17th Street Canal
Levee Was Doomed
Report Blames Army Corps
By Bob Marshall
The Times-Picayune
The floodwall on the 17th Street Canal levee was destined to fail long before it reached its maximum design load of 14 feet of water because the Army Corps of Engineers underestimated the weak soil layers 10 to 25 feet below the levee, the state's forensic levee investigation team concluded in a report to be released this week.
That miscalculation was so obvious and fundamental, investigators said, they "could not fathom" how the design team of engineers from the corps, local firm Eustis Engineering and the national firm Modjeski and Masters could have missed what is being termed the costliest engineering mistake in American history.
The failure of the wall and other breaches in the city's levee system flooded much of New Orleans when Hurricane Katrina slammed ashore Aug. 29, prompting investigations that have raised questions about the basic design and construction of the floodwalls.
"It's simply beyond me," said Billy Prochaska, a consulting engineer in the forensic group known as Team Louisiana. "This wasn't a complicated problem. This is something the corps, Eustis, and Modjeski and Masters do all the time. Yet everyone missed it -- everyone from the local offices all the way up to Washington."
Team Louisiana, which consists of six LSU professors and three independent engineers, reached its conclusions by plugging soil strength data available to the corps into the engineering equations used to determine whether a wall is strong enough to withstand the force of rising water caused by a hurricane.
"Using the data we have available from the corps, we did our own calculations on how much water that design could take in these soils before failure," said LSU professor Ivor van Heerden, a team member. "Our research shows it would fail at water levels between 11 and 12 feet -- which is just what happened" in Katrina.
Not deep enough
Several high-level academic and professional investigations have found that the sheet piling used in the design to support the floodwalls was too short for the 18.5-foot depth of the canal. In addition to holding up the concrete "cap" on the walls, the sheet piling is supposed to serve as a barrier preventing the migration of water from the canal through the porous soils to the land side of the levee, an event that rapidly weakens the soils supporting a wall and can cause it to shift substantially.
The corps has long claimed the sheet piling was driven to 17.5 feet deep, but Team Louisiana recently used sophisticated ground sonar to prove it was only 10 feet deep.
Van Heerden said Team Louisiana's latest calculations prove investigators' claims that a depth of 17 feet would have made little difference. He said the team ran the calculations for sheet piles at 17 feet and 16 feet deep, and the wall still would have failed at a load of 11 to 12 feet of water.
Investigators have been puzzled by the corps' design since it was made public in news reports. They said it was obvious the weak soils in the former swampland upon which the canal and levee were built clearly called for sheet piles driven much deeper than the canal bottom. It was not a challenging engineering problem, investigators said.
Prochaska said a rule of thumb is that the length of sheet piling below a canal bottom should be two to three times longer than the length extending above the canal bottom.
"That's if you have uniform soils, and we certainly don't have that in the New Orleans area," he said. "It kind of boggles the mind that they missed this, because it's so basic, and there were so many qualified engineers working on this."
Corps approved design
According to records, Eustis Engineering provided the detailed analyses of the ability of soils along the path of the levee to withstand water pressure once the wall was built on top. The information was provided to Modjeski and Masters, the contractor that designed the wall for the corps. If the project followed normal procedures, the engineers with those firms were using design criteria spelled out in various corps handbooks. "You use the corps cookbook, and you usually have to work it out using corps (computer) programs," Prochaska said.
Private-sector engineering work must be reviewed by corps personnel in relevant sections. In this case, legal documents show, the work was reviewed by engineers in the corps' geotechnical and structural engineering branches, as well as the flood control structures section. It was approved and accepted by the district's chief engineer at the time, Chester Ashley, according to the documents.
Robert Bea, a University of California, Berkeley professor who led a National Science Foundation investigation of the levee failures, said the mistakes made by the engineers on the project were hard to accept because the project was so "straightforward."
"It's hard to understand, because it seemed so simple, and because the failure has become so large," Bea said.
"This is the largest civil engineering disaster in the history of the United States. Nothing has come close to the $300 billion in damages and half-million people out of their homes and the lives lost," he said. "Nothing this big has ever happened before in civil engineering."



This Site Served by TheHostPros