Changing Course - Historical Avulsions and the Old River Control Structure

The Mississippi River has been delivering its sedimentary load to the Gulf of Mexico for the past 50 million years. The site of river’s delta during that time span has varied widely across the coastal plain from an area just southeast of the present day town of Denham Springs to a shelf edge delta 200 miles south of Lake Charles.  Since the end of the last ice age that demarks the Holocene Epoch the Mississippi delta system has spread out across present day southeast Louisiana.  The river channel has been meandered back and forth across the floodplain that stretches nearly 50 miles between Baton Rouge and Lafayette.  The colorful maps of the historical channels of the Mississippi River created by Harold Fisk for the Corps of Engineers in the 1940s and 1950s offer a striking visualization of the river channel moving freely across the flood plain over the past several thousand years of the late Holocene.  Fisk also showed that the delta of the Mississippi River moved freely across the coastal plain during this time period with major changes in course or “avulsions” of the river channel.

Historical Avulsions

The construction of the southeast Louisiana delta plain over the past 7,000 years was chronicled in the seminal work of David Frazier in 1967.  Using interpretation from hundreds of cores taken across the coastal plain Frazier reassembled the process by which the land area of coastal Louisiana was built up by historical deltas of the Mississippi River.  Frazier mapped and age-dated sixteen historical deltas that were active during this time period.  Each component delta followed a cyclical pattern from initiation when a change in course in the major channel of the Mississippi system brought its sedimentary load to a new area, thorough development and eventual abandonment when the river changed course to a new location.  Abandoned deltas progressively subsided below the surface creating an open body of water which would provide the site for an eventual reoccupation by the river and the building of a new active delta.  The formation of the deltaic wetlands of southeast Louisiana is therefore best understood as a continual switching back and forth of the river channel across the coastal plain like a garden hose spraying water across a patch of lawn.

Frazier logically grouped the sixteen historical deltas into four primary complexes, which he called the Teche, St. Bernard, Lafourche and Plaquemine Delta Complexes.  Each component delta is numbered in chronological order.  The chart below shows the life span of each delta, and the maps show the location of the delta complexes and each component delta.  The first major avulsion of the Mississippi River during the late Holocene occurred about 4800 years before present when the primary channel of the river at the time, which is represented today by the course of Bayou Teche, was abandoned in favor of a more easterly course that approximated the modern course of the river to an area near what is now New Orleans.  The site of this avulsion was very near to the current location of Old River.  This site appears to have allowed for the river to switch back and forth between the Teche and St. Bernard Complexes over the next few hundred years as Deltas number 4 and 5 developed.

About 3600 years before present a second major avulsion of the river occurred at what is the modern day location of Donaldsonville.  This event initiated the formation of the Lafourche Delta Complex and the development of the primary channels of the river that are represented today by Bayou Lafourche and Bayou Terrebonne.  The third major avulsion location developed at about this same time with formation of Delta 7 in the St. Bernard Complex.  From that point forward until the very recent past the Mississippi River switched back and forth between these two points.  Deltas number 10, 12, 14 and 15 fanned out below the avulsion point at Donaldsonville, and Deltas number 8, 9, 13 and 16 fanned out from the avulsion point just south and east of New Orleans.  Delta number 11, which formed the modern day marshes that surround the city of New Orleans, formed between the two avulsion points about 1900 years before present.  The entire process of building the coastal wetlands with delta deposits of the Mississippi River over the past 7,000 years was controlled by three principal avulsion nodes.  Each of these nodes was formed by a major avulsion of the river into a significantly new hydrologic basin.  These major avulsions occurred only three times during the late Holocene.  Once the node sites had been established they provided the sites of delta lobe switching within each of the major delta complexes.  The chart shows that delta lobe switching occurred about once every 500 years throughout the late Holocene.  Any given vertical line on the chart can be taken to represent a point in time in the past.  If, for example, one were to vertically trace a line equivalent to 2,500 years before present, it would intersect the lifespans of Deltas number 6, 7, and 8.  This means that not only was the river switching back and forth between delta locations at each of the nodes, but that the flow of the river must have also been divided between two or more delta systems at the same time.  In other words while the major avulsions tended to represent a complete shift of the flow of the river from one hydrologic basin to another, the smaller delta lobe switches may have been much more transitional in nature.

Avulsions and delta lobe channel switches occur at the time and place where three necessary initial conditions exist simultaneously:  

1) The change in course of the river must offer a significant gradient advantage, i.e. a route from the current elevation of the channel at the point of avulsion to sea level that is significantly shorter than the current route of the river,

2) the route of the new course must consist primarily of substrate that is easily erodible to allow for the initial scour at the site of avulsion and the development of the new channel by erosion in the direction of the gradient advantage,

3) a trigger event, such as a large flood, or possibly an earthquake, that facilitates the change in course.

The vestiges of the earlier river channels in the flood plain seen in Fisk’s map are the point bar and natural levee deposits that define the shape of the abandoned channels.  Most of the elements of the distributary channel networks of the abandoned deltas on the coastal plain are recognized today as bayous and tidal channels, which are also flanked by natural levee deposits.  The point bar and natural levee deposits of the abandoned elements of the Mississippi River system consist of significant percentages of sand and sandy soils.  This concentration of easily erodible sandy soils in the meander belts and distributary courses of the older river channels offers an explanation as to why the river has tended to continually reoccupy previous pathways by way of a select few avulsion nodes.  Changes in course of the river therefore tend to occur during flood events at the sites of one of the established avulsion nodes when there is a route from that node to sea level that offers a significant gradient advantage over the current course of the river.

The Old River Avulsion

The initial conditions that favored the next major avulsion on the Mississippi River began to develop in the 16th century at the site of the original late Holocene avulsion, which had occurred about 4,800 years before present.  The maps below show a time sequence of active channel configurations taken from Aslan et al. 2005 and overlain on Google Earth imagery.  The latest in the sequence is in the lower right corner and shows a representation of the active channels of the Mississippi River in red, the Red River in green and the Atchafalaya in orange.  The image is dated 1800 AD and shows the point in time at which a westerly-migrating meander loop of the Mississippi had fully intercepted the course of the Red River, and the two river were exchanging flows.  What had been the recently established southern course of the Red River that flowed into Atchafalaya Bay became recognized as the Atchafalaya River, which was by then receiving flow contributions from both the Mississippi and the Red River.  The northern and southern reaches of the meander loop of the Mississippi, which had begun interacting with the Red River 200 years earlier, were named Upper and Lower Old River.

The Lower Old River channel became the primary conduit of flow, and it maintained a bi-directional flow into the early 20th century.  Some of the time water from the Mississippi flowed westward through the Lower Old River channel and exited to the Gulf of Mexico through the Atchafalaya, and some of the time flow from the Red River flowed eastwardly through the channel and combined with the flow of the Mississippi in its course to the Gulf.  The maps above also show in the three other images the sequence that meander belts of these river systems had crisscrossed the point of avulsion several times in the past leaving behind sand-rich and easily erodible deposits that helped to facilitate the most recent change in course.  5,000 years before present the active channel of the Mississippi River shown in blue in the upper left image followed a course to the Gulf that is represented by Bayou Teche today.  The first major avulsion of the river occurred about 200 years later with a shift of the active channel to the east.  The general course of this channel is still shown to be the active channel 2,000 years before present represented as the green meander belt in the upper right image.  This channel was flowing into the St. Bernard Delta complex at this point in time, but the avulsion node at Donaldsonville had been established, and for the next 2,000 years the river switched back and forth between the St. Bernard/Plaquemine Complexes and the Lafourche Complex.  The image in the lower left shows that about 900 years before present the Red River shown as the purple channel crossed over the floodplain and joined the active channel of the Mississippi shown in orange.  At some point in the next few hundred years a crevasse in the Red River initiated a flow into the low-lying Atchafalaya Basin, an area between the elevated meander belts of the historical Bayou Teche and Mississippi channel courses.  This flow of the Red River into the Basin had a significant gradient advantage to the Gulf of Mexico, but it did not encounter the sandy soils of either of the bounding meander belts.  This lower arm of the Red primarily cut through clay-rich organic soils that had been deposited in the backswamps of the older river systems.  This accounts for the remarkably straight course of what is now the Atchafalaya River, and it may ultimately have been a factor in capturing the flow of the Mississippi River because the lower channel offered such a straight course to sea level.

It may also be true that the relative difficulty of eroding the clay-rich soils of the Atchafalaya Basin may have accounted for the extended transitional period over which the progression toward a complete avulsion developed.  The top graph below from Mossa, 2013 shows that the flow of the Atchafalaya measured as a percentage of the combined flows of the Atchafalaya and Mississippi Rivers at Old River progressively increased throughout the first half of the twentieth century.  At the same time the percentage of the total flow of the Mississippi that went through Old River went from oscillating between positive and negative during the period of bi-directional flow to progressively increasing in the direction of flow going down the Atchafalaya, as shown in the middle graph.  The bottom graph shows that the cross sectional area of the Atchafalaya measured at three stages of river elevation also grew steadily during this time period.  The combination of this data illustrates that progression of avulsion was dependent on the interrelationship between the amount of water that moved from the Mississippi River into the Atchafalaya and the size of the channel south of the conjunction at Old River.  A survey of the rivers in 1950 by Latimer and Schweizer for the Mississippi River Commission recognized the fundamentals of these trends, and they predicted that 43% of the flow of the Mississippi would be discharging into the Atchafalaya River by 1965.

Old River Control Structure

As a result of the survey of Latimer and Schweizer and the work of Harold Fisk, the U.S. Congress passed the Flood Control Act of 1954, which authorized the construction of the Old River Control Structure, which was completed in 1963.  The lower arm of the original meander loop of the Mississippi that had intercepted the course of the Red River had become the channel component known as Lower Old River.  The upper arm of this loop, which had been known as Upper Old River, was eventually cut off from the flow of the river primarily because of the dredging of a channel cut in 1831 called “Shreve’s Cut”, which offered a more direct course for flow of the river across the neck of the original meander loop.  The initial construction of the Control Structure included the installation of a lock and dam on Lower Old River that effectively prevented flow in either direction, but allowed for boat and barge traffic moving between the rivers.  It also included the creation of the Outflow Channel located north of the original Upper Old River Channel, and the installation of the Low Sill Structure, which was used to regulate flow from the Mississippi into the Atchafalaya.   It is the stated objective of the Old River Control Structure to maintain the “latitude flow” of both rivers measured at the latitude of Red River Landing (30°56’20.4”) below the Structure at a proportionate ratio of 70% in the Mississippi channel and 30% in the Atchafalaya channel.  The integrity of the Old River Control Structure was tested during the spring of 1973 in a major flood event.

One of the principal initial conditions for avulsion of the Mississippi River – the easy erodibility of the sand soils at the site of the avulsion – is also a principal reason for the near failure of the Old River Control Structure during the 1973 Flood, as well as the high probability that this historical avulsion node will be the site of the ultimate failure of the Structure and the change in course of the river.  The figure below is a profile across the site of the Old River Control Structure compiled using the interpretation from borings taken along the path of the Outflow Channel prior to its construction.  The profile documents the very sandy composition of the soils beneath the Structure down to depths of 100 feet and greater.  The upper layers of sandy soils are the natural levee and point bar deposits of the numerous channels of the river that have migrated back and forth across the flood plain.  The abandoned channels are evidenced by the clay-filled plugs they leave behind that dissect the plain.  The channels cut into a substratum of undifferentiated  sands  and  gravels  that  are the  deposits  of  the  early Holocene  that  filled  in  the Mississippi River valley as sea level rose with the melting of the last ice caps.  These layers of sandy soil are also the surface aquifers of the Mississippi River Valley alluvium.  There is good evidence that water from the Mississippi River flows into these surface aquifers, and that flow is increased during high river stages.  It may be possible that this subterranean flow of water from the river into the sandy soils formed as deposits of the meander belts of older river channels may precede, or even predict, the flow of the current of the river that cuts the path of a new avulsion of the river.

The flow of water from the Mississippi River reached its highest level since the construction of the Old River Control Structure during the flood of 1973.  The estimated rate was 2 million cubic feet per second – about twenty times the flow of Niagara Falls – and it stayed at that level for almost three months.  The flux of water from the river into the alluvial aquifers and the turbulence from the flow through the Low Sill Structure scoured out the foundation, and cut deeply into the sandy substratum beneath it.  The figure below shows that the current of the river completely undermined the Low Sill Structure cutting down to a depth of 130 feet below sea level.  The upper 50 feet of the 90-foot long steel pilings supporting the structure were exposed, and it came close to complete failure.  The U.S. Army Corps of Engineers constructed the Auxiliary Structure in an effort to reduce the flow across the Lower Sill Structure during a flood.  This structure shown in figure 6 was completed in 1986.  The Corps also dumped tons of rip rap into the scoured opening after the flood, but there is nothing in place to insure that the flow of water will not undermine the structure again in a future flood event.  A failure of this type at the Old River Control Structure during a major flood is the most likely occurrence that will precipitate an avulsion of the Mississippi River.

Impacts of avulsion at Old River

The immediate impacts of an avulsion of the Mississippi River caused by a failure of the Old River Control Structure would depend in large part of the nature of the failure and the severity of the triggering event whether it be a flood or an earthquake.  Scenarios for the impacts discussed by Johnson, 1990 include destruction of the Interstate 10 and U.S. Highway 190, as well as railroad bridges and natural gas pipelines crossing the Atchafalaya Basin.  Johnson also noted the potential for severe impacts due to flooding on Morgan City and the possibility for widespread flooding across much of coastal Louisiana south of New Orleans.  The longer term and sustained impacts of an avulsion would be seen along the channel of the Mississippi River between New Orleans and Old River.  The abandonment of the current channel of the Mississippi River would have profound long term impacts on the capacity of the channel as a navigational conduit and as a source of freshwater for industrial and residential use along all reaches of the river south of point of avulsion.

The figure below shows an elevation profile for the Mississippi River channel from the mouth of the river at the Southeast Pass jetty to Lake Providence in northern Louisiana.  The elevation of the base of the channel is below sea level for the full extent of the river from the mouth to a point north of Old River.  This means that without the flow of freshwater moving down the channel saltwater from the Gulf of Mexico would flow into all portions of the channel below sea level creating a “slackwater estuary”.  All points along the river south of Baton Rouge would be completely filled with saltwater, and there would be no short term source of fresh water for industrial and residential use along this stretch.  It is probable that without the flow of river water in the channel the elevation of the river surface would also drop to sea level for all points south of the point of avulsion.  With this drop in the elevation of the river and the loss of the scouring effects of the river’s current the channel north of Baton Rouge may become impassible to boat and barge traffic without program of regular dredging. 

REFERENCES

Aslan, A., Autin, W. J., Blum, M.,  2005, Causes of River Avulstion Insights from the Late Holocene Avulsion History of the Mississippi River, USA, Journal of Sedimentary Research v. 75, p. 650-664

Frazier, D.E., 1967, Recent deltaic deposits of the Mississippi River: their development and chronology, Trans. G.C.A.G.S., v. 17, p. 287‐315

Fisk, H. N., 1944, Geological Investigation of the Alluvial Valley of the Lower Mississippi River, Mississippi River Commission

Fisk, H. N., 1952, Geological investigation of the Atchafalaya basin and the problem of the Mississippi River diversion. Vicksburg, MS: US Army Waterways Experiment Station, Mississippi River Commission

Johnson, D. B., 1980, A Change in the Course of the Lower Mississippi River: Description and Analysis of Some Economic Consequences, LA Water Resources Research Institute, Bulletin 12B

Latimer, R. A., & Schweizer, C. W., 1951, The Atchafalaya River study: A report based upon engineering and geological studies of the enlargement of Old and Atchafalaya rivers. Vicksburg, MS: US Army Corps of Engineers

Mossa, J., 2013, Historical changes of a major juncture: Lower Old River, Louisiana, Physical Geology, v. 34, p. 315-334

Geological underpinnings of the Modern Mississippi Delta

On May 8, 1541 the expedition of Hernando de Soto encountered the Mississippi River near the current site of Memphis.  At the time the mouth of the river was just west of Grand Isle.  The modern “birdfoot delta” had not yet been formed.  The earliest navigational charts of coastal Louisiana postdate the inception of the latest delta lobe, so the timing is not certain.  It is likely that the river avulsed to a new course directed toward what is now lower Plaquemines Parish some time in the 18th century.  For the previous 4,000 years or so that river had changed is course roughly every 500 years.  These changes or the patterns of the delta lobes that they created were not random.  The river sought out accommodation capacity for its sedimentary load with the efficiency of a living organism.

Accommodation capacity in a delta is a direct function of subsidence.  For a delta lobe to have stayed in one place over the 500-year span that it took to develop, the geological structure of the area had to be able to give way to the accumulating sedimentary load.  A recreation of the history of delta lobes in southeastern Louisiana by David Frazier shows that the river methodically sought out the salt and fault basins.  The river tended to settle on sites underlain by geological synclines generally bounded by conjugate pairs of faults that down-dropped to the center of the syncline. 

Since the time of the Pyramids of Giza the river repeatedly returned to the elongate syncline called the Terrebonne Trough that runs beneath the coast of Louisiana.  Over time the accumulated thickness of the Holocene delta sediments revealed the shape of the Terrebonne Trough with its bounding conjugate faults.  At the eastern end of the trough the alignment of the faults and the axis of the sedimentary thick takes a distinct turn to a more north-south orientation.  This is the portion of the syncline sought out and occupied by the Mississippi River some time in the 18th century.

The fault and salt map of the area represents the salt domes with colors indicating the depth to top of salt.  The warmer colors indicate shallower depths.  The large Lake Washington dome in the northwest corner rises up to about 2,000 feet below the surface.  The major faults that reach the surface are represented by fault plane contour maps.  The blue surface fault traces coincide with the 0-depth contour of the fault plane.  The basin occupied by the modern delta is bounded by salt domes on its east and west sides.  The Bastian Bay, Main Pass and Garden Island Bay faults arc around the northern and eastern flanks to form very distinct boundaries.  At depth these faults are matched by conjugate sets of faults that create a graben structure.  These faults can be seen in profile view, but they do not extend to the surface and are not represented on this map.  The central axis of the graben clearly parallels the bounding faults to the east.

The river found maximum accommodation capacity along the central axis of the graben.  The course of the main channel is straight across the center of basin directly overlying the graben axis.  It appears that the tri-furcation of the main channel at Head of Passes occurred where the main channel encountered the down to the north Head of Passes fault.  Faults have been documented to have bathymetric expression at the surface in this area.  It is possible that an escarpment on the bottom of the channel caused by this fault could have provided enough resistance to flow to cause the river to separate its flow.

Two profiles across the basin reveal the geological underpinnings.  Profile 1 crosses the large Main Pass fault.  It can be seen extending to a depth of about 27,000 feet below the surface where it meets a smaller conjugate fault.  Together they form the central graben.  Sediments in the Upper Miocene, Pliocene and Quaternary are clearly thicker on the downthrown side of the Main Pass fault indicating that it has been continually active throughout that time.  The graben probably formed some time in during the Lower Miocene, and offset the underlying Paleogene sediments.  In the early stages of fault movement, the lateral component can be greater than the vertical component.  The lateral component of slip on the faults in this graben caused the Paleogene section to be offset to the extent that the central axis of the graben is defined by a belt in which no Paleogene section is found.  This belt directly coincides with the course of the main channel of the river.

Profile 2 crosses the Garden Island Bay fault.  The similarity in orientation and growth history of this fault with the Main Pass fault indicated that they are genetically related, but they are not the same fault.  As can be seen on the fault plane map, the contours do not connect. 

Three dimensional renderings of the delta show the relationships between the channels, lobes and faults from two different perspectives.   The first one from the northwest shows the main channel running parallel to the bounding faults of the central graben.  It also reveals how the channel spread out when it encountered the Head of Passes fault. 

The perspective from the southwest suggests a relationship between faults and each of the main sub-lobes/channels on the east side of the delta.  There is an undeniable relationship between the architecture of the delta and the subsurface geological structure.

Andrew Tweel and Eugene Turner at LSU’s College of the Coast and Environment studied the relationships between watershed land use, river engineering, wetlands formation and wetland loss in the Mississippi River delta.  In a 2012 study they reported an anthropogenically driven increase in mean suspended sediment concentrations in the Mississippi River below New Orleans between the end of the 18th century and late 19th century, followed by a sharp reduction, and then a period of stabilization after 1962.  The rise of agriculture across the Mississippi River watershed in the 18th and 19th centuries, and the inefficient practices associated with it caused an increase in erosion and a dramatic rise in the suspended sediment load of the river.  By the early part of the 20th century engineering of the river including levees, locks and dams reversed that dramatic rise.

Tweel and Turner noted importantly that “The rapid growth of the Mississippi River birdfoot delta until the 1930s, which has been used as a reference for one of the world’s largest wetland restoration efforts, may not be a suitable archetype for the majority of the coast.”  In other words the patterns and rates of land area growth in the modern delta were substantially impacted by anthropogenic factors, and are not indicative of the natural patterns and rates associated with deltas prior to the European occupation of North America.

Navigational charts from 1838 forward have allowed for a reasonable reconstruction of the modern delta of the Mississippi River in ten-year increments.  This reconstruction suggests the relationships between the patterns of formation of the delta and the underlying geology, as well as the influence of changing sediment supply throughout the time span.

REFERENCES

Frazier, D.E., 1967, Recent deltaic deposits of the Mississippi River: their development and chronology, Trans. G.C.A.G.S., v. 17, p. 287‐315

Tweel, A. W. and Turner, R. E., 2012, Watershed land use and river engineering drive wetland formation and loss in the Mississippi River birdfood delta, Association for the Sciences of Limnology and Oceanography, v. 57, p. 18-28