WSPC04651

• File Number: 8062
• Description: Federal Water Rights - Colorado Litigation - National Forest ISF Claims - Division 2
• State: CO
• Basin: Statewide
• Date: 9/17/1997
• Title: Materials of Interest - Effective Discharge and the Design of Channel Maintenance Flows for Gravel-Bed Rivers
• Water Supply Pro - Doc Type: Report/Study

0. () 2 \1 ~l1

a s11L1l1 quantity of sedimem over a period of years. The
very largest disch:ltges transpon sediment at an e~ceed.
ingly high rate. however. they occur so infrequently.
perhaps only a few hours per century. thar they carry only
. a small ponion of the to[31 quantity of sediment transponed
over a long period of time. The most effCl:tive sediment
transponing disch:ltges over a period of years are those
which transpon sediment at a moderately high rate and
occur as frequently as a few days per YeM.
The specific shape and characteristics of the sediment
transpon function. A. and the frequency of discharges. B.
vary greatly from river to river. Thus. the range and
occU1Tence of the most effective sediment transponing
discharges also vary greatly from one river to another. At
one extreme are Illose rivers which catTY some sediment ar
even the smallest discharge and do not have a panicu!arly
wide range of floods. In such streams. effective discharges
may oCCUr for several weeks to perhaps mouths per year.
At the other extreme are those rivers where only relatively
rare, extreme /loads are sufficient to move Ille available
sediment partide sizes. In such Streams, the range of
effective sediment transponing discharges may occur only
a few days per century.
The Wolman-Miller model for magnitude and frequency
of effective sediment transpOrt discharges has been
evaluated by numerous investigatioll5, [Pickup and Warner.
1976, Baker, 1977, Wolman and Gerson, 1978. Newson,
1980, and Miller. 1990]. Ntllh [1994] fOWld 138 citations
to Ille original paper between 1960 and 1990. The vast
majority of Ille insunces wbere the model has been
evaluated using field measurements have conSidered only
suspended sediment transpon. Measured suspended
sediment conoentration, in some C3Ses collected daily, are
available for sevet'll gaging sutioll5 in the United States
over periods of as long as 40 years. These comparisons
typically agreed quite well with the model. There is a well-
defIDed range of relatively frequent discharges which
tran5pOn most of the suspended sediment over a period of
years. Furthermore, Ille duration of the effeaive discharge
commonly varies from several days per year to perbaps
one day in several years. [see Wolman and Miller, 1960,
and Pickup and Warner, 1976].
Evaluations of streamflow magnitude and frequency
based upon suspended sediment tI1lD.<pon consider 90
percent or more of the sediment tran.sported out of a
drninage basin and, thus. are prohably a good approxima-
tion of the effe<;tive discharges for watersbed denudation.
Such evaluations. however. are not appropriate tests for the
second hypothesis of the Wolman-Miller model. namely
that the effective discharges are Ille cbanneHorming
discharge. because most of the suspended sediment is fmer

vOOd

0991 86v 0"6:131

ANDREWS AND NANKERVIS 153

than the materi31 composing river bed and banks. This
distinction is especially significant for gravel-bed rivers
where clay, silt, and fine sand eommonly represent 90
percent of the total sediment load and are transponed in
suspension even at relatively s11L1l1 discharges, whereas thc
river bed and banks consist of much coarser sized panieles
that may only be transponed by flows equalled or exceeded
a rew days or less per year.
Analysis of bed-material transpon magnitude and
frequency in gravel.bed riven; has been limited due to
substantial uncenainty in the transport rate at a speciflc
discharge. Extensive bedload transpon measurements bave
been made at relatively few sites, approximately 10, in
North America, and then only for a period of a few years
to at most a decade or so. [Milhous, 1973, Leopold and
Emmerr. 1976. Jones and Seirz:. 1980, and Andrews, 1994].
Thus, one must rely upon establishing a relation between
flow and bedload transpon rate, which can then be applied
over a longer period of reoorded discharges. For some of
the bedload measurement sites, however. the period of
record for discharges are insufficient to adequately defme
Ille frequeney of moderately large to extreme flows. Given
the very large oommitment of funds and time needed to
measure bedload transpon rates in a gravel-bed river over
a wide range of relatively large discharges, it is improbable
Illat there will be a significant number of sites wbere the
magnitude and frequency of measured bedload rranspon
will be determined solely from measurements.
Accordingly, the analysis of bedload transpon magnirude
and frequency must rely, to a substantial degree, upon
calculated transpOrt rates at stre3Illflow gaging stations with
long periods of reoord. This is the approach taken by the
investigation described herein. The approacb depends upon
Ille ability to predia these bedload tnlDspon rates in gravel-
bed rivers 10 a reasonable degree of acctU3CY.
The primary objective of this study is to formulate a
regime of streamflows sufficient to maintain the existing
bankfull channel for a given quantity and panicle size
distribution of sediment supplied to the cbatlllel. The
approach is to analyze the lD3gnirude and frequency of bed-
materialtranspon in gravel-bed riven; Illat are typieal of
the western United States, and determine wbether the
bankfull channel is related to the range of effective
transponing discharges. This approach is, in fact, an
evaluation of the Wolman-Miller hypotheses as they apply
to gravel-bed rivers. A seoond objective concerns only Ille
bankfull discbarge, Le. the conveyance of the naturally
formed cbannel. In many, perbaps most instances, other
channel anrlbu!es, e.g. channel sinuosity or relative pnol
depth, also may suppon significant or essential r050=5.
Maintaining a given bankfull channel by providing the

3JI^~3S lS3~Od vosn

H: 11 1031.11 L6 .d -d3S