By Jim Schutze
By Rachel Watts
By Lauren Drewes Daniels
By Anna Merlan
By Lee Escobedo
By Eric Nicholson
Trying hard to think how to bring proper perspective to the absolutely stunning pronouncement by the U.S. Army Corps of Engineers last week that the Trinity River in Dallas is not in danger of calamitous flooding after all.
We have been told over the last three years that the only way Dallas could protect itself from disastrous floods was through the construction of massive public works. It's as if the corps had come to us in the form of a mighty voice from above and said, "Make thee an ark of gopher wood; rooms shalt thou make in the ark, and shalt pitch it within and without with pitch."
Fine. We got gopher wood. Pitched with pitch. We're almost done, in fact. We have spent $18 million buying land needed for reinforcements to the Trinity River flood-control levees, $25 million for testing, another amount I can't get my hands on yet for the design of the reinforcements, and we are just now prepared to launch a $30 million construction campaign.
If we were Noah, we would have everybody on board now. Last week's sudden announcement, then, would be like the voice from the sky saying, "Make thee not an ark of gopher wood, after all. And be thee not sullen, Noah. You could use it for a restaurant."
I asked the corps to offer me some kind of explanation. They said they would. I checked back several times. Nothing. They wouldn't even try to explain it.
The Trinity River floods twice a year. The levees are big embankments or grassy dikes made of dirt, 23 miles long on both sides of the river, from a quarter to half a mile apart. When they fill up, they create a 23-mile-long lake from 20 to 50 feet deep.
If the levees ever broke, all of that water would go rampaging into downtown.
The corps told the city for three years that one of the big problems with the levees was the discovery of sand deposits beneath them. On the Mississippi where levees have been built on sand, the force of flood waters has been enough to scour through the sand and collapse the levees from beneath.
The corps said that problem and other shortcomings here meant the levees were not safe even to the minimal protection level of the so-called "100-year" flood.
Technically a 100-year flood is one that's likely to occur once in a century, but in fact the way the probabilities work is quite different. The actuaries say that if you live in a protected flood plain, you have about a 25 percent chance of getting hit by a 100-year flood during the life of a 30-year mortgage.
Even the 100-year-flood protection level — the minimum set by the corps — would be considered absurdly low in Europe. The Netherlands frets over the dunes area near the village of Ter Heijde on the North Sea because it's only safe to the level of a 10,000-year flood — too risky for the Dutch.
So not being safe even to the 100-year level in downtown Dallas is really bad. The fix we're about to embark on requires construction of massive underground walls all the way from the bottom of the levees down into the earth through strata of sand, clay and shale to bedrock far below.
Obviously we didn't decide to build underground walls beneath the levees on a lark. We did it because we were told by the U.S. Army Corps of Engineers that we had to do it, an opinion backed up by a private engineering firm we hired at great expense.
But at last week's City Council briefing, corps officials told the council that the levees, as they stand, are safe to a level somewhere between the 1,000-year flood and maybe the 5,000-year flood.
But then in response to questioning by Oak Cliff council member Scott Griggs, a corps official delivered the real stunner. In terms of under-seepage — the problem we are building the underground walls to correct — the levees are already safe, without the new walls, to the 100,000-year level.
To put that period of time in context, consider this: 95,000 years ago our own species, homo sapiens, still shared the earth with a sister human species called homo floresiensis, a much smaller creature with a brain about a third the size of our own.
Only 12,000 years ago did we finally figure out how to domesticate chickens.
I had a late-evening chat last week with council person Angela Hunt about the corps' surprise announcement of a 100,000-year safety guarantee for our levees. She suggested that 100,000 years from now we may be zipping about in space as disembodied intellects barely able to remember planet Earth, let alone worrying about floods. Which is to say, again, that something really does not add up here.
Griggs raised some of these questions in a letter to the city manager last week. He asked, among other things, "Given the corps' levee reassessment, is the cut-off wall required? If the levees today meet the 100,000-year standard for seepage, why is an expensive cut-off wall required to mitigate the risk of seepage?" He had no more luck than I did getting an explanation.
I did ask the corps a couple of more pointed questions — things that, at least in theory, should have had easy answers. For example, the corps has been carrying out a new national program of periodic levee inspections under what is called the "Corps of Engineers Levee Safety Program," funded by the American Recovery and Reinvestment Act of 2009. I noticed in looking at it that Texas is one of three areas that did not receive the ARRA funding.
So I asked if that new level of inspection was occurring here anyway. I also asked another more detailed question, but it went to the same point: If that new level of testing is not happening here yet, because it wasn't funded here, then where is the corps getting all this happy news about our levees?
In the absence of any response from the corps, I have to mention some unpleasant possibilities about where it might be coming from. I refer, of course, to the old poop-chute. And here is why.
In 2009 when the corps rated the Trinity River levees as "unacceptable," they also gave flunking grades to major levee systems on the Mississippi near St. Louis, in California and elsewhere. Dallas, like many of those communities, woke up to an unpleasant fact about levee maintenance: In most instances, the corps does the inspecting and has final say over design and construction, but local communities are responsible for repairs, especially where they have failed to do proper maintenance over the years.
Suddenly cities across the country were looking at repair costs that could potentially reach billions of dollars. Dallas joined several of those cities in lobbying the corps to back down from its standards.
Since then the corps has announced a whole new system for measuring flood risk, taking into account how much risk communities are willing to live with and how good their evacuation plans may be. The corps has invited the media to attend seminars on the new system. I attended one of them.
I suspect few of the journalists in the room that day were quick with long division, let alone probability. I think it's safe to say not a one of us left with a clear understanding of the algorithm for the number of life rafts divided by seepage rates, and I even wonder if that could have been the desired outcome.
The New York Times published a story last week on the new flood works in New Orleans describing "... the two-mile 'Great Wall' that can seal off the channel from Lake Borgne to the east, or the billion-dollar west closure complex, which features the biggest pumping station on the planet."
Amazing big stuff! But, except for one local official quoted saying the new system is "woefully inadequate," the story contained only scant discussion of what is known about flood risk and what the new works can do about it. Flood-control projects that look amazing when they're dry can be rendered laughable by a big enough flood.
Generally speaking, we journalists are slack in our coverage of this story, and the U.S. Army Corps of Engineers knows that. Meanwhile, they have been under intense lobbying pressure to back off their standards.
The almost casual tossing out of a 100,000-year guarantee in Dallas, followed by a stubborn unwillingness to provide any explanation, smells like a big backing down, cynically calibrated to coincide with the inability of anybody to catch them at it.
That leaves us about where Noah might have been. What was he supposed to do? Shake his fist at the heavens and say, "I demand to see your numbers?"
Perhaps it is all a part of yet another Dallas real estate development scheme. When the east levee breaks, the water wipes out Industrial (oops, I mean Riverfront), the jails, Stemmons Freeway, Reunion, Union Station, Dealey Plaza, the courthouses, and Griffin is renamed Waters Edge Avenue. Victory is redeveloped as a water park.
You think Mary and USACE will give you, along with Scott some correct numbers and straight answers? It will never happen.
Technically, a 100-year event (be it flood or drought) is one that has a one percent chance of occurrence in any one year. At some point in time, trying to simplify things (which did not work out very well) this became known as a 100-year event. The math was 1/0.01 (which is one divided by one percent ... 0.01). The so-called 100-year event has a one percent chance of occurrence this year, next year, the year after ... etc etc. It also has a 99 percent chance of NOT occurring this year, next year, the year after etc etc. The odds of it not occurring in all of the next 3 years is 0.99 * 0.99 * 0.99 which is 0.97 (97 percent of NOT occuring which is 3 percent of occurring in at least one year) .... if you do this for 30 years rather than 3 then you get 0.99 ^ 30 = 0.7387 (or 74% rounded) ... so the odds of it NOT occurring for the next 30 years is 74% and the odds of it occurring at least one time is 26%. To really illustrate the folly of calling a 1% event a "hundred year event" ... consider the odds of a so-called hundred year event occuring at least once in the next 100 years. Well, 0.99 ^ 100 = 0.366 (37% rounded) which is the odds of it NOT occuring in the next 100 years .... therefore the odds of it occurring at least once in the next 100 years is 63% (approx) Hope this helps - you would be better off dropping the term 100-year flood from your lexicon and rather use 1% event .