9708

• UCREFRP Catalog Number: 9708
• Author: Grand, T.C., S.F. Railsback, J.W. Hayse and K.E. LaGory.
• Title: A physical habitat model for predicting the effects of flow fluctuations in nursery habitats of the endangered Colorado pikeminnow ‎(Ptychocheilus lucius)‎.
• USFW Year: 2006.
• USFW - Doc Type: River Research and Applications
• Copyright Material: YES

FLUCTUATING FLOWS AND BACKWATER HABITATS

1141

orientation to the mainstem results in their becoming flow-through habitats at relatively high base flows (Rakowski
and Schmidt, 1997) may be affected more by flow fluctuations than the backwaters considered here. However, such
flow-through backwaters are relatively uncommon in our study area,
In order to predict the effects of river management on endangered fishes, it is necessary to first predict the effects
offlow on fish habitat (Petts, 1991; Osmundson et at" 2002), Typically, such predictions are made by creating a
model of the physical habitat and determining how the availability of suitable fish habitat varies with stream flow
(e.g, IFIMlPHABSIM; Bovee et at., 1998), While such habitat suitability models and their extensions have been
used with varied success on cold-water stream fishes, they are inappropriate for warm-water fishes that move over
large areas and are more likely to be limited by competition and food availability than by availability of habitat area
(Railsback, 1999; Railsback et a!., 2003),
As part of an em-going project to predict the effects of within-day flow fluctuations on the growth and survival of
endangered juvenile Colorado pikeminnow in the Green River, Vtah, we developed a model of the backwater
habitats preferred by these fish. Our model combines a cell-based representation of how backwater geometry and
mixing with the mainstem depend on flow, a pond-based temperature model and a model of invertebrate production,
Although relatively simple, our model makes several testable predictions about the effects of flow fluctuations on
backwater geometry, temperature and invertebrate availability. Specifically, our model predicts that as within-day
flow fluctuations increase, minimum daily wetted areas will shrink., summer and fall backwater temperatures will
tend to decline, and densities of invertebrates available for consumption by fish may be reduced. Depending on the
magnitude of the effect, such responses could affect fish growth and survival.
As with all models, we have made assumptions and simplifications that may affect the predictions generated, For
example, our model does not allow for the possibility that backwaters may become flow-through habitats at higher
flow. That is, we have assumed that backwaters are connected to the mainstem at only a single point (as is the case
for the backwaters considered here), :.md that water exchange and flushing of invertebrates into the main channel
will occur only in response to the rise and fall of backwater depth relative to mainstem stage. This assumption is
true for (1) the range of fluctuations we modelled and (2) the range of fluctuations typically observed during the
summer and autumn in the Green River study area, More extreme inundation events would likely increase the
proportion of invertebrate production flushed into the main channel and may even result in the complete elimination
of the backwater as a nursery habitat (Rakowski and Schmidt, 1997), Similarly, we have assumed that backwater
morphology remains constant for the duration of the simulation; in reality, flood events (which typicaIIy occur in
the spring, but may also occur later in the year) can transpOlt sediment, changing the topography of backwaters and,
in some cases, resulting in the elimination of existing backwaters and the creation of new ones (Rakowski and
Schmidt, 1997). Finally, we have necessarily simplified our calculations of invertebrate production and availability
by assuming that (1) the entire invertebrate community can be reasonably represented by a single species of benthic
invertebrate, (2) invertebrates produced but not eaten by fish on a given day are unavailable for consumption by fish
on subsequent days and (3) invertebrates can only be flushed out of the backwater, rather than into the backwater
from the mainstem, by fluctuating flows. While making the model more realistic, the addition of any of the above
processes would undoubtedly increase its overaII complexity and potentially reduce our ability to understand any
observed effects on pike minnow growth and survival.

ACKNO\VLEDGEMENTS

This research was sponsored by EPRI, Electric Power Research Institute, Inc., under agreement EP-P321S/C1529
with Lang, Railsback & Assoc., and by the Western Area Power Administration. Software SUPPOlt was provided by
Steve Jackson, Jackson Scientific Computing, McKinleyville, California, Argonne National Laboratory's work was
supported by the Western Area Power Administration under interagency agreement, through V,S. Department of
Energy contract W-31-109-Eng-38. We thank the V,S, Fish and Wildlife Service's Colorado River Fish Pr~ject and
the Ouray National Wildlife Refuge for logistical assistance provided during field studies. Special thanks to Gary
Burton, Clark Burbidge and Heather Patno (Western Area Power Administration), and August Deschais for their
assistance in collecting field data.

CopYlight t'o 2006 John Wiley & Sons, Ltd,

River Res, Applic. 22: \125-1142 (2006)
DOl: 1 (J,10021rra