9386

• UCREFRP Catalog Number: 9386
• Author: Ligon, F. K., W. E. Dietrich and W. J. Trush
• Title: Downstream Ecological Effects of Dams
• USFW Year: 1995
• USFW - Doc Type: A geomorphic perspective
• Copyright Material: YES

2,500
E
U
`r 2,000
N
aNi 1,500
t 1,000
N
z
a 500
c
0
0
co _
Distance from mouth (km)
Figure 4. Longitudinal profile of bound-
ary shear stress in the McKenzie River
for a ten-year recurrence-interval flood.
spawning habitat to accommodate
the nests (which are known as redds)
of the female salmon that return to
spawn.'
The number of female salmon per
redd counted in a reach of the
McKenzie for the years 1970-1986
ranges between 4 and 15, and aver-
ages approximately 8.5 (Figure 6).
If adequate spawning gravels were
available, the ratio of female salmon
to redds should be approximately
1:1. Because on average 8.5 times as
many females as redds are counted,
it is likely that spawning-gravel limi-
tations are resulting in redd super-
imposition (i.e., the building of one
redd on top of another).
Redd superimposition is a com-
mon source of density-dependent
mortality in salmonids (Hayes 1987,
McNeil 1964). When spawning grav-
els are saturated, subsequent spawn-
ers do not increase the total number
of eggs but merely replace those
already deposited with their own. If
there are sufficient spawners to satu-
rate the spawning gravels, the avail-
ability of spawning gravels deter-
mines the largest number of eggs
that can be seeded in a river, which
in turn can determine the equilib-
rium or average density at which the
population fluctuates-the fluctua-
tions are generally driven by den-
sity-independent mortality.
There appears to have been a
change in the population dynamics
of the salmon in the McKenzie from
the period 1945-1960 to 1969-1989
(Figure 7). In addition to smaller
fluctuations in 1969-1986, the av-
' F. K. Ligon and W. E. Dietrich, 1994, manu-
script in preparation.
erage population size has decreased
approximately 50%, from 17,000
to 8000. One of the two flood-con-
trol dams-Cougar Dam, built in
1963-eliminated access to a sig-
nificant amount of upstream spawn-
ing gravels (Howell et al. 1988).
The immediate reduction in the av-
erage population size after this re-
duction in spawning habitat is con-
sistent with our hypothesis that
spawning gravel limitations in the
McKenzie are the primary source of
density-dependent mortality.
As channel simplification due to
the reduced peak flows continues,
and further spawning gravels are
lost, we predict that the salmon
population will continue to decline.
The geomorphic evolution of the
McKenzie to a predominantly single-
thread channel is a gradual process,
occurring over decades. Because it
happens slowly, it goes unnoticed
even by fishing guides who have
been floating the river for many
years-the river still seems pristine.
Considerable research has been
conducted on the biology of McKen-
zie River salmon, such as examining
smolt mortality as a result of water
diversions. Variation in such mor-
186 BioScience Vol. 4S No. 3
0 50 100 150
Figure 5. This reach reflects the loss of islands between 1967 and 1990 in the
McKenzie River. River path is shown in dark blue for 1967 and in light blue for
1990. Islands in 1967 are green; there were no islands in 1990.