists. They need to be proficient in understanding and interpreting the data provided, to be able to adjust interpretations at a moment\'s notice. Second, and just as essential, they need to be able to communicate that information to the audience. This is tougher than it might seem: it means translating specialist data and language into accessible terms; it often means thinking on one\'s feet; and it demands a personality that people trust. It is no easy marriage of skills and the BBC is well served by a uniquely talented team.
The fog hovering over the Met Office has been deepened by press speculation that it has flunked its latest long-term forecasts. The talk of a " barbecue summer " and a "mild winter", it is insinuated, lay bare the Met Office\'s shortcomings. It is wrong to use long-term forecasts as a stick with which to beat the organisation. Predicting the weather for days ahead is a very different task to predicting an entire season. One is meteorology; the other is climatology. Producing forecasts for even one or two months ahead is a discipline in its infancy – and in my experience the Met Office has been reluctant to make public its long-term forecasts. The pressure to do so comes in large part from the very same parts of the media that now seek to discredit it, and the same parts of the media that so often exaggerate – or misrepresent – the information the Met Office provides.
Several years ago when I was still at the Met Office, I remember we produced a forecast for that December, pointing to a mild month of wet patches, with colder spells. Parts of the press managed to extrapolate that information into headlines that predicted a white Christmas. It bore almost no relation to the facts.
The Met Office, of course, should not be immune to criticism. But those who condemn it for wrongly predicting a "barbecue summer" would do well to revisit the data – measured over the three-month period, it was not the miserable summer that some imagine. And the mild winter? Well, it\'s not over yet. Let\'s wait until March, and measure the averages then, before we take the Met Office to task. My forecast? We\'re about to enter a period of mild weather.
For more information, please visit
http://www.guardian.co.uk/commentisfree/[...]air-weather-friends-met-office
Instagram is adding automatic in-app text translation - The Verge views: 1524
Here’s a ‘Brand-New’ Massive Multilingual Dataset for Machine Translation views: 1517