192
w,d,l
Q oc --w
s
w,l,s
Qs rx d,P
where Qw=discharge, O s =sediment load, w=width,
d=depth, !=meander wavelength, s=slope, and P=
sinuosity. For a few sets of rivers in specifi c climatic
ranges, some of these relationships have been quan�
tified, as summarized by Ethridge and Schumm
(1978). However, the data are inadequate to permit
generalizations that can be applied to all geological
conditions, a fact that considerably reduces the usefulness of quantitative geomorphology for geological reconstructions. Additional details on the
relationship between discharge and slope and the
braided-meandering threshold are given by )lridge
(1993b).
One of the primary prerequisites for braiding is a
high coarse-grained sediment load (bed load).
Leopold and Wolman (1957, p. 50) stated:
"Braiding is developed by sorting as the stream leaves
behind those sizes of the load which it is incompetent to
ha11dle ... if the stream is competent to move all sizes
comprising the load but is unable to move the total quantity provided to it, then aggradation may take place without braiding."
In addition to the relationship between discharge
and channel style shown in Fig. 8.1, Miall (1977) and
Rust (1978a) noted that braided rivers are commonly characterized by high discharge variability;
for example, rivers in alpfue and arctic areas, with
highly seasonal discharge variations, tend to be
braided, as are ephemeral rivers in arid regions.
Miall (1977, p. 7) argued from these and other observations that discharge variability is a primary control on. the. braiding pattern:
"In rivers of highly variable discharge competency will be
similarly variable, and there will be long periods of time
throughout which the river will be unable to move at least
the coarsest part of its bed-load. The incidence of bar
initiation, flow diversion and the creation of new channels
(braiding) will thus be high."
On the scale of the alluvial basin the nature of the
vegetation cover has an important effect on discharge characteristics, sediment load, and fluvial
style. In vegetated areas presently, runoff following
major rainfalls is rarely catastrophic because the
precipitation is absorbed by soil and plants and released slowly. Similarly, sediment is stabilized by
roots, and sediment yields are relatively low. VegFluvial Styles and Facies Models
eta ted banks are highly resistant to erosion, as demonstrated in an experiment by D.G. Smith (1976).
This limits the supply of sediment to the river, and
also reduces the bank widening that commonly accompanies braiding. In vegetated areas, therefore,
because of the effects on di,scharge variability, bank
stability and sediment supply, braiding is inhibited.
Jlaker (1978a) quoted examples in tropical Africa
and South America where dense bank vegetation
inhibits braiding, even where abundant coarse sediment is present in the banks. Stanistreet et al. (1993)
described the channels of the Okavango fan in
Botswana, which are low-sinuosity to meandering
rivers, in which banks are stabilized by peat levees.
The changes in river behavior when forests in upland
catchment areas are destroyed by fire or deforestation are marked and well known. Runoff becomes
more fl ashy, the sediment load increases, and a tendency may develop for debris fl ows to occur.
Rivers are sensitive to longitudinal changes in the
controlling variables and to temporal changes in
these variables. Carson (1984a,b,c) showed from his
studies of rivers in the Canterbury Plains area of New
Zealand that the transition from meandering to
braided takes place at higher slopes as the caliber of
the sediment load is increased. He also showed that
the development of the braided pattern is very sensitive to the local rate of supply ofbed load to the river.
A large bed load, supplied, for example, from easily
erodible banks, leads to channel shoaling and local
flooding, bank incision, and avulsion. N.D. Smith
and D.G. Smith (1984) described the William River
in northern Saskatchewan, a river that changes
downstream from a single-channel to a multiplechannel, braided pattern where a large sand bed load
is introduced by eolian processes. Stollhofen and
Stanistreet (1994) demonstrated a change in fluvial
style from meandering to braided when volcanoes
erupted in a fluvial plain and provided a large supply
of pyroclastic debris. Friend and Sinha (1993) recorded much local variablity in sinuosity and braiding character in the large rivers of India. They
confirmed the conclusions of Carson (1984a,b,c),
and noted that the nature of the local alluvial substrate, and the sediment input from tributaries, will
also contribute to continuing downstream adjustment and consequent change in the configuration of
the river. Important experimental work that confirms these trends has been reported by Ashmore
(1991), Leddy et al. (1993), and Germanoski and
Schumm (1993), as noted in Sects. 6.3 and 1!.2.2.
Temporal changes may be exemplified by several
studies of rivers in the United States. Nadler and
w,d,l
Q oc --w
s
w,l,s
Qs rx d,P
where Qw=discharge, O s =sediment load, w=width,
d=depth, !=meander wavelength, s=slope, and P=
sinuosity. For a few sets of rivers in specifi c climatic
ranges, some of these relationships have been quan�
tified, as summarized by Ethridge and Schumm
(1978). However, the data are inadequate to permit
generalizations that can be applied to all geological
conditions, a fact that considerably reduces the usefulness of quantitative geomorphology for geological reconstructions. Additional details on the
relationship between discharge and slope and the
braided-meandering threshold are given by )lridge
(1993b).
One of the primary prerequisites for braiding is a
high coarse-grained sediment load (bed load).
Leopold and Wolman (1957, p. 50) stated:
"Braiding is developed by sorting as the stream leaves
behind those sizes of the load which it is incompetent to
ha11dle ... if the stream is competent to move all sizes
comprising the load but is unable to move the total quantity provided to it, then aggradation may take place without braiding."
In addition to the relationship between discharge
and channel style shown in Fig. 8.1, Miall (1977) and
Rust (1978a) noted that braided rivers are commonly characterized by high discharge variability;
for example, rivers in alpfue and arctic areas, with
highly seasonal discharge variations, tend to be
braided, as are ephemeral rivers in arid regions.
Miall (1977, p. 7) argued from these and other observations that discharge variability is a primary control on. the. braiding pattern:
"In rivers of highly variable discharge competency will be
similarly variable, and there will be long periods of time
throughout which the river will be unable to move at least
the coarsest part of its bed-load. The incidence of bar
initiation, flow diversion and the creation of new channels
(braiding) will thus be high."
On the scale of the alluvial basin the nature of the
vegetation cover has an important effect on discharge characteristics, sediment load, and fluvial
style. In vegetated areas presently, runoff following
major rainfalls is rarely catastrophic because the
precipitation is absorbed by soil and plants and released slowly. Similarly, sediment is stabilized by
roots, and sediment yields are relatively low. VegFluvial Styles and Facies Models
eta ted banks are highly resistant to erosion, as demonstrated in an experiment by D.G. Smith (1976).
This limits the supply of sediment to the river, and
also reduces the bank widening that commonly accompanies braiding. In vegetated areas, therefore,
because of the effects on di,scharge variability, bank
stability and sediment supply, braiding is inhibited.
Jlaker (1978a) quoted examples in tropical Africa
and South America where dense bank vegetation
inhibits braiding, even where abundant coarse sediment is present in the banks. Stanistreet et al. (1993)
described the channels of the Okavango fan in
Botswana, which are low-sinuosity to meandering
rivers, in which banks are stabilized by peat levees.
The changes in river behavior when forests in upland
catchment areas are destroyed by fire or deforestation are marked and well known. Runoff becomes
more fl ashy, the sediment load increases, and a tendency may develop for debris fl ows to occur.
Rivers are sensitive to longitudinal changes in the
controlling variables and to temporal changes in
these variables. Carson (1984a,b,c) showed from his
studies of rivers in the Canterbury Plains area of New
Zealand that the transition from meandering to
braided takes place at higher slopes as the caliber of
the sediment load is increased. He also showed that
the development of the braided pattern is very sensitive to the local rate of supply ofbed load to the river.
A large bed load, supplied, for example, from easily
erodible banks, leads to channel shoaling and local
flooding, bank incision, and avulsion. N.D. Smith
and D.G. Smith (1984) described the William River
in northern Saskatchewan, a river that changes
downstream from a single-channel to a multiplechannel, braided pattern where a large sand bed load
is introduced by eolian processes. Stollhofen and
Stanistreet (1994) demonstrated a change in fluvial
style from meandering to braided when volcanoes
erupted in a fluvial plain and provided a large supply
of pyroclastic debris. Friend and Sinha (1993) recorded much local variablity in sinuosity and braiding character in the large rivers of India. They
confirmed the conclusions of Carson (1984a,b,c),
and noted that the nature of the local alluvial substrate, and the sediment input from tributaries, will
also contribute to continuing downstream adjustment and consequent change in the configuration of
the river. Important experimental work that confirms these trends has been reported by Ashmore
(1991), Leddy et al. (1993), and Germanoski and
Schumm (1993), as noted in Sects. 6.3 and 1!.2.2.
Temporal changes may be exemplified by several
studies of rivers in the United States. Nadler and