By Anthony R. Wood,Knight Ridder Newspapers

PHILADELPHIA – The numbers of powerful hurricanes worldwide have increased significantly in the last 35 years, evidently incited by warm ocean waters brewed by Earth’s rising temperatures, a research team reports.

However, the researchers, led by Georgia Tech atmospheric scientist Peter J. Webster, cautioned that the study period – and the overall knowledge of hurricanes – were so limited that they couldn’t blame global warming for certain.

And while Katrina, the deadliest U.S. hurricane in 78 years and the strongest ever to affect the United States, has recharged the debate over whether humans are making the planet hotter, Webster said it would be “impossible” to blame global warming for any one storm.

In the paper published Thursday in the journal Science, the researchers said they had documented about a 60 percent jump in intense hurricanes – those with winds of 131 mph or more – and a 1-degree increase in tropical sea-surface temperature, or SST. They said both findings appeared to transcend the effects of mere natural variability.

“The fact that we do see the trend in SST, which is relentlessly rising, the hurricane intensity … there’s probably a substantial contribution of greenhouse warming,” co-author Judith A. Curry, a global warming expert at Georgia Tech, said during a media conference call Wednesday.

The researchers said their results posed riddles that would require further study. For example, the total number of all hurricanes actually declined slightly. “We don’t have a simple explanation for it right now,” said Curry.

And the researchers were surprised by something they didn’t find. Experts had forecast that global warming would lead to an increase in maximum hurricane wind speeds, but the study documented no such increase.

Christopher Landsea, a researcher at the National Hurricane Center and an expert on tropical-storm cycles, questioned the study’s findings. “This is an exercise that says to me there’s problems with the data,” he said.

The data came from satellites, unavailable before 1970. The researchers used it to interpret wind speeds to identify Category 4 and 5 storms in six ocean basins, comparing 1975-89 numbers with the 1990-2004 period.

William M. Gray, the hurricane expert at Colorado State University, also took issue with the data. He said that overall improvements in technology and better interpretation might explain the higher numbers in the 1990-2004 period. Outside of the North Atlantic and eastern Pacific, the data wouldn’t have been as good in the earlier period, Gray said.

Webster and his co-authors said they had confidence in the satellite data because the results matched well with data from aircraft that have measured winds in Atlantic hurricanes.

But the team acknowledged that their study wasn’t perfect. “In providing a definitive answer we’re frustrated by several things,” said Curry. “The data record isn’t long enough. The record isn’t homogenous.”

Gray, who believes that hurricane-season variations are part of natural cycles, said, “It’s a good paper, except the main point is wrong.”

Aside from the data, Gray said he also had problems with the linkage of higher sea-surface temperatures with intense storms. He argued that the overall rise in SSTs in the study was unimpressive and that no one has proved that warmer surface water leads to stronger hurricanes. “We have October storms where the SSTs have gone down,” he said. “It’s not so much the SST, it’s the deep SST.”

Katrina, the largest intense hurricane ever to make U.S. landfall, according to the insurance modeling firm AIR Worldwide, drew its strength from deep and warm Gulf waters.

The Atlantic surface temperatures have warmed about a degree, the study noted. But Landsea said the increase shouldn’t be surprising because the waters were emerging from a cool period at the onset of the study.

He said that the increase in intense hurricanes found in the Atlantic Basin was part of a natural oceanic cycle known as the Atlantic Multidecadal Oscillation – not global warming. The 1970s and 1980s were part of a hurricane-lull era. The lull ended in 1995.

Whatever the study’s weaknesses, the team’s overarching finding was unmistakable, said study co-author Greg Holland, a researcher at the National Center for Atmospheric Research, who, like Landsea, is a former student of Gray’s.

“There is no doubt that there’s a substantial increase in Category 4 and 5 hurricanes,” he said.

“The increase is consistent with what we would expect in global change.”