Forecast is grim for the Met Office, but are they at fault for missing Big Gail?

BY 6AM on Tuesday morning, the wild winds had already woken Mary O’Connor. There was a “horrendous noise” coming from the roof of her Possilpark tower block. Her bed, on the seventh floor, was shaking. The whole building was shaking. By 7am, the roof had blown off. O’Connor found this out when her grandson’s pal rang her up. “See your roof’s blew off, Mary,” he informed her in a cheery voice. “It’s lying up in Crowhill Street.”

Yet as gusts estimated at 100mph were laying waste to this pocket of north Glasgow – two streets were cordoned off, families evacuated, three fire crews attended, firefighters expressed astonishment that no one was killed – the Met Office’s warning was still at amber. It remained at the “be prepared” level, anticipating winds of between 70mph and 80mph until 8.15am, when it was finally upgraded to red.

This was not a blooper on the scale of Michael Fish’s 1987 pronouncement, a few hours before the worst storm since 1703 hit south-east England: “Apparently, a woman rang the BBC and said she heard there was a hurricane on the way… well, if you’re watching, don’t worry, there isn’t.” In fact the Met Office had been predicting severe weather since last Sunday. But somehow, in the fag end of the festive season, the message got lost.

So how reliable are our weather forecasters? And how much of our faith do they deserve? The Meteorological Office, to give it its Sunday name, is an offshoot of the Ministry of Defence with an annual budget of around £170 million, 1,700 staff around the world and a £30 million supercomputer capable of 1,000 billion calculations a second. Yet it appears to be less reliable in its predictions than a grandmother looking out of the window of her council flat. It predicted the “barbecue summer” of cool and dreary 2009 and the “mild winter” of 2010, the coldest in living memory.

According to Met weather forecaster Dave Britton: “We had been forecasting windy conditions since before New Year. By Sunday we had a yellow warning for very windy weather across Scotland. This was upgraded to an amber warning on Monday, predicting winds of 70-80mph, gusting up to 90mph. On Monday our weather advisors met with the Scottish Government’s emergency team and warned them about the strength of the winds. By Tuesday morning we updated the status to red to the areas most at risk when we had the information from overnight.”

The difficulty of getting precise advance information creates a dilemma for government, local authorities and businesses. Having been caught on the back foot in the big freeze of 2010, when the country ran out of grit, salt and snow ploughs, there has been been a move towards excessive caution. Before Christmas, in the week after Hurricane Bawbag, with further warnings of blizzards, floods and high winds, many public bodies and companies closed early, sent staff home and battened down the hatches. The predicted storm failed to materialise. Then there was last week’s almighty scramble as the winds gusted even higher than predicted.

“Neither the Met Office nor the Scottish Government can stop bad weather,” says Britton. “The best we can do is be as prepared as we can be.”

For science blogger Mike Haseler, this is not good enough. He thinks the Met Office’s “PR spinsters and IT modellers have taken over from time-served expert forecasters. What they offer is, he says, “defensive forecasting – forecasting what they know they can get right, such as, it will rain tomorrow,” rather than specifying when and where. “They make a forecast, then a few hours before the snow storm or wind, they change the forecast to reflect what is actually happening, and then the next day their spin doctors are everywhere telling us how wonderful their forecasts were, when we all know they got it wrong.”

This leaves government forced to over-react or take the flak, and the public confused about what to expect. “Computers should be there to help the experts, instead they seem to have taken over from the experts. Good graphics should help present the forecast; now the forecast seems to be all graphics, with no real information which allows us to make decisions.”

Haseler blames the London-centric weather forecasting system used by the Met Office for the inconsistent quality of their predictions. “So many of the forecasters are based in London, which is one of the lowest wind speed areas in the UK. Scotland has some of the highest winds. In London, the question is whether to take a brolly to work. In Scotland, the real question is whether the rain is going to be falling downwards, or sidewards ... and if it’s falling sidewards, it doesn’t matter if you take a brolly or not because it will be blown away.

“Scotland is by and large a country of hills. In the central belt, we are exposed to the south-west. So wind from the north-west or north-east doesn’t affect us, but when it blows from the south-west, even moderate winds result in wheelie bins dancing down the road.”

London, however, is built on a flat plain. “There are no rocky hills inside the M25. Any hills there are smooth, so wind direction hardly matter.” This means the whole region will get the same weather conditions. Not so in Scotland. “We get isolated pockets affected by particular local geography, creating complex local weather. We need very detailed local forecasts in Scotland whereas in London you can have a very broadbrush approach.”

The Met is investing in new radar technology that measures rainfall and wind speed. But shouldn’t we just learn to relax a little and accept whatever weather is thrown at us? “Anyone can predict it is going to rain in Glasgow and get it right most of the time,” Haseler says. “It doesn’t need several million pounds of super computer to do it.” «