Geoscience Reference
In-Depth Information
particular during floods, yet their possible contri-
bution to the avulsion processes on the braid plain
itself is not tackled by the experimental studies
mentioned in this section.
local reduction in channel capacity or by local
obstruction. Flow blockage may also be caused
by storm surge migrating up river (backwater
effect).
For low-gradient, subcritical (Froude < 1), flow-
ing rivers, the backwater effect is defined by the
distance L at which the water level has adapted to
67% of its upstream normal flow depth and is
estimated by
Low-gradient rivers
The autogenic behaviour in the low-gradient
river category includes the behaviour of all sin-
gle thread leveed anastomosing and meandering
channel systems. Experimental studies for this
category focusing on effective aggradation rates
and flow occupancy are almost non-existent,
with the exception of the revolutionary cohesive-
delta experiments by Hoyal & Sheets (2009).
They find, on the basis of their experiments for
cohesive delta plains, that avulsion of channels
(and their lobes) happens in three steps: The first
step involves bar aggradation above the point
where the incipient topography affects the flow
which leads to flow widening and flow bifurca-
tion leaving a V-shaped, subaerial region on the
bar surface and ending the bar cycle. The second
step, of negative feedback, involves a morpho-
dynamically mediated backwater effect that is
created by the mid-channel bar. As the bar grows,
a hydraulic backwater effect propagates slowly
upstream in the delta distributaries and is fol-
lowed immediately by a wave of channel bed
aggradation. As the lobe continues to grow and
channel bed aggradation increases, overbank
flow drives accelerated subaerial levee growth.
This drives the system to step 3, where the com-
bined effect of bed aggradation and progressively
upstream levee growth leads to super-elevation
of the channel and ultimately to the 'discovery'
of a more favourable path to the shoreline, i.e.
avulsion.
Additional insight into the autogenic behav-
iour of low gradient rivers is mainly based on
historical and sedimentological reconstructions
and on numerical modelling. These reconstruc-
tions have led to the common belief that avul-
sion of single thread rivers is driven by 1) local
super elevation of some part of a channel or
channel complex above its surroundings by the
ratio between cross-valley and down-valley gra-
dient (gradient advantage) and 2) the occurrence
of a trigger event, commonly a flood (see review
by Jones & Schumm, 1999; Stouthamer &
Berendsen, 2007) or storm surge, the latter being
important in delta distributaries. The river flood
may cause avulsion by blockage of the flow by
h
s
L
= 3
[m
(1)
with h = flow depth [m] and s = channel slope
[-] (e.g. Van Rijn, 1994). Hoyal & Sheets (2009)
found in their experiments that the real morpho-
dynamic backwater effect may easily be twice as
much of the calculated effect, which could bring
the avulsion node that much farther upstream,
theoretically. Several important examples of
avulsions triggered by various means of channel
blockages have been documented by King &
Martini (1984), Schumann (1989), McCarthy
et al . (1992) and Harwood & Brown (1993). The
interaction of both  drivers for autogenic change
(i.e. gradient advantage and triggering events)
was tested by numerical modelling of river
behaviour (e.g. Mackey & Bridge, 1995; Törnqvist
& Bridge, 2002; Karssenberg & Bridge, 2008).
Recent numerical modelling by Kleinhans et al .
(2008) demonstrated that during the initial bifur-
cation of the river, when water and sediment are
split over two branches, the choice of which
bifurcate channel becomes more important than
the other is determined by a number of factors of
which local gradient advantage is just one. The
other factors are the position of the avulsion node
relative to the upstream meander bend (Kleinhans
et al ., 2008), the channel width-depth ratio of the
bifurcate channels or the breach (e.g. Slingerland
& Smith, 1998), the grain size sorting and the
presence of local obstructions (bars and bank
irregularities, see Kleinhans et al ., 2008). The
factors together offer an explanation of why some
bifurcations were destabilised in decades and
others in centuries in the Rhine Meuse system
(Kleinhans, 2010).
Although avulsion drives the single thread riv-
ers to distribute their sediment evenly over the
coastal lowlands, bank stability and differential
compaction rates between the fine grained and
peaty floodplains and silty to sandy channel belts
makes the surface area of such systems highly
irregular, even at high avulsion rates (e.g.
Search WWH ::




Custom Search