Future Forecast: Stronger Hurricanes?

Researchers debate whether global warming will make storms more destructive

Special report: The Gulf Coast's colleges begin to grasp the damage done

Article tools

Printer friendly	E-mail article	Subscribe	Order reprints
	Discuss any Chronicle article in our forums
Latest Headlines New Mexico State U. President Fires Off New Allegations in Dispute With Professors Two whose contracts were not renewed by the university, in what they say was a case of discrimination, now say they are also the victims of baseless allegations by the university's president. U. of Iowa Puts Flood Damage at More Than $230-Million Professors Cry Foul as Kean U. Increases Required Office Hours Portfolios Replace Qualifying Exams as Step Toward Dissertations Robots Hit the Ivies Commentary Updating Higher Education's Past: 1940 to 2005

By RICHARD MONASTERSKY

When it came to global warming and hurricanes, Kerry A. Emanuel used to be a skeptic. In fact, as one of the foremost theorists who studies such storms, Mr. Emanuel helped write a paper last year dismissing the idea that climate change would make hurricanes significantly more dangerous.

That paper will soon be published in a meteorological journal. But Mr. Emanuel's name will not be on it.

While looking at historical records, the atmospheric scientist from the Massachusetts Institute of Technology found that the total power released by storms had dramatically increased -- more than doubling in the Atlantic in the past 30 years. The evidence was so overwhelming, he couldn't stand by his earlier statements.

"I wasn't even looking for it," says Mr. Emanuel. "The trend was just so big that it stood out like a sore thumb."

He withdrew his name from the forthcoming paper that plays down global warming's influence on hurricanes. Then he published a new study in Nature last month, proclaiming the opposite conclusion.

"I didn't feel comfortable saying what we said a year ago," he says. "I think I see a strong global-warming signal."

That statement puts him at odds with many meteorologists, who have searched in vain for hints of such a signal in the damage records from past storms. And looking forward, the new evidence fits into a growing debate -- argued loudly in the wake of Hurricane Katrina -- about whether global warming will lead to more-destructive storms.

On one side stand Mr. Emanuel and other researchers who use computer models to predict storm behavior. They see signs that a hotter climate will brew more-damaging storms. On the other side, Mr. Emanuel's former co-authors argue that global warming will have little or no influence on storms.

"It seems pretty clear looking back in time from the perspective of damages, we're not going to find a large change in the behavior of storms," asserts Roger A. Pielke Jr., an associate professor of environmental studies at the University of Colorado at Boulder and lead author of the paper that Mr. Emanuel once supported.

Mr. Emanuel's reversal -- and the reaction it drew from his former collaborators -- provides a curious view into the way science progresses in the politically charged atmosphere surrounding global-warming research, where pundits and press releases often dominate the public discussion.

In the wake of any natural disaster, commentators invariably face off on the airwaves, in op-eds, and in cyberspace. Some blame the incident on humans. Others categorically denounce any potential connection.

Last week, for example, Ross Gelbspan wrote in The Boston Globe that Katrina's "real name is global warming." James K. Glassman struck back in his blog, TechCentralStation.com, writing that "there is no evidence that hurricanes are intensifying" -- a statement that ignores Mr. Emanuel's paper.

Last year's hurricane season spawned a particularly vigorous debate, in part because the Atlantic served up an unusually large number of storms, including four hurricanes that battered Florida. Those four storms were the most that any state has endured since four hit Texas in 1886. In 2004 hurricanes killed 60 people in the United States and caused an estimated $45-billion in damages, a record until this year.

Last October three scientists held a press conference to respond to public statements denying any possible connection between the storms and global warming. Kevin E. Trenberth, head of the climate-analysis section at the National Center for Atmospheric Research, in Boulder, and two colleagues said it was impossible to tie one storm, or even four hurricanes, to global warming. But the storms fit into a larger pattern of unusual weather around the globe that indicated the climate is indeed changing, according to the researchers.

Looking ahead to future hurricanes, said Mr. Trenberth, "a key consequence, I think, is certainly perhaps increased damage from winds. But I think the biggest consequence is likely to be more heavy rains and flooding."

That press conference and other statements in the news media motivated Mr. Emanuel, Mr. Pielke, and several colleagues from the National Oceanic and Atmospheric Administration to respond. Like many other researchers who had studied storm statistics, they didn't see any evidence that global warming was changing hurricane behavior. And even if storms did respond to increasing temperatures, it was unlikely that the change would have important societal implications, the scientists reasoned. Given the steady flow of people settling in coastal areas and developing vulnerable regions, the researchers concluded that societal forces -- rather than climate change -- would play the dominant role in determining how much damage came from future storms.

The researchers submitted the paper unsuccessfully to Science in the fall of 2004. They sent it to the Bulletin of the American Meteorological Society in the winter.

While the paper was going through review, however, Mr. Emanuel was continuing his research into hurricane behavior. He developed a new way to measure storms, which he called the "total power dissipation." That measurement is a mathematical combination of the peak wind speed and the storm's duration. This marker, he says, provides a better indicator of the threat from tropical storms than do other numbers, such as their frequency or peak intensity.

When Mr. Emanuel plotted the power output of storms since 1930, he found that the value had cycled up and down from one decade to the next. But in the past 30 years, the numbers had shot upward, suggesting that storms have grown stronger or lasted longer or both.

Although he cannot say for certain, Mr. Emanuel sees global warming as the prime suspect. In any given ocean region, the power output of storms closely tracks water temperatures, and records of the tropical oceans show that surface waters have warmed by almost 1 degree Fahrenheit in the past 50 years. That change, possibly unprecedented in the past few thousand years, helps connect the recent hurricane intensification to global warming, he says.

Mr. Emanuel admits that his argument is controversial and rests on different types of data that are hard to put together: "It's one of those things where every single piece of evidence has holes and flaws. But all of these pieces of evidence point to the same culprit."

Revving Up the Engine

In theory, at least, it is not surprising that tropical storms might strengthen as the world warms up. Hurricanes are basically heat engines, transforming the warmth of the tropical ocean into a swirling vortex of winds and rain. If the sea surface gets warmer, that provides more fuel to stir up tropical storms.

But on its own, the ocean's warming could not explain why Atlantic storms would put out twice as much power now as in the 1970s. Other factors, like the stability of the atmosphere and the status of the ocean's subsurface waters, might also have helped punch up the power of storms, speculates Mr. Emanuel.

"Whatever the cause, the near doubling of power dissipation over the period of record should be a matter of some concern, as it is a measure of the destructive potential of tropical cyclones," he writes in Nature.

Given such conclusions, Mr. Emanuel says, he could no longer support the statements he had signed months before suggesting that global warming's impact on hurricanes would be tiny.

Nonetheless, he agrees with his former co-authors that the most important factor for increased damage in the near term is coastal development and related societal forces. "There's this huge migration to the coastline," he says. At the same time, people are destroying wetlands and other natural barriers that help protect against storm surge. "These trends are the ones that are really nailing us. The global warming sits on top of it and is more serious the further in time you look out," says Mr. Emanuel.

He says he had a "friendly parting of the ways" with his former co-authors, who are understandably skeptical of his findings. "They're not going to believe what I did until they get into the data" and examine it for themselves, he says.

One who has started to do that is Christopher W. Landsea, science-and-operations officer at the National Hurricane Center, in Miami. He says that Mr. Emanuel made unnecessary adjustments to the wind speeds for past storms, thereby making older hurricanes look too small relative to those of the past decade. For example, the second-strongest hurricane on record was Camille in 1969, which had peak recorded winds of 195 miles per hour. But Mr. Emanuel's method incorrectly reduced Camille's winds to 160 miles per hour, says Mr. Landsea.

He also takes issue with a graph in the Nature paper, which shows a dramatic peak in the power output of hurricanes in the last 10 years. According to Mr. Landsea, the spike is an artifact of the mathematical procedure used to produce the graph. When looked at properly, the hurricanes of the past decade or two have not been unprecedented, says Mr. Landsea. With respect to Mr. Emanuel's paper, Mr. Landsea says simply, "he's wrong."

Mr. Emanuel responds that he gets similar results when he uses other methods to correct for the wind speeds of older hurricanes. He admits making a mistake in the graphs but says the error overemphasizes changes in some ocean areas and underemphasizes them in others.

Mr. Pielke plays down any disagreement between Mr. Emanuel and his former co-authors, saying that they all agree on the critical point of what society should do to protect itself. "No matter where you come down on hurricanes and global warming, the consequences for action are identical," he says. "The major driver for impacts are human vulnerabilities."

He suggests that the measure of power dissipation studied by Mr. Emanuel is not an important factor for determining how storms affect people. Even though the power output has surged in 30 years, he says, "that increase has not been correlated with an increase in damage. That tells me that index is not a good metric for anticipating damage."

Not yet at least, acknowledges Mr. Emanuel. Hurricanes release most of their energy over the open ocean and some die before they ever reach land. Any changes in the power of hurricanes while they are at sea would not increase the damage they cause on land. At present, the increasing power would not have made a discernible difference because too few storms hit the coastline each year. "It's just chance," he says. "The numbers are far too small and the element of chance is far too large to expect to see any real trend in the data."

On top of that, natural swings in climate make a far more noticeable difference in the numbers of storms hitting the U.S. coastline, says Mr. Emanuel. For example, the eastern United States and the Gulf States passed through a relatively quiet period between 1970 and 1994 when fewer hurricanes hit these regions. Since 1995 the pace has picked up considerably.

But look into the future 50 years and global warming may boost the destructiveness of storms by a noticeable amount, he says. Not only will storms unleash more power, but sea levels are expected to climb even faster than in the past, enhancing the risk of coastal flooding, he says.

As Mr. Emanuel has predicted, hurricane researchers are greeting his numbers with some skepticism. "It's a difficult problem he's taken on," says Thomas R. Knutson, a research meteorologist at NOAA's Geophysical Fluid Dynamics Laboratory. "It's a very interesting result, and if they are really borne out, they are an alarming result."

Mr. Knutson has tried to project how global warming might affect the behavior of storms. In one recent study, published in the Journal of Climate last year, he and Robert E. Tuleya, of Old Dominion University, used the results of nine different computer climate models that projected how the ocean and atmosphere might warm over the next 80 years. They fed those results into four different hurricane models to see what kind of storms might form.

On average, the storm intensity went up by a half of a category, and the amount of rain increased by 18 percent, the researchers found. With stronger winds and bigger storm surges, more-intense storms could cause more damage, says Mr. Knutson. "We've seen the devastation with Katrina; imagine if that storm had been half a category stronger."

But Mr. Pielke maintains that whatever global warming does to storms will pale next to the way society puts itself in danger. For every dollar of future damage caused by climate change, he says, "we should expect between $22 and $60 of damages from societal effects."

The results point to a clear course of action, he says. "If we want to lower the human impacts of hurricanes, we have to address the economic vulnerability."

But the United States seems particularly uninterested in tackling that problem. Hurricane experts have been warning for decades that without monumental investments to protect the city, a major storm would eventually sink New Orleans. Only now are people starting to listen.