WSP12249

• File Number: 8240.200.10.B
• Description: UCRBRIP - Riverine Fish Flow Investigations
• Basin: Colorado Mainstem
• Date: 6/1/2004
• Author: CDOW
• Title: Riverine Fish Flow Investigations 2004
• Water Supply Pro - Doc Type: Publication

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

METHODS

Fish sampling was performed by electro-shocking from a IS ft Achilles raft from
1998 to 2000 and from a 16-ft Hyside self-bailer raft in 2001 and 2003, A Smith-Root
electro-tisher, SOOO-watt generator and anode array mounted on a forward boom was used in
all years. The raft was maneuvered by either oars or by a battery powered 40- to 7S-pound
thrust trolling motor, Two netters caught as many fish as they could while the shocker was in
operation. All fish were measured to the nearest millimeter. Density estimates were made
for each study site on the Yampa, Colorado and Dolores rivers, Only fish over 150 mm were
marked, and therefore used for mark and recapture population estimates.

The Darroch multiple mark method (Everhart and Youngs 1981) was used to make
the population estimate with ninety-five percent confidence intervals. A total fish estimate
was made for all species and for each species, Recapture rates generally varied between
species and size-groups, In our sampling larger suckers had the highest recapture
probabilities and the species with appreciably lower recapture probabilities included catfish,
bass, pike and carp (the lower group), The total fish estimate represented a blend of
recapture probabilities. but were expected to produce reliable comparisons for total fish
abundance between years. when species and size composition was consistent. For rare
species (pikeminnow, etc) with zero or one recapture in the sample, abundance was estimated
by dividing the number in the sample by the mean recapture probability of the lower group,

The z-test, with an alpha of 0,05 (z = 1,96) was used to test for significant differences
in density estimates between years at each station, At stations with three or more years of
sampling. the Bonferroni inequality was also used to control the overall significance level
(,OS) for the simultaneous comparison of all pairs of years (Dr. David Bowden, CSU, pers.
communication). At stations with four years of data, (Duffy and Sevens) the z value (2,631)
corresponded to an alpha of 0,05 divided by six (O,0083).

RESULTS

This report compares data collected in 2003 (post 2002 !lows) to data from earlier
years, A more complete synthesis of fish data from prior years (1998, 1999, 2000 and 200 I)
was presented by Anderson (2002), making it unnecessary to re-present all fish data collected
prior to 2003, Each annual progress report between 1999 and 2002 compared species
composition, size and density data to the prior year or years. Length frequency histograms
for each species at each site were presented in prior annual progress reports. Anderson and
Stewart (2000) presented length data for fish sampled in 1998 and 1999, Anderson and
Stewart (200 I) contained length data for the 2000 sample. The 200 I sample length data was
in Anderson (2002). Meso habitat preferences for depths and velocities determined for
flannelmouth sucker and bluehead sucker were given in the projects completion report
(Anderson and Stewart 2003),

15