The Met Office and its seasonal problems
As Britain remains cold and snowy, an interesting little dispute has boiled up between the Global Warming Policy Foundation (GWPF) and the Met Office over the quality of longer-range weather forecasting.
And this is illuminated by documents obtained by the BBC under freedom of information from the Met Office. These shed new light on the problems faced by the Met Office in its public communications and the strategies it has adopted for tackling them.
The Met Office is under attack from the GWPF, for its "poor advice" on the likelihood of a harsh and cold winter.
The GWPF is drawing attention to a map published on the Met Office website in October which indicated that the UK was likely to experience above-normal temperatures in the ensuing three-month period.
For the GPWF, which is sceptical of the Met Office and other mainstream analysis of global warming, this is evidence of a Met Office tendency to under-predict cold weather and over-predict mild winters.
The Met Office replies that these maps, which feature in the scientific research section of its website, are probabilistic estimates of the chances of a range of outcomes and are not to be taken as simple weather forecasts that can be right or wrong. It tried to squash news stories in October that it was predicting a mild winter.
It should also be noted that, according to the Quarmby report on transport and winter resilience earlier this week, the Met Office did give "early indications of the onset of a cold spell from late November at the end of October".
This argument is linked to views about climate change, but part of the background is the major difficulty the Met Office has faced for some time over forecasting seasonal weather and conveying its views to the public.
It goes back to the well-publicised (and in due course much ridiculed) Met Office forecast of "a barbecue summer" for summer 2009, which turned out to be true if you use your barbecue for collecting rainwater. It became one of the wettest summers in the past century. The widespread derision that resulted left the Met Office feeling badly burnt (while the nation's sausages were not).
The documents we requested show that scientists within the Met Office were uneasy about the language of this prediction. One internal report states:
"The strapline 'odds on for a barbeque summer' was created by the operations and communications teams to reflect the probability of a good summer. Concern over the use of the strapline and its relationship to the scientific information available was expressed by the scientific community, who were not consulted prior to the media release."
The Met Office then resolved to use "more conservative terminology" in future. But its seasonal prediction for last winter was also awry, failing to signal sufficiently the long and severe cold spell.
An internal executive paper noted the impact as follows:
"Unfortunately, less 'intelligent' (and potentially hostile) sections of the press, competitors and politicos have been able to maintain a sustained attack on the Met Office ... The opprobrium is leaking across to areas where we have much higher skill such as in short range forecasting and climate change - our brand is coming under pressure and there is some evidence we are losing the respect of the public."
This report argued that one downside of the seasonal forecasts was that they remained on the website and could easily be later compared to reality. It said:
"One of the weaknesses of the presentation of seasonal forecasts is that they were issued with much media involvement and then remain, unchanged, on our website for extended lengths of time - making us a hostage to fortune if the public perception is that the forecast is wrong for a long time before it is updated."
In contrast it noted that the "medium range forecast (out to 15 days ahead) is updated daily on the website which means that no single forecast is ever seen as 'wrong' because long before the weather happens, the forecast has been updated many times."
The intense embarrassment over the seasonal calculations prompted the Met Office to rethink its approach to predictions for several months ahead. It stopped publishing a seasonal forecast for the UK for public consumption (although it added a rolling 30-day view to its main forecast page). Instead it decided to put probabilistic seasonal data on the scientific pages of its website where, in the words of a board paper, such figures can be "more targeted towards users who appreciate their value and limitations".
As another document put it, "'Intelligent' customers (such as the Cabinet Office) find probabilistic forecasts helpful in planning their resource deployment."
A communications plan in February 2010 instructed staff that "interested customers" should be told the three-month outlook will be available on the research pages of the website but that "this message should not be used with our mainstream audiences".
Met Office staff clearly feel the general British public find it difficult to cope with probabilistic statements. A board paper from September 2009 states: "Feedback from Met Office surveys suggests that users would rather receive a deterministic forecast."
It adds: "It is considered that the task of educating the UK public in interpreting probabilistic information will be neither a short-term, nor simple task." It compares this unfavourably with the apparently greater ability of the US public to grasp such material.
Its location and temperate climate mean that predicting the British weather is a tricky task - especially when your audience is members of the general British public, who don't like probabilities and who may not be the most "intelligent customers" or able to cope with understanding the uncertainty of the longer-range predictions. Thus the problem for the Met Office is not only the variability of the British weather, it's also coping with the intellectual capacity of the British public. That, anyway, seems to be the view within its staff according to these FOI disclosures.
Are they right? Your answers can either be expressed deterministically or probabilistically, according to your own taste.