Dr Corell heads a team of some 300 scientists who have spent the past four years investigating the matter in a process known as the Arctic Climate Impact Assessment (ACIA). The group, drawn from the eight countries with territories inside the Arctic Circle, has just issued a report called 'Impacts of a Warming Arctic', a lengthy summary of the principal scientific findings. Scientists have long suspected that several factors lead to greater temperature swings at the poles than elsewhere on the planet. One is albedo (反照率)-the posh scientific name for how much sunlight is reflected by a planet's suce, and how much is reflected. Most of the polar regions are covered in snow and ice, which are much more reflective than soil or ocean. If that snow melts, the exposure of dark earth (which absorbs heat) acts as a feedback loop that accelerates warming. A second factor that makes the poles special is that the atmosphere is thinner there than at the equator, and so less energy is required to warm it up. A third factor is that less solar energy is lost in evaporation at the frigid poles than in the steamy tropics. Arctic warming may influence the global climate in several ways. One is that huge amounts of methane, a particularly potent greenhouse gas, are stored in the permafrost of the tundra. Although a thaw would allow forests to invade the tundra, which would tend to ameliorate any global warming that is going on (since trees capture carbon dioxide, the greenhouse gas most talked about in the context of climate change), a melting of the permafrost might also lead to a lot of trapped methane being released into the atmosphere, more than offsetting the cooling effects of the new forests. Another worry is that Arctic warming will influence ocean circulation in ways that are not fully understood. One link in the chain is the salinity of sea water, which is decreasing in the north Atlantic thanks to an increase in glacial meltwaters. Because fresh water and salt water have different densities, this 'freshening' of the ocean could change circulation patterns. The most celebrated risk is to the mid-Atlantic Conveyor Belt, a current which brings warm water from the tropics to north-western Europe, and which is responsible for that region's unusually mild winters. Some of the ACIA's experts are fretting over evidence of reduced density and salinity in waters near the Arctic that could adversely affect this current. The biggest popular worry, though, is that melting Arctic ice could lead to a dramatic rise in sea level. Here, a few caveats are needed. For a start, much of the ice in the Arctic is floating in the sea already. Archimedes's principle shows that the melting of this ice will make no immediate difference to the sea's level, although it would change its albedo. Second, if land ice, such as that covering Greenland, does melt in large quantities, the process will take centuries. And third, although the experts are indeed worried that global warming might cause the oceans to rise, the main way they believe this will happen is by thermal expansion of the water itself. Nevertheless, there is some cause for nervousness. As the ACIA researchers document, there are signs that the massive Greenland ice sheet might be melting more rapidly than was thought a few years ago. Cracks in the sheet appear to be allowing melt water to trickle to its base, explains Michael Oppenheimer, a climatologist at Princeton University who was not one of the report's authors. That water may act as a lubricant, speeding up the sheet's movement into the sea. If the entire sheet melted, the sea might rise by 6-7 metres. While acknowledging that disintegration this century is still an unlikely outcome, Dr Oppenheimer argues that the evidence of the past few years suggests it is more likely to happen over the next few centuries if the world does not reduce emissions of greenhouse gases. He worries that an accelerat