First author

Tropical cyclones go by many names — hurricanes, typhoons and tropical storms, as well as human names such as Katrina and Albert. Worldwide, there is an average of 86 such storms a year, more than half of which reach or exceed hurricane wind strength of 119 kilometres per hour. During the past 30 years, tropical cyclones over the Atlantic have strengthened, and some climate researchers have speculated that global warming may be to blame. Climatologist James Elsner of Florida State University in Tallahassee and his colleagues studied 26 years' worth of satellite data, and their findings may give more power to this argument (see page 92). Elsner tells Nature why stronger tropical cyclones need more than just hot air.

How was global warming implicated?

Kerry Emanuel discussed the possibility in a 2005 Nature paper (Nature 436, 686–688; 2005), which made a lot of noise both in the scientific community and elsewhere. The standing theory on what drives a tropical cyclone is the 'heat engine' theory, whereby warm surface water evaporating from the ocean provides the 'fuel' that ignites the storm, which spirals similarly to a rotary motor. Kerry linked the increase in storm intensity to rising sea-surface temperatures — as waters warm, there is more energy available to convert to tropical cyclone wind.

What did your results show?

We speculated that if the theory is correct, you might not see a trend in the average intensity of tropical cyclones as oceans warm, but you should see a trend in the strongest storms. Our breakthrough came by thinking of the problem in terms of the subset of storms closest to their maximum potential intensity. We used satellite observations to create a consistent data set of storm intensity for all tropical cyclones around the globe. We looked at the maximum wind speeds in the strongest 30% of storms each year and showed that the strongest tropical cyclones are getting stronger, particularly in the North Atlantic. We also identified ocean temperature as an important factor in driving this trend by taking measurements in the tropics worldwide.

Is this an open and shut case, or are there other contributing factors?

Our results do not prove Kerry's theory. We've just shown that the data are consistent with his theory. We don't fully understand why some storms intensify and others don't. The heat engine theory is that you need a warm ocean — but you also need a cold upper atmosphere. We have an upcoming paper indicating that changes in solar activity that affect upper-air temperatures might also have an impact on tropical cyclone intensity.