Hurricane Expert Dr. Neil Frank Issues An Ominous Storm Warning for Millions of Americans
08/29/1983 at 01:00 AM EDT
A tropical storm in the Gulf of Mexico grew into Hurricane Alicia last week, indicating that hurricane season again is upon us. And once again, Dr. Neil Frank, the director of the prestigious National Hurricane Center in Miami, is worried. It's his contention that America has never been more vulnerable to a killer storm. Indeed, Frank, 51, thinks we might be headed toward a cataclysm which could dwarf the Mount St. Helens eruption, the Great Chicago Fire or even the Great San Francisco Earthquake in its toll of human lives. Weather satellites and modern long-range forecasting techniques are little help, Frank says, to the millions of Americans who have been encouraged by "unwise" land-use policies to move to dangerous, unprotected coastal areas throughout the southern and eastern parts of the U.S. Most of these new residents are woefully ignorant of how destructive hurricanes actually can be. And that, Dr. Frank told PEOPLE'S Kent Demaret, creates a scenario for what could be the worst disaster in U.S. history.
Which was the worst hurricane ever?
We can't be sure, but in 1970 in Bangladesh one million people were drowned or swept out to sea. The worst hurricane in this country's history took place in Galveston, Texas in 1900, when 6,000 people died. To put the destructive capacities of these storms into perspective, think of it this way: In one hour a large hurricane expends the energy of thousands of atomic bombs.
Could something like the Galveston tragedy happen now?
Unfortunately, yes. It's ironic, but we're far more vulnerable to hurricanes now than we have ever been. We have discovered a beautiful way of life along our waterfronts, and we now have more than 40 million people living in coastal areas where hurricanes might strike. There are so many people out there that I'm not sure we could evacuate everybody to high ground with the lead time we currently can provide, or with the existing road systems.
How can we avoid a Galveston-type catastrophe?
For one thing, every hurricane-prone community should have an evacuation plan coordinated with other communities that may also be evacuating at the same time. Only five coastal areas in traditionally dangerous hurricane lanes have studied the problem, mostly because it wasn't until 1979 that we first developed the necessary computer models. It's a very complicated business, which takes into account the elevation of a community, the congestion in the area, the available evacuation routes—which can change as the water rises—and, most important, the storm surge.
What exactly is a storm surge?
It's the killer element in any hurricane. Essentially, it's an enormous dome of water, perhaps 50 miles wide, and when it hits—sometimes traveling miles inland in low-lying areas—it can be as devastating as a tidal wave. A hurricane may release torrential rains and winds of up to 200 mph, as well as dozens of tornadoes, but nine of 10 deaths are caused by the surge.
What causes the surge?
Two factors—the force of the winds and the depths of water offshore. In places where the water is shallow, the storm surge will be relatively high. For example, the same storm that produces a 12-foot surge at Palm Beach could produce a 30-foot surge along the Texas or Louisiana coast.
How many hurricanes can be expected per season?
Between June 1 and Nov. 30 there will be an average of eight to 10 tropical storms. Five or six will intensify into hurricanes.
Why don't satellites and long-range forecasting offer more protection ?
Satellites, radar and airplanes are only observing tools. They tell us what's happening now. But determining what will happen involves knowledge of wind currents over vast areas, far into the Atlantic, and we simply don't have that kind of detailed information. Also, hurricanes are extremely unpredictable. They can gather strength or turn back on themselves. By the time we become certain where a storm will land, it may be 3 a.m., and that too becomes a factor. This happened in 1957 in Cameron, La., when Hurricane Audrey arrived in the middle of the night. Few heard the warnings, and about 500 people were killed.
Don't such tragedies have a discouraging effect on coastal development?
Unfortunately, human memory is both short and selective. I've had people tell me that they went through a bad hurricane in such and such an area—and I know that all that really happened was that a weak storm passed close by. In 1969, as Hurricane Camille was about to hit Mississippi, a civil defense director tried to get people who lived on the waterfront to evacuate their homes. One couple who lived in a three-story brick apartment building had been in Jacksonville, Fla. in 1964 when Hurricane Dora, a relatively minor storm, passed nearby. Being "veterans," they decided to ride out Camille in their apartment building. Everybody talked to everybody else, and eventually 25 residents decided to ignore the warning and have a "hurricane party." A 25-foot storm surge interrupted the party, literally leveling the building. There was nothing left but the foundation. It's still there. Twenty-three of the people at the party died. It's a terrible story, but the point is they decided to ride it out because they thought they knew about the fury of hurricanes. They didn't. Another problem is the federal flood-insurance program, which encourages people to develop in coastal areas where they shouldn't be. If federal insurance protection were eliminated, there's no question that such development would be slowed down. We'd be saying to people, "If you want to build there, go right ahead. But it's your loss if you get washed away."
In August the weather service changed the way it issues warnings. How?
The new prediction system will issue what some people have called "odds" on where a storm will strike. As the storm moves toward land, we will increase or decrease the odds. A 12 percent probability is the highest percentage that will be predicted for a specific area 72 hours in advance. An 18 percent probability is the highest percentage 48 hours in advance, and in a 36-hour time frame we will issue probability reports of up to 26 percent. If it is 24 hours away, we can assign up to a 45 percent probability.
Isn't there a danger that some people will relate that to rain probabilities, which are so often wrong, and not react as they should?
That danger certainly does exist, but I think that most people are aware that there's a crucial difference between a rain forecast and a hurricane forecast. If you ignore the latter, it could cost you your life.