Storm clouds are gathering over the long hot summer. Or so the Met Office says. Tom Wilkie looks at the difficulty of predicting our capricious weather
Tomorrow, according to the Meterological Office, the long, hot summer is finally going to break. Two days of showers and disorder are predicted for the British Isles, bringing welcome relief to a country that has sweltered for so long under unaccustomed sun. But even if it does break, the length of our hot spell raises a tantalising question: will global warming deprive the British of one of the staples of their conversation? Our weather for the past couple of months has not just been hot and dry, which is terribly un-British; it's been predictable, which is completely foreign to our experience.

The Continentals are used to this sort of thing: apart from short seasons of transition in spring and autumn, during the long summer the weather is hot and stays hot, while in the winter it is very cold and it stays very cold. For the French and Germans and Austrians and Italians, the best weather forecast for tomorrow is usually "very much like today". Hardly promising material for a conversation: no wonder the Continentals get excited about love and politics, for they have nothing better to talk about.

The glory of Britain's weather has been its variability: we do not need to trouble ourselves about politics or international relations when our weather offers endless scope for conversational gambits. Country vicars used to agonise over the question: "Will it be dry for the summer fete?" In the days before cheap package holidays to Majorca, families would anxiously consult past weather records when planning their next fortnight by the seaside.

The entire country took Michael Fish to its televisual bosom when, in October 1987, he so spectacularly failed to predict the worst storms to hit southern England in living memory. In front of the whole nation, he remarked that a woman had phoned up to say she had heard there was a hurricane on the way. There wasn't, he said. Ah, but there was. Within hours.

In the United States in the Sixties, the Weathermen were a militant revolutionary group. Few countries other than Britain make heroes of their weathermen and women.

Most importantly, the weather has served that greatest of British passions, emotional detachment from one's neighbours. Casual acquaintances, instead of having to talk about each other at the risk of revealing something of their personal lives and becoming emotionally more involved, could devote long periods to discussion of the climate. The unpredictability of British weather meant that one day could provide enough material for a short conversation: if more topics were needed, then variation over the past week and the futility of predicting the next week were readily available.

Now, even the grey accountants who scrutinise the spending of public money, the National Audit Office, are taking up the national passion. On Friday, the NAO will publish a report on the performance of the Meteorological Office. The NAO is tight-lipped about the contents, but a spokeswoman did concede that it will examine the accuracy of the Met Office's forecasts. Bizarrely, the Met Office is part of the Ministry of Defence, from which it was partly hived off as an "Agency" in 1990. The NAO report scrutinised the extent to which the Met Office is providing its users with value for money, following its transition to agency status.

Unofficially, it is believed that the NAO accepts Met Office claims that the forecasts delivered through television and radio are 90 per cent accurate (Michael Fish's 1987 debacle is past history). It is the more specific and the long-range reports provided to the military and commercially to businesses such as British Gas (which needs to know the weather in advance to predict likely demand) that are in dispute.

Academic experts on climate are in little doubt about the excellence of the work done by the Met Office and its associated Hadley Centre for Climate Prediction and Research. Professor Alan O'Neill, from the meteorology department at the University of Reading, described the Met Office as "one of the world's leading agencies".

But all this begs two questions: why has our weather recently adopted this settled, continental-style pattern? And is, say, Manchester's weather inherently less predictable than that of Munich? These questions are answerable; there is a third - is this a harbinger of global warming? - whose answer is less clear.

We in Britain happen to be sitting under a blocking area of high pressure at present. Normally, our weather progresses from west to east, bringing successions of rain and fine weather. However, occasionally, flows in the upper atmosphere take on a more north-south direction, which blocks the normal progression. The situation persists in part because the jet stream, the river of fast-flowing air about 25,000ft up which is one of the main energy conveyors round the planet, has moved to the north of us; its position is the main determinant of whether we get high or low pressure in summer.

According to Derek Hardy from the Met Office's information department, this high pressure regime will break as the sun wanes in the northern skies and our latitudes cool during the autumn. The trick is to predict when the pattern will break. In the great drought year of 1976, Mr Hardy recalls, "it happened with a vengeance on the August Bank Holiday Monday and then it hardly stopped raining for three months." That event was triggered by the remnants of an old hurricane coming across the Atlantic. Perhaps Hurricane Felix might do the same for us this year?

And there lies one of the reasons Britain's weather is difficult to predict. We are the first to be hit by the weather coming in off the Atlantic. "The people in Munich have the benefit of nearly 1,000 miles of data gathered upwind of them. We haven't," Mr Hardy points out. Fewer (because larger) ships cross the Atlantic these days, sending back less comprehensive data on the weather they are experiencing, which Manchester will get a couple of days later. Munich has the benefit of Manchester's experience.

But the basic problem is that, just as the wrong sort of snow falls on British Rail tracks, so Britain is the wrong sort of shape for the weathermen's computers. It is long and thin and hilly and nowhere is far from the sea. Both hills and sea are a major influence on the local weather. In Central Europe, the complicating influence of the sea is much, much farther away, but "Norway's weather is just as doubtful as ours," Mr Hardy remarks by way of consolation.

Though this autumn in Manchester may be unpredictable, researchers are increasingly confident about their predictions of the world's climate in the year 2050. Although the increase may not be quite as much as the most pessimistic forecasts once indicated, better computer models of the climate are still showing that the temperature will increase as a result of humanity's burning fossil fuels. On 10 August, according to Professor O'Neill, the journal Nature published "the best predictions [researchers] can make at present" and the evidence for a man-made global temperature change "is increasingly robust".

Global warming will disrupt patterns of agriculture around the world and, Professor O'Neill warned, "Britain is operating in a global economy, so we must be concerned about the global changes."

Perhaps then, we should not switch to sex and politics: the weather and the climate will provide an even better topic of conversation in future.