As Hurricane Ida lashed New Orleans on Sunday, the Category 4 storm pushed almost a dozen feet of water from the Gulf of Mexico up toward the Crescent City, threatening to send the contents of the ocean spilling down into a metropolis that for the most part sits below sea level. The result would have been catastrophic: The city would have filled up with water like a bowl, trapping thousands of residents in their homes.
That was what happened during Hurricane Katrina in 2005, but it did not happen this time. As Ida’s storm surge pushed toward New Orleans, the water ran up against an intricate system of levees and flood walls built by the federal government after Katrina, and the system stopped the water from entering the city. Although the hurricane destroyed many buildings in New Orleans and knocked out the city’s entire power supply, the worst potential damage was averted.
The main takeaway for the city, at least according to major news headlines, is that the new levee system passed the test of Ida, with the Washington Post saying the system stood “firm” against the storm. It’s true that the levees worked, but the levee system’s success this time by no means guarantees it will stand firm next time. Rather than assure us that the residents of Louisiana are safe, this week’s close call should make us rethink our reliance on levees and floodwalls — not just in Louisiana but along all the nation’s waterways.
The New Orleans levee system, rebuilt at a cost of $14 billion after Katrina, featured numerous upgrades: The new flood walls are stronger, they’re rooted deeper in the ground, and they’re designed to hold up even if water goes over them. The Army Corps of Engineers also erected a massive storm surge barrier around Lake Borgne and closed a canal outlet between the river and the gulf, which had funneled storm surge into the city during Katrina. A Corps spokesperson told Curbed on Wednesday that the system “performed as designed — it did its thing.”
For close observers of the city’s history, though, this isn’t enough.
“There’s nothing about the fact that the levees held in this storm that changes the reality that metropolitan New Orleans is sinking into the Gulf of Mexico, that the Gulf of Mexico is rising, that the warming water and climate are making storms more severe,” said Andy Horowitz, a professor at Tulane who has written a book on Hurricane Katrina. “All of those things make it inevitable that, at some point, the Gulf will come back into the city.”
The levee system was only designed to protect against a so-called hundred-year storm, or a storm that has a one percent chance of happening in any given year. Despite its numerous design flaws, the old system was designed to protect against a two-hundred-year storm—a level of protection Congress didn’t see fit to match after Katrina. The Army Corps hasn’t determined yet what size storm Ida was, but we know that many Category 5 hurricanes would exceed the theoretical capacity of the new system, as would weaker storms if they approached New Orleans from the right direction. Recent history shows that storms of such a severity are far from uncommon: In Texas, for instance, the past five years have seen the arrival of not one, not two, but three five-hundred year floods, including Hurricane Harvey.
The new system in New Orleans protected the city’s residents this time around, but other Louisianans weren’t so lucky: Multiple levees outside New Orleans failed when Ida hit with disastrous consequences.
The most notable failure was in the town of Lafitte, just south of the city. As Ida rolled north, a massive storm surge of around 12 feet overtopped the ring of seven-foot-tall levees that surround the town of around 2,000 residents, inundating almost every home and business. The town’s locally maintained levee ring, which was the product of a sustained lobbying campaign by local leadership, had been completed only last year and had been intended to attract long-term investment to the small shrimping town. Ida revealed in the space of hours that the town was much more vulnerable than engineers had assumed: At its peak, the storm surge was almost twice as high as the levees that were supposed to contain it.
Farther to the east, in Plaquemines Parish, there were multiple instances of water overtopping levees. The parish looks like a long snake, with walled land on both sides of the Mississippi, and as the wind blew the river back and forth over the course of Sunday night, the water spilled over on both banks. On the west bank, it surged through a temporary levee near the Alliance Phillips 66 plant, inundating the historic African American town of Ironton and two other subdivisions. On the opposite bank it breached another levee and rolled through the village of Braithwaite, which was devastated by storm surge from Hurricane Isaac in 2012.
If the catastrophe of Katrina was the result of shoddy engineering, the levee failures during Ida were the result of a more fundamental truth about levees, one that is true to some extent of all engineering projects that attempt to control nature. No matter how tall you build a wall, nature can always overtop it. The levees in Lafitte and Plaquemines Parish were little more than earthen hills built to a certain height: that height represented a kind of compromise between engineers’ expectations of future flooding and politicians’ willingness to spend money on flood protection. As sea levels rise and the ocean warm, and as hurricanes get more intense with each passing year, it is almost inevitable that, at some point, water will overtop the earthen walls in such places.
For a town like Lafitte, whose seven-foot levee proved no match for Ida, it might seem like the best response is just to build a taller levee: Revise the storm projections upward, figure out the new height, and pile on more dirt. This has been the standard response at most levels of government for much of the past century and a half. As a result, there are now more than 3,000 miles of levees in Louisiana, and around 3 percent of all people in the United States live in an area protected by an Army Corps flood control structure. In the lower parishes that were slammed by Ida, the Corps has been working for decades to build a chain of levees and flood gates known as the “Morganza to the Gulf” system. When finished, this system will span almost the entirety of Louisiana’s bayou coast. The highest of these walls will be around 13 feet — tall enough to withstand most floods, yes, but not all of them. The Corps and other agencies have proposed similar storm-surge barriers in several other major cities, including New York, Charleston, Miami, and Houston.
Securing the funds required for adequate protection, though, is easier said than done: In order for the Corps to justify a new project to Congress, it must prove that the total financial benefits outweigh the costs — in other words, that the structures inside the levee are worth more than the dirt used to build it. For the small and shrinking towns along the Gulf Coast, this calculus is not getting any easier, and indeed there are a few places along the coast, including the Indigenous community of Ile de Jean Charles, that the Corps did not see fit to include inside the Morganza to the Gulf.
For the communities that do receive protection from levees, the problem is that levees are a textbook definition of what economists call a perverse incentive — instead of reducing flood risk, they often increase it by enticing more people to live in those areas. If a town springs up next to a flood-prone river, and the local government builds a levee along the river, more people will move to the town because they believe the levee protects them from flooding.
Horowitz views the risk of a levee failure like the one that occurred in Plaquemines Parish as part of life on an unpredictable planet. As he sees it, living in the shadow of a levee is rational for the same reason getting on an airplane is rational: There’s a certain amount of risk involved in doing anything or living anywhere.
“Would you like to cross a river? There’s only a certain number of ways to do it,” he told me. “You can go on a bridge you hope doesn’t fall, a boat you hope doesn’t sink, a helicopter you hope doesn’t crash — at a certain point, you’ve got to get across the water.”
But there is one permanent way to escape the kind of fate that befell the residents of Lafitte: Move out of the way altogether, which in Horowitz’s metaphor would be akin to staying on dry land and refusing to cross the river. This kind of coordinated relocation is known among climate experts as “managed retreat,” and it’s gaining more currency among climate experts as disasters grow more expensive.
“As sea level rises, more and more cities are going to be asking for us to spend all kinds of federal money on levee systems,” says Orrin Pilkey, a retired coastal geographer at Duke who is known as the father of managed retreat studies. Pilkey has been critical of the Corps for decades. “It seems silly to me that we would keep building rather than respond in a more flexible way — and in some cases, that means moving,” Pilkey said.
Pilkey acknowledges that it’s not feasible to relocate a major city like New Orleans, but he insists that in other places, retreat is the only just solution. Unless you encourage people to move away from risky areas, he says, they’re going to keep losing their homes to floods and fires. The government has already started to pursue such relocation policies in cities like Houston, where it has purchased thousands of flood-prone homes so that residents can relocate to higher ground nearby, and in areas like coastal Alaska, where several Indigenous communities have moved in their entirety away from collapsing shorelines.
These policies are controversial by their very nature, since they fly in the face of people’s attachment to their homes and of politicians’ need to encourage growth. The alternative, though, is to rely on structures like the ones that protect dig the canals a little deeper, and hope the system works the next time, knowing that someday it won’t.
“I fear their time is coming,” Pilkey says of New Orleans. It just wasn’t this time.