The follow-up story that I am most interested in seeing, even though I doubt it will get much national media, is the engineering analysis of the New Orleans levee failure:
- Were the levees known to have structural damage, cracks, or other physical signs of stress concentration at or near the points where they eventually broke?
- Did they appear to fail suddenly, without warning, or did they begin to fail gradually?
- Did they fail in response to a sudden fluctuation in mean water pressure or did they bear a constant stress for some time, invisibly weakening until they could not bear it any more?
- Did the first visible point of failure occur deep underwater or near the surface?
- Did a floating object strike the levees? (In three places?!?)
- Did the designers allow for the possibility that a massive, floating object might strike the levee in a flood, or did they design it only to hold back water, mud, and air?
- How well did the maintainers follow the designers’ recommended repair schedule?
- Was the designers’ recommended repair schedule reasonable in the first place? (Did the designers allow for the VERY PREDICTABLE possibility of future funding constraints in this tax-supported project?)
- What were they made of, anyway?
- Were they made of material designed to absorb water and remain strong, or only to resist water?
- How was stress distributed along the length of each levee at full loading?
- How were the levees designed to fail? Can a levee be designed to transfer stress away from a weakened place to a strong place, as truss bridges can? Or is it just a fact of life that when the levees start to break they must keep breaking?
That’s just a start, from someone whose training is in chemical engineering, not civil or mechanical engineering. Somewhere, someone who knows how to answer these questions is asking them, I hope, and asking others that I don’t know enough to ask.
There will be plenty of highly visible people, in the next days, weeks, and months, asking the question Why did the levees fail? But the vast majority will be seeking political answers. Their question is really why didn’t somebody spend more money on the levees? As if wads of cash, dropped from helicopters, could plug up the hole; as if stacks of coins could resist pressure fluctuations that earthen berms could not.
Here’s an example from August 31, in Editor and Publisher:
…[W]ith the 2004 hurricane season starting, the Corps’ project manager Al Naomi went before a local agency, the East Jefferson Levee Authority, and essentially begged for $2 million for urgent work that Washington was now unable to pay for. From the June 18, 2004 Times-Picayune:
"The system is in great shape, but the levees are sinking. Everything is sinking, and if we don’t get the money fast enough to raise them, then we can’t stay ahead of the settlement," he said. "The problem that we have isn’t that the levee is low, but that the federal funds have dried up so that we can’t raise them."
Obviously, fixing stuff costs money. But to say that the problem you have isn’t the physical problem is to live in an abstract political world where the solution to every problem is to throw money at it and hope it sticks. Projects can be lavishly funded and yet poorly engineered.
Contrast that article with this one, after a levee break in 1997, which details the kind of thinking I am looking forward to seeing:
Meehan said the doomed levee [in the 1997 failure—E.] was "generally similar" to two nearby levees that failed in previous disasters – the Christmas Eve 1955 levee break on Shanghai Bend south of Yuba City that killed 38 people, and a break near the Yuba County town of Linda in 1986 that produced $500 million in damage claims.
Both of those levees failed, Meehan contended, because flood water had saturated a subsurface layer of gravel, then flowed underground far to the landward side of the levees and erupted, geyser-like, undermining the structures and causing their collapse.
The engineer quoted in that article doesn’t say a word about funding. Instead he testifies about pressure relief wells, underground migration of flood water, subsurface gravel layers, and premonitory sand boils.
Let’s hear some more talk like that.
In one very real sense, funding is no more than another engineering constraint. The pylons can be no wider than this, or the largest barge cannot pass under the bridge. The floating roof of the storage tank can be allowed to fluctuate no more than this, or the emissions will exceed federal standards. The materials can cost no more than this, or the project will run out of money.
The only engineering difference between the properties of money and the properties of stuff like, oh, I don’t know, the shear strength of concrete, or the percolation rate of water through gravel, is the persistent, mesmerizing illusion that the supply of money is theoretically infinite. Running out? You can always get more from somewhere. Raise taxes! Divert it from somebody else’s project! Full speed ahead!
It may very well be true that lack of funds was a prime cause, if a remote cause, of the New Orleans levee failure. If so, then it will be legitimate to ask whose decisions failed to provide the necessary funds.
But neither question can be answered until we understand the proximate cause for the breach in the levees. Not just one, but three levees, simultaneously. By "the proximate cause" I mean "the cause of pieces falling off of them in this particular spot at this particular moment."
"Lack of funding" doesn’t explain why the levees failed here instead of there, as if someone laid the cash on with a trowel and missed a spot at 17th street.
"Lack of funding" doesn’t explain why the levees failed on Tuesday instead of Saturday.
"Lack of funding" doesn’t even explain why the drearily predictable scenario, New Orleans underwater, occurred after a levee failure instead of after a levee flood, in some other, stronger hurricane, some other day in the future.
But if we figure out why the levees failed here instead of there, if we figure out why the levee failed sooner instead of later, if we figure out why the levees failed in this storm instead of holding in this storm, then we can begin to understand the degree to which more money could have made a difference; then we can begin to understand the degree to which competent foresight could have predicted that need; then we can begin to understand the degree to which the political climate enabled decisionmakers to advocate for that need in the timespans of their offices.
We might even be able to answer the question of whether the levees were competently designed and competently built at all—that is, whether the levees were up to the standards you should expect even given the particular funding constraint. In other words, did we get our money’s worth?
Henry Petroski, where are you?
bearing blog 