8256

• UCREFRP Catalog Number: 8256
• Author: Holly, F. M. and R. Ettema
• Title: Sediment Imbalance in Rivers
• USFW Year: n.d.
• USFW - Doc Type: Simulation Possibilities and Problems
• Copyright Material: YES

r

f: ..

;1
iI,
~.:
;~

416 BIOLOGICAL REPoRT 19

':f.
Ji
-,J
if

changes on a river and its ecosystem, and generally
in a very short time (months to years). But the
response of the river and its ecosystem to these
perturbations is not necessarily coITespondingly
short. Several decades, perhaps even centuries,
may be required before the river returns to a new
(or former) equilibrium channel and planform ge-
ometry. Ironically, such natural response times
often span at least the typical professional career
of a river engineer. Yet this seemingly slow re-
sponse may, in just one generation, exterminate a
species that has otherwise survived from the age
of the dinosaurs.
Nature's deliberate but definite response to
man's rapid (and equally definite) intervention
may deprive the river engineer of an immediate-
feedback mechanism for evaluating the effective-
ness (and indirect habitat effects) of engineering
works. Design evaluations thus have relied on
physical scale-model studies for local, rather short-
term effects. Yet the scaling of physical models can
be inaccurate, especially when extended lengths of
river or periods of time are modeled. For this
reason, and because of the inordinate physical size
of laboratory space required, it rarely has been
practical to use physical models for those situ-
ations.
Computational simulation of extended lengths
of rivers and periods of time began in the 1960's.
At that time fixed-bed computational modelling
(that is, numerical simulation of flood propagation
and backwater effects in immobile-bed channels)
began to achieve a certain engineering maturity,
and the next logical step was to apply the same
techniques to mobile-bed simulation. However,
whereas fIXed-bed modelling evolved into a reli-
able, mature engineering tool in the space of about
10 years, mobile-bed modelling is still in its in-
fancy, groping for the statUB of its fIXed-bed pro-
genitor.
The problem in mobile-bed computational mod-
elling is one of applying effective numerical tech-
niques to frustratingly limited, crude schematiza-
tions of complex water-sediment processes, with
the goal of mAking reliable long-term predictions
of a river's response to changes imposed upon it.
That some progress has been made toward this
goal is a tribute more to the skill, intuition, and
experience of numerical modelers than to any in-
herent reliability of the modelling techniques em-
ployed.-
In the remainder of this paper we provide a brief
overview of mobile-bed computational modelling,

i
I
~
II
'"
,
~

I

~
I'
'I

discuss challenges and opportunities for the fu-
ture, and present an example application to the
Missouri River. Our purpose is to provide general
background on the capabilities and limitations of
such modelling in support of river-development
decisions.

Sediment Non-equilibrium

An early recognition of the notion of the mecha-
nisms by which an alluvial river is always trying
to achieve an equilibrium was presented by Lane
(1957) and depicted in a classic diagram, a recent
version of which is shown in Fig. 1. The diagram
reflects the essential role of shear stress. exerted
on a river bed by flowing water and relates the
sediment transport to the flow hydraulics. On the
water side of the scale, bed shear stress increases
as the channel slope and water discharge increase.
On the sediment side of the scale, the amount of
sediment transported increases with the shear
stress, and finer sediment is more easily trans-
ported than coarser sediment. Thus excessive or
insufficient shear (with regard to equilibrium con-
ditions) results in degradation (scour) or aggrada-
tion (deposition) toward a new equilibrium, or bal-
ance. This can result from adjustments of sediment
load, sediment size, channel slope (as in planform
meandering), or channel cross section (not explic-
itly represented in the diagram). Of course even
natural rivers are seldom in a state of true equilib-
rium, but rather are constantly adjusting. The
balances of Fig. 1 reflect both natural and per-
turbed adjustment tendencies.
The qualitative understanding of the processes
represented in Fig. 1 has entered the river-engi-
neering lexicon through the term "hungry water,"
refemng to the inevitable bed scour and degrada-
tion immediately downstream of dams that trap
sediments, releasing clear water with its potential
for unmitigated scour.
Figure 1 is thought-provoking and summarizes
the essential features of dynamic river equilibrium.
However, there are complexities and subtleties in
the way equilibrium can be obtained. For example,
hydraulic sorting, that is, the preferential trans-
port of finer material in a mixture, can result in a
coarSening of the bed; indeed, this is the essential
process by which the sediment-size adjustments of
Fig. 1 occur. But accumulation of a relatively small
amount of coarse, nonmoving material in an inter-
locking pattern on the bed can suppress virtually