The LA Times today published a lengthy report on the preliminary National Science Foundation findings on the levee failures in New Orleans following Hurricane Katrina.
The article gives details on the construction of the levees, their lack of maintenance, and their underdesign. Despite all of these warning signs of massive and perhaps criminal incompetence, the very organization that constructed the levees is investigating their failure. The failure of New Orleans’ levee system is the single most expensive engineering design failure both in terms of material losses and loss of life in the history of our nation. Congress and the President demanded independent investigations into the losses of NASA’s Challenger and Columbia orbiters, and should absolutely do the same in this case. Thorough investigation and understanding of engineering failures is crucial to protecting others who rely on similar systems nationwide.
| The collapse of New Orleans’ levee system is the single most expensive engineering failure in the history of our nation. |
Also, the article does not exactly raise but suggests a more fundamental question; even if the Corps is capable of constructing the levees to a sufficient strength for a Category 3 storm, the city of New Orleans will again be flooded. It may not flood next year, and maybe not even this century, but we are at a critical juncture where we must question whether the city should be rebuilt (as I did last August). But having visited New Orleans, I know that the culture of that city is an American treasure, and should not be lightly abandoned. If it is rebuilt, then dramatic changes to construction methods will be needed, including improved evacuation routes, mandatory housing-on-stilts constructions in areas adjacent to Lake Ponchartrain (like the lower 9th ward), and top-floor escape routes.
The LA Times article summarizes the basic engineering failure issues in this case: 1) the levee failure may not be a design failure if conditions were worse than were designed for, and 2) it appears that in at least several cases, the levee failed in a way that designers should have accounted for. The levees were designed to handle a Category 3 hurricane, yet they failed anyway. This alone should suggest that the levees were either underdesigned or badly maintained. Yet there may be specific engineering issues not taken into account with a Category 3 storm that are outside the purview of the Army Corps. Even if this is the case, engineering safety margins were razor thin. Typically, civil engineering safety margins operate near a factor of two: if levees need to withstand 2 million pounds of pressure, then they should be designed to handle four. This may be an outdated rule of thumb to be sure, but razor thin safety margins demand superior analysis in the design phase which was clearly not demonstrated in this case.
But, in a number of cases, the levees failed where they should not have, and failed catastrophically when they should have remained in place. The breach of the 17th street canal resulted from failure in the engineering design process itself. The waters were 3 to 5 feet beneath the top of the levees when they failed. Prior to their collapse, the walls were pushed backward 45 feet because their foundations had not been sunk deep enough into the slippery Mississippi delta soils. The weakness of the soil was known, but perhaps due to loss of expertise in an downsizing Corps during the 90s the sheet piling foundations were driven only a fraction of the distance into the slippery soils. Levees along the Industrial Canal were overtopped by floodwaters that then scoured away the foundations of the levee on the dry side. This should have been prevented by a common practice called armoring involving the lining of the backs of the levees with large blocks of concrete. Finally, near the 17th street canal, large trees were allowed to grow on the earthen portion of the levees themselves. When one large tree overturned, it likely brought up a large plug of earth from the levee, further weakening an already underdesigned structure.
There is enough meat in this first NSF report of the New Orleans levee failures for Congress to demand a complete and independent investigation of the disaster. We cannot afford to whitewash this problem, especially when billions are being spent to construct a new set of levees. If New Orleans is to remain occupied, those paying for their safety should demand, as am I, a complete accounting for how their taxes are being spent. I do not want to waste billions on a failure that will endanger the lives of thousands sometime in the future. The fact that no one has lost their jobs over the levee failure indicates that the Corps is in no mood to open its records and practices to a real investigation. It’s time for Congress and the President to stand up and demand that the Corps be cleansed.

Very good post on this difficult subject. There’s an interesting article in the Feb. 2006 issue of Popular Science on five innovative technology solutions (from various experts) that could provide a “category-5 strength defense.” But the costs would be enormous and the timetable for some of them very long.
-Bruce
Bruce,
Thanks for the heads up on that, I will make sure to check it out.
Thanks for writing about our plight in New Orleans. I’m living and blogging here, and I can tell you that I hope we learned our lesson. The Corps has been using the 100-year storm for hurricane protection, while the Dutch and others use a 10,000-year standard. Would you get on an airplane if you knew you had 1 in 100 odds of getting killed? It’s time we provided appropriate protection for our citizens. Over 1,100 are dead–have we learned our lesson?
Peace,
Tim