More storms are developing into potentially devastating hurricanes, say researchers, who link the finding to increasing ocean temperatures.
They caution, however, that although hurricanes in the strongest categories — 4 and 5 — have doubled in number since the 1970s, longer-term data is needed to say conclusively whether climate change is to blame.
The term hurricane denotes a tropical storm (cyclone) that travels faster than 33 metres per second. Scientists classify them according to their speed, with category 5 hurricanes — such as Hurricane Katrina, which devastated New Orleans last month — being the fastest.
Cyclones can only form when the sea's surface temperature rises above 26 degrees Celsius. Because oceans have been warming up in recent decades, some climate scientists have speculated that hurricanes could become more common.
Research published today (16 September) in Science by a team led by Peter Webster of the Georgia Institute of Technology in the United States appears to confirm this.
Webster's team used satellite data to assess the numbers and intensity of cyclones that occurred between 1970 and 2004 in ocean basins worldwide.
They found that the number of cyclones did not change significantly over that period except in the North Atlantic. The results showed, however, that the number developing into powerful hurricanes increased, especially in the North Pacific, Indian and Southwest Pacific Oceans.
The increased number of North Atlantic cyclones in recent years and the rising sea surface temperatures there have led scientists to suggest that global warming is causing both phenomena.
Last month, a group of climate scientists led by Kerry Emanuel at the Massachusetts Institute of Technology, United States, published a paper in Nature, suggesting a link between the rise in hurricane intensity and sea surface temperatures in the Atlantic Ocean to the fact that average global temperatures rose by more than one degree Celsius during this period.
Emanuel told SciDev.Net that the research by Webster's team was "very much consistent" with that of his group.
Reference: Science 309, 1844 (2005)