PAST,
PRESENT, AND
FUTURE FISHERY MANAGEMENT
IN CABINET GORGE AND NOXON
RAPIDS RESERVOIRS
Scott
S. Rumsey and Joe E. Huston
Montana
Department of Fish, Wildlife and Parks
Kalispell. MT 59901
Abstract--Noxon
Rapids and Cabinet Gorge Reservoirs are run-of-the-river hydroelectric
impoundments on the Clark Fork River in western Montana. Operations at Noxon
Rapids Reservoir changed in 1961 and 1978. The first change in operations
increased average annual spring drawdown from less than 10 to more than 30 feet.
The second eliminated drawdowns of more than 6 feet, except for unusual power
demands. Establishment and maintenance of a satisfactory sport fishery has been
largely unsuccessful in both reservoirs since the 1950's. Rainbow trout (Salmo
gairdneri) exhibited potential from 1958 through 1960 when they produced an
excellent fishery, followed by a dramatic decline in 1961. Continued planting of
rainbow trout never reestablished a substantial fishery. Other fish species
planted produced similar results with the exception of brown trout (Salmo
trutta) and small mouth bass (Micropterus dolomieui), which presently
provide a modest fishery. Largemouth bass (Micropterus salmoides) were
present prior to impoundment and appear to be increasing. Operation changes,
combined with the brief retention times, have encouraged the downstream movement
of most introduced salmonids in Noxon Rapids and Cabinet Gorge Reservoirs.
INTRODUCTION
Noxon Rapids and Cabinet Gorge
Reservoirs are two run-of-the-river hydro- electric impoundments on the Clark
Fork River in western Montana (fig. 1). Both dams are owned and operated by
Washington Water Power Company of Spokane, Washington. The company has provided
the Department of Fish, Wildlife and Parks with funds and personnel for ongoing
fishery projects since construction. Cabinet Gorge was completed and filled
during 1952 and Noxon Rapids in 1959. Major activities affecting the watershed
within the immediate area include logging and forest road construction, mining,
and municipal, agricultural, and urban development. The primary objective of
work conducted since impoundment of both reservoirs was to provide an adequate
sport fishery. This was attempted through fish introduction.
DESCRIPTION
OF PROJECT AREA
Cabinet Gorge Reservoir has a drainage area of approximately 22,000 mi2 and
surface area of 3,200 acres at full pool. The water exchange rate during
Operations
Since completion of Noxon Rapids
Dam immediately upstream, Cabinet Gorge Reservoir has been primarily used as a
baseload reregulating impoundment for fluctuating flows from Noxon Rapids
Reservoir and provides a 3,OOO-ft3/s minimum flow to the Clark Fork River.
Before construction of Noxon Rapids Dam in 1959, the weekly fluctuations of the
Cabinet forebay ranged from 3 to 14 feet, with 15 feet being maximum drawdown.
Since that time, fluctuations have rarely exceeded 5 feet.
Noxon
Rapids Reser
Noxon Rapids Reservoir contains
7,940 surface acres at full pool, with an actual maximum drawdown of 54 feet.
The water exchange period during average annual flows of 21,470 ft3/S is
approximately 12 days. Flows vary seasonally from 4,000 to 120,000 ft3/S. Noxon
Rapids Reservoir can be divided into two distinct habitats: (1) the upper,
slow-moving river area extending from the Vermilion River upstream 20 miles; and
(2) the lower 18-mile lake area downstream from the Vermilion River to the dam (fig.
3). The upper portion averages one-fifth mile in width, has a maximum
depth of 75 feet, and has visible current the entire year. Shorelines are
generally steep with limited areas less than 30 feet deep. The lower section is
wider, averaging one-half mile in width with a maximum depth of approximately
225 feet. Shoreline slopes are moderate with numerous littoral areas less than
30 feet. Aquatic vegetation is extremely abundant in these areas as water levels
become more stabilized. Tributaries that have served as spawning habitat for
fish are Swamp Creek, Martin Creek, Vermilion River, Graves Creek, and Prospect
Creek. Prospect Creek is the largest, with a mean annual discharge of 260 ft3/S.
Noxon Rapids Reservoir is presently used as a storage facility with an active
storage capacity of over 230,000 acre-feet. Immediately following construction
in 1959 until 1961, the dam was used for base load and peaking power production,
drafting approximately 1-2 feet daily and not more than 10 feet annually. From
1961 until 1978 the dam was operated in a similar manner; however, early spring
drawdowns averaging 35 feet were implemented for power production and flood
control. In 1979, operations reverted to previous conditions (1959-61) where
early spring drafting was eliminated and minimum drawdown typically did not
exceed 12 feet (fig. 4).
Operational Jurisdiction of Cabinet Gorge and Noxon Rapids Reservoirs
Both reservoirs fall under jurisdiction of the Northwest Power Agreement of
1960. Basically, this requires private utilities to honor requests from other
utilities for electrical energy. The delivery utility can provide the requested
energy from its own resources or from outside resources. The Army Corps of
Engineers may also request assistance for flood control. The Pacific Northwest
Electric Power Planning and Conservation Act of 1980 also required plans to
protect and enhance fish wildlife and to mitigate their losses due to
hydroelectric development on the Columbia River and its tributaries.
METHODS
Numerous surveys have been conducted intermittently since the mid-1960's by
State fishery and Company personnel to inventory habitat and determine species
abundance and composition. Techniques included: gill netting, beach seining,
electrofishing, explosives, hook-and-line sampling, snorkeling, angler
interviews and creel census, visual observations and historic accounts from
local residents.
Stocking of various fish species was conducted using hatchery and wild sources.
Juvenile. subcatchable (1- to 6-inch) and catchable (greater than 6-inch) fish
were used throughout the report period. In some cases, eyed eggs were also
planted to induce homing to natal areas as spawning adults. Vibert boxes or
artificial redds were used in four tributary streams of Noxon Rapids Reservoir:
Prospect Creek. Graves Creek. Martin Creek, and Tuscor Creek. Particular species
introduced were selected for compatibility with characteristics of these Clark
Fork reservoirs and were dependent on available hatchery stocks.
An artificial incubation channel was constructed on a spring
channel of Prospect Creek during the fall of 1966 and planted with cutthroat and
brown trout eggs. The channel was previously used by Northern Pacific Railroad
to raise trout in the late 1920's and early 1930's for use in their diner cars.
Silt and debris were excavated from the channel, and 85 yd3 of 1/2-inch washed
gravel was placed 6-12 inches deep. When completed, 3,000 ft2 of riffle area
were available for egg plants. The outlet structure on the ponds was also
modified to monitor numbers of out-migrating fry.
Experimental gill nets were used as the primary means of determining fish
population trends throughout the period. Nets were either 125 or 250 feet in
length, 6 feet deep, and contained graded, square mesh from 3/4 to 2 inches.
Interviews, creek checks, log books, and catch data were
obtained from anglers to further determine trends on both reservoirs.
Fish were fin-clipped or jaw-tagged at various times to
determine the success of plants and to provide harvest and movement information.
Monetary rewards were offered for return of these tags.
Age
and Growth
Scales to determine age and growth were taken from selected fish species in Noxon Rapids Reservoir. In computing growth rates, a nomograph was used with a zero intercept, assuming a straight-line proportion between scale radius and total fish length.
Temperature and dissolved oxygen measurements were made in Noxon Rapids and
Cabinet Gorge Reservoirs to determine if either of those factors restricted
trout production. Temperatures were measured with Foxboro resistance
thermometers. Oxygen concentrations were determined using the Modified Winkler
method.
RESULTS
During August and September 1958, chemical treatment (rotenone) of the Clark
Fork River from Thompson Falls Reservoir to Cabinet Gorge Dam was conducted.
This was done to eradicate undesirable rough fish species thought to compete
with game fish. Success of the treatment was considered to be limited, due to
higher flows than expected.
Prior to completion of Cabinet Gorge Reservoir, the section
of the Clark Fork River between Lake Pend Oreille, Idaho, and Thompson Falls Dam
provided spawning, rearing, and resident habitat for numerous fish species (Table
1). Cabinet Gorge Dam blocked spawning runs of fish from Lake Pend
Oreille, Idaho. Game fish spawning runs were maintained or developed in few
tributary streams or tail water areas of Cabinet Gorge Reservoir. A spawning
area below Cabinet Gorge Dam has been used by bull trout, brown trout, kamloops,
cutthroat trout, and kokanee. The Bull River, Rock Creek, and the tail water
below Noxon Rapids Dam presently have limited spawning use by brown trout and
bull trout from Cabinet Gorge (fig. 2).
Numerous tributaries with spawning habitat enter Noxon Rapids
Reservoir (fig. 3). Rainbow were only observed in small numbers in Prospect,
Tuscor, Martin, and Graves Creeks, and the Vermilion River. Graves Creek and
Prospect Creek were primarily used by brown trout. Counts in 1962 showed
approximately 250 pairs of fish spawning in Graves Creek. In 1965, this stream
developed an underground section near its mouth, blocking upstream fish passage.
This section began to flow above ground again during 1983. Prospect Creek and
the Vermilion River were the principal streams used by bull trout; however,
spawners were not counted.
The first introductions of fish occurred during 1953 in Cabinet Gorge Reservoir
and in 1958 in Noxon Rapids Reservoir. Fish have subsequently been planted
nearly every year until present (table
2). From 1953 to 1984, the following
species were planted: rainbow trout--3,865,000; kokanee salmon-- 1,807,000; west
slope cutthroat--1,346,000; Yellowstone cutthroat--1,184,000; brown trout--689,000; small mouth bass--150,000; coho salmon--100,000; and ling--420.
All of these plants were fry or juveniles with the exception of ling, which were
primarily adult fish. All plants were made directly in the reservoirs or
tributary streams. The incubation channels constructed on Prospect Creek were
apparently unsuccessful for brown trout because adults never returned in
significant numbers.
Fish Population Trends
Gill net catches from 1955 on both reservoirs indicate extremely low numbers of game fish except during 1960 (table 4). Results are expressed using an average catch-per-net night based on a 125-foot net. Rainbow trout caught during 1960 were a direct result from large plants made 1 and 2 years earlier. Subsequent rainbow plants made in 1960-64 and 1981 never contributed significantly to gill net catches.
The number of suckers in both reservoirs has decreased
substantially during the netting period. Chemical treatment in 1958 may have
temporarily decreased suckers in Cabinet Gorge, but had little effect on Noxon
Rapids' populations. Northern squawfish numbers dropped slightly during the
early portion of the period and now appear to be increasing. Peamouth and yellow
perch have steadily increased.
Small mouth bass planted in 1982 and 1983 in Noxon Rapids
appear to be spreading throughout the reservoir in good numbers from 1983 to
1984 surveys (fig. 5). Scale aging showed the majority of fish resulting from
the 1982 plant averaged 6.9 inches in June and 8.2 inches in August. Small mouth
bass were not located in Cabinet Gorge during 1984 sampling, suggesting
downstream movement was not extensive.
Largemouth bass appear to be increasing in both reservoirs since flow stabilization in 1978 (figs. 5 and 6). Gill netting is not an efficient method to capture largemouth; however, they were observed and caught by fishing and observed during snorkeling. Increased growth of aquatic vegetation, particularly in Noxon Rapids, appears to be providing more suitable habitat for largemouth bass. Redside shiners, a prey species for both largemouth and small mouth bass, also prefer and benefit from this type of habitat. Ling have not been caught during sampling since introduction in 1971, but three were reported caught by anglers.
Creel
Census
Information compiled from anglers is similar to the results from gill netting on both reservoirs when considering game fish (table 5). An elaborate creel census was conducted from 1959 to 1962 when large numbers of rainbow trout were tagged and released. Fingerling rainbow primarily entered the catch 2 years after stocking, and this occurred during 1960 and 1961. During 1962, catches were primarily 3- and 4-year-old fish, indicating poor survival on the later plants.
Fish marking was conducted from 1958 until 1962 in Noxon Rapids and from 1959 until 1961 in Cabinet Gorge Reservoir. This effort was directed toward rainbow trout and was used to determine fish movement. Returns from gill net catches and anglers demonstrated that rainbow trout, with few exceptions, moved downstream. Downstream movement in both reservoirs was also found through gill netting in the spring of 1961 when operations changed from a limited daily drawdown of 1-2 feet to an annual spring flood control drawdown of approximately 35 feet (table 6). There was a considerable decrease in the number of rainbow trout taken from gill nets after drawdown, whereas other fish were not greatly affected.
Scales for age and growth determination were collected from gill netting and creel census from 1958 to 1961 and 1966 to 1976 from Noxon Rapids (table 7). Data listed represent back-calculated annulus growth increments in inches from fish before and after the drawdown change.
Growth rates for Noxon Rapids' rainbow
were slow before the drawdown change in comparison to other Montana reservoirs.
Sample sizes for rainbow were inadequate, and no conclusions were drawn from
after the drawdown change. Growth for bull and brown trout was better after the
drawdown change and probably resulted from an increased forage base in the
reservoir.
Growth
rates for largemouth bass inhabiting Noxon Rapids Reservoir were slow when
compared with other reservoirs. Fish reach a catchable size of 10 inches at the
start of their fifth year of life.
Neither Noxon Rapids nor Cabinet Gorge Reservoirs forms an abrupt thermo-cline.
Surface temperatures during July seldom rise above 74 of and average 72 of
except for the shallow littoral areas. Temperatures 100 feet deep average 65 of,
while those at 175 feet were about 55 of. Dissolved oxygen concentrations ranged
between 7.5 and 9.0 ppm at the surface, and never fell below 5.0 ppm at the
175-foot. level. Temperature and oxygen levels in the deeper areas are within
the range of game fish tolerance during the warm months.
Complete ice cover on Noxon Rapids Reservoir occurs from 2
to 3 months during a typical year, where Cabinet Gorge Reservoir rarely exceeds
30-40 per' cent cover for 1-2 months.
Productivity within both reservoirs is expected to be limited due to the rapid
exchange rates. Reservoir drawdowns further reduce production and disrupt food
chains due to the variations in habitat.
Gas saturation levels were determined for water entering and
leaving Thompson Falls Dam, Noxon Rapids Dam, and Cabinet Gorge Dam in 1974. The
Thompson Falls forebay averaged about 100 percent from April through August.
During spill discharge, each dam added about 10 percent to gas saturation
levels. No equilibrium of supersaturated water occurred in Noxon Rapids, Cabinet
Gorge, or the Clark Fork River downstream. Fish bore few scars symptomatic of
gas bubble disease; however, the effects of super saturation on bio- logical
communities were not determined.
CONCLUSION
Development and
maintenance of a sport fishery in these two run-of-the- river reservoirs has not
been successful. Downstream movement by rainbow trout has been attributed to
changing flow regimes, rapid exchange rates, and limited productivity.
PLANS
FOR FUTURE MANAGEMENT
Cooperation
with Washington Water Power Company has established more stable reservoir levels
in recent years (1978-present). Under this reservoir-level regime, species
showing past success will receive the greatest management emphasis as well as
fish with potential in similar reservoir situations. These will include small
mouth and largemouth bass, ling, brown and bull trout, kamloops, and the
McCounaghy strain of rainbow trout that have proven successful in other
fluctuating reservoirs. Bass management for 1985 includes planting 100,000 small
mouth in Noxon Rapids Reservoir. Ling will be collected from the Kootenai River
and reintroduced during 1985. Brown trout eggs are scheduled to be taken from
Willow Creek Reservoir in the fall of 1985. Eggs and fry will be imprinted into
Noxon Rapids Reservoir tributaries in the spring of 1986. Bull trout will also
be planted when an egg source becomes available. Kamloops rainbow are also
scheduled for planting in 1985. Follow up surveys will be necessary to determine
success of the above introductions. Changes and development within the Clark
Fork drainage will also warrant future investigation.
LITERATURE
CITED
1.
Gaffney, 3.3. 1956. A survey of the fishery resources in a section of the Clark
Fork River in western Montana. Progress report, Project 29-E-1, Mont. Fish &
Game, Helena, MT. 12 p., mimeo.
2.
Gaffney, J.J. 1959. Partial rehabilitation of a section of the Clark Fork River.
Progress report, Project 29-E-1, Mont. Fish & Game, Helena, MT. 19 p.,
mimeo.
3.
Huston, J.E. 1965. Investigation of two Clark Fork River hydroelectric
impoundments. Proc. Mon~. Acad. Sci., 25: 20-40.
4.
Huston, J.E. 1968-1983. Noxon Rapids and Cabinet Gorge Reservoirs study. Fed.
Aid Project F-34-R-l through 16, Job 1, Mont. Dept Fish, Wildlife and Parks.
Mimeo reports.
5.
Huston. J.E.; and Vaughn. T. 1968. Temporal movement of rainbow trout in
reservoirs. Proc. West. Assoc. State Game and Fish Comm.. Reno, NV.
6.
Pfeiffer, D.E. 1985. Studies of Micropterus species in two Clark Fork
River hydroelectric impoundments in western Montana. Washington Water Power Co.,
Spokane, WA.