Nashville is nearly three months dry, two months shave-and-shower-sanctioned, and still — unsurprisingly yet nevertheless frustratingly — quite a ways from a truly satisfying explanation of what went wrong.
Late last month, the Army Corps of Engineers released the “draft May 2010 Flood Event in the Cumberland River Basin After-Action Report,” a bulky, esoteric document written mostly in long acronyms; then there was a congressional hearing, in which Democrat Rep. Jim Cooper, of Nashville, fiercely criticized the corps for what he believes was a lack of planning before and coordination during the flood. “I know Boy Scout troops that are better prepared than this,” Cooper said in his testimony.
In an interview with The City Paper, Cooper said the report, while lengthy and fairly detailed is still a work in progress.
“It’s taken a while for people to digest that report. It’s a lot,” he said. “I spent the better part of a night reading it.”
The narrative is still far from complete and likely won’t be for many months until a final report is eventually released, he added.
“They’ve asked us to help edit this report because we’re pretty good at finding mistakes,” Cooper said. “It will be well into the next year before we get a post-flood report. My guess is a number of people will have retired or been promoted by the time a report comes out. They’ll make it sound ultra-technical — a bunch of math equations, but I think we’ll find some interesting information in it.”
Much of what we’ve gleaned from the initial report is basically what we knew before. What went wrong was that it rained too much for way too long. Why we didn’t know it was going to rain that much until it was already raining is because no one could have known it was going to rain quite that much. Also, there were some pretty serious communication problems between and within the agencies — specifically the corps and the National Weather Service — tasked with dealing with this type of disaster.
The first two points are self-evident and don’t require a large document to explain them. Most of the big revelations in the report deal with the communications issues and their ramifications. The report highlights 18 problems, most of them related to communications.
We know now, for example, the reason that the corps’ Nashville District offices experienced a nearly 12-hour Internet outage (a Verizon T1 line was underwater); why this was such a big problem (it meant they couldn’t easily receive weather data or post dam release data to its site except by using a staff member’s Internet-ready cell phone); and that they had no contingency plan like wireless-equipped laptops or more than one Internet-capable phone.
We also know that in a Sunday, May 2, conference call between the National Weather Service and the Nashville corps offices, the NWS misunderstood waterflow information provided about Old Hickory Dam, mistaking “forecasted” outflow for the intended actual waterflow figures. That’s why there was some confusion as to when and how high the Cumberland would crest.
Cooper sees this as the corps placing fault with the National Weather Service. He places blame on the corps itself, again pointing to what he sees as its penchant for complicated technical language.
“Absolutely, that’s what they do. Part of it is the nature of engineers. Part of it is the military. But part of it is that it’s a protective shield, but that can be very dangerous — for example their inability to communicate the huge releases from Old Hickory on May 2, and how that was just a huge, devastating thing for Nashville,” Cooper said. “It’s just assumed that the weather service realized that Old Hickory didn’t exist. They were releasing so much water that the dam might as well not exist. To assume that for 12 hours, 15 hours on the crucial day. That’s a killer assumption, literally.”
The report also reveals the existence of an April 29 email from the corps’ Great Lakes and Ohio River Division Offices Water Management Department warning the local office of an increased flood forecast. On its face, it’s simply an embarrassing admission. The email was ignored, possibly due to the high-volume of incoming mail, according to the report.
“You would think people would read emails from emergency agencies during an emergency, and this is while the Internet was up, well before it went down. And we don’t even know what the content was,” Cooper said, adding that he has made a call for a release of all documentation pertinent to the flood, hopefully to be included in the later post-flood report.
The City Paper was able to obtain a copy of the email, sent at 10:54 a.m. and marked high importance. In fact, though it does note the possibility of flash flooding in the region, its rain forecast is still significantly below the actual total. The email warns of “widespread rainfall amounts of 2 to 4 inches” over the May 1-3 weekend, with “isolated rainfall amounts of up to 7 inches.” It notes the then-current forecast of 37 feet for the Cumberland’s crest in southern Kentucky, adding that rainfall could push it higher. Finally, it warns that pool rises at Wolf Creek and Center Hill dams are expected.
The report says that the email resulted in the pools at Old Hickory and Cordell Hull being lowered half a foot, Cheatham by one foot “which placed the projects in a better position to manage for the rainfall totals.” That was done by May 1. Not included, though, is any information as to how long it was ignored before those draw-downs began.
Here the email admission may raise some questions.
In the days immediately leading up to the flood, Cordell Hull, J. Percy Priest and Old Hickory were each maintaining near maximum
pool levels, at 504, 490 and 445 feet, respectively. While Cordell Hull, for example, was lowered by half a foot by Saturday morning, its minimum level is 499 feet, with a surcharge storage pool up to 508. Floodgate releases didn’t begin there until Saturday, reaching 50,000 cubic feet per second rushing toward Old Hickory — which was already above capacity — by midnight.
In a teleconference following the congressional hearing last month, Gen. John Peabody, commander of the Great Lakes and Ohio River Divisions of the corps, said that the dam releases were crucial to avoid overtopping the dams, causing even greater destruction.
“If the water overtops the dam, it will enter into the projects themselves. It will short out all the electrical components. It will destroy the hydraulic elements. It will destroy the operating mechanisms. And it will likely seriously injure, or more likely kill, any project personnel that are in the office,” Peabody said. “More importantly for the city of Nashville, it will result in a free-flow situation, in which water will flow unhampered by anything we can do until the dam is repaired.”
Peabody added that dam releases reduced the Cumberland’s crest by 5 feet. That higher level, he said, would have breeched the MetroCenter levee.
But if an advance warning hadn’t been ignored, could or would the corps have lowered lake levels even more before the rain, perhaps to their minimum levels, potentially avoiding such massive mid- and post-rain releases and, maybe, making Nashville less vulnerable to such severe flooding?
“It’s certainly possible,” Cooper said. “The release from upstream to minimize flow downstream, I can’t tell that it was done at all. Now, I don’t have specific proof, we don’t have full data yet, but Cordell Hull is a big piece of the puzzle, as is Old Hickory, as is Percy Priest. Those have the most direct bearing.
“The corps, their answer will be, one, that they do not pre-condition the dams. They do not trust any forecast no matter how good to release water. Then they’ll tell you, ‘Well, we lowered Old Hickory a half a foot.’ ”
According to Hershel Whitworth, a hydraulic engineer at the corps’ Nashville office, the warning, late and unspecific as he says it was, wouldn’t have made much of a difference. In fact, Whitworth said that when his office modeled the outcome if Old Hickory had been drawn down 2 feet, the crest would have only lowered 3 inches more.
“The weather warnings we got were not specific enough as to location to make a good decision about whether a pre-draw would not actually make the situation worse,” Whitworth wrote in an email. “For a pre-draw to work, it has to be done with enough time to get the water completely out of the Cumberland River system; otherwise, you would just move the problem to another location. This takes about three days to do depending on which project it is.”
But Cooper believes the lesson here is that water should be routinely lowered, seasonally rather than based on specific forecasts, a proposal that would require new legislation.
“The lakes in East Tennessee are routinely lowered,” he said. “We’re just not used to seeing that in Middle Tennessee.”
Of course, East Tennessee is a different agency. It is managed by Tennessee Valley Authority, not the Army Corps of Engineers, a point that reminds Cooper of another major issue brought up in the corps’ report, that a storm that big did not immediately trigger round-the-clock staffing in the Nashville office.
“TVA, they manage their water 24-7,” he said. “That’s another area where the corps is different.”