Meteorologists routinely tell us what next week's weather is likely to he, and climate scientists discuss what might happen in 100 years. Christoph Schar, though, ventures dangerously close to that middle realm, where previously only the Farmer's Almanac dared go: what will next summer's weather be like? Following last year's tragic heat wave, which directly caused the death of tens of thousands of people, the question is of burning interest to Europeans. Schar asserts that last summer's sweltering temperatures should no longer be thought of as extraordinary. 'The situation in 2002 and 2003 in Europe, where we had a summer with extreme rainfall and record flooding followed by the hottest summer in hundreds of years, is going to be typical for future weather patterns,' he says. Most Europeans have probably never read Schar's report (not least because it was published in the scientific journal Nature in the dead of winter) but they seem to be bracing themselves for the worst. As part of its new national 'heat-wave plan', France issued a level-three alert when temperatures in Provence reached 34 degrees Celsius three days in a row hospital and rescue workers were asked to prepare for an influx of patients. Italian government officials have proposed creating a national registry of people over 65 so they can be herded into air-conditioned supermarkets in the event of another heat wave. London's mayor has offered a 100,000 pound reward for anybody who can come up with a practical way of cooling the city's underground trains, where temperatures have lately reached nearly 40 degrees Celsius. (The money hasn't been claimed.) Global warming seems to have permanently entered the European psyche. If the public is more aware, though, experts are more confused. When the Intergovernmental Panel on Climate Change hammered out its last assessment in 2001, scientists pulled together the latest research and made their best estimate of how much the Earth's atmosphere would warm during the next century. There was a lot they didn't know, but they were confident they'd be able to plug the gaps in time for the next report, due out in 2007. When they explored the fundamental physics and chemistry of the atmosphere, though, they found something unexpected: the way the atmosphere—and, in particular, clouds—respond to increasing levels of carbon is far more complex and difficult to predict than they had expected. 'We thought we'd reduce the uncertainty, but that hasn't happened,' says Kevin Trenberth, a climate scientist at the National Center for Atmospheric Research and a lead author of the next IPCC report. 'As we delve further and further into the science and gain a better understanding of the true complexity of the atmosphere, the uncertainties have gotten deeper.' This doesn't mean, of course, that the world isn't warming. Only the biased or the deluded deny that temperatures have risen, and that human activity has something to do with it. The big question that scientists have struggled with is how much warming will occur over the next century? With so much still unknown in the climate equation, there's no way of telling whether warnings of catastrophe are overblown or if things are even more dire than we thought. Why do scientists like Schar make predictions? Because, like economists, it's their job to hazard a best guess with the resources at hand-namely, vast computer programs that simulate what the Earth's atmosphere will do in certain circumstances. These models incorporate all the latest research into how the Earth's atmosphere behaves. But there are problems with the computer models. The atmosphere is very big, but also consists of a multitude of tiny interactions among particles of dust, soot, cloud droplets and trace gases that cannot be safely ignored. Current models don't have nearly the resolution they need to capture what goes on at such small scales. Scientists got an inkling that somethin