8019

• UCREFRP Catalog Number: 8019
• Author: Montana Department of Fish, W., and Parks.
• Title: Kootenai River Fisheries Investigations, Final Completion Report.
• USFW Year: 1983.
• USFW - Doc Type: \
• Copyright Material: NO

Ordination techniques were applied to the data with the use of two
computer programs from the Cornell Ecology Program series. DECORANA,
a Fortran program which was used for detrended correspondence analysis
(Hill 1979) and ORDIFLEX, which was utilized for its programs for polar
ordination and principal components analysis (Gauch 1977). DECORANA
was performed with no transformation of the data and no downweighting
of rare species. Polar ordination was run with both automatic and user
selected samples as endpoints. Percentage distance was the measure used
for the computation of similarity of species composition among the various
samples, which is required for polar ordination. A major fu~~tion of
ordination is identification of groups of similar samples, The equation
for the percentage distance (PO) similarity measure is:
I
200 . L m in (D.., D. k )
PDjk = IA - PSjk, where PSjk = i = 1 lJ 1
I
L (Dij + Dik)
i = 1
where IA is the internal association, PS is the percentage similarity,
where the summations are over all species (I), Dij and Dik are the abundances
of species i in samples j and k, and Sj and Sk are the numbers of species
in samples j and k. The data were log transformed before principal components
analysis was appl ied. The output from PCA was centered and standardized.

The reduced biological data were then related to environmental predictor
variables using multiple regression and correlation analysis. These
statistical methods were util ized to assess the importance of such factors
as rates of flow change, temperature, substrate heterogeneity, and altered
autochthonous (periphyton) and allochthonous (seston) carbon resources
in determining the composition of benthic communities downstream from
dams.

RESUL TS

Density and Biomass Estimates

Data collected from October, 1979 to September, 1980 (1980 water
year) were used for comparisons of insect densities (number/m2) and biomass
(cc/m2) at the four sample stations. Accurate quantification of numbers
and biomass in the Kootenai River was compl icated by the fact that discharge
from Libby Dam was reduced to 4,000 cfs just prior to invertebrate sampling.
Insect drift is induced by reductions in flow and these flow reductions
may have concentrated insects along the shoreline where samples were
taken.

Annual mean densities (numberjm2) of all invertebrates ranged about
1.5 to 2.5 times qreater in the Kootenai River than in the Fisher River
(Table 1, Figure 3). Monthly mean densities for each insect order at
each station are listed in Appendix 1.

A one-way analysis of variance (ANOVA) at the four stations was
run on log transformed data of the monthly mean densities of all invertebrates.
Densities were significantly different (p<.05) for pairwise comparisons

-7-