Fluvial Processes of Braided Streams: Rakaia River, South Island New Zealand


Chris Below
Geography 401
Professor Harry Jol


Table of Contents

Abstract                                                                     1

Introduction                                                               1

Braided River Systems                                               3

Outside Factors Dealing with Stream Braiding            4

Indications of Braiding                                               6

The Study of Theoretical Stability                              7

Influence on Sediment Supply                                   9

Conclusion                                                               14

Braided river systems are some of geography’s most unique land features.  Studies being conducted on various braided river systems around the world are always coming up with new evidence as to why these unique river systems form and what factors influence their formation.  This paper is written to provide an overview of the mechanisms that drive the formation and causes of braided river patterns, specifically the Rakaia River located on the South Island of New Zealand.   Much is known about these systems with many studies still being conducted to find more information about these amazing river systems.  Most research conducted on braided rivers such as the Rakaia River has been done within the last one hundred years, with most of the advances coming in the last thirty years due to more studies on various different rivers and the beginning process of braided streams along with the increase in the level of technology.  Many of these studies are conducted at the headwaters of the systems, where features such as large glaciers form high in the mountains producing the mass amounts of water and sediment that feed braided streams such as the Rakaia. 

Forming out of the glaciers of the Southern Alps and flowing easterly one hundred and fifty kilometers across the Canterbury Plains on the South Island of New Zealand and entering into the Pacific Ocean approximately fifty kilometers south of Christchurch, New Zealand is the Rakaia River (Figure 1).  The Rakaia River, which has a mean annual flow of two hundred and three cubic meters per second, is mainly a braided river system; it becomes briefly confined in a narrow canyon known as the Rakaia Gorge roughly sixty-one kilometers from the east coast of the South Island.  Through this paper, I am going to go explain the processes that form a braided stream system.  Explanation’s about what a braided river system is and what specific features it has and needs, to be considered a braided river system, reasons such as why most braided river systems are most commonly found wherever drastic reduction in stream gradient causes the rapid deposition of a streams sediment load.  Once we know, what a braided system is we will go over the many external factors involved with braided streams.  Reasoning behind the most commonly accepted fact that what influence’s channel patterns of braided systems are the varying supply of water, sediment supply, and in short term, the factor of the slope of the channel.  My final objective will be to explain braiding indexes, which are the levels of the degree of braiding in a system, and the glacial influence on sediment supply to a braided river system.  Factors to sediment supply such as high sediment change in a natural glacial systems from the erosion of the earth’s surface due to abrasion caused by the advancement and retreat of a glacier. 

Figure 1. Location of the Rakaia River, shown inset of the South Island of New Zealand.

Braided River Systems
There are three general categories as to the theories on the cause of braiding in alluvial systems.  First is a practical explanation relating the braiding to a combination of external forces, environmental factors such as discharge and the supply of sediment.  Second is the formation of braids to theoretical stability analyses of channel bars in two dimensional flow regimes.  Finally is the focus on physical processes such as sedimentary and hydraulic conditions that begin the braiding process (Ashmore, 1991).    A braided river system like the Rakaia is a river channel that consists of a network of small channels separated by sediment deposits known as braid bars.  Braided river channels are unique geographic features.  Compared to other types of river formations, braided rivers are more complex and less easy to understand, unlike a system such as a meandering river system.  Braided streams are most common wherever drastic reduction in stream gradient causes the rapid deposition of a streams sediment load.  Channels and braid bars within a braided river system are highly mobile (Figure 2).  Channel formation changes regularly, most significantly during major floods.  Braided river channels can flow within an entire valley floor or in an area defined by stable banks.  The Rakaia River has eroded a channel up to one hundred meters deep into the Canterbury Plain.  Some conditions, which promote the formation of a braided channel, are erodible banks, an abundant supply of sediment, and the frequent and rapid changes in the discharge of water in the drainage basin.  Braided systems are commonly located in upland and proglacial settings. Braided river systems have a broad size range, varying from ten meters up to twenty kilometers in larger alluvial systems.  These river systems are distinct because of their high stream power and rates of erosion, deposition, and channel change compared to other types of river systems (Bridge, 1993). 

Figure 2. A diagram of the features that make up a braided river system. (Mount, 1995)

When comparing braided rivers to differently classified systems such as meandering rivers there has been little studying of braided river systems.  The reason for lack of study is that braided rivers are hard to study due to the difficulty of measuring their flow, sediment transport, and their morphology in a quickly shifting environment of a braided river. There are a number of issues when it comes to braided rivers.  First off there is the means of starting a braid bar, the flowing together and splitting of the channel, the control of flow stage and the period where the river is going through an increase in the storage of sediment on the depositional pattern (Bristow and Best, 1993). 

Outside Factors Dealing with Stream Braiding

Many of the theories behind explaining how braiding occurs are involved in establishing a relationship between river braiding and the traits of the environment. Factors such as watershed area, discharge, sediment supply and size, channel and valley slope, width to depth, and bank resistance (Ashmore, 1991).  It is commonly accepted that the channel pattern of a braided river is controlled by the amount of water and sediment supplied, and in the short term the factor of the slope of the channel.  The quantity of sediment supply and the slope of the channel can be more easily measured and therefore are usually the basis of the telling the difference between the beginning of a meandering or braided system.  The relationship between slope and discharge usually works well within the regional context, but often fails when trying to categorize the different ranges of braided river systems.   There is an opposite connection between slope and discharge.  A greater than predicted slope puts the river into a braided system, and a less than predicted slope puts the river into a meandering system.  Discharge is the same way; patterns of braided streams are predicted for a discharge greater than a given slope.  When slope or discharge becomes greater, the value of the other factor necessary for braiding lessens (Parker, 1976). 
Channels slopes with high values are commonly referred to as the qualification of a braided stream.  Braiding will occur in an area with low slopes and high discharges.  High discharge and low slope are a prime example of the Rakaia River, which has high water discharge from the Southern Alps and a low slope across the Canterbury Plains.  Distribution of the size of the bed material also plays a major role in the level of braiding.  Smaller grain sizes are less resistance, which allows braiding to happen at a lower discharge and slope.  A difference in high discharge has a tendency for braiding to occur because of rapid fluctuations in the discharge that produces a high sediment supply and width to depth relation.  Discharges that are excessive, lead to erosion of the banks and irregular bedload movement, which is a key factor in the formation of a braided stream (Knighton, 1998).  High discharge systems with inconsistencies are very common in proglacial systems, meaning that not all braided rivers are the same and they have different factors affecting them.  Inconsistent discharge is not necessary for braiding, flume experiments have been conducted have shown that braiding will occur at a discharge that is constant and braided reaches can be in intervals with meandering systems. 
Bank erodibilty is a major component of braided river systems.  Braided systems that have banks of readily erodible material, like those found in the environment of the Rakaia River, are factors in the source of sediment along with the necessary factor to be able to widen the stream channel.  Braided streams need ample power to effectively erode the stream banks and to accomplish a high mobility of their streambed (Knighton, 1998), meaning that without a high stream power braided streams would not be able to change so drastically due to the need of water power required to move the streams bedload.  The vegetation in a braided river system also determines a river’s channel pattern.  A positive reaction of the vegetation is that it stabilizes the cut banks and bar surfaces of the system, it will increase the resistance of the banks and reduce channel widening.  Bank stabilization is not a major factor in the formation and change of the Rakaia River due to the abundant amount of pastureland containing non-native vegetation in the surrounding area.  Pastureland increases the erodibilty because of its shallow root systems, which does not stabilize the soils around the river. Rapidly shifting braided river systems serve as a major disturbance in vegetation establishment once they established themselves because of their nature of constant course shifting. This is very noticeable in the Rakaia system, which has a very wide channel and has cut deeply into the landscape.

Indications of Braiding
Establishing what degree a braided system has developed into and measuring how the pattern changes through time, you first have to gain a gauge of the degree of braiding that has taken place.  A consistent development of repeatable braiding indications is a difficult task because of the rapid change of channels and the difficulty of being able to gain accurate measurements of a large number of braided stream channels.  The hydraulic properties of a braided stream are very dependent on their stage and cannot be determined for different flows in a repeatable method.  The indexes of braiding usually are in two categories.  Ones that count a mean number of braid bars or active channels per channel transect, and those that account for the ratio of the sum of channels lengths in a reach to the total reach length (Bridge 1993).  Bridge (1993) states, “the first category is preferable because it relates braiding to the number of alternate bars or channels, which stability analyses have shown are a fundamental component of braid development.  Total sinuosity can be an indicator of braiding, but is not necessarily a controlling variable in braid establishment.”  Meaning it is easier to count if a river has a high amount of active channels within the channel transect determining whether or not a river is braided.

The Study of Theoretical Stability
To define values logically for the threshold of meandering or braided streams, theoretical stability analyses disturbance techniques for motion equations are used.  Natural instability of flow and sediment transport is what the theoretical stability analysis are based on.  Simply, the main factor in the cause of the instability is the phase difference between shear stress gradient and the bed form gradient.  Necessary conditions for instability are sediment transport and bed friction.  One of the early pioneering stability analyses by Engelund and Skovgaard (1973) found that the primary control of braiding is a width to depth ration greater than fifty shown in Figure 3 (Parker 1976).   Braided rivers cannot be stable and aggradation will occur until a higher stability slope is attained at which point the braiding in the river will end.  Rivers do also have a tendency to form bars and braids even though they can be in a state of stability.  Formations of bars will lead to braiding of the river if the slope and the width to depth ratio are sufficiently high at formative discharges.  Sediment loads that are excessive along with successive aggradation will increase braiding tendencies by increasing slope and forcing a river channel out of its banks.   

Figure 3. Meander and braiding threshold based on Parker’s (1976)stability analysis. The star indicates the results from Engelund and Skovgaard (1973) where primary control was found. (Parker 1976)

Erosional and depositional processes both can be inducive to braiding.  Flume studies conducted by Ashmore (1991) found that two different depositional mechanisms could induce braiding; central bar deposition and transverse bar conversion.  He identifies that there are two erosional mechanisms responsible for braiding: chute cutoff of lone or irregular point bars and dissection of multiple bars shown in Figure 4.   Instability of rapidly changing flow and sediment transport during flow expansion are the primary response mechanisms.  Stream build up is caused by the instability in the flow expansion.  Ability to transport sediment usually declines during the development of the flow expansion, this leads to coarse particle build up at the start of braiding.  Establishing braiding allows all of the processes to occur together which will promote braiding further.  Ashmore found that chute cutoff was the primary mechanism and assumed that conditions of the mobility of sediment and the channel formation will determine which is more common (Ashmore 1991).

Figure 4: Illustration of sedimentary mechanisms for braid bar initiation (Knighton 1998).


Influence on Sediment Supply from the Influence of Glaciers
High sediment change is very natural for a glacial system.  Glaciers advance down slope during accumulation season and retreat during ablation season on an annual basis. Significant erosion caused by abrasion is due to the constant movement of the glacier ice across the landscape. Glaciers can strip away anywhere from 0.01 millimeters per year to 100 millimeters per year beneath the ice sheet.  Additional sediment is received by the glacier from mass wasting of hill slopes, sheet wash which is the erosion of the hill slopes from rain, and from tributary inputs shown in figure 6. Measuring sediment for a glacier is difficult because the majority of the sediment that is being transferred is inaccessible for a detailed measurement.  Depending on the glacier, the location, and amounts of sediment input, output, and storage varies.  Glaciers contribute about twenty three percent of their total sediment loads at the basin outlet and about ninety five percent comes from the erosion of the valley floor (Warburton 1991). 
The production of sediment from a glacier varies widely; the processes that operate beneath them and the material that they flow over all have an impact (Figure 5).  Abrading bedrock compared to an unconsolidated sediment surface will erode much slower than the unconsolidated surface, also warm quickly moving glaciers produce a higher sediment yield than slow moving, cold glaciers (Knight 1999).  Yearly sediment production for large glaciers varies very little, whereas small glaciers have large fluctuations in their sediment production due to intermittent removal of high proportions of stored sediment that leads to the exhaustion of the glacier. 

Figure 5. Bedrock that was once located beneath the Franz Joseph Glacier that has abraded by the Glacier.


Figure 6: Glacial Sediment Budget (Knight 1999).

Hallet (1996) found that “basins with extensive glaciations produced much higher loads of sediment.”  Discharge of sediment that takes place in the area beyond the terminal glacial edge (proglacial) is all dependent on the amount of sediment made available by the glacier and from secondary zones like moraines, and is also dependent on the amount of melt water from the glacier that is made available to transport the sediment.   The discharge from a glacier is not always reflected in downstream proglacial fluvial sediment flows, sediment production and the periodic release and storage of sediment from traps at the margin of the glacier all have a reflection on the discharge.  Local topography and melt water routing are the major factors that control the sediment discharge into water channels past the marginal zone.  When a glacier ends past a topographic obstruction, sediment can be released directly to the proglacial zone from the edge of the ice.  Sediment storage is common when glaciers end behind bordering moraines; they are likely to have their sediment stored in traps such as lakes and other closed basins.  Depending on whether or not the underlying topography of the glacier concentrates the melt water or diffuses the water has a major impact on the transport capacity.


Figure 7. Headwall of the Franz Josef Glacier, notice the large amounts of sediment and debris located within the ice to supply the river system. 

A common glacial mechanism for focusing the melt water discharge and increasing the transport capacity is moraine ridges. Discharge from sediment to a stream is overall complex and is a locally variable balance between the transport capacity of fluvial features, sediment supply, and the topography of the pathways between the glacial and proglacial environment. 
Advancing and retreating of a glacier is an important variable in the controlling of melt water and sediment supply.  Change in a glacier is determined by the net gain or loss in the volume of ice per year.  Sediment discharge varies with the advance and retreat of a glacier due to the interactions of the glacier with sediment sources on the edge and also with the changes in melt water discharge and flow capabilities (figure 7).  Advancing glacier sediment is less likely to run into topographic obstructions and sediment traps, which leads to a higher likelihood that the sediment will reach the channel. Conversely, if the glacier advanced beyond the current stage at another time, moraines and other sediment traps are more than likely in the proglacial zone (Schumm, 1979).   Retreating glacier sediment is going to be subjected by moraines and deposit prior to reaching its channel. Glaciers that are advancing can draw in sediment from a proglacial moraine.  However, the retreating glaciers melt water has access to the recently exposed moraines.
Fluctuation of glaciers on channel aggradation or degradation is somewhat unclear but both may be associated with the advance or retreat of the glacier.  Hallet and Schumm both explain that they believe maximum aggradation of the channels occurs during the advance of the glacier because of enhanced glacial erosion; when advancing ice is abrading the floor and supplying melt water streams with high sediment loads.  Others say that aggradation occurs during the retreat of the glacier when high quantities of debris are available to melt water streams by the exposure of deposits as ice retreats and the increase of melt water due to the mass melting of the ice (Flint 1971).  Figure 8 shows the results from a study conducted by Flint of the Bossons Glacier in France over a six year period as the glacier was in a stage of advance. 

Figure 8. Results from Flint’s study of the Bosson Glacier in France (Flint 1971)


Braided rivers like the Rakaia River are very complex systems. While there has been significant studies conducted on many glacially influenced braided streams around the world, there is still more research that needs to be conducted to fully understand their full processes.  The Rakaia River has received very little in terms of studies being conducted to explain its processes.  Location of the Rakaia River is optimal for a braided river system.  Supply of sediment from glaciers located in the Southern Alps and a course to the Pacific Ocean make the Rakaia River a classic braided system.  Although the reason why a river forms a braided bed is n not fully understood, a high bed load supply, erodible bank material, and high stream power are essential factors for a braided system.



Ashmore, P.E., 1982, How do gravel-bed rivers braid?: Canadian Journal of Earth Sciences,   
v.28, p. 326-341.

Ashmore, P.E., 1991, Channel Morphology and Bed Load Pulses in Braided, Gravel-Bed
Streams:  Geografiska Annaler, Series A, Physical Geography, v.73, No. 1. pg. 37-52.

Bridge, J. S. 1993, The interaction between channel geometry, water flow, sediment
transport and deposition in braided rivers: In: Best, J.L. and Bristow, C.S.  Braided Rivers, Geological Society of London, Special Publication, v.75, p.13-71.

Bristow, C.S. and Best, J. L. 1993, Braided rivers: perspectives and problems: In: Best, J.L.
and Bristow, C.S (eds) Braided Rivers, Geological Society of London, Special
Publication v.75, p.1-12.

Ferguson, R. I. 1993. Understanding braiding processes in gravel-bed rivers: progress and
unsolved problems: In: Best, J.L. and Bristow, C.S (eds) Braided Rivers, Geological
Society of London, Special Publication v.75, p.73-88.

Flint, R. F. 1971, Glacial and Quaternary Geology, New York, John Wiley and Sons, Inc.

Hallet, B., Hunter, C. and Bogen, J. 1996, Rates of erosion and sediment evacuation by
glaciers: A review of field data and their implications, Global and Planetary Change v.12,

Knight, P.G., 1997, Glaciers: Cheltenham, Stanley Thornes Ltd.

Knighton, D. 1998, Fluvial Forms and Processes, New York, John Wiley and Sons, Inc.

Parker, G. 1976, On the cause and characteristic scales of meandering and braiding in  
Rivers: Journal of Fluid Mechanics v.76, p.457-480.

Schumm, S.A. 1979, Geomorphic Thresholds: The Concept and Its Applications,
Transactions of the Institute of British Geographers, New Series, v. 4, No. 4, p. 485-515.

Mount, Jeffrey, California Rivers and Streams: Chapter Four. University of California Press, 5   
May 2008 <http://www.ucpress.edu/books/