Copyright 2002 The Times

June 23, 2002 Sunday

SECTION: NATIONAL; Pg. J12

LENGTH: 2863 words

HEADLINE: Evolving danger;
Experts know we face a greater threat from hurricanes than previously suspected. But because the land is sinking and the coastline is disappearing, scientists can't say just how vulnerable we are.

BYLINE: By John McQuaid and Mark Schleifstein; Staff writers

BODY:

The New Orleans area's last line of defense against hurricane flooding is a 475-mile-long system of levees, locks, sea walls and floodgates averaging about 16 feet high. The Army Corps of Engineers says the system will protect the city and suburbs from a Category 3 hurricane that pushes in enough seawater to raise Lake Pontchartrain 11.5 feet above sea level -- high over the head of anyone standing on the other side of a levee.

That margin of error is critical because a storm that pushes the lake any higher can force water over the top of the levees and inundate the city. The water could quickly rise 20 feet or higher. People would drown, possibly in great numbers.

The corps doesn't know what that safety margin is anymore. Generally speaking, the corps says the powerful, slow-moving storms capable of overwhelming the system are rare and the levees are safe. But corps engineers say their own safety estimates are out of date, and an independent analysis done for The Times-Picayune suggests some levees may provide less protection than the corps maintains.

The corps' original levee specifications are based on calculations made in the early 1960s using the low-tech tools of the day -- manual calculators, pencils and slide rules -- and may never have been exactly right, corps officials say. Even if they were, corps officials and outside scientists say levees may provide less protection today than they were designed for because subsidence and coastal erosion have altered the landscape on which they were built.

Experts dispute corps' estimates

According to the rough statistical analysis done by engineering consultant Lee Butler, the risk of levee overtopping in some areas -- St. Bernard Parish, eastern New Orleans and the Lower 9th Ward -- may actually be close to double what the corps once thought it was. The corps disputes Butler's numbers but has no current alternative figures.

The agency is undertaking a new study to reassess the level of protection and another to determine whether the levees need to be raised still higher.

Measuring the risks of disaster is a technical feat that few understand. But such exercises are critical to the future of New Orleans. If the new corps study confirms that protection is less than previously thought, the answers could have major effects on issues such as flood insurance rates, future levee expansions, emergency planning, evacuation and long-term business decisions.

Thanks to its low, flat profile and its location on the Gulf of Mexico, south Louisiana is more at risk from a major natural disaster than most other places in the country. The risk of a catastrophic levee-topping flood in New Orleans is roughly comparable to the risk of a major earthquake in Los Angeles. Because of coastal erosion and subsidence, that risk is growing.

But judging that risk and how to protect against it can be difficult. Recent experience tends to confirm the idea that catastrophic hurricane floods are rare. Even if a powerful hurricane comes close to New Orleans, only certain storm tracks could flood all or part of the city and suburbs. Twelve storms rated Category 3 and above have hit the Louisiana coast in the past 100 years, but only four produced major flooding in the New Orleans area. The levee system was built largely in response to those storms, to prevent or reduce flooding in similar events.

Analysts quantify the unthinkable

Statisticians typically define the risk that something bad will occur by "return periods," the amount of time it takes on average for a given event to recur. The more often something occurs, the higher the risk. For example, many flood-protection projects are designed to shield people from the "100-year flood," which occurs in a given place on average every 100 years. That means that in any single year, the risk of that occurring is 1 in a 100, or 1 percent. If the average time between floods is 50 years, then the annual risk is double that: 1 in 50, or 2 percent.

Such statistical methods have a central role in levee design. The Corps of Engineers' original specifications say the lakefront levees would protect the city from a 300-year flood, defined as 11.5 feet above sea level, not including waves that crest even higher. The levees surrounding St. Bernard and the east side of eastern New Orleans are rated for a 200-year flood. The hurricane levee in Lafourche Parish is designed for a 100-year flood.

At first glance, those risks appear remote: Less than a 1-in-200 chance in a given year sounds like pretty acceptable odds. But they actually conform to risks that most people consider relatively common. For U.S. residents, for example, the annual odds of being attacked with a deadly weapon are 1 in 261, according to statistics compiled by Larry Laudan, a philosophy of science researcher now at the National Autonomous University of Mexico, in "The Book of Risks." The odds of someone older than 35 having a heart attack in a single year are 1 in 77. The odds of injuring oneself on a chair or bed are 1 in 400. By contrast, the chance of dying in an auto accident in a given year is 1 in 5,000, and the chance of dying in an airplane crash, 1 in 250,000.

The risk of hurricane flooding also is much higher than for river flooding, although the government has committed remarkable resources to protect the region from the river. The Mississippi River levees average more than 25 feet high and are rated for an 800-year flood. "The city is exposed to as much as four times the risk of hurricane flooding as it is to river flooding," said Louisiana State University engineering professor Joseph Suhayda. "That's always been an odd issue to me. Why would the government think that water from the lake is less dangerous than water from the river"

Scientists rely on the past

To design a hurricane-protection system, engineers must look at historical data and try to figure out how high water will rise at certain points. Then they can figure out how high to build the levees.

It sounds straightforward, but it is a complex challenge. Accurate hurricane data go back only 100 to 150 years. Statistically speaking, not very many hurricanes have hit the New Orleans area -- at least not enough to allow a solid projection into the future.

And the recent past isn't always an accurate basis for predicting the future. A Science magazine paper written last year by meteorologists William Gray, Christopher Landsea and Stanley Goldenberg predicted that based on long-term trends in sea-surface temperature, the Atlantic Ocean is entering a 10- to 40-year period of more intense hurricane activity. That means more big storms may menace areas that are more heavily populated than during the previous such cycle, from 1920 to 1960.

Storm surges are even harder to analyze. Flooding can vary dramatically mile by mile, even lot by lot, depending on the storm, rainfall, land elevation, levee heights and proximity to waterways and drainage pumps. Storm surges flowing into Lake Pontchartrain literally slosh around, first raising water heights to the north and west, then on the south shore. A record-setting rainfall could swell water heights by a foot or more, something that could turn a relatively weak storm into a killer.

Hurricane flood statistics are even spottier because scientists often did not have the equipment positioned in enough places to measure high water during past storms. The landscape also is changing because of coastal erosion, sinking and even levee building. So a flood height from the past wouldn't be the same today.

These were the challenges the corps faced in the early 1960s when it determined most of the current levee heights around New Orleans without computers or modern knowledge of hurricane dynamics.

Engineers wanted to prevent a repeat of the flooding that hurricanes in 1915 and 1947 had caused in the city, according to Jay Combe, the coastal engineering chief for the New Orleans district of the corps. They needed a single, hypothetical storm to use in the design process, something that embodied the worst flooding conditions the area had experienced.

So they mixed and matched features of both earlier storms to devise something called the Standard Project Hurricane. The levees would be built to protect against that imaginary storm.

Models envision the perfect storm

Meteorologists today say the Standard Project Hurricane could not exist in nature. It had a barometric pressure of 27.6 inches, the equivalent of a powerful Category 4 hurricane on today's Saffir-Simpson scale. But its maximum sustained winds were 100 mph, the equivalent of a relatively weak Category 2 hurricane. After running computer simulations in recent years, corps officials say the Standard Project Hurricane corresponds to a fast-moving Category 3 storm.

When they tried to predict how high a storm surge their imaginary storm would generate, engineers found the answers didn't match up with reality, Combe said. An estimate based on a statistical analysis of real floods showed higher storm surges were likely. So the engineers combined those two results. They decided the levees would protect against a potential flood of 11.5 feet above sea level. On top of that, they added several feet of "freeboard" to block higher waves from washing over the top, along with calibrated sloping, rocks and other features to reduce wave heights.

Combe defends the methods even while admitting that modern technology, and perhaps changes in the landscape, make them outdated. "Given the state of the art, the computing resources of the time, they did a crackerjack job," he said.

Butler basically agrees. "Those estimates are outdated, but they were very conservative," he said.

Today, engineers can fill in many of the gaps in the hurricane and flood data that challenged designers 40 years ago. Using computer models that incorporate current knowledge of hurricane and flooding dynamics, they can simulate past hurricanes in a modern landscape.

Then they can set levee heights to whatever level they find necessary. corps engineers are using such methods to reanalyze the levee system in the agency's new study. Butler, the engineer retained by The Times-Picayune, helped pioneer such methods as a division chief for the corps' Waterways Experiment Station in Vicksburg, Miss. He is now a principal partner of VeriTech Inc., an engineering consulting firm, also in Vicksburg.

Butler estimated flood risks around New Orleans and its neighboring suburbs.

He used historical flood information, his own modeling data and figures from a 1996 corps modeling study. The study was not completed in part because of disagreements between corps engineers and modelers over some results, which the modelers say are accurate. He crunched these in a statistical program to generate predictions of flooding at five points around the New Orleans area and compared that with levee heights to get a rough estimate of the risk of water coming over.

The results tend to confirm the corps' estimate of the level of protection along the lakefront, Butler said, although he said the exact risk could not be pinned down without more study.

He said the risks may be significantly higher than the corps maintains -- perhaps double -- on the east side along levees protecting eastern New Orleans, the Lower 9th Ward, Arabi and Chalmette. Where the corps says the levees protect against a 200-year flood, Butler says it's more like a 100-year flood.

Levee heights along the Mississippi River-Gulf Outlet and Intracoastal Waterway in the area range from 17.5 to 19 feet. Butler's estimates put the 100-year flood level at 16.3 feet above sea level, meaning waves on top of that would wash over the top and flood areas inside.

The historical record tends to confirm these results, Butler said. "All along the levee, there has been very high water measured there for several storms, certainly in Betsy. If you had the right kind of storm come in there, you'd really be in trouble."

Model's accuracy under fire

The corps agrees this is a weak spot because the area is closer and more open to the Gulf of Mexico than Lake Pontchartrain. "Flooding from a storm coming in on a track critical to New Orleans is more likely to occur in these areas outside the city," Combe said. "The MR-GO levee is more likely to be affected than the area in the lake itself."

Another reason flood heights tend to be higher there, Butler said, is that the levees protecting eastern New Orleans and St. Bernard converge in the shape of a V. When a storm pushes water into a narrowing space like that, the water tends to pile up and rise higher, increasing the risk of overtopping.

Combe and other corps officials disagree with this assessment. They have a scientific dispute about the accuracy of the computer model both Butler and the corps use to simulate storm surges, called AdCirc for "Advanced Circulation Hydrodynamic Model."

corps engineers think it tends to overestimate flood heights in some areas, especially where two levees meet at an angle, Combe said. But computer modelers, including Butler and others, defend their results and say they have been corroborated by historical data.

Combe suggested the model doesn't account for a flow away from the levee at the bottom of the water column, something that would reduce the volume of water next to the levee -- and the height.

University of Notre Dame engineer Joannes Westerink, one of the modelers Combe hired to work on the corps' current project, said he thought this effect would be relatively small. "Levees, land, a solid wall of buildings all cause storm surge to build up. But there is a return current: That effect does reduce it somewhat. . . . Is it 5 percent Is it 2 percent Is it 10 percent Our best estimates are that it's on the low end," Westerink said.

Regardless, scientists in and out of the corps say a new study is necessary because of advances in technology and changes in the landscape. "We have lost acres and acres, square miles of land out there near the Gulf. So conditions and bathymetry (water depths) are different now," Combe said.

Why has it taken the corps this long to evaluate the problem No clear bureaucratic mandate exists for reassessing the blueprints once levees are built. Congress appropriates money for levee construction based on corps studies that take years to complete. Dramatic changes or reassessments typically occur after major disasters, when political momentum generates for preventing a repeat.

"The government sort of does things strange," Combe said. "We do things in response to the direction of Congress and the president. Local sponsors say we need something done here, they pass a law, the president signs it, and we go to work. Going back later and looking at pieces of a project is something we have to look at, and we are looking at it and doing a more up-to-date analysis. We are in the process of redoing it. But the government wheels grind slow."

The disagreement over the computer analysis also dragged things out. A 1996 attempt to study Lake Pontchartrain-area levees broke down over that dispute and because of bureaucratic disagreements, according to Combe and others involved.

Meanwhile, sinking, erosion and sea-level rise mean that the odds of getting flooded have been getting worse across south Louisiana. "The frequency of flooding is increasing at all levels," Suhayda said. "You might find in 50 years that the risk of these infrequent events doubled. The 50-year event became a 25-year event, the 100-year event became a 50-year event."

'I'd get out of Dodge'

New avenues have opened for floodwaters entering the New Orleans area. The marshes and barrier islands of St. Bernard Parish have gradually disappeared, though not as much as in areas south of the city. As a result, more water can flow across them and into Lake Pontchartrain faster than 40 years ago. Since it opened in 1963, the MR-GO has eroded and widened to more than four times its original width in some areas. It now forms a giant sluice leading straight up to the city from the southeast, into the Intracoastal Waterway and ultimately into Lake Pontchartrain through the Inner Harbor Navigation Canal. Proponents of closing and filling in MRGO say it has evolved into a shotgun pointed straight at New Orleans, should a major hurricane approach from that angle.

In any case, scientists say they want to know what the corps study finds, in part because they want to get a better grip on what the risks are for the sake of the city -- and for their own peace of mind.

"I think everyone familiar with this is sitting on pins and needles because nothing has happened in that lake for 50 to 60 years and you start to think, are we due" Butler said. "And the answer I think is yes, statistically you're due. And that's scary. Based on my knowledge of hurricanes, I'd watch what happens very closely -- and I'd get out of Dodge."

. . . . . . .

John McQuaid can be reached at john.mcquaid@newhouse.com or (202) 383-7889. Mark Schleifstein can be reached at mschleifstein@timespicayune.com or (504) 826-3327.