A tropical tempest: The storm over hurricanes and climate change
Stunning as Katrina’s impact was on the Louisiana and Mississippi coasts, the hurricane wasn’t completely unexpected. Forecasters at NOAA pegged the behemoth’s strength and path more than two days out, and for years physical and social scientists had warned of New Orleans’ vulnerability. Just as striking as Katrina itself was the sheer scope of 2005’s Atlantic hurricane season: a record-smashing 28 tropical and subtropical storms and hurricanes. Four of those systems—Cindy, Katrina, Rita, and Wilma—reached Category 5 status, the most ever in a single Atlantic season.
Could the influence of human-produced greenhouse gases be lurking in the hyperactive hurricanes of recent years? Although gaps in the data cloud the picture somewhat, recent work at UCAR universities and NCAR suggests that global warming is indeed boosting the odds of intense hurricanes, especially in the Atlantic.
Warm water fuels hurricanes, and the world’s tropical oceans have warmed in recent decades. Still, until recently, scientists had uncovered little evidence that our warming planet might be influencing tropical cyclones (tropical storms and hurricanes, also called typhoons or cyclones in some parts of the world). Although the global count of tropical cyclones holds remarkably steady from year to year, some oceans are busier than others in a given year. Activity in the Atlantic picked up dramatically and consistently beginning in 1995. A number of scientists, including William Gray (Colorado State University), attributed this shift to a natural multidecadal cycle.
NCAR’s Kevin Trenberth suspected there was something more going on. At a 2004 press conference, Trenberth pointed to global increases in sea-surface temperature, sea level, and atmospheric moisture. These factors, he said, could produce more-intense hurricanes and more-severe flooding. “The North Atlantic hurricane season of 2004 may well be a harbinger of the future,” he said.
To see if Trenberth’s claims held up under scrutiny, NCAR’s Greg Holland teamed up with Peter Webster and Judith Curry (both of Georgia Institute of Technology) to look at global records of tropical cyclones since the 1970s. “To be brutally honest, I didn’t think we’d find any change,” recalls Holland. Instead, the team found a near-doubling in frequency of the most intense cyclones, those with winds topping 209 kilometers per hour (130 mph).
At about the same time, Kerry Emanuel (Massachusetts Institute of Technology) discovered that tropical cyclone wind energy since the 1970s across the Atlantic and northwest Pacific had increased by more than 50% due to stronger and longer-lasting systems. Like Holland, Emanuel had long believed that climate change wasn’t having any major effect on tropical cyclones. “I changed my mind in a big way,” he now says. “I don’t think there’s any crime in changing your mind when presented with evidence to the contrary.”
The new findings stirred a hornet’s nest of debate. Among others, Christopher Landsea (NOAA)—who leads a multiyear project to examine centuries of Atlantic hurricane data—emphasized weaknesses in the global record. There’s little doubt that some tropical cyclones were missed or underrated, he notes, before satellite monitoring became routine in the 1970s and 1980s.
Critics also pointed to the up-and-down rhythm of Atlantic sea-surface
temperatures and the accompanying multidecadal shifts in tropical cyclones.
Major hurricanes were frequent from the 1930s to 1960s, then more sparse
from the 1970s until the most recent surge began in the mid-1990s.
Climate change, in contrast, was assumed to be a slow, steady process.
Holland and Webster added more fuel to the fire with a 2007 study that linked global warming to sharp ramp-ups in tropical cyclone activity around 1930 and 1995, both following multiyear rises in sea-surface temperatures. “We believe the sharpness of the transitions in several different variables points toward physical changes rather than analysis errors,” says Holland. Meanwhile, Landsea and colleagues continued to emphasize gaps in the historical record. “Large, long-term ‘trends’ in tropical cyclone frequency are primarily manifestations of increased monitoring capabilities and likely not related to any real change in the climate in which they develop,” wrote Landsea in a 2007 article for Eos.
Up to now, computer models haven’t been much help in resolving the conflict. Some global models can depict weather elements as small as 20 kilometers (12 miles) across, but that’s still not detailed enough to portray the cores of most hurricanes. Increased computing power is coming to the rescue, though. NCAR is now using a super-sharp, research- oriented version of the Weather Research and Forecasting model to simulate 2005’s Atlantic hurricane season hour by hour at resolutions of 12 and 4 km (7.4 and 2.5 mi). The idea isn’t to precisely replicate the likes of Katrina and Rita, but to see if—and how—the warm sea-surface temperatures in the model lead to a harvest of cyberhurricanes that resembles the actual bumper crop.
“If the model gets the statistics right, and if we can work out
why, then we have a substantial step forward in understanding the current
trends,” says Holland.
How to talk about climate change | Oceans and carbon dioxide | A tropical tempest