NEW ORLEANS—Engineers investigating the flooding of New Orleans from Hurricane Katrina have found "dozens and dozens" of places where a patchwork of mismatched floodwalls, other design flaws and even a missing floodgate helped drown the city.
The extent of the problems is far greater than has been generally acknowledged. It suggests that officials planning the city's rebuilding not only will have to restore the levees but also will need to revise construction techniques and overhaul the complicated governmental network that oversees the system.
An apparent lack of coordination between the many jurisdictions that are responsible for different parts of the flood protective system contributed to some of the weaknesses, the engineers believe. Among the jurisdictions are levee districts, sewerage and water boards, the federal government, state transportation and highway departments, and private industry.
Two groups of engineers, one funded by the National Science Foundation and the other by the American Society of Civil Engineers, have been working together to find out what went wrong and devise better ways of protecting the city as it rebuilds.
The team has yet to issue a comprehensive report or make recommendations, but it has announced one key finding: that floodwalls along the 17th Street and London Avenue canals failed when soil beneath the walls gave way, not because the storm waters were so high they flooded over the walls. Those failures were responsible for most of the flooding in central and western New Orleans.
Interviews with engineers involved in the investigation indicate that problems go far beyond those two canals and the other most-discussed breach, on the Industrial Canal, which allowed floodwaters to flow into the city's Lower Ninth Ward.
"There are many more breaks than the three that the media are concentrating on," said Peter Nicholson, a University of Hawaii civil engineering professor who is leading the ASCE group. "I cannot even count them. There were dozens and dozens."
Most of those problems were found along the Industrial Canal, which links Lake Pontchartrain with the Mississippi River, and the Mississippi River-Gulf Outlet Canal, which links the Industrial Canal with the Gulf of Mexico.
Among the problems, Nicholson said, were walls built at different times and by different agencies that were of different heights and different materials, setting the stage for erosion and flooding where the lower level was made of weaker material.
He described one location where a railroad track and a highway cut through a berm to cross a canal. The transition between the differing materials—concrete to earthen berm, concrete to steel—allowed the water to wear away the weaker material when the storm hit and scour out the area behind the walls, causing their collapse.
"It was a mess," he said.
Robert Bea, professor of civil engineering at the University of California at Berkeley and a member of the NSF team, said investigators have documented at least 100 places where floodwalls and levees of different heights and materials joined one another, causing erosion.
"Almost every levee we visited had this disease," he said. While many caused only minor problems, "my guess is of the 100, half were big, substantial volumes of water coming through that ended up in the neighborhoods," he said.
Even in parts of the system that did not fail Bea saw reason for concern. For example, the Orleans canal, which lies between and runs parallel to the 17th Street and London Avenue canals, did not fail. But Bea said he examined one section of the canal where the floodwall had been built to five different heights because five different agencies were responsible for various sections of the wall.
"Each one had a different working elevation," Beas said. "In such a case, the lowest one wins," he noted—meaning no matter how well constructed the sections were, they were only as effective as the shortest one, because that's where water would overflow.
Bea said he thinks only one agency should be responsible for the entire levee system, a sentiment echoed by other disaster experts in Louisiana.
But changing the way Louisiana administers its flood protection system could be fraught with political challenges. Many of the levees are governed by district boards whose members are either elected or appointed and control budgets in the hundreds of millions of dollars. One Louisiana-based disaster expert declined to discuss his views on changing the governance system on the record because of the politics involved.
Engineers also are debating how best to deal with other design issues that have surfaced in the wake of Hurricane Katrina. Key is settling on a single approach to what to do where different flood structures meet.
"There're different possibilities and different schools of thought on that," said Paul Mlarker, an engineer with the Army Corps of Engineers' Engineering Research and Development Center. "There was considerable discussion amongst the teams over whether it was better practice to have the concrete lower, because it could withstand overtopping more robustly than the earthen section."
Some problems appear related to repair and maintenance. For example, engineers found a floodgate missing where train tracks crossed a levee to reach a harbor facility near the Industrial Canal.
The floodgate was supposed to be closed, except when trains are passing through. But the floodgate was missing entirely, leaving an open space in the levee.
Engineers said they are trying to confirm a report that a train derailment had taken the gate off a month or two before Hurricane Katrina.
Nicholson said that engineers found sandbags nearby, but do not know whether they had been placed there before or after the storm.
Engineers also are curious about what role in the flooding might have been played by several huge oak trees whose roots were embedded in the soil underlying the floodwalls along the 17th Street Canal.
The trees were toppled during the storm, and some engineers wonder whether the holes created when the roots were pulled from the ground helped pipe water under the levees, contributing to their failure.
But Nicholson cautioned that investigators don't know if the trees fell before or after the breaches occurred and that the tentative conclusion announced recently that massive soil failure caused the walls' collapse wouldn't be confirmed until soil samples and other data are analyzed.
"The walls failed without overtopping, so obviously something in the system failed to meet the design," he said.
(Carey reports for the San Jose Mercury News.)
(c) 2005, Knight Ridder/Tribune Information Services.
ARCHIVE PHOTOS on KRT Direct (from KRT Photo Service, 202-383-6099): levee
ARCHIVE GRAPHICS on KRT Direct (from KRT Graphics, 202-383-6064): 20050831 KATRINA levee, 20010424 Flood control
GRAPHIC (from KRT Graphics, 202-383-6064): 20051014 STORMS LEVEES
Need to map