9668

• UCREFRP Catalog Number: 9668
• Author: Crowl, T. A. and e. al.
• Title: Bonytail Draft 1998 Annual Report, February 2000.
• USFW Year: 2000.
• USFW - Doc Type: Logan, UT.
• Copyright Material: NO

Statistical Analysis
All analyses were performed using the SAS statistical package (1988) for personal
computers. All tests were performed on original data since the observations conformed
to the normality assumption.
To determine if bonytail differentially utilized substrates and if use occurred differently
during the day and night, R x C analyses were performed. We also compared the
position in the water column (bottom, middle, or surface) using a R x C test with position
and time (day and night) as the two categorical variables.
In order to determine whether flow selection was different between day and night
observations a one-way analysis of variance (ANOVA) test was done for each velocity
treatment separately. Because we observed no difference in flow selection as a
function of time of observation, day and night observations were pooled for the rest of
the analysis. A T-test was done for each velocity treatment (low, medium, high) to
compare nose velocities of bonytail with the available mean velocity. R x C tests were
done for each velocity treatment in order to determine whether bonytail were selecting
different depths as a function of time of observation (i.e. day versus night). Finally,
another R x C test was used to determine whether depth selection was dependent or
independent of the velocity treatment. R x C analyses were used to determine if flow
regimes affected substrate selection.
To determine how the presence of red shiners affected substrate and depth selection of
bonytail, R x C tests were conducted. The specific hypothesis tested was whether or
not substrate or depth selection were independent of the presence/absence of red
shiners. T-tests were used to compare average velocities at observed bonytail
positions when red shiners were present versus when they were absent from the tanks.
Initially, analyses were performed for each experimental flow regime separately. To
determine whether flow x shiner interactions occurred, a two-way ANOVA was
performed with the presence/absence of red shiners as a main effect, and low, medium,
14