WSP05604

• File Number: 8220.101.10.D
• Description: Glen Canyon Dam/Lake Powell
• State: AZ
• Basin: Colorado Mainstem
• Date: 1/1/1999
• Author: Topping et al.
• Title: Linkage Between Grain-Size Evolution and Sediment Depletion During Colorado River Floods
• Water Supply Pro - Doc Type: Report/Study

72 GRAIN-SIZE EVOLUTION DURING FLOODS

malion is known for a particular flow event, the results may
have link generality, as the pre.c;(isting sediment volume
and size-class conditions in the channel are likely (0 he
different for subsequent high-now events. Therefore, it
would be advantageous 10 develop a different methodology
for determining the state of the sediment supply,
In Ihis paper, we describe the basis for a different
methodology for extracting information about the state of
the supply of fine sediment in Marble and Grand Canyons,
Rather than Irying 10 assess Ihe actual volumes and sizes of
sediment present on the bed of the channel. our approach
views the problem indireclly, Specifically, by tracking lhe
temporal evolution of the concentration and grain~size
distribution of sediment in suspension. it is possible to
inverllhe problem and determine the state of the supply of
fine sediment in Marble and Grand Canyons, As a first step
in solving this problem, this paper documents the linkage
between the grain-size distribution of the bed, the concen-
Iration of each size class of suspended sediment in the main
channel and in a lateral recirculating eddy, and the grain-
size distribution in flood deposits in eddies,
One of lhe primary strengths of this perspective on
sediment-supply limilalion is Ihal il allows amalgamation
of both modern and historical suspended-sediment dala
with the observed sedimenlOlogy of both pre-dam and
recent channel deposits. This combination of seemingly
disparale dal3 produces an excellent conlext for evaluating
the behavior of the river with respect to different supplies of
sediment. In addition, this view of the sediment-supply
problem has implications for river management. For
example, this perspective illustrates how future controlled
floods released from Glen Canyon Dam (U.S, Deparrmellt
of Ihe IlIlerior, 1995J could be designed to compensate for
very different supplies of sediment.

2, PRE- AND POST-DAM SEDIMENT SUPPLY

Sediment tran.port in the Colorado River i. charac-
terized by (Wo key fealUres lhat are similar in both the pre.
and post-dam systems in Marble and Grand Canyons, This
similarity allows these systenls to be linked in the data
analysis presented in this paper, Fi"t, both pre- and post-
dam systems are characterized by annual supply limitation
with respect to fine sediment (i.e., sand, silt. and clay),
meaning that, over (he course of a year, the river has the
capacity to transport more fin~ sediment than is lypically
supplied 10 II by its lribulilnes. Second, both (he pre- anLl
post-dam systems display a mismatch hetween (he liming
of high sedimenL supply to the channel ~nd the timing of
high sediment-transport evenlS in the channel.

For (he pre-dam river, annual supply limilalion and (he
mismatch in liming of sediment supply and mainslcm
transport are closely related, Before completion of the dam.
010" of the water that passed through Grand Canyon origi-
nated in the mountains of Colorado, Utah, and Wyoming.
whereas most of the sediment was supplied by Colorado
Plateau tributaries that drained direclly into the canyons of
the Colorado River downstream of the confluence of the
Green and Colorado Rivers [Smith et ai"~ 1960; Alldrews,
1990, 1991], High discharges in Ihe Colorado were
primarily associated with lhe annual snowmelt floods that
typically occurred from late March lhrough early July, In
contrast, the highest sediment inputs to Ihe Colorado River
occurred during floods on the Colorado Plateau tributaries
during the summer thunderSlOrm season (July through
October) and also during late-winter and early-spring floods
on two of the largest tributaries, the San Juan and Lillie
Colorado Rivers, Mainstem discharges during times of high
..diment supply were typically low, so some ponion of the
sediment supplied during this period was stored in the
channel and eddies of the Colorado River until Ihe next
snowmelt season. Over lhe course of the snowmell flood,
sediment stored in the river during the previous months
gradually became depleted. such that, at the same discharge
of water, sediment concentrations decreased over time. This
hysteresis in suspended-sediment concentration. which is a
typical characteristic of a supply-limited system, was first
recognized and inv",ligated at lhe Grand Canyon gage by
Leopold alld Maddock [I95:\J, Hysteresis in sediment
concentration in the Colorado River has been related both
to seasonal scour of the bed [Leopold WId Maddock, 1953;
Brooks, 1958; Burkhalll, /9861 and seasonal coarsening of
the bed [Colby, 1964; Burkham, 1986; Rubin era/., 1998J,
After completion of Glen Canyon Dam in 1963, a similar
situation arose in the Colorado River below the dam. but
for somewhat differenl reason., Although the dam dramati-
cally reduced the frequency and magnitude of relatively
high sediment-transporting flows, it also virtually elimi-
nated the sediment supply to the portion of Glen Canyon
below the dam and greatly reduced the supply of fine
sediment to Marble and Grand Canyons. Furthermore, the
dam introduced wide-ranging daily discharge fluctuations
(hal sometimes exceeded 820 O13/s in a single day, while
greatly diminishing the seasonal discharge variation
relative to pre-dam conditions. The dam effectively
replaced relatively high and low flows with a greater
frequency of moderate flows that range from :\00 and 600
mJjs; these moderate discharges have significant sediment-
lransporling capabililies l U.S. Departmew of the Iwerior,
19951, The nel result of these alterations in the hydrograph
is that tl1l2 sl:dimenl-transporl capacity of the Colorado