The debate over whether global warming is making hurricanes worse has been nothing if not stormy.

The issue came to a head in January, when leading US meteorologist Chris Landsea resigned from the Intergovernmental Panel on Climate Change, complaining that a colleague on the panel, Kevin Trenberth, had supported a link between warming and storms in a press conference. Now, just in time for the 2005 hurricane season, Trenberth has clarified his views in print (Science 308, 1753–1754; 2005). He argues that the intensity, if not the frequency, of hurricanes and typhoons will increase as the oceans warm.

Open season: the latest hurricane analysis will wind up those who say that worsening storms are not related to global warming.

The hurricane seasons from 1995 to 2004 have been far above the long-term average in terms of the number of storms and accompanying rainfall. However, most scientists are still extremely cautious about connecting hurricane activity and global warming. Since satellite detection started 35 years ago, there has been no detectable trend in hurricane frequency, points out modeller Kerry Emanuel of the Massachusetts Institute of Technology in Cambridge, Massachusetts.

But Trenberth, head of the climate analysis section of the National Center for Atmospheric Research in Boulder, Colorado, argues that because the number of hurricanes is relatively small, and fluctuates in cycles of various lengths, proving the existence of a trend from weather records is extremely difficult.

He has looked instead at how hurricanes form. “Trends in human-influenced environmental changes are expected to affect hurricane intensity and rainfall,” he concludes.

One simulation that Trenberth reviewed suggests that warming tropical oceans will stretch the upper limit of cyclones' potential strength.

“Most storms may actually not reach the limit,” says Tom Knutson, a co-author of the simulation based at the US National Oceanic and Atmospheric Administration in Princeton, New Jersey. “But in principle, Trenberth's conclusions are consistent with our studies.”

Trenberth also argues that higher sea surface temperatures in the Atlantic Ocean and increased water vapour in the lower atmosphere — caused by global warming — are to blame for the past decade's intense storms.

His conclusions will not please some in the meteorology community. In an upcoming paper in the Bulletin of the American Meteorological Society, Landsea, Emanuel and colleagues argue that there is no proven link between greenhouse-gas emissions and hurricane behaviour. They point out that even if there were a trend that had been missed in weather records, the change would have to be quite small relative to the year-to-year variability that already exists.

Trenberth counters that sceptics are ignoring the evidence. “I am trying to get people to think about things in a different fashion,” he says. “The point is that all meteorological events around the world are influenced in some way by global warming.”

In any case, everyone is hoping that there will be fewer severe storms this summer than last — when four strong hurricanes struck Florida, and Japan was hit by a record ten typhoons. They also hope that the storm that came ashore unexpectedly in Brazil on 28 March 2004 will not be a harbinger of things to come: Catarina, as it was christened, was the first ever recorded hurricane to develop in the southern Atlantic Ocean.