7618

• UCREFRP Catalog Number: 7618
• Author: Rose, K. L. and D. R. Hann.
• Title: Summary of Historic Habitat Modeling on the Yampa River Using the Physical Habitat Simulation System
• USFW Year: 1989.
• USFW - Doc Type: Final Report.
• Copyright Material: NO

accurately reflect habitat conditions actually used by the fish. Further.
because gear capabilities limit our ability to effectively sample all habitats
(particularly deep. swift areas). utilization data used to construct SI curves
can be biased toward shallow. low velocity habitats that are more easily
sampled. For more detailed information on curve development. see Bovee
(1986).

Depth. velocity and substrate are weighted equally and treated independently
in the PHABSIM model. however. the model is most sensitive to the velocity
criteria. As a result. velocity 51 curves which peak at zero. (Valdez et al.
1987. Attachment 1) account for predictions of optimal WUA at close to zero
flow for adults. For adult Colorado squawfish the velocity curve has a peak
utilization value of 1.00 at 0.00 ft/sec; which is not surprising considering
that 76.4 percent of the data used to construct the curve came from pools
and/or eddies. Similarly. the adult humpback chub curve has a peak value of
1.00 at 0.23 feet per second. with 49.6 percent of the data from pools and/or
eddy habitats.

Winter requirements of Colorado squawfish and humpback chub in the Yampa River
are not completely understood. and were not included in this analysis.
Continued research and SI curve development for the Yampa River are in
progress. however, there is some question that the PHABSIM models may not work
well under ice. Winter habitat requirements for endangered fish may be quite
different from summer habitat requirements.

Conclusions

Ideally a species model must include important ecosystem variables that allow
for a direct connection to a response by the population. i.e. growth.
recruitment. survival. standing crop or some other quantifiable species
response. otherwise. results will not be biologically meaningful. In lieu of
such a model. critical limiting habitats and/or other quantifiable variables
need to be identified. and related to species habitat utilizations. In
addition. detailed field data specific to those habitats and/or variables
should be collected to develoD more definitive habitat versus flow
relationships. .

In a "natural" system like the Yampa River (no major dams or diversions). the
availability of physical habitats above a certain unknown minimum threshold.
mayor may not be limiting to endangered fishes. It must be recognized.
however. that there are many more environmental factors potentially
influencing population size than just the availability of gross physical
habitat. The physical microhabitat component of IFIM requires the assumption
that fish population response has some correlation to changes in WUA. This
assumption has never been validated for large. turbid rivers. It has not been
demonstrated that WUA. as defined by depth. velocity. and substrate has any
relationship to endangered fish populations whatsoever. Determining the
degree of accuracy of existing endangered fish SI curves and the corresponding
predictive reliability of PHABSIH. and what it means to the overall health of
fish populations will require many years of extensive field validation and
hypotheses testing. Until species criteria can be validated. they are of
limited value. Model outputs based on existing SI curves for the Yampa River
do not, in most instances. seem to accurately predict habitat versus flow

17