Waiting For Hurricane X

By ERIK LARSON

Anyone trying to recover in the wake of last week's visit by Hurricane Bonnie probably isn't feeling especially lucky at the moment. Good fortune isn't the first thing you think of when your living room is full of mud, your roof is missing, your power has been out for days on end. But considering the destruction that often accompanies storms of this magnitude, residents of North and South Carolina and Virginia got off remarkably lightly. Only three people died. Property damage was far lower than it might have been. Beaches remained largely intact.

If anything, the storm reinforced the popular belief that hurricanes are so thoroughly tracked, probed and forecast these days that they cannot possibly cause great loss of life. Scientists don't share that optimism, however. Many believe we're entering a cycle in which violent storms are going to be more frequent, and in which the likelihood of a disastrous strike will be greater than ever. The scientists' pet nightmare is of the Big One--a catastrophic storm that could do $100 billion dollars' worth of damage and kill thousands of people. No one knows when or where the Big One will hit, but the certainty is growing that it will.

Even a Little One like Bonnie, of course, can do plenty of harm. Some half a million people were forced to flee inland last week, as the 400-mile-wide storm--mammoth in size even by hurricane standards--swirled toward Cape Fear, N.C. And though Bonnie's 115-m.p.h. winds slowed rapidly as she lumbered inland, her forward progress slowed too, with the result that the storm hovered over the state and pummeled it for more than a day. Downed power lines robbed over 240,000 people of electricity. Even worse than the winds were the rains--more than 12 in. in some places--which caused flooding in North and South Carolina. When the crisis seemed to be over, Bonnie regained some of her fury to pound Virginia before heading out to sea.

But the Big One will come eventually. Many researchers think we are entering a period of increased hurricane intensity, more like the period from 1940 through 1969 when monster storms swept ashore with greater frequency. Anyone who lived through Hurricane Andrew in 1992 might disagree, but the experts say that for the past quarter-century America has got off easy. The last Category 5, or "catastrophic," hurricane was Camille, which struck the Gulf Coast in 1969 with winds over 200 m.p.h. and a storm surge 24 ft. high. Since then, hurricane activity has been mild.

Such cycles occur every several decades. Researchers emphasize that both phases, active and quiet, are normal, but that's not very reassuring. Even if the next high-intensity phase of hurricane activity is simply a replay of the last such period, it will wreak far more destruction. Reason: a frenzy of coastal construction has brought huge populations to live at America's beaches and barrier islands--people with no conception of what it's like to sustain a direct hit from a truly powerful hurricane.

It's hard to say precisely when the shift to more frequent hurricanes began. It probably started with the exceptionally intense seasons of 1995 and 1996. The past year, to be sure, was exceptionally quiet, possibly due to the recent El Nino, which tends to suppress Atlantic hurricanes. But now things are hopping again. Just days before Bonnie hit, a tropical storm struck Texas and caused extensive flooding. Even as Bonnie ran out of steam, a new hurricane, Danielle, was barreling across the Atlantic behind her. Meanwhile, by the end of last week, hurricane forecasters had begun watching a new tropical disturbance whirling in the western Caribbean.

No one knows why these cycles occur. According to Bill Gray, a hurricane expert from Colorado State University, one reason may be a phenomenon known as the "Atlantic conveyor." The subject of much recent research, the conveyor is a gigantic oceanic flywheel that transports cold water from the seas off Iceland and Greenland in a majestic, slow current along the bottom of the ocean to Antarctica, where it surfaces several decades later and flows back north, absorbing heat as it passes the equator. The conveyor seems to have kicked into a faster gear lately, bringing warm equatorial water north before it can cool. Hurricanes draw their energy from warm water.

Whether this turns out to be the key factor or just one of many, the trend toward more damaging hurricanes is clear. The reason was made explicit in a study done by Christopher Landsea, a research meteorologist with National Oceanic and Atmospheric Administration's Hurricane Research Division in Miami, and Roger A. Pielke Jr., a social scientist with the National Center for Atmospheric Research in Boulder, Colo. They looked at the most destructive hurricanes in U.S. history and then, says Pielke, posed the question: "If history repeats itself, and it certainly will, what might we expect?" To answer it, the researchers did not simply adjust the original damage totals for inflation. They also included data on the great increase in population, wealth and development that had occurred over the decades in the places where the storms struck.

Looked at in this light, 1992's Hurricane Andrew, officially the costliest hurricane on record with damage, in today's dollars, of $35.5 billion, dropped to second place. The first was a Category 4 hurricane (wind speeds above 130 m.p.h.) that struck southeast Florida in 1926 and skipped into Alabama. (It has no name because the custom of naming storms began only in the early '50s.) If that storm took the same path today, it would cause damage totaling $77.5 billion. Of the 10 costliest hurricanes of the century, nine occurred before 1970. The only recent hurricane to make Pielke's and Landsea's Top-10 list was Andrew.

In all cases, what boosted those early storms to Top-10 status was population growth. In 1995, for example, Pielke says, the population of two Florida counties alone, Dade and Broward, was greater than the 1930 population of the entire coast from Texas through Virginia. Like compulsive gamblers betting the mortgage, Americans have pressed their luck to the limit. There has been so much development on barrier islands and beaches along the Atlantic and Gulf coasts, researchers say, that a hurricane of even modest intensity can cause a multibillion-dollar disaster. In Florida the value of insured coastal property rose from $566 billion in 1988 to $1 trillion in 1995. Consider Pinellas County, Fla. The last hurricane struck there in 1921, when the county had 28,000 people. Today Pinellas County has nearly 870,000 people.

All this development occurred in plain view, of course, but until the morning of Aug. 24, 1992, no one seemed to recognize its implications. Hurricane Andrew was not merely a wake-up call; it was a stick of dynamite under the pillow. Prior to Andrew, no one envisioned more than $7 billion in insured losses for a single storm. But after Andrew's landfall, Karen Clark, founder of Applied Insurance Research Inc., in Boston, one of a new breed of "catastrophe modelers," sent an audacious message to her clients estimating insured losses at $9 billion. If Andrew proved to be more intense than first estimated, she added, the damages could hit $13.5 billion. "Nobody believed it," she says. A client remarked, "A few mobile homes and an Air Force base--how much could it be?"

In fact, insured losses topped $18 billion. In Dade County alone, the storm destroyed 63,000 homes and damaged 110,000 others. Nine small insurance companies failed. Large companies raised rates, dumped policies and tried to pull out of coastal areas, but regulators forced them to stay. The "reinsurance" companies, which in effect provide insurance to insurance companies, also got queasy and sharply limited coverage.

The most catastrophic consequence of the Big One, though, won't be property damage; it will be loss of life. It could easily have happened already, in fact, except for pure dumb luck. The Weather Channel and CNN's round-the-clock coverage notwithstanding, hurricane forecasting is not as precise as people like to believe. Storms are capricious. Indeed, the National Hurricane Center's warnings, issued 24 hours before landfall, are subject to a 90-mile error in either direction.

For that reason, even a well-tracked storm like Hurricane Andrew could have caused death on a huge scale, just by zigging a few miles to the north at the last minute. Had it made landfall on Miami Beach, where a third of residents didn't evacuate, Andrew could have killed many more than the 15 people whose lives it claimed.

Another potentially lethal factor: hurricanes can suddenly change in intensity. For Jerry Jarrell, director of the federal Tropical Prediction Center (which includes the National Hurricane Center), the most frightening near miss was not Andrew but Hurricane Opal, which hit the Gulf Coast in October 1995. Opal had been a weak storm, but just before it struck, it underwent what forecasters call "rapid deepening," leaping from Category 2 to nearly Category 5, with winds at 150 m.p.h. It also started moving faster. Such rapid change is the thing emergency managers most fear. Says Tom Millwee, coordinator of the Texas Division of Emergency Management: "You go to bed thinking you've got a Cat 1 moving at 10 m.p.h. You wake up, it's a Cat 5 moving at 20 m.p.h."

The sudden change threw evacuation plans into chaos. Most people waited until morning to evacuate, and did so in a vast crush of cars. Road construction further slowed progress. By morning the scene along Interstate 10 outside Pensacola was like something from a grade-B disaster film. Jarrell estimates 10,000 people were stranded on the highways, listening to ever more urgent broadcasts on their radios. Some drivers abandoned their cars and fled for high ground.

Then, just before landfall, Opal had another change of heart. She fizzled. By the time the storm crossed the coast, its winds diminished by at least a third. Until then, however, it fit Jarrell's vision of what the Big One will be like: traffic jams and cars blown wholesale into storm-surge waters. "I'm afraid you're going to drown hundreds if not thousands of people," says Jarrell. "And it's going to happen some time."

One of the darkest scenarios, however, envisions a powerful storm of Category 4 or higher making a direct hit on a major city like New Orleans or Miami. Surprisingly, hurricane researchers now consider one of the most vulnerable targets to be downtown New York City. They made the discovery by accident, in the course of a routine storm-evacuation study begun in 1990 by the Army Corps of Engineers, the kind of study done for every large community on the nation's hurricane-prone coasts. "We were all shocked," says Allan McDuffie, the Corps' study manager.

They discovered that the city has some unique and potentially lethal features. First, they realized that its major bridges, like the Verrazano Narrows and the George Washington, were so high they would experience the advance winds of an approaching hurricane several hours before winds of the same velocity were felt at ground level. These critical escape routes would have to be closed well before ground-level highways.

Even more surprising, however, were the results of computer models done with a federal computer program called SLOSH, which stands for Sea, Lake and Overland Surge from Hurricanes. The program, the backbone of all evacuation studies, takes into account storm tracks, local landmarks and coastal geography to calculate the effects of a hurricane storm surge, the dome of water pushed ashore by strong winds. Such surges can be the deadliest aspect of a hurricane. An immense surge overflowed the city of Galveston, Texas, in 1900 and killed more than 8,000 people, and possibly as many as 12,000. A surge raised on Florida's Lake Okeechobee in 1928 killed 1,836.

The SLOSH analysis of New York City revealed that the sharp bend in the Atlantic coastline where New York and New Jersey meet, the New York Bight, would amplify the effects of a storm surge to the point where even a modest hurricane could generate deadly flooding in lower Manhattan. "That right angle, believe it or not, can cause 30 ft. of storm surge above normal tide conditions," says Donald Lewis, a hurricane-evacuation expert based in Miami who worked on the New York City study. "The same storm in other parts of the country might cause only a 14- or 15-ft. surge."

Suddenly the project took on a new urgency. The researchers concluded that the surge from a Category 4 storm would put John F. Kennedy International Airport under 20 ft. of water. Seawater would pour through the Holland and Brooklyn-Battery tunnels and into the city's subways throughout lower Manhattan. The flooding would be especially disastrous if people seeking to escape torrential rains and falling debris were to take shelter in subway stations. The report didn't estimate casualties, but observed that storms "that would present low to moderate hazards in other regions of the country could result in heavy loss of life."

For now, at least, disaster scenarios like this remain only hypothetical. Perhaps that's one reason most people don't pay much attention. Homes still rise on the barrier islands off North Carolina. On Galveston's westernmost beaches, where the land is barely above sea level, luxurious new mansions stand atop stilts so tall the scene is almost comical. Just a few minutes up the road, however, there's a poignant monument to this sort of denial. Hard by bright blue signs marking Galveston's primary evacuation route, a small plaque commemorates the site where the hurricane of 1900 destroyed an orphanage and took the lives of 87 children. Just across the street stands a brand new Wal-Mart.