While coastal residents from Maryland to the Carolinas pick up the wreckage from Isabel, meteorologists are warning East Coast communities to prepare for busier hurricane seasons for the next 10 to 40 years.
Higher sea surface temperatures and favorable wind conditions in the Atlantic since 1995 have more than doubled the number of major storms roaring out of the tropics each year. Those affecting the Caribbean have quintupled, and the rare ones that strike in October or later have jumped tenfold.
This is the same confluence of weather forces that spawned a barrage of big hurricanes in the 1950s and 1960s, storms that older Marylanders recall with a shudder. Among the most damaging here were Hazel (1954), Connie and Diane (1955), and Donna (1960).
Hazel brought storm tides 2 to 6 feet above normal to the Chesapeake, with 73-mph winds. It killed six Marylanders, flooded streets around Baltimore Harbor, blew roofs away and floated homes off their foundations.
"If you look back at the 1950s ... it was year after year of these major hurricanes ramming into the East Coast. And we would not be shocked to see that take place again," said Stanley B. Goldenberg, a meteorologist with the National Oceanic and Atmospheric Administration's hurricane research division.
Population increases and intense coastal development since the 1960s are likely to produce a correspondingly higher toll in lives and dollars, he said.
Goldenberg was the lead author of a 2001 paper in the journal Science that first noted the return of higher hurricane activity in the Atlantic. The pattern, which began in 1995, is striking - and it has persisted.
"You have to be deliberately blind not to see that something is going on here," he said.
This is not the result of global warming. There is evidence that these long-term hurricane cycles - which can last up to 40 years from peak to peak - stretch at least to the 1600s, Goldenberg said.
They're even clearer in the more abundant data gathered since 1900, which show a period of low activity through the mid-1920s followed by higher activity through the 1960s.
But reliable data on Atlantic hurricanes did not became available until 1944, when aircraft - and later satellites - began tracking storms at sea that had previously gone unnoticed.
Goldenberg and his colleagues found an average of 2.7 "major" hurricanes a year in the active period from 1944 to 1970.
These intense storms have sustained winds above 110 mph - Category 3, 4 or 5 on the Saffir-Simpson scale. Although they constitute 20 percent of all tropical storms, they cause 80 percent of the damage.
In 1970, something changed. The hurricane seasons became much quieter, and the average number of major storms fell to 1.5 per year. There were no Category 4 or 5 storms for 15 years.
But in 1995 the pendulum swung back. That year saw one of the most active Atlantic storm seasons on record, with 19 named storms, five of them major. And a pattern was set. The six years from 1995 to 2000 averaged 3.8 major storms per year - even busier than the 1950s and 1960s.
"We don't expect the system to switch back anytime soon," Goldenberg said.
None of this activity is perfectly consistent. Even a busy hurricane era will produce slow years, Goldenberg said, just as the "low-activity" era that ended in 1994 produced some ferocious storms. Those included Andrew, which devastated south Florida in 1992, and Hugo, which battered Charleston, S.C., in 1989.
Nor will every hurricane season during an active era produce East Coast landfalls. Goldenberg said Eastern states have been shielded by a persistent atmospheric pattern that has steered many storms out to sea.
But "we expect things to shift," he said. Florida was a frequent target in the 1940s, before the bull's-eye shifted to the East Coast during the 1950s and to the Gulf Coast in the 1960s.
With two months left to go, this season has produced only an average number of hurricanes. But thanks to Fabian and Isabel, it has been an above-average season on indexes that factor in the number of days with intense storms on the radar.
Isabel and Fabian, which raked Bermuda with 120-mph winds Sept. 5, were "among the top five, longest-lasting intense hurricanes of all time," said Phil Klotzbach, a research associate at Colorado State University's Tropical Meteorology Project. The project is headed by William Gray, a co-author of the 2001 paper in Science.
Isabel was a Category 3 storm or stronger for eight days, spinning in a perfect, terrifying spiral. "It was the best-looking storm I've ever seen in the Atlantic," Klotzbach said. "Thankfully, by the time it made landfall it had weakened a lot. You wouldn't want to see what would happen if it came ashore like that."
Conditions in the Atlantic are hurricane-friendly now. One key ingredient is high sea-surface temperatures, which provide the energy and instability needed to spawn storms.
Another is low "wind shear," a measure of the difference in speed and/or direction between winds near the surface and those 40,000 or 50,000 feet up.
"A hurricane has to organize in the vertical, like a chimney," Goldenberg said. "If you have strong differences, you knock the chimney over."
Andrew was a struggling tropical storm that was being "decapitated" by wind shear as it approached Florida in 1992. "But as soon as the shear relaxed, Andrew snapped to and rapidly intensified from a tropical storm to a Category 5 in 36 hours," Goldenberg said.
The storm crashed ashore in Florida and demolished much of southern Dade County where Goldenberg, his children and thousands of others clung to the wreckage of their homes. "I'm not some scientist in some ivory tower saying this stuff," he said.
In their paper in Science, Goldenberg, Gray, Christopher Landsea and Alberto M. Mestas-Nunez pointed out decades-long cycles in sea-surface temperatures and wind shear in the tropical Atlantic. Those factors closely matched the cycle of major hurricane formation.
But there are other players in the cause-effect game. Over the past decade, meteorologists have tied warm-water El Nino conditions in the eastern tropical Pacific to increases in vertical wind shear in the Atlantic. That wind shear has tended to suppress the formation of major hurricanes.
In 1997, the strongest Pacific El Nino of the 20th century emerged in the midst of the resurgent Atlantic activity that had prevailed since 1995. The Atlantic hurricane season fizzled that year. It fizzled again when a weaker El Nino returned last year.
While most meteorologists have accepted the existence of a hurricane formation cycle that spans decades, there is less unanimity about what drives it.
The most prominent theory, espoused by Gray and others, is something called the "thermohaline circulation." This is a global current - a sort of oceanic conveyor belt driven by changes in temperature and salinity - that speeds up and slows down in cycles of roughly 25 years.
When that circulation is strong, Goldenberg said, the North Atlantic - including the hurricane regions - warms up. When the flow weakens, the water cools.
"Not everybody is in agreement on that," he said, "but the Atlantic multidecadal mode [the hurricane cycle] is very clear, and its effect is very clear."
Coastal residents and emergency managers need to be prepared, Goldenberg said. "We can expect to see continued deadly hits in the Caribbean, a continued increase in overall activity and more landfalls than even we've been seeing on the East Coast."
Scientists can't predict exactly when the East Coast will again come under more persistent assault from the tropics, he said: "Maybe this year is the beginning of it."