8157

• UCREFRP Catalog Number: 8157
• Author: Pitlick, J., M. V. Steeter, B. Barkett, R. Cress and M. Franseen.
• Title: Geomorphology and Hydrology of the Colorado and Gunnison Rivers and Implications for Habitats Used by Endangered Fishes.
• USFW Year: 1999.
• USFW - Doc Type: \
• Copyright Material: NO

Another behavior pattern of Colorado squawfish that is common to both river systems is their use
of backwaters. "Backwaters" are ephemeral, low-velocity embayments that form along the shore,
downstream of islands, or at the mouths of secondary (side) channels. In terms of areal extent,
backwaters constitute a small amount of the total riverine habitat in the upper Colorado River, but
adult squawfish are found relatively often in these low velocity habitats, especially in spring
(Osmundson and Kaeding, 1991; Osmundson et al., 1995). Apparently, the fishes seek out these
habitats because they provide areas for resting which are close to areas used for foraging.
Rationale Behind This Study
The prevailing thought among biologists is that adult Colorado squawfish prefer "complex" river
reaches with a multithread channel pattern (Osmundson and Kaeding, 1991; Osmundson et al.,
1995). Complex river reaches offer diverse and heterogeneous habitats, among which fish can
select according to their particular needs. Our work is motivated from the point of view that the
Colorado River should be managed to improve the widest range of habitats in the widest range of
places. The most recent data indicate that while there is a clear tendency for adult squawfish to
congregate in the Grand Junction area during the spawning season, they do not use the same sites
year in and year out. Thus we suggest that until specific spawning sites are identified, and it is
established that these sites are used repeatedly, the most reasonable approach for managing
spawning habitats is one that improves the quality of gravel and cobble substrates in many places.
The solution to this problem involves specifying a discharge or range of discharges that will initiate
gravel transport on a widespread basis, and thereby prevent fine sediment from accumulating on
the bed. Fine sediment has probably always been a major constituent of the sediment load of the
Colorado River, but there is a tendency for silt and sand to build up on the bed during periods of
low flow (c.f. Milhous, 1998). It has been shown in many studies that fine sediment cannot be
winnowed from appreciable depths within the bed unless the framework particles themselves are
moved (Diplas, 1994; Kondolf and Wilcock, 1996); thus, periodic movement of gravel particles is
a key requirement for maintaining spawning substrates.
Another major goal of our work was to specify conditions under which backwater habitats are
formed and maintained. The physical characteristics of these features vary widely throughout the
study area, however, we do know that they tend to fill in with fine sediment during periods of low
flow (Osmundson et al., 1995; Van Steeter, 1996). To prevent this from occurring, a balance
must be maintained between the sediment supplied to the reach and the sediment carried out of the
reach. Any sediment that is not carried out of a particular reach will be deposited somewhere. It is
well established in the sediment-transport literature that materi al moving in suspension will be
deposited on the bed if either (a) the sediment-laden water enters an area of lower flow velocity,
such as a backwater, or if (b) the sediment concentration increases. Thus we provide an analysis
of sediment concentration data to evaluate historic trends in sediment loads, and determine which
flows carry the majority of the annual sediment load through the Grand Valley.
Finally, we consider what discharges would be required to increase channel complexity and form
new backwater habitats. To do this, the channel must become wider to create the space for new
bars and side channels to form. Using a physically based theory developed by Parker (1979) we
show that an approximate threshold for channel widening can be defined in terms of a bed load
transport criterion. Parker's results suggest that a channel formed in noncohesive sediment (sand
or gravel) will begin to widen once the average boundary shear stress, ; exceeds the critical shear
stress for bed load transport, r,, by about 20%. Using field measurements of bankfull depth, and
reach-average values of slope and grain size, we show that the bankfull r is consistently about 1.5
times the r, through the entire 90-km study reach.