© Rich Sugg
"Ooh, man. Ooh." Schaeffer followed and looked down in disbelief. A riot of water roiled where he'd spent a lifetime of lazy fishing.
"I've never seen anything like it," he said. "Nothing even close."
Eight years out of 10, the 14 flood gates, 40 feet wide, spill not so much as a bucket of the brown water into the Missouri River.
Now enough is barreling out of Lewis & Clark Lake to cover a football field 3½ feet deep every second. Water will race through the dam at that record rate, ultimately swamping farms and towns for hundreds of miles downstream, through August.
"When your bathtub is full, you just can't put any more water in it," said Dave Becker, the operations manager for the U.S. Army Corps of Engineers at Gavins Point. "Water is going to spill over."
But how did the bathtub get so full? Why did the six huge Missouri River reservoirs - Gavins Point is the farthest downstream - fill to the brim and force the months-long release of floodwater?
The short answer: The corps could have prevented or drastically held down flooding by opening flood gates sooner. The reasons it didn't - reasons putting government water managers on the spot this summer - rest in a tangle of history, physics, meteorology and politics.
We had ample warning last winter that snow was piling on the Rockies. Consequently, the corps made room in its man-made lakes for the coming runoff. Just not enough.
It chose not to make more room, its engineers point out, because it was unaware of the torrents of rain that would deluge the Missouri basin in May. As the river now rises in downtown Kansas City and floods soybean fields and hamlets to the north, the corps insists it couldn't have predicted those storms.
The agency also says it was simply following orders - from us. Over lifetimes and through our politicians, we've said we don't just want those dams to protect us against cataclysm.
We want cheap electricity, and the system gives us plenty of hydro power.
The sparsely populated Dakotas want to keep the reservoirs close to full to draw boaters and sports fishermen, and to irrigate the lower reaches of their river valleys.
Downstream, farming interests want enough water to keep the Missouri River barge industry - a steadily shrinking business - alive.
All that means storing water in the reservoirs in the spring, not leaving empty space to protect against flooding.
"There's a natural tension there," said Brig. Gen. John McMahon, who commands the corps division that manages the Missouri. "You can't say I'm only for navigation, I'm not for flood control or recreation or whatever. ...
"They're wicked problems."
Come the Snows
As snow fluttered from the sky late last year, the corps got ready.
In its December 2010 plan it set aside 22 percent of the reservoirs' capacity to collect snowmelt and normal spring rains.
"The entire system flood control zone will be available to store surplus runoff," the plan promised.
That 22 percent, the corps thought, would be plenty. A century-plus of weather data studied by hydrologists suggested it would provide room for the coming moisture even after a bad winter.
Creating more empty space in the reservoirs was certainly possible, but the competing needs for water - for navigation, power, recreation and other uses required by a mountain of regulations directed by Congress and the courts - explain why the corps couldn't. The reservoirs aren't just catch basins, they're also places to bank water to feed those needs on a not-so-rainy day.
"Releasing at higher-than-normal rates early in the season ... is inconsistent with all system purposes other than flood control," the corps' controlling river manual says.
In 1997 the corps did release extra water starting in February in anticipation of heavy snowmelt. But that year's snowfall showed up early enough to make the need for late winter releases more obvious, the corps says. And, it adds, winter releases are chancy because ice floes can dam the river unpredictably.
This year water managers kept almost 57 million acre-feet of water in the reservoirs.
And they left empty the remaining 16.3 million acre-feet - an acre-foot equals one acre covered a foot deep in water - to absorb the spring runoff and rains.
Snowfall in the Rocky Mountains came gentle early in the year. Then it got heavy, piling foot upon foot of frozen water in remote valleys and ravines.
By February, hints of coming problems deepened, with snow estimated as much as 50 percent above normal. One South Dakota official warned in February of a "biblical" springtime flood, according to the Pierre Capital Journal. Even today, nearly a third of the snow expected to melt has yet to trickle into the river.
By March 2, the Missouri River Basin Forecast Center warned of "major (spring) flooding at many locations" from saturated soil, rain, and "significant snowpack" in the areas above the river.
Still, in offices in Omaha, Kansas City and elsewhere along the river, the corps remained confident its reservoirs would have adequate room. The snow had come gradually and was piling up later than usual.
But in late April, the corps opened flood gates at Gavins Point - although not as wide as they are now - to flush away what was beginning to look like an eerie surplus of water. Small releases at the other dams also started.
Then came the unprecedented surprise: Torrential rains began falling in Montana and the upper Midwest. The rain was a surprise partly because eight weeks earlier the National Weather Service had foreseen a spring likely to be normal in the basin's upper reaches and, if anything, on the dry side farther downstream.
But even if those forecasts had been more accurate, it might have made little difference, corps documents show. That's because the corps appears to distrust long-range precipitation forecasts, consistently downplaying their role in decisions.
"Rainfall events ... cannot be predicted," the corps' river manual says.
If it relies just on long-range forecasts, the corps thinks, it's more likely to make a wrong decision on water flows. If it releases lots of water in anticipation of heavy rains and those rains don't come, too little water will be left in the reservoirs to float barges through Missouri in the summer.
Instead, the corps places much more confidence in history - its entire river flood plan relies, in large part, on an 1881 Missouri River flood. Each year it prepares five runoff and release models based on weather statistics back to the 19th century. That "eliminates the need to forecast future precipitation, which is very difficult," this year's plan says.
The models factor in past droughts and extreme precipitation, of course, but they only include things that have actually happened. Engineers concede there's always a chance for events they've never seen before - in fact, this year's plan set a 10 percent chance of runoff greater than the worst case of the five models.
That chance came true in mid-May, when rainfall exceeded anything the corps had seen in the last century.
Violent rainstorms soaked Montana, Wyoming and the Dakotas week after soggy week. They dumped trillions of unexpected gallons of water into the basin. The upper Yellowstone River basin got more than double its average rainfall in May. The lower Yellowstone got more than 3 1/2 times the normal. Other tributaries in Montana and the Dakotas swelled, too.
And the snowpack had grown, ominously. Snow in the Bighorn River basin piled up more than 2 1/2 times over normal. In the Wind River basin, it was triple the average.
"That's when we got concerned," said McMahon, the corps general. "That's when we started ... realizing that the rain was the wild card and was taking away the storage space that we had."
In fact, the 16.3 million acre-feet of storage had turned laughably inadequate. Runoff in May alone reached 10.5 million acre-feet above Sioux City, Iowa, smashing the previous record by 30 percent. The upper Missouri basin got as much rain as it usually does in an entire year in just the last three weeks of May.
By June the annual runoff forecast had reached 54.6 million acre-feet, more than three times the runoff the system normally holds.
The corps "saw rain in the upper part of the basin they just weren't prepared to handle," said Tim Cowman of the Missouri River Institute at the University of South Dakota.
The only answer: Open the floodgates on the six dams, pouring all the excess water into the Missouri.
The 2011 Missouri River flood was fully under way.
Assessing Blame
The rising water has only increased the way some people along the river and their politicians second-guess the corps.
"The corps ... isn't very dynamic," said Diane Oerly, president of a Missouri-based group called Friends of Big Muddy. "And they're trying to deal with a dynamic being - the river. If all you do is look to the past to define reality, you can't be too accurate."
Kansas Gov. Sam Brownback has called for a federal investigation of the corps' decisions. Missouri state Sen. Brad Lager, a Republican who represents hundreds of flooded-out constituents, said the corps' decisions border on the criminal because it failed to respond adequately to growing snowpack.
"Someone who does this for a living knew this was going to be a problem," Lager said.
McMahon said the criticism was misplaced: "Nobody likes to have their livelihoods taken away. We had a whole bunch of unprecedented rain in the upper basin, and that rain fundamentally took away the flexibility that we had built into the system."
Even Lager, when pressed, concedes the corps' ability to stop the rain was limited.
"You always encounter a time when Mother Nature deals you a hand you just cannot manage," he said. "But I think they could have planned better" by releasing more water in the winter.
'Its Prayer Fulfilled!'
Although the floods may have raised the volume of the argument over the Missouri, it hasn't changed the content.
Indeed, upstream and downstream states have been stuck in almost constant conflict for nearly 90 years, each trying to bend the river to its needs.
Kansas City businessmen - led by developer J.C. Nichols - began the push in the 1920s to build dams and deepen the river to aid shipping from Sioux City to the Gulf of Mexico.
The first dam, in Montana, forms Fort Peck Lake, the fifth-largest man-made lake in the country.
"Modern science has harnessed this wild and difficult stream," Nichols said in a 1935 speech. "The longest haul agricultural territory in the world has its prayer fulfilled!"
Today some students of the river say Nichols' hubris was folly. Barge traffic has all but disappeared. The river was open to navigation as far upstream as Leavenworth last week, yet only one towboat was moving barges.
At the same time, there are limits to how man can tame a river as powerful and unpredictable as the Missouri. Without the man-made reservoirs in place, flooding in Missouri, Kansas, Iowa and Nebraska would be far more ruinous than what we're seeing menacing residents from Sioux City to Leavenworth to Levasy and beyond this summer.
In fact, those cities and others saw a calamitous flood in 1993 that differed from this year's in important ways. That disaster was caused by torrential rain far below the dams, on the plains of Kansas, Nebraska, Iowa and Missouri.
And while the reservoirs this year started out relatively full, a long drought left plenty of storage space in the reservoirs in 1993. At the crest of that flood, the dam system released an average of just 8,000 cubic feet of water per second, a relative drip.
Today the release at Gavins Point - 160,000 cubic feet a second - is more than twice the previous dam-era record. But consider that in 1952, two years before the dam was built, flooding sent three times as much water raging down the channel.
Still, there would have been virtually no flooding in 2011 had the corps let loose far more water earlier. But to do so would have invited criticism from the commercial users of the river worried about drier times later in the year.
"The Missouri River is not a reflection of democratic values. It does not provide equal benefits to all valley residents," said Robert Schneiders, author of "Unruly River." "Instead, the bulk of its benefits go to a few individuals and interest groups in Missouri."
Others take an even broader view, suggesting climate change has contributed to the high water.
"The corps has not taken climate change into account in its master manual for the Missouri River," said blogger and environmentalist Brad Johnson.
Evidence of the role of possible climate change in the flood is mixed. The basin flooded more than a dozen times in the last century, and droughts come regularly. The basin has set records for runoff, however, in two of the last 15 years.
The American Rivers environmental group has long been critical of the corps management of the river - and of the squabbling over water by upstream and downstream states.
It sees this latest round of flooding as a chance to rethink the river. It concedes it would be impractical to remove the dams - too much commerce depends on having a mostly tamed river.
But perhaps, the organization suggests, it's time to move levees farther back in less populated areas. Move homes out of the flood plain. Give the Missouri more elbow room.
"That way you get more natural storage below the dams," said Shana Udvardy, who studies flood management policy for the group.
Sen. Tom Harkin, a Democrat from Iowa, didn't blame the flooding on the corps.
"Let's face it. In the last 50 years we've plowed up more land, and we've put more land into cultivation near the river," he said at a stop last week near flooding in Hamburg, Iowa. "We might need to rethink that."
How to Use the River
That rethinking has already started.
Congressional hearings are almost a certainty. Brownback wants a "9/11-style" commission. In 2009 Congress authorized a five-year study of the current uses for the river, a $25 million effort that's underway but not yet fully funded.
Naturally, downstream politicians have fought the study, attributing nefarious motives to its supporters. A new Senate working group formed last week.
For some 40 years David Pope has been involved in the basin's debates, for a time as the Kansas water director and more recently as the executive director of the Missouri River Association of States and Tribes.
Pope said this year's flood was sure to lead to some revisions in river management. The corps might gain flexibility to make flood protection a greater priority. That, in turn, would mean the waterway would breach fewer levees.
But doing so would risk the outrage of barge operators or municipal water systems left high and too dry at the end of a summer, Pope said. Had this year's rains not been the heaviest ever recorded in parts of the Missouri River basin, and the corps had let out more water, upstream states likely would have sued for the damage done to their fishing guides and roadside motels.
"You just make a bigger flood pool and have less water in the storage for all the other uses," Pope said. "There is a direct conflict between flood control and navigation. Those are really direct trade-offs, and you can't have it both ways."
For its part, the corps promises to take another hard look at its five prediction models - with this year's muddy cataclysm now part of that math.
"Now we have a new data point," McMahon said. "Now...we go back and ask if the flood control space is adequate. Should it be more? And if it should be more, at what cost to those other uses?"
'Natural tension there'
Here is an edited transcript of The Star's interview with Brig. Gen. John McMahon, commander of the Army Corps of Engineers' Northwestern division. Questions and answers have been edited for clarity and length.
The Star: How would you grade the corps for its river management?
McMahon: We've done as well as we could under the circumstances, but we will go back and scrutinize and review everything we've done.
Q: Did the corps manipulate releases for reasons other than flood control, like buying up property?
A: That's utter nonsense.
Q: Did flooding in the Mississippi River play any part?
A: No. We were asked in early May to look at that. We did look at it, but how we manage the Missouri is mandated in law.
Q: It's clear the corps has an enormous number of competing interests in how it manages water in the river. How do you juggle those interests?
A: There are eight authorized purposes. ... Flood control requires empty space. The other purposes require water. So there's a natural tension there. That balancing act is very delicate.
Q: But why not set aside more space for flood control?
A: The (space) is a product of this public process, where you get the science and engineering combined with the politics - that's the master manual process.
Q: Some people say it was a mistake to try to control the river.
A: That's a great other question. What if we hadn't had the dams and levees? What would have been the extent of the damage? History is the teacher, and we'll do some modeling, but 1952, 1844, 1881 we had floods, too.
Q: What happens now?
A: The first step is, get the water out of the system while we minimize damage and avoid loss of life. We can't really go back into the dams and levees until the water gets down to a certain level and we can do the inspections. Given the high water throughout the system, even through late fall ... as soon we can get into the dams and look at the levees, we will do that. It'll probably be late fall, early winter.
Q: Do you anticipate a great expense for repairs?
A: We do.
Pulses and faux flooding
The same water specialists trying to wrestle control of Missouri River flooding in most years actually set out to make the channel mimic natural floods.
Technicians at the Army Corps of Engineers usually release relatively small pulses of water into the riverbed along the South Dakota-Nebraska border for a handful of days in March and May.
Those small gushes are meant to mimic natural spring rises that would make the once-meandering waterway stray from its banks before dams and dredges mostly tamed its behavior.
That faux flooding discourages shore birds - like the endangered piping plover and least tern - from nesting too close to the water's edge. The pulsing also clouds up the river, helping fish such as the endangered pallid sturgeon that evolved to survive in murky waters.
The pulses also demonstrate - often overlooked in this spring of troublesome glut - how precious water in the Missouri River can be. The water pouring from Gavins Point now runs at 32 times the rate of the routine spring releases. So even if the corps had released 5,000 cubic feet per second down the riverbed in March, the difference in the reservoirs would be negligible. The dams would still be gushing floodwaters today.
Yet fishing and boating advocates upstream loathe the small pulses, seeing them as stealing from scarce reservoir reserves. Barge interests despise them for the same reason, fearful that water won't be around when they need it to move autumn's harvest.
And Capital Sand Co., based in Jefferson City, is dreading the possibility that higher river levels might take its three dredging boats off the water for a few days.
"We need to be out there working every minute we can," said Steve Bohlken, Capital's vice president. "We just hope the crest passes quickly."
Reader Comments
to our Newsletter