HOUSTON — New evidence has emerged that supports the view that man-induced warming of the world's oceans may be stoking stronger storms.
After analyzing the strength of hurricanes around the globe between 1970 and 2005, U.S. climate scientists have found a steady increase in the number of the most powerful storms, Category 4 and Category 5 hurricanes. Hurricane Katrina hit the Gulf Coast as a Category 4 storm.
In 1970, the scientists found, these most powerful storms only made up about one-sixth of all hurricanes. In recent years, they say, the proportion of major storms has risen to one-third of all hurricanes.
During the same time period the average temperature of the world's oceans has increased by about 1 degree Fahrenheit.
"With some confidence, we can say these two things must be connected," said Judith Curry, a researcher at the Georgia Institute of Technology and author of the research paper, which appears today in the journal Science.
Warm seas are essential to tropical storms — they cannot form unless surface temperatures reach nearly 80 degrees. Warmer water causes more evaporation, which rises into the atmosphere and condenses, releasing the energy that drives hurricanes.
There are good theoretical reasons, then, to believe that if global warming continues, and seas warm more, hurricanes might become more violent. What the new study provides is some evidence this may already be happening, evidence which hadn't existed until now.
That doesn't mean global warming caused Katrina, the scientists said, but it may make the environment more favorable for Katrina-like storms to form.
"Of course it's difficult to attribute any particular hurricane or hurricane season to global warming," said Jim Lawrence, an associate professor of geosciences at the University of Houston who studies hurricanes. "But there's reason to believe there may be a trend here, and we ought to study it some more."
Some climate scientists greeted the new study with skepticism. Chris Landsea, a meteorologist at the National Hurricane Center, said the study is premised on flawed data.
For example, he said, methods for estimating storm intensity vary widely around the globe. In the Northern Atlantic Ocean planes fly into storms to take precise measurements. Forecasters in other areas often rely on interpretations of satellite observations, a method not widely adopted until the mid-1970s. That's after the time period the Science paper's measurements began.
Moreover, the researchers did not find any increase in the maximum wind speed of the strongest global storms, Landsea said. Climate models suggest warm seas should also increase the intensity of the very strongest storms, meaning the researchers should have found some effect.
Landsea said it's unlikely global warming would already be increasing hurricane intensity. He cited the work of Thomas Knutson, a meteorologist with the National Oceanic and Atmospheric Administration, who studies the long-term effects of warming seas on hurricanes.
If global oceans warm between 2 and 5 degrees Fahrenheit in the next century, Knutson found, major storms would see a 5 percent to 12 percent increase in wind speeds. Such an effect would simply be too small to measure today.
"I'm not saying global warming won't have an effect," Landsea said. "Yes, hurricanes may become stronger, but it's going to pretty darn tiny, and it's a long way into the future. I wouldn't expect to see any global warming signal in the hurricane record for decades."
Gregory Holland, a scientist at the National Center for Atmospheric Research and another author of the new study, defended his group's research. The data is reliable, he said, and the trend is clear.
"There is no doubt there is a substantial increase here in the number of category-4 and category-5 storms," Holland said.
The research group studied storms on a worldwide scale because, in individual basins, there are substantial, decades-long ups and down in the number of storms. The Atlantic has experienced such a trend since about 1995, when the number and intensity of storms has increased, following a widely accepted trend known as the Atlantic Multidecadal Oscillation.
Between 1970 and 1994 there were an average of 0.8 major storms per year. The decade ending last year saw an average of 2.3 major storms a year, nearly a three-to-one increase.
A researcher at Colorado State University's hurricane forecasting program, Phil Klotzbach, said these periodic oscillations, more than any global warming trend, may explain the data in the new study.