By June in most years, the water height of the Mississippi River in New Orleans, La., has peaked and is falling. But 2019 was not like most years. Winter and spring 2019 brought many stories of flooding in the Midwest, notably on the Missouri River and its tributaries, which eventually flow into the Mississippi. Even before the 2019 Midwest floods made the news, though, Mississippi River water levels were higher than usual in the New Orleans area.
The U.S. Army Corps of Engineers (the corps) monitors the levees regularly when river stage (water height) hits 4.6 meters (15 feet) in New Orleans, about 0.6 meter below flood stage. Water height on the Mississippi River remained high throughout water year 2019 (which ran from 1 October 2018 through 30 September 2019), exceeding the 4.6-meter level in New Orleans on 164 days (Figure 1). In contrast, there were just 49 such days in 1991, the previous record holder among data going back to 1990.
The vulnerabilities of current water management practices on the Mississippi River were readily apparent in water year 2019, when the unprecedented amount of water had a variety of effects, including stressing ecosystems and contributing to shipping accidents and disruptions. The water level was still elevated in July 2019 when Hurricane Barry moved into the Gulf and threatened to compound the situation with storm surge, which could have been catastrophic. The high water has continued into this year. In fact, as of 28 February, the Mississippi River had already exceeded the critical 4.6-meter monitoring threshold on 21 days in water year 2020, compared with 16 days by the same date in water year 2019. Since 1990, there have been only five water years (1991, 2005, 2016, 2019, and 2020) with more than 1 day above the 4.6-meter stage by 28 February. And with extremely high antecedent soil moisture and abnormally high snowpack throughout the Missouri River Basin this year, along with record precipitation regionally, the expectation is for yet more flooding along the lower Mississippi.
To protect people and industry, the corps has engineered the river to a large extent. The deltaic Mississippi River, or the last 540 kilometers of the Mississippi before it enters the Gulf of Mexico, is lined with more than 483 kilometers of concrete and rock revetments that prevent channel migration and the formation of new distributary channels [Smith and Winkley, 1996]. The Mississippi River and Tributaries Project includes an extensive system of levees that help keep the river and its tributaries in place and prevent flooding. There are levees on the deltaic Mississippi River all the way to Venice, La. (about 16 kilometers from where the river flows into the Gulf of Mexico), and to date, no levee built to the Mississippi River Commission’s current standards, implemented in 1978, has ever failed [U.S. Army Corps of Engineers, 2014].
Despite all the engineering of the river—including control structures, spillways, and a stabilized channel—there are now new challenges for managing the deltaic Mississippi River. Historically, river management practices have been designed assuming stationarity, or the concept that the mean and window of variability in river flow are not changing [Milly et al., 2008]. However, because of the effects of climate change, including increasing annual precipitation and land use changes in the watershed that deliver more water downstream faster, the past is no longer a good indicator of the future when it comes to river flow. In other words, designing flood control practices using historical flow information, such as a static, 100-year flood estimate, is no longer reasonable.
When a river overtops or breaks its levees, water inundates the surrounding floodplain, which decreases the flow, and the flood hazard, downstream. In fact, purposefully breaking levees—long an illicit means of flood control, as in the Great Mississippi Flood of 1927 when people blew up levees around New Orleans to save themselves—can be part of an effective water management plan. But the deltaic Mississippi River needs more than broken levees to make it sustainable.
Upstream of where the Mississippi River enters the Gulf of Mexico, the river becomes part of a distributary delta system. In its natural state, the river split into multiple branches, or distributaries, and the main distributaries periodically switched locations. There have been five main distributary systems of the Mississippi Delta in the past 5,000 years [Coleman, 1988]. Today, the Mississippi River distributary system splits into two main branches. The larger branch is the 540-kilometer stretch that we refer to as the deltaic Mississippi River; the secondary branch is the Atchafalaya River, which flows about 230 kilometers to the Gulf (Figure 2). Although these two branches started naturally, they are now in a highly engineered state.
McPhee  poetically documented the story of how the deltaic Mississippi River became one of the most engineered rivers in the world in his book The Control of Nature. Key to the story—which, like the river, has many twists and turns—is that the corps determined that the distribution of flow between the two distributaries as it was in 1950 should remain permanently: 70% down the Mississippi and 30% down the Atchafalaya [Mossa, 1996].
In 1954, Congress acted on the corps’ recommendation and authorized the Old River Control Project. Today, the corps continues to manage the Old River Control Structure, which controls this flow distribution on a minute to minute basis (Figure 2). The current system of structures, which was upgraded and added to after the catastrophic Mississippi flood of 1973, can handle discharge up to about 20,000 cubic meters (700,000 cubic feet) per second into the Atchafalaya River. For comparison, the average flow in the Niagara River in Buffalo, N.Y., upstream of Niagara Falls, is about 5,700 cubic meters (200,000 cubic feet) per second. Had the Old River Control Project not been in place during the flood of 1973, most of the flow would have likely switched to the Atchafalaya River.
Diverting High Flows on the Mississippi
There are numerous reasons why maintaining flow within a range that sustains beneficial uses while limiting flooding in the deltaic Mississippi River is critical. Both New Orleans and, farther upstream, Baton Rouge, Louisiana’s capital, have ports with extensive shipping and train yards and are surrounded by billions of dollars of industry, including oil refineries and chemical plants that depend on the deltaic Mississippi River.
To protect these areas, the corps also controls two major spillways to release water away from the deltaic Mississippi River during floods (Figure 2). Upstream of New Orleans but downstream of Baton Rouge is the Bonnet Carré Spillway, which was built in response to the Great Mississippi Flood of 1927 and became operational in 1931. This spillway is designed to release up to about 7,000 cubic meters (250,000 cubic feet) per second of water into Lake Pontchartrain, a brackish lake connected to the Gulf of Mexico, offering relief primarily for the Greater New Orleans area and downstream.
Between 1931 and 2008, the Bonnet Carré was opened eight times. Between 2008 and 2018, it was opened four times. Last year became the first year in which the Bonnet Carré was opened twice. Even with the Bonnet Carré open, flow in the deltaic Mississippi River remained high in 2019 (Figure 1). Opening the spillway comes with costs. For example, the oyster season has historically been less productive when the Bonnet Carré Spillway is opened because the brackish waters of Lake Pontchartrain may freshen significantly. Further, the river water entering Lake Pontchartrain often has very high nutrient loads—enriched by farm runoff from 32 states—that promote toxic algae blooms, which are especially deadly to benthic marine animals like oysters.
The bigger of the two spillways is the Morganza, situated downstream of the Old River Control Structure and upstream of Baton Rouge. Morganza is designed to take up to 17,000 cubic meters (600,000 cubic feet) per second of flow from the deltaic Mississippi River and drain it into the Atchafalaya Basin. Because there are homes and farms in the Atchafalaya floodplain that can be affected by such drainage, opening the Morganza is more controversial than opening the Bonnet Carré, and it has been opened only twice in its history: during the 1973 and 2011 floods. There was talk of opening it again in June 2019, but that did not happen.
Conservation of mass means we cannot make water disappear. So sustained high water means greater potential for flooding—whether it is purposely induced by humans via spillways or not—and more water on the floodplains of both the deltaic Mississippi River and the Atchafalaya River, which endangers people, crops, and wildlife like the indigenous Louisiana black bear.
High Stage Raises Concerns
High stage on the deltaic Mississippi River throughout the summer has implications for water management. Hurricane season officially begins on 1 June, when the average river stage in New Orleans is about 3.4 meters (11 feet) and dropping on the basis of data from 1990 to 2018 (Figure 1). On 1 June 2019, river stage was at 5 meters (16.5 feet). Hurricane-induced storm surge from the Gulf of Mexico can increase the stage in the deltaic Mississippi River in New Orleans and even Baton Rouge. For example, Hurricane Katrina drove the river stage up by about 3.4 meters in just 1 day (28–29 August 2005; Figure 1).
The levees all along the deltaic Mississippi have been improved since Hurricane Katrina. Because peak flood season in the deltaic Mississippi River and peak hurricane season do not generally coincide, however, the river levees are not designed to accommodate storm surge on a flooded river. When Hurricane Barry threatened to hit New Orleans in mid-July 2019, the river stage was about 4.9 meters (Figure 1). Levee heights vary, but the lowest levees are at about 6.1 meters. Luckily, the storm impacts were not as great as initially forecasted, levees on the deltaic Mississippi River did not overtop, and a Katrina-scale disaster was averted.
High stage also means that the levees are at greater risk of failing. As river water piles up on levees, the added weight increases the susceptibility of the underlying soil to seepage and formation of sand boils that may undermine the integrity of levees over time.
Even when levees do not overtop or fail, prolonged flooding has impacts on industry. Navigation on the winding deltaic Mississippi River is always complicated and requires different ship captains with localized knowledge in different stretches of the river even in the best circumstances. However, higher stage means swifter flow, making the river much harder for ships to navigate. There are more shipping accidents, sometimes resulting in deaths, when the river is high.
Stronger currents also mean that ships at anchorage can be unmoored and drag anchors. More space is required between anchored vessels, resulting in fewer spots for anchorage, and some ports become entirely unusable in high water. Tugboats also push smaller loads at higher discharge, further resulting in reduced shipping. In 2019, the total load of commodities shipped on the Mississippi River was 25% lower than the 10-year average. Downstream trips are faster, but upstream trips are slower and require more fuel. And dredging efforts cannot keep pace with siltation at the mouth of the river, limiting the size of ship that can come up the river.
A Silver Lining
Amid the challenges of prolonged flooding is the silver lining that higher flows on the deltaic Mississippi River could contribute to land building. Coastal wetlands protect New Orleans and all of southern Louisiana from the impacts of hurricanes. An often-cited statistic is that Louisiana loses the equivalent of a football field of wetland every hour, or 42.9 square kilometers per year, because of natural delta deterioration processes exacerbated by subsurface fluid withdrawal and construction of canals [Couvillion et al., 2011]. The $50 billion Louisiana Coastal Master Plan aims to reduce this land loss, and one of its proposed strategies involves harnessing natural delta-building processes to create land through sediment diversions from the deltaic Mississippi River into selected areas of shallow deteriorating marshes.
Optimized sediment diversions would occur only during high-stage, sediment-laden discharges to maximize the amount of river sediment entering receiving basins in the coastal marshes while minimizing the introduction of fresh water into saltwater ecosystems. The state of Louisiana is currently designing two large sediment diversions, each with a flow conveyance capacity of about 2,100 cubic meters (74,000 cubic feet) per second, to be located roughly 50 kilometers downstream of New Orleans (Figure 2). Ultimately, the land-building potential of these projects will depend on sea level rise; however, numerical modeling suggests they could produce 20–60 square kilometers of new marshland, or the equivalent area of the modern Wax Lake Delta, over 5 decades of operation.
There are caveats to this approach, however: As we have learned from spillway operations, altering salinity in the estuaries surrounding the delta may negatively affect habitat for species such as oysters, brown shrimp, blue crab, and bottlenose dolphins. In addition, a wide range of stakeholders managing or making a living from coastal resources would likely be affected by sediment diversions, creating many legal and economic constraints that may require selective use of diversions and limit potential land building.
An Uncertain Future
The future of the deltaic Mississippi River remains uncertain, in part because it is surrounded by many changing systems. Over the past century, the upstream network that delivers water and sediment to the deltaic Mississippi River has been highly engineered. Dams control flooding and reduce sediment supply, whereas increased channelization of the network changes flow patterns and travel times. The watershed has also seen extensive urban and agricultural development, which lead to faster and greater-volume deliveries of water to the Mississippi network.
Climate change will also continue to have multiple impacts. Comparing the period from 1986 to 2015 with that from 1901 to 1960, much of the upper Mississippi River watershed has experienced increases in precipitation ranging from 5% to more than 15% [Easterling et al., 2017]. On the downstream end, the number of hurricanes overall and the number of very intense hurricanes are predicted to increase [Kossin et al., 2017]. Further, relative sea level rise will drive the terminus of the river upstream.
The design of flood management infrastructure has typically relied on historical precedent, such as flood frequency records. Going forward, these designs must account for the potential of future conditions looking significantly different from those of the past. Efforts to modernize planning are helped by the fact that researchers have made great strides in improving abilities to predict future conditions through the development of high-quality community hydrologic and climate numerical models [e.g., Kauffeldt et al., 2016]. Ambitious physical laboratory models are also providing opportunities to test the effects of proposed experimental river management projects such as those related to sediment diversions.
Ultimately, the only real certainty is that past recipes for managing the river through hard-structure engineering will not be adequate given all the stressors on the system. After large floods, there are often discussions about relocating people away from floodplains and about changing zoning laws, but as the amount of time since a disaster increases, the sense of urgency for such changes dwindles. The urgency is now here to stay.
Creative, although possibly unpopular, solutions beyond infrastructure are also required to manage the deltaic Mississippi River. The most effective options for the long term are nature-based solutions that leverage ecosystem functions, such as fostering vegetation growth to dissipate storm surge [Barbier et al., 2013], and that are more adaptable to changing environmental conditions than concrete and steel structures. Tough decisions, including abandoning some areas where people live, may be part of the answer as well, and these discussions are already occurring. Solutions must not put the entire burden of change on marginalized socioeconomic communities, however, as has occurred with other development projects in the region, such as the siting of new industrial plants. An equitable solution that relies on sound science should be the priority.
Many groups have roles to play in addressing the future of the deltaic Mississippi River, from scientists, engineers, and river managers to stakeholders, politicians, and the public. The question now is whether these groups will be brave enough to embrace innovative and perhaps yet to be designed solutions.
We thank R. Waid, A. Ferchmin, and R. Perez for information on shipping conditions and boating accidents during Mississippi River floods and J. Corbino for information on dredging practices.