Cold open
About fifty miles northwest of Baton Rouge, out where the levee road runs for miles without a town, there is a row of concrete and steel gates sitting across a channel that most Americans have never heard of. It does not look like much from the highway. No soaring towers. No famous skyline. Just gate piers, a road deck, and brown water churning on the far side.
But those gates are holding back one of the most powerful natural processes on the continent. The Mississippi River, the river that drains a huge share of the lower forty eight states, wants to leave its channel. It wants to abandon Baton Rouge and New Orleans and take a shorter, steeper path to the Gulf of Mexico, down a river called the Atchafalaya.
Rivers do this. Geologists have a word for it: avulsion. Over the last several thousand years, the lower Mississippi has jumped from one course to another again and again, building a new delta each time. The Army Corps of Engineers counts at least seven of those shifts across Louisiana. The river is not being stubborn. It is doing exactly what physics tells it to do.
And since the early 1960s, a complex of structures called Old River Control has been telling it no.
This is the story of how that happened. How a steamboat captain's shortcut in 1831 set the whole thing in motion. How engineers in the 1950s designed a system to freeze a river in place. How a flood in 1973 nearly tore one of those structures out of the ground. And how the fix turned this lonely spot in Louisiana into one of the most carefully watched pieces of infrastructure in America.
A river that moves
To understand Old River, you have to start with what kind of river the Mississippi is.
The Mississippi is an alluvial river. That means it flows through material it deposited itself: sand, silt and clay, carried down from half a continent and dropped wherever the water slows down. The river builds its own banks. In floods, it spills over and lays down sediment on either side, so over centuries the channel ends up running along a low ridge, a little higher than the land around it.
Down near the coast, the same thing happens at the mouth. The river dumps sediment into the Gulf and builds land outward. Every mile of new delta makes the river's path to the sea a little longer. A longer path means a gentler slope. A gentler slope means slower water. And slower water means the river carries less and drops more.
Eventually, the river reaches a point where its current route is long and nearly flat, while some other path to the sea is shorter and steeper. When a flood breaks through to that steeper path, the water takes it. The flow speeds up, the new channel scours deeper, and over decades the river moves house.
According to the Corps' New Orleans District, the Mississippi south of Baton Rouge began to take shape about five to six thousand years ago, as sea level approached where it is today. Since then, the river has migrated back and forth across Louisiana at least seven times, each time building a new lobe of delta. Some researchers describe the cycle as happening roughly once every thousand years. The exact interval is debated, but the pattern is not.
Now look at the map. The Mississippi's current route from the Old River area to the Gulf runs about three hundred and fifteen miles, through Baton Rouge, past New Orleans, out to the bird's foot delta. The Atchafalaya offers another way down, a much shorter one. The Corps puts it at about one hundred and forty two miles.
Same starting point. Same sea level at the end. Less than half the distance. To a river, that is an enormous temptation.
Turnbull's Bend and Shreve's shortcut
For most of recorded history, the Red River and the Mississippi were separate. Then, around the fifteenth century, the Mississippi swung a big westward loop that broke into the Red River's valley. That loop later got a name: Turnbull's Bend. When it formed, it captured the Red River, and it also connected to a small stream flowing south out of the area. That stream became the Atchafalaya.
When the first Europeans arrived, they found the Red River emptying into the Mississippi at Turnbull's Bend, and the Atchafalaya flowing out of the bend a few miles farther south. A river coming in, a river going out, all tied together in one muddy knot.
At that point the Atchafalaya was not a threat. And here is the reason: its head was jammed shut. A massive natural log raft, a pileup of driftwood the Corps describes as about thirty miles long, blocked the Atchafalaya and kept it small.
Then came the steamboat era.
Turnbull's Bend was a time sink. A boat had to steam around a loop of roughly twenty miles, only to end up about a mile from where it started. In 1831, Captain Henry Miller Shreve, a steamboat man and river engineer, the namesake of Shreveport, cut a channel across the narrow neck of the bend.
The river accepted the shortcut. At the next high water, the Mississippi poured through Shreve's cut and abandoned the loop. The upper arm of the old bend slowly silted up. The lower arm stayed open, and people started calling it Old River.
So now Old River was a short channel linking the Mississippi to the Red and the Atchafalaya. Water could move either way through it, depending on which river was higher.
Then, in 1839, the state of Louisiana began clearing that thirty mile log raft from the Atchafalaya, to open it as a navigable stream. The work was finished around 1840.
That was the moment the problem began. With the plug gone, the Atchafalaya started to deepen and widen. Every year it could carry a little more water. Every year it pulled a little harder on the Mississippi through Old River. The days when Old River flowed east into the Mississippi got fewer. The days it flowed west, away from the Mississippi, got more. By the late 1800s, it rarely flowed east at all.
Measuring a slow motion capture
How do you measure a river being stolen?
The Corps used a concept called latitude flow. Draw an imaginary east to west line at the latitude of Red River Landing, a few miles downstream of Old River. Add up everything crossing that line: the Mississippi on one side, the Atchafalaya on the other. Then ask what share is going down the Atchafalaya.
Around 1850, that share was under ten percent. By 1950, it was about thirty percent.
Thirty percent might not sound like a takeover. But the trend was the point. The Atchafalaya was getting more efficient every decade, and every bit of extra flow made it scour deeper, which let it take more flow next time. It was a feedback loop. The Corps' own history says that by 1951 it was apparent that unless something was done soon, the Mississippi would take the course of the Atchafalaya. One widely repeated estimate from that era put the full switch somewhere around 1990 if nothing was done. We will treat that specific date as an estimate, not a prediction carved in stone.
What would that switch mean? The Corps spelled it out plainly. If the Mississippi changed course, its old channel below the split would turn into a saltwater estuary. Along both banks between Baton Rouge and New Orleans sit refineries, petrochemical plants, grain elevators and power plants, many of which depend on fresh river water for their processes. Cities below Baton Rouge, including New Orleans, would struggle to find drinking water. Barge traffic between the nation's heartland and the ports of Baton Rouge and New Orleans would be disrupted, and the Atchafalaya Basin's own channels and locks would have to be rebuilt to handle a river many times bigger.
In engineering terms, this was not a flood risk. It was a geography risk. The river was not threatening to overflow. It was threatening to relocate.
The plan
In 1953, the Mississippi River Commission recommended that the flow from the Mississippi into the Atchafalaya be controlled by a complex of structures at Old River.
The concept was elegant. Do not try to cut the Atchafalaya off completely. That would be impossible, and it would starve the Atchafalaya Basin, one of the largest river swamps in the country, of fresh water. Instead, freeze the split. Keep the distribution of flow and sediment between the two rivers roughly where it was in 1950: about seventy percent down the Mississippi, about thirty percent down the Atchafalaya, measured as an annual average of latitude flow.
Here is the plan the design engineers proposed. First, dam the natural Old River channel. Close the open connection completely. Second, build two control structures just upstream, in new channels cut between the rivers. One would operate at all times and all river stages. The other would only operate in floods. Third, build a navigation lock, so boats could still move between the Mississippi and the Red and Atchafalaya system without the river getting a free path.
On September 3, 1954, Congress authorized the Old River Control Project.
Building it in a dry hole
Construction began in 1955. The Corps' New Orleans District started roughly fifteen million dollars of work on the two main structures: the Low Sill Structure and the Overbank Structure.
They were built in a dry hole. That is a common trick for big river structures. Instead of fighting water, you build the structure on dry land, inside an excavation, next to the river. Then, when the concrete is done, you dredge channels to connect it to the water on both sides. The river only meets your structure once it is finished.
The structures were completed in 1959. After the Corps dredged and stabilized the connecting channels to the Mississippi and the Red, they began operating. The Corps' brochure gives that start as 1962, while other summaries say the complex was completed in 1963. We will call it the early 1960s and note that the sources differ by a year.
Let's look at the workhorse. The Low Sill Structure is about five hundred and sixty six feet long, with eleven gates, each forty four feet wide. The name tells you what makes it special. Its weir crest, the floor that water flows over, is low. The Corps lists it ranging from five feet below sea level to ten feet above. A low sill means it can pass water even when the river is low. That is why it is the structure that runs every day, in every season, adjusting the split.
And it does not sit on rock. There is no rock to sit on down here. It stands on steel piles driven deep into the river sediments. The Corps' history describes those piles as ninety feet long. Remember that number. It matters later.
The second structure, the Overbank Structure, is a different animal. It is long, about three thousand three hundred and fifty six feet, with seventy three bays, each forty four feet wide. Its weir crest sits high, at fifty two feet above sea level. Most of the time, the ground in front of it is dry. It only comes into play when the Mississippi rises out of its banks in a major flood, and the extra water needs somewhere controlled to go.
Then there is the lock. The Old River Lock sits about eleven miles downstream of the gates, roughly where the Mississippi and Atchafalaya used to connect directly. It was completed in 1963 at a cost of about fifteen million dollars. Its chamber is seventy five feet wide and one thousand one hundred and eighty five feet long. The Corps says about fifteen tows pass through it on a typical day, carrying mostly petroleum, chemicals, agricultural products and aggregates. Just north of the lock, a dam closes the natural Old River channel for good.
All in, the Corps puts the original complex at a total project cost of sixty seven million dollars.
How you steer a river
So how does this thing actually work day to day?
The Corps is not setting a fixed gate opening and walking away. The river changes constantly. The Mississippi rises and falls with rain and snowmelt across half the continent. The Red River has its own floods. The Atchafalaya responds to both. So the operators adjust the gates to keep the annual latitude flow at roughly seventy thirty.
The key variable is head: the difference in water level between the high side, the Mississippi, and the low side, toward the Atchafalaya. Water falling through a gate converts that height difference into speed. More head, faster water, more force on the structure, and more energy that has to be dissipated in the stilling basin downstream so it does not scour out the channel bed.
The Corps' history says the Low Sill Structure was designed to withstand a thirty seven foot difference in water levels. That design number would soon be tested in a way nobody wanted.
1973
The fall of 1972 was wet. So were the winter and spring that followed. Persistent heavy rain across the central plains and the Mississippi and Ohio valleys pushed tributary after tributary above flood stage. The Mississippi crested several times that spring.
At Old River, the Low Sill Structure was taking the brunt of it. Instrument records from Corps prototype tests in early April 1973 show total flows through the structure around four hundred and forty five to four hundred and fifty thousand cubic feet per second.
Now think about what that does to a structure standing on sediment. Water racing through the gates is turbulent. Turbulence near a structure finds the edges, the corners, the places where flow wraps around a wall, and it digs. An eddy formed near the south end of the structure, at the wing wall that guided water into the gates.
On April 12, 1973, a major portion of that south wing wall collapsed. The Corps' history describes it as a sixty seven foot high concrete wall. Its foundation had been scoured out from under it.
But the wing wall was not the scary part. The scary part was underneath. A scour hole opened up in front of the structure and extended under it. Later engineering studies describe that hole as roughly three hundred feet across, with a maximum depth around sixty feet. The Corps' own account says it exposed about fifty feet of the ninety foot steel piles supporting the structure.
Picture that. A concrete structure holding back the Mississippi in flood, and more than half the length of its foundation piles is now standing in moving water instead of packed sediment.
If the Low Sill Structure had failed, the river would have had an open channel straight toward the Atchafalaya, in the middle of a major flood, through a hole that would have deepened by the hour. That is exactly the scenario the whole complex was built to prevent.
To take the pressure off, the Corps turned to another structure farther downstream: the Morganza Spillway. Morganza was built in 1954 to divert Mississippi floodwater into the Atchafalaya Basin in an emergency. It had never been used. On April 17, 1973, the Corps opened forty two of its one hundred and twenty five gates for the first time, passing roughly one hundred and ninety four thousand cubic feet per second. That drew down the Mississippi and reduced the head across the damaged Low Sill Structure.
The structure held. The flood passed. And the engineers were left with a damaged, partly undermined control structure that the entire lower river depended on.
Patching a foundation you can't see
The emergency repairs started immediately, and they are a lesson in working on something you cannot take out of service.
First, the collapsed wing wall was replaced with a rock dike, to guide the flow and stop the erosion at the left abutment. Second, the deep scour hole in front of the structure was filled with stone, to stop the erosion from connecting under the structure to the stilling basin downstream. Third, the void beneath the structure, under gates five through eleven, was filled by pumping in a specially developed grout. A later Corps engineering synthesis puts the emergency grouting at roughly thirty three thousand cubic yards.
Fourth, and this one is subtle, the operators changed how they opened the gates. Instead of concentrating flow through a few bays, they spread it more evenly across the structure. More uniform flow means fewer concentrated jets and eddies, and less scour at any one spot.
Then came the hard engineering question: how much could the structure safely take now?
The Corps ran stability analyses after the repairs. The answer was a strict limit on head. Immediately after 1973, the safe head was set at thirteen feet. After further repair, studies and an eight year rehabilitation program for the Low Sill and Overbank structures completed in 1981, the limit was raised to twenty two feet. That is still well below the thirty seven feet the structure was originally designed to withstand.
The Corps' history puts it bluntly: the structure had been seriously and permanently damaged, but its remaining capability was substantial. The floods of 1974, 1975 and 1979 kept testing it. It kept working. But it could no longer be the only line of defense.
The Auxiliary Structure
The fix was redundancy. If one structure might be compromised, build another one that can share the load, or take all of it if the first one has to be closed.
The Corps' New Orleans District recommended an auxiliary structure with its own new inflow channel. The Chief of Engineers approved it. Construction started in July 1981, about a mile downstream of the Low Sill Structure, and the Old River Control Auxiliary Structure was completed in December 1986, at a total cost the Corps lists as two hundred and six million dollars.
It is a heavier piece of equipment than the original. Six gates instead of eleven, but each gate is sixty two feet wide. A hydraulic model report describes them as tainter gates, curved radial gates, about seventy feet high, separated by five fourteen foot piers, for a total width of four hundred and forty two feet. Like the Low Sill, it is pile founded and made of reinforced concrete. And it was tested extensively in scale models before construction, including the stilling basin downstream, which had to safely absorb design flows of about ninety thousand cubic feet per second passing through a single bay.
With the Auxiliary Structure online, the system changed character. Now, if a barge strike, a gate problem or maintenance takes one structure out, the other can carry the job, and either can be closed safely. The Corps says the complex was not just restored to its original capability, but substantially improved beyond its original design. With all of the structures together, it lists a maximum discharge capacity of about seven hundred thousand cubic feet per second. That's roughly three hundred million gallons a minute.
A power plant built in a shipyard
There is one more structure in the complex, and it is the strangest one.
In 1977, the mayor of Vidalia, Louisiana, Sidney A. Murray Junior, started looking at hydroelectric power as a way to bring down electricity costs. Old River had something rare in flat Louisiana: a guaranteed, managed drop in water level. Every day, a large share of the Mississippi was being pushed from a higher river to a lower one. That head was being thrown away in stilling basins. Why not run it through turbines?
The result was the Sidney A. Murray Junior Hydroelectric Station, a one hundred ninety two megawatt run of river plant just upstream of the control structures. Construction began in 1985, and the plant went into full operation in 1990.
How it was built is the remarkable part. Instead of pouring the powerhouse on site, the main structure was prefabricated at Avondale Shipyards in New Orleans, while the foundation was built in the dry at the site. The Corps describes the finished structure as about twelve stories tall and twenty five thousand tons. It was towed two hundred and eight miles up the Mississippi from New Orleans, and the Corps says it remains the largest vessel ever towed up the river past Baton Rouge. They nicknamed it the "Merrimac." The Corps calls it the largest prefabricated power plant in the world.
Inside are eight bulb turbines. In a bulb unit, the generator sits in a sealed, streamlined pod right in the water flow, with the turbine runner behind it. It's the go to design for low head, high flow sites, and Old River is the definition of low head. Engineering papers from the time describe runners about twenty seven feet in diameter, an operating head that swings between about eight and twenty feet, and a combined discharge of up to one hundred fifty thousand cubic feet per second.
And here's the important part: the power plant does not get its own share of the river. It uses part of the water the Corps was already diverting to hold the seventy thirty split. A 1989 design paper estimated the plant would pass, on average, about sixty percent of the diverted flow. The Corps adjusts the plant's allocation every day, based on conditions on the Mississippi, the Atchafalaya and the Red. The turbines become another set of valves in the control system.
What the river is still doing
So the structures are built, backed up, and producing electricity. Is the problem solved?
Not exactly. Old River Control does not stop the river's pressure. It manages it. And managing it means watching constantly.
The Corps and its research center have surveyed the channel bottom in front of the Low Sill Structure, the forebay, over and over since 1963. Those surveys track how sediment moves, how the channel deepens or fills, and whether scour is creeping back toward the structure. A study published by Corps researchers reviewed those surveys across more than five decades, from 1963 to 2019.
Then there is the structure itself. In recent years, the Corps' Engineer Research and Development Center ran a new stability analysis of the Low Sill Structure, this time using decades of real monitoring data from instruments in and under the structure. The finding: uplift conditions under the structure were not as severe as earlier analyses had assumed, and the structure's measured performance supports its current operating limits. In other words, half a century of data says the 1973 repairs have held up.
Under normal conditions, the Corps says the head across the structures ranges from about four to nineteen feet, depending on the stages of the three rivers. The limit is twenty two. The greater the head, the greater the stress. That gap between normal operations and the limit is the margin the whole system lives on.
And the upstream pressure has not gone away. The Atchafalaya is still the shorter route. The Mississippi still slopes more gently on its current path. Every flood still pushes. The difference now is that the push meets steel gates, concrete, rock, grout and redundancy instead of an open channel.
Why this matters
There's a detail about this place that captures it perfectly. Louisiana Highway 15 runs along the top of the whole complex. Starting at the power plant, it crosses the Overbank Structure, then the Low Sill Structure, then the Auxiliary Structure, then the island between the rivers, and finally the Old River Lock. You can drive across the entire machine in a few minutes, on a rural two lane road, and never see a sign telling you how much depends on it.
Below Old River, along the Mississippi, sit some of the busiest port and industrial corridors in the United States. Fresh water for cities. Process water for refineries and chemical plants. A barge route that ties farms in the Midwest to export terminals on the Gulf. All of it assumes the Mississippi will be where it is today.
Old River Control is the reason that assumption holds.
It is not the biggest dam in America. It is not the tallest, or the newest, or the most photographed. But it may be one of the purest examples of what engineering actually is. Not conquering nature. Negotiating with it. Measuring a force too big to stop, deciding how much of it to let through, and building something strong enough, and redundant enough, to hold that line every single day.
A steamboat captain cut a shortcut in 1831. A log jam was cleared in 1839. A river noticed. And a century later, engineers had to build one of the most important control structures in the country to keep the river from finishing what those decisions started.
The Mississippi is still pushing. The gates are still holding.
Next time on Steel and Speed: a highway that runs seventeen miles across the mouth of the Chesapeake Bay, and then dives under the ocean. Twice.