Rain continues to fall (as it has for the past month) in record-breaking amounts across the middle Mississippi and Ohio river valleys, swelling the two waterways and their tributaries. As some residents evacuate and others await word on whether they must flee, the U.S. Army Corps of Engineers is considering its increasingly limited options for containing a major catastrophe already washing away homes and farmland.

Most of those options depend on the volume of water and the length of time that water stresses the man-made infrastructure of levees, dikes and spillways built along the Mississippi.

The Corps partially opened Louisiana's Bonnet Carre Spillway earlier this week to reduce the volume of water heading toward New Orleans. By the end of Friday a total of 264 spillway bays (out of 350) will be opened, according to the Corps, which plans to open additional bays as needed. They are also considering opening some portion of the 125 bays at Morganza Spillway near Baton Rouge as early as this weekend—for the first time in 38 years—to divert Mississippi River floodwaters heading down from the Midwest and Ohio River into the Atchafalaya River Delta, according to The Times-Picayune. If Morganza is opened even partially several thousand homes as well as farms, a wildlife refuge and a small oil refinery in the Atchafalaya Basin would be at risk of flooding in order to prevent a deluge from hitting densely populated Louisiana state capital Baton Rouge and New Orleans.

Heavy rains aside, the current flood conditions can be traced back to a number of factors, including the buildup of shallow flood barriers atop loose soil along the Mississippi and its tributaries over time that narrowed their pathways. This has led to more water running faster through narrower spaces, thereby placing more pressure on the region's levees.

To gain a better understanding of the precarious situation along the Mississippi as well as how such flooding might be mitigated in the future, Scientific American spoke with J. David Rogers, a professor of engineering at the Missouri University of Science and Technology, who also teaches flood control courses for the Corps. Whereas floods are inevitable, their impact can be mitigated through well-designed and constructed flood barriers, proper land management along major bodies of water, and a better understanding of how floods work, according to Rogers.

[An edited transcript of the interview follows.]

How serious is the situation facing communities along the Mississippi (in particular, Memphis, Vicksburg, Baton Rouge, New Orleans)?
The forecasted flows should be within the design capacity of the Mississippi River and Tributaries [MR&T] flood control project constructed by the Army Corps of Engineers between 1931 and 1980. That doesn't mean there won't be any levee failures or other problems, but that it should be within the design capacity to handle it.

Levees and other man-made attempts to regulate water levels often become an issue when the Mississippi threatens to flood its banks. Are such floods an inevitable fact of life along the river, or is it possible to protect all of the communities that have sprung up along the river over time?
We have probably diked off too much of the river's active flood plain; in many cases close to 90 percent of the historic flood plain. We probably shouldn't have diked off more than 75 percent, because the dikes located close to the low-flow channel (pdf) [designed to concentrate flows and increase channel velocity and depth during low-flow periods] are those most prone to foundation problems, due to under-seepage through pervious materials. Floods are an inevitability, and we would need 1,000 years of hydrographic records to accurately estimate a flood with a 100-year recurrence frequency. We have about 160 years of records, so there is a lot of guesswork involved.

Are there certain areas along a waterway that are more suitable for levees to be built, or is it a rule of thumb that any community living along a major waterway should protect itself using a levee system?
We shouldn't be constructing levees intended to protect high-value assets close to the low-flow channel without substantive "design-in-depth" protective measures against under-seepage (jpg). This includes slurry cutoff walls, sheetpile cutoffs, under-drains (pdf) and plenty of uplift relief wells (pdf). The closer we construct the levees to the low-flow channel the more pervious the foundation materials beneath such levees. Virtually all non-overtopping-induced levee failures occur due to under-seepage along old channels and sloughs, which have increased permeability. The farther away from the low-flow channel, the foundations tend to become less and less pervious. So, protective levees should really be set back a considerable distance from the low-flow channel if we desire robust and survivable protective systems. This means giving back more of the natural flood plain to the rivers—that is, increasing the "river's share" from 8 to 10 percent of the flood plain to something more like 20 to 25 percent. The levees that are set back from the river channels also tend to comprise more clay, which provides increased soil cohesion, which is critical to the levees' ability to survive overtopping.

What is the purpose of opening up levees—for example, using explosives as the Army Corps of Engineers did on May 2 at Missouri's Birds Point levee?
This was a scheme worked out with physical models designed and monitored by First Lt. Herbert D. Vogel at the Corps Waterways Experiment Station at Vicksburg, Miss., in early 1931. The idea is not new—explosives had been used numerous times since the American Civil War to effect cut-offs. Gen. Grant used these to cut a bypass channel during the Siege of Vicksburg in 1863. And, of course, nine different levee demolitions [were] effected below New Orleans in the Great [Mississippi] 1927 Flood. The Birds Point levee scheme originally envisioned the use of explosives-laden barges, but after the 1993 floods the Corps of Engineers installed embedded pipe charges of a much more reliable and sophisticated design, to minimize unintended damages. When we "open up" a levee using explosives, its only intended to be used once, or at most twice, in 100 years. The whole idea is to save money, in comparison with constructing a multimillion-dollar flow-control structure to divert floodwaters, such as those constructed at Bonnet Carre, Morganza and Old River. Those were very expensive structures and can be expected to be used maybe 10 times per century, much more often than a "fuse plug levee," [which is lower than adjacent levees and designed to be torn down by the rushing water if the pressure becomes too great. As designed, no explosives are needed to remove these levees].

What is a spillway and what impact will opening some in New Orleans have on flood waters there? Where are spillways best placed?
A spillway is simply a "flow-control structure," whereby excess floodwaters can be safely diverted out of a river or around a dam. They come in two varieties. The first is a controllable spillway, which employs movable gates to regulate the volume of flow being discharged. All dams are constructed with spillways, and the vast majority of engineered dams are also equipped with outlet works, essentially conduits which allow the lake levels to be maintained at any desired level, and can draw the reservoir level below the spillway sill level, to provide for flood storage during extra-normal runoff years [like 2011].

The second spillway variety is the uncontrolled weir, or crest spillway, like those used for many dams, where the total volume of water passing through the spillways is uncontrolled once the lake level reaches the sill elevation of the spillway. In my opinion, every large reclamation district along the modern flood plains of major rivers that employs earthen levees—like the Mississippi River—should be equipped with spillways that can be opened before their levees are overtopped and destroyed. These same protective districts should also be equipped with outlet works at their distal downstream ends to safely discharge whatever floodwaters enter the protected basin via its spillways. If we added those two things, we'd prevent an enormous amount of unnecessary flood damage because we'd give the excess water somewhere to be "stored."
How does the present situation compare with the storm surge that New Orleans and surrounding areas faced in the summer of 2005 with Hurricane Katrina?
That's like comparing apples to oranges: not much in common, but both are round—in this case, both are wet. Hurricanes present an entirely different set of problems and load protective structures almost instantaneously, whereas duration, flow cycles (up and down) and the total volume of flood water are the three major variables that require weeks or months of careful observation and emergency mitigation. It's much more difficult to manage a major flood. The worst time to let your guard down is after the flood begins to subside, because that's when most of the failures occur, due to rapid drawdown of pore water pressures built up in the levees during sustained high flows.

Why is the Army Corps of Engineers using high-density polyethylene [HDPE] in Vicksburg? What is the significance of using this on the backside of the Yazoo Backwater Levee?
HDPE sheeting is simply a cheap waterproof membrane that can be employed with sandbags and small bulldozers pushing soil to create temporary "push-ups" on top of existing levees, to increase their effective heights by three to six feet [one or two meters]. They have been increasingly deployed since the 1993 floods with considerable success—not 100 percent success, maybe 55 percent. They have to be removed after the flood waters subside by the local districts—without federal assistance. FEMA usually pays for their initial deployment during the flood fight.

How has the construction of levees along the Mississippi River impacted the river's path?
We reclaimed more of the natural flood plains than we probably should have, because seepage is an invisible phenomenon, which takes time. The more volume of water that seeps though the foundation beneath a dike, the more chance there exists of internal erosion, known colloquially as "hydraulic piping".  So, the longer these unseen pipes exist, the more invasive they become, eventually resulting in a levee breach. In many cases those breaches don't occur for five, 10 or even 20 years after the seeps and sand boils are first noted.

Overall, has the system of levees along the Mississippi made communities along the river safer from flooding or have they exacerbated the problem?
This depends on where you look. Certainly, in a grandiose view of the entire MR&T system, it is much safer than it was in 1927 in terms of the flood threat posed to those areas that were then under cultivation, like the cotton fields surrounding Greenville, Mississippi. But, high-value areas like Memphis and places where engineering infrastructure crosses the rivers—bridges for example—we have constrained the channels with considerable severity, for which we later pay an exacting price. One of the biggest challenges in flood routing is shifting flow regimes. The Mississippi River has been narrowed down to just a 3,000-foot [900-meter] width where at passes downtown Saint Louis, when the natural low-flow channel, prior to Corps of Engineers involvement, was about 5,000 feet [1,500 meters] wide. When you squeeze the channels adjacent to densely populated urban centers, the result is usually back-eddy under-scour problems in the flow transitions upstream and downstream of such constructions. And, you heighten the flood depths in the very areas you are seeking most to protect. You have to "store" the excess floodwater somewhere. The big debate is always about where the somewhere else is going to be.

Are earthen levees still being built, or have they largely been replaced by newer technology? What are some of the alternatives—for example soil reinforcement?
Earthen levees are still being built because they are cheap, not because they are good. The average levee in the United States is 10,000 times more likely to fail than the average dam. That's not a matter of expert opinion. It's simply the actuarial figure. We can make levees more survivable by engineering them—hardly any levees are "engineered," per se—which means examining their respective foundation conditions and designing the dikes accordingly, instead of employing a one-size-fits-all mentality, which had been the norm until Katrina. There are all sorts of engineered measures that could be undertaken to make levees more resilient and survivable, if the will exists to do so.

That said, it is very tough to construct robust levees in the Mississippi–Atchafalaya delta because of grievous and pervasive ground settlement and low soil strength. When you stack up three feet [one meter] of earth to begin a levee, it settles about two and a half feet, due to exceedingly low bearing capacity. In those kinds of conditions, we need to develop methods and techniques that are more cost-effective than simply piling up earth like they did during the Civil War. This includes using clay as a binder. Clay wasn't used in New Orleans's levees prior to Katrina, so they didn't have any soil cohesion within the levees [making them more susceptible to erosion once they were overtopped].

What progress has been made in building levees more as energy-dissipation systems as opposed to simply being used as barriers?
I have been one of the advocates of energy dissipation systems because that is the type of system that has been developed by coastal engineers to design resilient breakwater systems. It could easily be applied to protective structures for hurricanes and tsunamis, which are much easier engineering problems than 100-year floods because of their very short duration. You only have to survive six to eight hours of overtopping, even in the largest hurricane.

What are "green" levees?

This is a term I've coined to describe a new type of protective barrier system, which would employ woody vegetation as a critical defensive component, along with a series of low-weight reinforced soil berms, which would be crafted to enhance recreation opportunities—hiking, biking and boating paths, for example—and be capable of surviving overtopping by the largest hurricanes.

View a slide show of the Army Corps of Engineers opening up the Bonnet Carre Spillway and blowing up a portion of the Birds Point Floodway.