Click here to login if you're an NAE Member
Recover Your Account Information
Author: Robert G. Dean
The greatest uncertainties for wetland restoration will be political will and stakeholder response.
The New Orleans-Mississippi River and associated wetland system are central to the well-being of the nation. The Mississippi River drains 41 percent of the water in the continental United States, and every year the system provides some one billion pounds of seafood and 25 percent of the nation’s oil and gas. In addition, $15 billion worth of cargo passes through the Port of New Orleans. Barrier islands and wetland systems2 provide habitat for much of the finfish and shellfish consumed in the United States as well as a degree of protection for New Orleans against the effects of hurricanes.
Estimates of the size of the wetlands range from 6,000 to 12,000 square miles (in this paper, I use 9,000 square miles). This article is about (1) the degradation of these valuable wetlands and the diminishment of their protection against hurricanes and (2) the prognosis for restoring them and rebuilding New Orleans. To understand and, ideally, reverse wetland loss, three salient features of the area must be kept in mind: (1) the immensity of the area of concern; (2) the pervasiveness of the natural forces that tend to degrade the wetlands (particularly the rate of subsidence and the rise in sea level); and (3) the interests of entrenched stakeholders.
The wetlands of southern Louisiana, the largest in the United States, which provide a buffer against storm surges and waves, are being lost at an unprecedented rate—approximately 1,900 square miles since 1930. Figure 1 (see PDF version) shows the documented historic and projected future losses of wetlands. Reduction in the loss rate from 1990 to 2000, an average of 25 square miles per year in 1995, is probably the result of several factors: (1) the more readily eroded wetlands had already been lost, and the remaining wetlands are more resistant; (2) there were fewer major storms in recent years; and (3) limited remedial measures have had some beneficial results. A preliminary assessment by the U.S. Geological Survey established that 100 square miles of wetlands were lost due to Hurricanes Katrina and Rita (NWRC, 2005), and it is not known if some of these wetlands will recover, as they have after other major storms.
The Natural System
In its natural condition, the Mississippi River can be considered a vast, “leaky” plumbing system; the leaks increase during higher flows (Figure 2 shows the associated river distribution systems in 1880 and 1990). Under the influence of natural forces (Figure 2a), Mississippi River waters spilled over its banks during periods of high flow, depositing coarser sediments close to the river and finer sediments farther inland. The sediments deposited near the river formed natural (elevated) levees. The finer sediments provided a rich matrix for agriculture over broad flooded areas, as well as sediment that enabled the land to maintain its elevation in the presence of rising relative sea level.3
In some areas, the overflow was more or less uniform; in other areas, the overflow was concentrated into channels with deposits forming “splays” on the lee sides of the natural levees. After the river returned to its normal flow, the incipient channels formed during high flows would usually heal, leaving the sediments in place.
During high flows, the river also deposited sediments in the vicinity of its terminus. If this terminus were in relatively shallow water, the river mouth would advance toward the Gulf through deposition and the formation of natural levees. Figure 3 shows the advancement of Southwest Pass (historically, the main shipping channel) by 15 kilometers in 215 years, an average rate of about 70 meters per year! Today, the terminus of Southwest Pass is at the edge of the continental shelf where the sediment load is discharged into deep water, considered a favorable process by those responsible for maintaining navigation depths.
As the length of the channel increases, it becomes less efficient hydraulically and during high flows more susceptible to channel “switching,” when a breach can become the predominant channel that delivers water and sediments a shorter distance to the Gulf. The creation of a new, hydraulically preferred channel and the abandonment of a previous channel has been documented to occur, on average, approximately once every thousand years. When channel switching occurs, the sediments and fresh waters that nourished the previous channel levees and wetlands are “switched” to the new area where wetlands formation, natural levee building, and river channel extension are active processes. Thus, the area of previous wetland growth and maintenance is subject to rapid erosion and wetland drowning as a result of relative sea level rise, and the area of new sediment delivery undergoes vigorous wetland formation. It is possible to observe this natural process of wetland growth in the modern delta system (albeit in a controlled setting) where sediments delivered by the Atchafalaya River result in the annual net production of 180 acres of wetlands and would produce more if it weren’t for the dredging and disposal in deep water of 8 million cubic meters per year of sediment to maintain the Atchafalaya River Navigational Channel.4
In summary, under natural conditions, the uncontrolled Mississippi River delivered its waters and sediments toward the Gulf of Mexico through channels that were subject to overflow during floods, thereby providing the nutrients and sediments necessary to maintaining wetland elevations in the presence of relative sea level rise. Channel switching caused the sediments to be routed to shallow waters, resulting in rapid wetland growth in the areas of new sediment supply and rapid wetland degradation in areas of channel abandonment. Thus, under natural conditions, the wetland systems were dynamic. However, to the best of our ability to determine, the net wetland area was stable over the long term.
The Altered System
A comparison of Figures 2a and 2b shows how the “leakiness” of the Mississippi River has changed under modern systems. The flows through Bayou La Fourche have been essentially eliminated, and practically all of the flows over and through the natural levees have ceased as a result of levee construction for flood protection. Thus, the supply of sediments that formerly spilled over the banks along the length of the river has been all but eliminated. Similarly, the river now discharges most of its sediment load into deep water rather than depositing it where it can offset the effects of relative sea level rise. This has had a significant negative impact on the barrier islands and wetlands.
The extraction of large quantities of hydrocarbons has also had a significant effect. The first successful oil well in southern Louisiana was drilled in 1901. The earliest oil fields were developed on land, but as those fields became depleted and larger and more productive offshore fields were discovered, and as new technologies became available, the offshore resources were developed.
Unless water is injected to replace the oil, gas, and water extracted, the pressure in oil fields decreases substantially, which can contribute to subsidence in the area. Morton and colleagues (2002, 2003) have conducted careful studies correlating areas of high subsidence with decreases in bottom-hole pressures in oil fields. Their studies strongly support a link between hydrocarbon extraction and local subsidence. Not everyone agrees, however. Gagliano et al. (2003), for example, have argued that local subsidence effects are due to natural thrust faults; even without hydrocarbon extraction, they say, it is well known from geological evidence that faulting (called “thrust faults”) in the region has been caused by sediment loading of the delta region.
Other anthropogenic causes of wetland loss include channels cut for hydrocarbon exploration and production, the consumption of wetland vegetation by nutria (fur-bearing animals introduced into the region), the alteration of wetland areas for agriculture, and so on.
In my view, and the view of many, but not all, other investigators, the dominant causes of wetland loss are the channelization of the Mississippi River to ensure that sediments are discharged offshore into deep water and the restraint of the river to eliminate natural over-bank flooding and channel switching. Thus, substantial progress in large-scale restoration of the wetlands will require that large-scale sediment processes be reinstated to the degree possible.
Subsidence, Sediment, and Relative Sea Level Rise
Two unique physical features of wetland loss in southern Louisiana are the pervasiveness of natural “forces” and the vastness of the affected area. The dominant natural force is subsidence, which ranges from approximately 0.25 centimeters per year near the southern limits of New Orleans to nearly 2 centimeters per year near the terminus of the river. This is 2 to 16 times the worldwide average rate of sea level rise (based on a rate of 1.2 millimeters per year). The wetlands must accrete vertically at the rate of local relative sea level rise to maintain their viability, which requires substantial quantities of sediment. Keep in mind that the wetlands cover approximately 9,000 square miles, with an east-west dimension of 300 miles from the Mississippi state boundary on the east to the Sabine River at the Texas border on the west. The north-south dimension averages 30 miles.
Because essentially all of this area is experiencing some degree of subsidence, to maintain the present wetland area, mineral (i.e., inorganic) and organic sediment volumes must be sufficient to maintain the elevation against relative sea level rise. However, the sediment delivery of the Mississippi River has been reduced in historic times by the upstream construction of impoundments and by agricultural practices to control erosion. In fact, these reductions raise the question of whether the river now delivers enough sediment to stabilize or increase the present wetland area.
To address this question, we consider an annual mineral sediment delivery of approximately 100 million cubic meters and that the proportions of mineral to organic sediment volumes necessary to form wetlands vary from 1:10 to 1:4. A simple calculation suggests that to maintain elevation of this 9,000-square-mile area in the presence of an (assumed) average 0.25 centimeter per year relative sea level rise, the present rate of mineral sediment delivery is approximately 8 to 20 times the amount required—but only if the sediment can be delivered to the areas in need.
Unfortunately, targeted delivery is not economically possible for several reasons, such as the energy costs and long distances of these areas from the Mississippi River, which is the primary source of sediment. In addition, many communities have been built on the natural levees of the river, making it almost impossible to raise the general elevations of these areas; the building of man-made levees remains the only means of protecting them against flooding.
A great many stakeholders will be involved in a wetland restoration plan that incorporates the optimum characteristics of a natural system. The stakeholder most likely to be resistant to large-scale sediment delivery to the areas in need and to channel switching is the navigation industry. At present, the shunting of sediments offshore through Southwest Pass to deep water provides substantial benefits for the industry. First, it reduces the amount of dredging necessary for channel maintenance. Second, vessels now enter the protective shelter of the levees of Southwest Pass immediately after leaving the deeper waters of the Gulf of Mexico. Thus, the present system provides both economic and safety benefits that the navigation industry will certainly not want to compromise. Third, the long distance from New Orleans to the Gulf of Mexico effectively limits the encroachment of saltwater from the Gulf up the river to the municipal water intakes of New Orleans.
In addition to the navigation industry and, perhaps, the city of New Orleans, the many communities located on natural levees, the elevations of which are decreasing, are facing difficult choices. They can (1) do nothing to change the situation, thus awaiting another disaster, (2) eventually abandon their communities, or (3) build levees to protect themselves. Agricultural interests will also require substantial compensation, if they agree at all, to participate in a plan that floods their lands with the sediments necessary to maintain elevations equal to rising relative sea level.
We can now identify the conceptual elements of a viable wetland restoration plan, as well as some of the uncertainties. Capturing and delivering large quantities of sediment to areas conducive to the development of barrier islands and wetlands must be the centerpiece of the plan. If the sediment is made available, the opportunistic nature of the wetlands will ensure their formation. To accomplish this objective, the plan must mimic the natural system of sediment delivery as closely as possible. The greatest uncertainties will be the willingness of the nation to provide adequate financial resources and the acceptance or resistance of stakeholders, such as communities built on natural levees, to meaningful action.
A wetlands restoration plan
must mimic the natural
system of sediment delivery
as closely as possible.
Previous efforts to reduce the rate of wetland loss have been partly successful. In 1990, Congress passed the Coastal Wetlands Protection and Preservation Act or CWPPRA (also known as the Breaux Act) to provide federal funding (approximately $50 million per year) for coastal restoration. This program has consisted primarily of the diversion of fresh water and sediments to allow for wetland building. Partly because of limited resources, however, none of the CWPPRA projects has attempted to capture a major portion of the river sediments. The Louisiana Coastal Area (LCA) plan (developed by the U.S. Army Corps of Engineers [USACE] and the Louisiana Department of Natural Resources and reviewed by the National Research Council [NRC]) is a $2 billion proposal that includes five major projects to be constructed within a 10-year period—the initiation of a science and technology program, several smaller scale demonstration projects, and preliminary evaluations of a few large-scale long-term projects, one of which is discussed below. The purpose of the science and technology program is to provide an organizational framework for the monitoring and evaluation of constructed projects and to assist in the selection and design of future projects. Table 1 shows estimates of the reduction in wetland-loss rates as a result of CWPPRA and other projects and the predicted effects of implementation of the proposed LCA plan.
A major component of the overall LCA plan that would be evaluated, but not constructed during the 10-year project, is the so-called “third delta,” originally proposed by Gagliano and van Beek (1999). A channel would be constructed from near Donaldsville, Louisiana, some 55 miles from the Gulf of Mexico. The channel would be aligned approximately parallel to Bayou La Fourche, the natural conveyance that, under natural conditions, carried approximately 15 percent of the Mississippi River discharge (see Figure 2a). This project appears to be feasible from an engineering perspective and would mimic to some degree the natural sediment and water delivery processes of the river. Stakeholders may be the major limitation to this plan, although a substantial educational program might persuade them to consider the idea.
In the NRC report, Drawing Louisiana’s New Map: Addressing Land Loss in Coastal Louisiana, a committee of experts recommends some modifications to the LCA plan (NRC, 2005). Instead of locating the point of diversion near Donaldsville, the committee suggests locating it farther south (north of Head of Passes),5 thus diverting water and sediments to the west. In this way, almost all of the discharge could be diverted without affecting the navigational function of Southwest Pass. The smaller sediment fractions would flow predominantly to the west under natural forces; coarser sediments (called “bed load”) would require substantial dredging from the river bed with predominant placement to the west.
The NRC committee also suggests that the LCA plan proceed with the aid of an evolving map of future landscapes for various time horizons under alternative options for remediation. Predictions of future landscapes, although somewhat limited, can be made with a reasonable degree of confidence without additional restoration efforts. Capabilities for predicting the effectiveness of various restoration options are already sufficient to initiate a program, and these capabilities will improve with future monitoring of the wetlands and their response to remedial measures.
With the modifications recommended in the report, the LCA plan would focus first on large-scale projects that could provide economies of scale in delivering sediments to the areas in need. If stakeholder concerns and political consequences render the first choices infeasible, expectations could be adjusted accordingly through revisions to the map.
To evaluate this approach and demonstrate its benefits (and perhaps gain broad approval), a scaled-down version could be put in place. The diversion could be limited in magnitude with dredging still capturing and delivering coarser sediments to the west. Ideally, these sediments could be discharged a sufficient distance to the west that the waves, through natural processes, would continue to carry them in that direction, thus nourishing the present barrier islands.
Rationale for an Integrated Plan
Whatever protective system is selected, two essential components must be levee construction and wetland and barrier island restoration. Without the latter component, the protective capability of levees will continue to decrease with time. Wetlands reduce hurricane storm surge and wave impact partly by bottom friction, which dampens waves and reduces storm surge from rapidly moving hurricanes. Robust wetlands preserve their buffering capability and contribute to the effectiveness of the levees protecting New Orleans and other areas by maintaining vertical elevation in the presence of relative sea level rise. Although various “rules of thumb” have been developed for gauging the effectiveness of wetlands in reducing storm surges and waves, this is a complicated process that depends on the translational speed of the hurricane, among other factors.
However, deferring wetland restoration because of a quantitative uncertainty of their effectiveness would be ill advised. With further wetland degradation and continued relative sea level rise, future hurricane-generated waves and surges in the New Orleans area are certain to be larger and more destructive. The degree of protection will certainly diminish with time as the wetlands degrade, and restoration from an even more degraded condition will be more difficult and more costly than maintaining present conditions.
Hurricanes Katrina and Rita and the associated tragedies have raised several questions for the engineering, scientific, and political communities, especially the best way to protect New Orleans from a recurrence of flooding. The design of a new levee system could benefit substantially from lessons learned from recent events. First, partitioning the areas at risk with additional levees would make it less likely that a single failure would cause widespread flooding. Second, for areas where ground elevations are reasonably high, low surrounding levees should be considered to isolate these areas from flooding. Existing roads and other infrastructure could provide rights of way for this purpose.
A general question for the engineering and political communities is how to identify areas of extreme vulnerability and address these vulnerabilities before disasters occur. It was well known that New Orleans was vulnerable to hurricanes of magnitudes that had occurred in the past and would recur in the future. In fact, if Hurricane Ivan in 2004 had maintained its course, it is likely that severe flooding of New Orleans would have occurred then. Thus, the scientific/engineering question was not whether, but when, such a disaster would occur.
If the warnings had been heeded, the estimated cost of prevention would have been less than 1 percent of the cost of reconstruction, and many lives would have been saved. The cost of debris removal alone as of November 2005 was more than the preventive cost of upgrading the levees and restoring the wetlands.
In the wake of Hurricanes Katrina and Rita, many questions remain to be answered. Might the tragedy have been prevented if the technical community had conveyed realistic, convincing scenarios of the impending disaster? Or, was the political community unwilling to take proactive action to avoid certain disaster because of the uncertainty of its timing? More important, will appropriate steps be taken to avoid preventable disasters in the future?
I believe a program to identify vulnerable areas throughout the United States is warranted. Once these areas have been identified, the associated risks and failure consequences can be ranked objectively and the costs of disaster prevention evaluated. Finally, the agencies and organizations responsible for the safety and financial future of our nation should establish schedules for upgrading infrastructure and risk management plans to avoid future disasters.
Coleman, J.M. 1988. Dynamic changes and processes in the Mississippi River Delta. Geological Society of America Bulletin 100(7): 999–1015.
Gagliano, S.M., and J.L. van Beek. 1999. The Third Delta Conveyance Channel Project: Proposed Mississippi River Diversion Channel and Subdelta Building in the Barataria-Terrebonne Area of Coastal Louisiana. Section 6 of Appendix B of Coast 2050: Toward a Sustainable Coastal Louisiana. Baton Rouge, La.: Louisiana Department of Natural Resources.
Gagliano, S.M., E.B. Kemp III, K.M. Wicker, and K.S. Wiltenmuth. 2003. Active Geological Faults and Land Change in Southeastern Louisiana. Prepared for the U.S. Army Corps of Engineers. Baton Rouge, La.: Coastal Environments Inc.
Gould, H.R. 1970. The Mississippi Delta Complex. Pp. 3–30 in Deltaic Sedimentation: Modern and Ancient, J.P.
Morgan, ed. SEPM Special Publication 15. Tulsa, Okla.: Society of Economic Paleontologists and Mineralogists.
Kesel, R.H. 2003. Human modifications to the sediment regime of the lower Mississippi River flood plain. Geomorphology 56: 325–334.
Louisiana Coastal Wetlands Conservation and Restoration Task Force and the Wetlands Conservation and Restoration Authority. 1998. Coast 2050: Toward a Sustainable Coastal Louisiana. Baton Rouge, La.: Louisiana Department of Natural Resources.
Morton, R.A., N.A. Buster, and M.D. Krohn. 2002. Subsurface controls on historical subsidence rates and associated wetland loss in south central Louisiana. Gulf Coast Association of Geological Societies Transactions 52: 767–778.
Morton, R.A., G. Tiling, and N.F. Ferina. 2003. Causes of hot-spot wetland loss in the Mississippi Delta Plain. Environmental Geosciences 10(2): 71–80 (chapters 2 and 7).
NRC (National Research Council). 2005. Drawing Louisiana’s New Map: Addressing Land Loss in Coastal Louisiana. Washington, D.C.: National Academies Press.
NWRC (National Wetlands Research Center). 2005. USGS Reports Preliminary Wetland Loss Estimates for Southeastern Louisiana from Hurricanes Katrina and Rita. Available online at: http://www.nwrc.usgs.gov/releases/pr05_007.htm.
USACE (U.S. Army Corps of Engineers). 2004. Scoping Report: Louisiana Coastal Area (LCA) Ecosystem Restoration Study. New Orleans: USACE, New Orleans District.
1The author recently chaired the National Research Council (NRC) Committee on Restoration and Protection of Coastal Louisiana, which evaluated the so-called “LCA plan” a near-term, $2-billion program formulated by the U.S. Army Corps of Engineers and the state of Louisiana Department of Natural Resources. The plan was a scaled-down version of two much more comprehensive and ambitious earlier plans. Unless otherwise indicated, the views expressed in this article are those of the author and do not necessarily reflect those of the committee or the NRC.
2Barrier islands in this area, elongate features generally formed by sediments from the Mississippi River, are coarser than similar features in the landward wetlands. In this article, the term wetland systems includes the barrier islands.
3Relative sea level rise is the sum of the eustatic (worldwide average sea level rise) component and local effects, which include subsidence. Land subsidence in the delta region, both natural and anthropogenically induced, can be as much as 2 centimeters per year, approximately 16 times the eustatic rate.
4As seen in Figure 2, the Atchafalaya River, a distributary of the Mississippi River, discharges into Atchafalaya Bay. The flow volumes in the Atchafalaya River are controlled at 30 percent of the total flow in the Mississippi River. Without control, the Atchafalaya River would by now have captured the main flow of the Mississippi River.
5Head of Passes is the location shown in Figure 2b where the main
river divides into three main distributaries.
TABLE 1 Land-Change Projections for the Next 50 Years with the CWPPRA and Other Projects and the LCA Plan