AN ABSTRACT OF THE THESIS OF
Jack David Sleeper for the degree of Master of Science in Fisheries Science
presented on September 7. 1993.
Title: Seasonal Changes in Distribution and Abundance of Salmonids and Habitat
Availability in a Coastal Oregon Basin
Redacted for Privacy
Abstract approved:
Stanley V. Grego
Visual estimation techniques were used to quantify habitat characteristics,
habitat type (pool, riffle) use and longitudinal distribution of steelhead
(Oncorhynchus mykiss), cutthroat trout (0. clarki), and coho salmon (0. kisutch)
in spring, summer and fall in 8.8 km of Cummins Creek, a basin in the central coast
of Oregon. Fish were distributed significantly different than habitat type
availability in most samples. Pool habitats contained a disproportionate percent of
the salmonid assemblage and 1+ fish in each sample, and the percentage of fish in
pools increased as flow decreased. In spring, coho salmon fry were concentrated
in side channels and valley floor tributary habitats. Large woody debris formed 5768% of pool habitats and was significantly correlated with pool volume, maximum
pool depth, slow surface velocity in pools, and pieces of small woody debris.
Longitudinal distribution of the salmonid assemblage did not differ from
habitat distribution seasonally or between years, even though certain species
differed Coho salmon and cutthroat trout were distributed in proportion to
longitudinal habitat availability only when fish abundance was relatively high and
streatnflow was low. In most samples, both 0+ and 1+ steelhead were distributed
in proportion to longitudinal habitat availability. Differences in coho salmon
abundance between years appeared to influence longitudinal distribution of each
species and age class. Certain reaches had consistent numbers of fish between
years while the number of fish in other reaches varied widely. In most samples,
reaches with highest abundance for steelhead were in the lower basin, cutthroat
trout in the upper basin and coho salmon between the two other species.
Timing of reduction in number of fish varied among species. Fifty-five
percent of 0+ steelhead and 73% of 1+ steelhead lost between August 1988 and
April 1989 were lost between August and October during low flow conditions.
However, only 18% of the losses, for 0+ coho salmon, occurred between August
and October with the remaining losses occurring after October.
This study illustrates that habitat availability is not a good index of fish
distribution when fish abundance is low, and it highlights the importance of habitat
in the lower portions of basins when fish abundance is high. It also demonstrates
that the basin wide distribution of salmonids varies among species, age classes,
seasons, and years and suggests that our understanding of salmonid distribution
and abundance could be greatly enhanced by adopting a basin-wide, community,
and seasonal perspective. In addition, the methods used in this study offer one
way to assess the seasonal distribution and abundance of salmonids in a relatively
quick, inexpensive, and non-destructive manner.
Seasonal Changes in Distribution and Abundance of Salmonids and Habitat
Availability in a Coastal Oregon Basin
by
Jack David Sleeper
A THESIS
submitted to
Oregon State University
in partial fulfillment of
the requirements for the
degree of
Master of Science
Completed September 7, 1993
Commencement June 1994
APPROVED:
Redacted for Privacy
Associate Professor of Fish
s in ch,
e of major
Redacted for Privacy
Head of Department of Fisheries and Wildlife
Redacted for Privacy
Dean of Graduat School
Date thesis is presented
Typed by
September 7, 1993
Jack David Sleeper
ACKNOWLEDGMENTS
I would like to thank Dr. Gordie Reeves, Dr. Jim Sedell, and Dr. Fred
Everest for giving me the opportunity to work on this project and for their
enthusiasm, encouragement, guidance and support throughout all aspects of it. I
owe a special thanks to Dr. Gordie Reeves for reviewing and editing the drafts of
this manuscript. I would also like to thank Dr. Stan Gregory, Dr. Bill Liss, Dr.
Bob Beschta, Dr. Richard Johnston, and Dr. Hiram Li for providing constructive
criticism and serving on my graduate committee. I would like to thank Dr. Bob
Brandon at Eastern Oregon State University for his assistance in my statistical
analysis and interpretation. A special thank you goes to Brad Lovatt, Charlie
Dewberry, Tom Mendenhall, Dave Price, Miranda Raugh, and Darcy Tickner for
collecting the data with exceptional skill and great attitudes, and for stimulating
many discussions that helped me grow in many ways. I would like to thank Miles
Hemstrom for sharing his data on Cummins Creek and Tom Smith for providing
the precipitation data from Tenmile Beach. I would also like to thank Don
Niskanen for sharing his home with me so that I could be closer to the study site
and for sharing his knowledge of the ecosystem. I would like to thank the many
fine people at Oregon State University that helped me weave my way through the
process. Most importantly, I would like to thank my parents, Jack and Tess
Sleeper, for providing unending encouragement and support, and my wife, Robyn
Sleeper, for her assistance, patience, encouragement, and confidence throughout
all aspects of this project. Finally, I would like to thank my son, Joshua Sleeper,
for those brief moments of solitude you gave me so I could complete this
manuscript and the many hours of joy you provided in between. This project was
conducted as part of the Coastal Oregon Productivity Enhancement (COPE)
Program. Financial support was provided by USDA Forest Service Pacific
Northwest Research Station, Corvallis OR.
TABLE OF CONTENTS
Page
INTRODUCTION
1
STUDY SITE
4
METHODS
9
RESULTS
16
Habitat Type Use
Longitudinal Distribution
16
19
Survival
26
DISCUSSION
28
LITERATURE CITED
42
APPENDIX
50
LIST OF FIGURES
Figure
Page
1.
Location of Cummins Creek basin and reach designations.
4
2.
Hydrograph for Big Creek, Oregon and stream temperature
for Cummins Creek with sample dates marked.
8
Abundance of each species, age class and total salmonids
by habitat type and date.
18
4.
Habitat volume by habitat type, reach and date.
22
5.
Longitudinal distribution of the salmonid assemblage in September
1987 and October 1988.
24
Longitudinal distribution of the salmonid assemblage in April 1988
and April 1989.
25
3.
6.
LIST OF TABLES
Table
1.
2.
3.
Page
Sampling dates, percent units snorkeled by habitat type,
and temperature and streamflow.
6
Percent habitat type availability and use by each species,
and associated Chi-square value and significance.
17
Percent longitudinal habitat availability and use by each
species, and associated Chi-square value and significance.
20
LIST OF APPENDIX TABLES
Table
A.1.
A.2.
A.3.
A.4.
A.S.
A.6.
Page
Mean densities of each species and age class sampled by habitat
type and reach in Cummins Creek in September 1987.
50
Mean densities of each species and age class sampled by habitat
type and reach in Cummins Creek in April 1988.
51
Mean densities of each species and age class sampled by habitat
type and reach in Cummins Creek in June 1988.
52
Mean densities of each species and age class sampled by habitat
type and reach in Cummins Creek in August 1988.
53
Mean densities of each species and age class sampled by habitat
type and reach in Cummins Creek in October 1988.
54
Mean densities of each species and age class sampled by habitat
type and reach in Cummins Creek in April 1989.
55
SEASONAL CHANGES IN DISTRIBUTION AND ABUNDANCE OF
SALMONIDS AND HABITAT AVAILABILITY IN A COASTAL OREGON
BASIN
INTRODUCTION
Distribution of salmonids has primarily been investigated on the scale of
habitat types (e.g., pool, riffle, etc.). Particular species and age classes have been
found to prefer specific habitat types (Saunders and Gee 1964; Hartman 1965;
Everest and Chapman 1972; Bustard and Narver 1975; Bisson et al. 1982;
Tschaplinski and Hartman 1983; Bisson et al. 1988; Murphy et al. 1989; Hicks
1990; Schwartz 1991; Frissell 1992; Nickelson et al. 1992). However, habitat type
use by salmonids has been observed to change seasonally as fish grow (Lister and
Genoe 1970; Everest and Chapman 1972) and habitat conditions change (Bjornn
and Mallet 1964; Saunders and Gee 1964; Skeesick 1970; Bjornn 1971; Elliott and
Reed 1974; Cederholm and Scarlett 1982; Peterson 1982; Hartman and Brown
1987; Hillman et al. 1987). In addition, interactions among species (Hartman
1965; Hartman and Gill 1968; Glova 1984, 1986, 1987) and age classes (Bohlin
1978) have also been shown to influence habitat type use. Few studies have
considered the effect that seasonal habitat type use and interactions among species
and age classes may have on the longitudinal distribution of fish.
Most longitudinal distribution studies have focused on differences in
species diversity and abundance (Hartman 1965; Sheldon 1968; Evans and Noble
1979; Hawkins and Sedell 1981; Solomon 1982; Culp and Davies 1982; Hughes
and Gammon 1987). Some studies have related longitudinal distribution patterns
of salmonids to the distribution of spawning habitat (Hartman and Gill 1968),
2
habitat types (Bisson and Sedell 1984; Bilby and Bisson 1987; Schwartz 1991),
food availability (Hawkins et al. 1983; Bisson and Sedell 1984; Wilzbach 1985;
Bilby and Bisson 1987), and feeding efficiency (Wilzbach and Hall 1985). In most
of these studies, the small number of sample sites and the large distances between
them have prevented the researchers from determining where these differences
begin and end, or how they are related to the basin wide distribution of the fish
assemblage and habitat characteristics. The few studies that have assessed fish
abundance in contiguous reaches have found that abrupt changes in species
abundance and composition can occur over relatively short distances (Cederholm
and Scarlett 1982; Newman and Waters 1989; Schwartz 1991; Decker and Erman
1992). These patterns appear relatively consistent between years and seem to be
related to habitat characteristics and life history patterns.
Seasonal changes in longitudinal fish distribution are well recognized. Both
juvenile and adult salmonids have been observed to undergo extensive longitudinal
migrations (Bjornn and Mallet 1964; Peterson 1982; Meyers et al. 1992). Changes
in longitudinal distribution occur as adults seek suitable spawning areas (Bjornn
and Mallet 1964; Johnston 1982; Trotter 1989), fry disperse from spawning areas
and feeding territories are established (Chapman 1962), and habitat conditions
change (Bjornn and Mallet 1964; Bjornn 1971; Bilby and Bisson 1987; Meyers et
al. 1992). Few studies have assessed how seasonal migration patterns influence
the longitudinal distribution of the assemblage as a whole or have related seasonal
patterns in longitudinal fish distribution with seasonal habitat characteristics.
The objective of this study is to determine seasonal abundance and
distribution of juvenile salmonids in a nearly-pristine coastal Oregon basin, and to
relate fish distribution to habitat availability and characteristics. This study is
3
intended to improve our understanding of the seasonal dynamics of juvenile
salmonid habitat use relative to habitat availability and to identify factors that
influence the distribution of salmonids. In addition, the methods used in this study
could be used in other basins to identify seasonally important habitat areas, assess
limiting factors, and locate areas of greatest potential to increase fish production in
a relatively quick, inexpensive and non-destructive manner.
4
STUDY SITE
Cummins Creek is located about 48 km south of Newport, Oregon on the
central coast (Fig. 1). This 14 km2 basin is in the volcanic geology of the Yachats
Basalt Formation and drains directly into the Pacific Ocean. The elevation ranges
from 754 meters at Cummins Peak to sea level at the mouth. Topography is steep
with side slopes averaging 25-35% in the lower basin and 15-25% in the upper
basin. The gradient of Cummins Creek averages 2.5% in the lower 7 km and then
gradually increases to 4.5% in the upper sections of the study site (M. Hemstrom,
USDA Forest Service, P.O. Box 25127, Lakewood, CO 80225, unpubl. data).
Pacific
Ocean
Fig. 1. Location of Cummins Creek basin and reach designations.
5
Most major tributaries are located in the upper basin, except for Little
Cummins Creek. Only the lower sections of a few upper basin tributaries support
fish, primarily steelhead and cutthroat trout, because upper sections are too steep.
Little Cummins Creek is the only major tributary that supports the same salmonids
as Cummins Creek and it enters Cummins Creek within a few meters of the ocean.
Only the main channel of Cummins Creek and its associated floodplain habitats
were sampled for this study.
Several of the tributaries have experienced debris torrents that delivered
large amounts of large woody debris (LWD) to Cummins Creek. The largest
deposits are located at 4.2, 5.0, and 8.8 km upstream from the mouth. The
Cummins Creek basin is mostly pristine and less than 1% of the basin had timber
harvest activity. Sitka spruce (Picea sitchensis), western hemlock (Tsuga
heterophylla), Douglas fir (Pseudotsuga menziesii), red alder (Alnus rubra) and
big-leaf maple (Acer macrophyllum) dominate the overstory while salmonberry
(Rubus spectabilis), swordfern (Polystichum munitum), and salal (Gaultheria
shallon) dominate the understory.
Average stream temperature and discharge for each sampling period and
dischage between periods are presented in Table 1. Stream temperatures are mild
throughout the year, rarely going below 7°C in winter or above 16°C in the
summer (Fig. 2). Precipitation in the Cummins Creek basin falls as rain, primarily
between November and April. Precipitation during the study period averaged 200
cm/yr at nearby Tenmile Beach (T. Smith, 95295 Highway 101, Yachats, OR
97498, unpubl. data), which is approximately 6 km south of Cummins Creek. A
hydrograph for Big Creek, a 31 km2 basin about 10 km south of Cummins Creek,
is included to present the relative flow conditions that Cummins Creek experienced
6
during the study period (Fig. 2). Big Creek flow is highly correlated with
Cummins Creek stage height (r = 0.95, 9-df, P < 0.01) and volume (r = 0.97, 3-df,
P = 0.03). Estimated streamflow in Cummins Creek ranged from 0.1 to 13.5 m3/s
during the 19 month sampling period. Streamflow was highest and most variable
from November to March. Lowest flows usually occurred in early fall.
Substrate throughout Cummins Creek is primarily cobble (8-15 cm),
intermixed with gravel (2-8 cm). A few small boulders (>15 cm) can be found in
the upper basin and at tributary junctions. Large concentrations of spawning
gravels are found above debris deposits at 4.2 and 5.0 km, and smaller patches are
found in pool tail-outs throughout the basin.
The fish assemblage in Cummins Creek consists of steelhead trout
(Oncorhynchus mykiss), cutthroat trout (0. clarki), coho salmon (0. kisutch),
sculpins (Coitus spp.), and Pacific Lamprey (Lampetra tridentata). Three
Columbia River smelt (Thaleichthys pacificus) were captured in a smolt trap from
10-23 April 1988, yet none were observed during the sample periods.
Table 1.
Sampling dates, percent units snorkeled by habitat type, average stream
temperature, and estimated streamflow for Cummins Creek, Oregon.
Sampling Period
Sept. 28 - Oct. 5, 1987
April 11-13, 1988
June 8-9, 1988
August 15-18, 1988
October 21-22, 1988
April 13-16, 1989
P=Pool,
Surveyed
Habitat Fish
x
x
x
x
x
x
x
x
x
x
% Units Snorkeled
Average
G
R
SC VFT Temp °C
20
12
12
0
0
P
33
33
33
33
33
12
12
12
12
20
5
5
5
5
5
0
50
50
50
50
0
50
50
50
50
R=Riffle, SC=Side Channel, VFT=Valley Floor Tributary
10
11
Estimated Streamflow (m3/s)
Between Between
Survey Surveys
Surveys
Average
a e Average
Peak
0.1
1.6
1.7
12
0.2
0.2
11
1.1
14
1.8
1.6
0.5
0.2
3.1
11.3
3.2
1.6
0.5
13.5
7
The Cummins Basin supports a variety of wildlife species. Species
observed during the sampling period include black-tailed deer (Odocoileus
hemionus hemionus), Roosevelt elk (Cervus canadensis), beaver (Castor
canadensis), mink (Mustela vison), river otter (Lutra canadensis), coyote (Canis
latrans), black bear (Ursus americanus), great blue heron (Ardea herodias), green
heron (Butorides striatus), and rough-skinned newt (Taricha granulosa). Beaver
had a strong influence on fish habitat availability and quality (Leidholt-Bruner et al.
1992), particularly in fall (September and October) when beaver dams were most
extensive in the main channel. Most beaver dams in Cummins Creek were washed
out by the first major storms in November, except those that were out of the main
channel on the floodplain which often remained intact.
8
40
20
7 35
-15
a.)
30
-10
25
5
ro
a)
E-
20
-0
ro
15
10
C3
.z..7
5
O
Oct
Fig. 2.
Feb
June
1987
Oct
Feb
June
1988
Oct
Feb
June
1989
Hydrograph for Big Creek, Oregon (USGS station number 14306900)
and stream temperature for Cummins Creek with sample dates marked as
solid squares.
9
METHODS
Habitat and fish surveys were conducted in September 1987, April and
August 1988, and April 1989 (Table 1). In June and October 1988, fish
abundance was surveyed but the habitat was not. For both June and October
1988, the habitat was assumed to be identical to the previous sample (i.e., April
and August 1988, respectively). This assumption is based on similar stage height
of Cummins Creek in April and June, and in August and October; and it is
supported by the high correlation between stage height and habitat volume (r =
0.99, 2-df, P = 0.07). In addition, United States Geological Survey gaged
streamflow in nearby Big Creek basin (31 km2) had similar streamflow during the
sample periods in April and June, and in August and October.
Habitat characteristics were determined using methods of Hankin and
Reeves (1988). Habitat units were classified as pool, glide, riffle, side channel
(Bisson et al. 1982), or valley floor tributary. Valley floor tributaries were the
low-gradient portions of tributaries that were located along the floodplain of
Cummins Creek; in general, they resembled side channels. Split flows, areas where
the main channel of the stream split into two or more channels, were considered
main channel habitats and were separated into pool, glide, or riffle habitat types.
In summer when very little flow went through one side of a split flow it was
classified as a side channel.
During each habitat survey, average length, width, and depth of a given
habitat type were visually estimated. Dimensions of each habitat unit in the basin
was estimated and recorded, beginning downstream. To correct for estimator bias,
10
dimensions in approximately 5% of the units were measured to develop correction
factors for each habitat type (Hankin and Reeves 1988).
Habitat characteristics that may influence fish distribution were also
recorded during habitat surveys. The two dimensional (flat-plane) area of LWD
accumulations was estimated for main channel units in August 1988 and April
1989. Pieces of small woody debris (<1 m long and <8 cm diameter), which add
habitat complexity and cover, was estimated for each unit using three abundance
classes (5-15; 16-25; >25 pcs) in April 1989. Only the portion of wood that
provided potential fish cover at the flow level when sampled was included in these
estimates. Maximum pool depth was measured for all pools in August 1988, but
was too deep to be measured at higher flows in April and June. Since low velocity
areas are scarce during moderate to high streamflow and velocity has been shown
to influence fish distribution (e.g., Bustard and Narver 1975), a visual estimate of
the pool area that had a surface velocity <0.5 m/s was taken at all pools in April
1989.
Fish numbers were sampled in systematically selected habitat units (Table
1). Divers using mask and snorkel counted the number of fish in each sampled unit
by beginning downstream and proceeding slowly upstream. Areas providing cover
were carefully searched to locate hiding fish. Two divers were used in larger units.
Divers partitioned the unit, each counting fish in part of the unit. The estimate for
the unit was the sum of the individual diver's estimates. In small units, a single
diver made the counts. All fish counts were made between 0900 and 1600 hrs
when visibility was good, averaging approximately 4 m.
Fish were classified by species and size-class (estimated age-class). Species
identified and counted included 0+ and 1+ steelhead, cutthroat trout, and coho
11
salmon. It was assumed that both coho salmon and steelhead changed age classes
just prior to the April sampling period. Coho salmon and steelhead that were
expected to migrate to the ocean as smolts from April through June were classified
as pre-smolts during those sampling periods. Age 0+ steelhead and cutthroat trout
could not be differentiated and thus were lumped together and classified as 0+
steelhead. It was assumed that most 0+ trout were steelhead, since 1+ steelhead
were 2 to 8 times more abundant than 1+ cutthroat trout in all sampling periods.
In addition, assuming all 0+ trout were steelhead facilitates data interpretation
when assessing seasonal changes in abundance and distribution. The length where
0+ and 1+ steelhead were separated varied seasonally and gradually increased as
fish grew.
Divers used incorrect length classes to separate 1+ and pre-smolt steelhead
in April 1988. The data for these fish was adjusted by assuming that only 87 1+
steelhead were lost from April to June 1988 and then back calculating the April
numbers from the June data.
In September 1987 and April 1988 only cutthroat trout larger than 20 cm
in length were classified as cutthroat trout with smaller 1+ cutthroat trout
considered 1+ steelhead. However, it was determined during the first two
sampling periods that cutthroat trout between 9-20 cm could be distinguished from
steelhead, so from June 1988 through April 1989, both small (9-20 cm) and large
(>20 cm) 1+ cutthroat trout were identified.
Since the relationship between diver counts and the actual number of fish
present was not known, the estimated populations are relative estimates, not
absolute. It was assumed that a constant fraction of each species and age class
were observed in each habitat type during each sampling period. Hankin and
12
Reeves (1988) found strong correlations (r > 0.94) between diver counts and
electrofishing estimates of abundance for 1+ steelhead and coho salmon in riffles
and coho salmon in pools in Cummins Creek in July 1985. The relationship was
not as strong (r = 0.61) for 1+ steelhead in pools. They estimated ratios of
electrofishing to diver counts between 0.97 and 1.05 for 1+ steelhead and coho
salmon in pools and 1+ steelhead in riffles, and 1.36 for coho salmon in riffles.
Thus, diver counts and electrofishing produced similar estimates of fish abundance.
The same divers were used, when possible, to minimize variance of
estimates of fish abundance among sampling periods. Three people were used to
estimate fish abundance during the study. The same two individuals made all
estimates of fish abundance from June 1988 through April 1989.
Habitat availability and fish population estimates were summarized for
eight contiguous reaches, each measuring 1.1 kilometer of stream length (Fig. 1).
Equal length reaches were used to compare habitat and fish abundance among
reaches.
The fish sampling scheme was designed to concentrate sampling effort in
pool habitats, since Hankin and Reeves (1988) found that fish were concentrated in
them. For pool habitats, a mean relative density of a given species or age class
was calculated for each reach. That density was then multiplied by the total
available pool habitat within that reach to obtain a population estimate in pools
within that reach.
Fish abundance in pools/reach:
Example for reach x:
Nprx = (Dprx)(Hprx)
13
where,
Nprx = estimated number of fish in pools in reach x
Dpi = mean relative density of fish in pools in reach x
Hp
total pool habitat in reach x
Since glides, riffles, side channels, and valley floor tributaries were not
sampled frequently enough to obtain mean fish densities in each reach, a mean
relative density was calculated for the lower (reaches 1-4) and upper (reaches 5-8)
halves of the basin. Estimating half basin densities will act to lessen reach level
differences in fish abundance if they existed. Each density was then multiplied by
the total available habitat of that particular type for each reach. Summing
population estimates from each habitat type in each reach produced reach level
estimates of fish abundance. Within a given habitat type and reach, it was assumed
that fish densities observed in units sampled was characteristic of all units in that
reach (or half basin depending on the habitat type).
Fish abundance in glides, riffles, side channels, valley floor tributaries/reach
Example for glides in reach x:
Ngrx
(Dgrx) (Hgrx)
where,
Ng
estimated number of fish in glides in reach x
D gnS = for reaches 1-4 or 5-8
mean relative density of fish in glides in reaches 1-4 or 5-8
Hgrx = total glide habitat in reach x
14
Total fish in reach x:
Nrx = Nprx 4- Ngrx + Nrrx + NSCDC + Nvftrx
Total fish in basin:
Nb = Nrl ± Nr2
+ Nr8
where,
Nrl = estimated number of fish in reach 1
Nb = estimated number of fish in the basin
Coefficient of variation for each species and age class was used to
determine which measure of unit size should be used to estimate fish abundance.
In general, variation was lowest with area in April through June samples and with
volume in August through October samples. For this reason, fish abundance was
estimated by area in April and June 1988 and April 1989, and by volume in
September 1987 and August and October 1988.
Reaches 1 through 7 were sampled during each sampling period. All of
reach 8 was sampled in the summer and fall of both years. However, only the
lower half was sampled in April and June 1988, and the lower third sampled in
April 1989 because of time constraints and presence of few fish. Fish and habitat
abundance was calculated for only that part of reach 8 that was surveyed.
A chi-square goodness of fit statistic was used to determine if fish were
distributed in proportion to habitat, both longitudinally and by habitat type. The
chi-square test compared the percent of fish in a given habitat type or reach with
the percent of habitat. An independent Student's t-test was used to determine
differences in habitat characteristics between the upper and lower halves of the
15
basin. All habitat and fish data were log transformed to meet the assumption of
normality when calculating probabilities.
16
RESULTS
Habitat Type Use
In each sampling period except August 1988, the salmonid assemblage
used habitat types significantly different than the availability of habitat types (P <
0.01; Table 2). Age 0+ steelhead was the only species or age class to use habitat
types in proportion to their availability in some sampling periods (P > 0.1; Table
2), and to use riffle habitats in greater proportion than riffle availability (Table 2).
Disproportionate use of riffle habitats by 0+ steelhead occurred in August 1988,
when fish abundance was highest and streamflow was near its annual low.
Pool habitats contained a disproportionate percent of the salmonid
assemblage and older age classes (1+ steelhead, 1+ cutthroat trout, steelhead presmolts and coho salmon pre-smolts) in each sampling period (Table 2). More than
79% of 1+ cutthroat trout (>10 cm) were located in pools in each sampling period.
The proportion of salmonids located in pool habitats was highest during low flow
periods in September 1987 and August and October 1988. As low flow conditions
persisted from August to October 1988, the assemblage abundance decreased 9%
in pool habitats and 55% in riffle habitats (Fig. 3H).
Side channels and valley floor tributaries contained a disproportionate
number of coho salmon fry in June 1988 and April 1989 (Table 2). During these
samples, coho salmon fry densities in these habitats were 8 to 31 times greater than
any main channel habitat. Although these habitats contributed only 5% of the total
habitat available in April and June, 20-60% of the total coho salmon fry in the
basin were found in them (Table 2). The majority of fish found in floodplain
habitats were concentrated in off -channel pools. Floodplain habitats provide small
fry with persistent slow velocity areas during spring when main channel flow
17
Table 2.
Percent habitat type availability and use by each species, and associated
Chi-square value and significance. Habitat type availability in
September 1987 and August and October 1988 is based on volume, and
in April and June 1988 and April 1989 is based on area. It is assumed
that all 0+ fish become 1+ or Pre-smolts in April.
Steelhead
Date
Sept 87
Hab Type
Pool
Glide
Riffle
Side Chan
VF Trib
Habitat
0+
a-
54
25
60
20
20
70
20
10
-
-
21
X2
1.71
0.42
Prob
April 88
Pool
Glide
Riffle
Side Chan
VF Trib
24
22
13
60
3
2
X2
Pool
Glide
Riffle
Side Chan
NT Trib
46.74
<0.01
11.93
<0.01
8
84
16
36
-
0
8
0
59
18
23
91.78
<0.01
41.35
<0.01
181.81
212.29
76.72
<0.01
4.01
49
92
8
0
0
0
60
16
4
10
10
100
21
166.93
<0.01
30
0
0
-
4.01
82
18
79
21
0
0
0
0
0
0
354.55
230.56
67.57
<11.01
217.61
<0.01
226.00
<0.01
<0.01
4.01
86
6
80
6
13
64
1
2
81
19
0
0
0
51
16
23
5
5
43
75
13
0
85
9
2
4
73
7
19
3
7
18
0
1
8
0
13.06
<0.01
38.14
<0.01
9.56
0.02
33.12
<0.01
20.11
<0.01
1.82
0.61
72
12
16
92
8
79
0
10
0
0
0
76
8
14
2
12.61
<0.01
52.33
23.15
4.01
4.01
14.08
<0.01
83
14
3
94
6
86
57
12
2
0
0
15
15
9
4
140.81
<0.01
57.47
44
58
56
10
12
82
10
29
32
8
3
0
0
3.78
0.29
28.14
Prob
Prob
70
18
12
58
10
29
X2
X2
88
7
5
58.06
<0.01
15.54
Pool
Glide
Riffle
Side Chan
VF Trib
-
Assemblage
8
12
52
4
3
<11.01
April 89
-
Sahnonid
>10 cm
76
24
29
Prob
Pool
Glide
Riffle
Side Chan
VF Trib
8
8
8
51
13
60.94
<0.01
X2
Oct 88
10-20 cm >20 cm
85
15
0
4.20
0.38
Prob
Aug 88
Cutthroat Trout
0+ Pre-smolt
44.51
<0.01
62
19
19
Prob
Pool
Glide
Riffle
Side Chan
VF Trib
13.79
<0.01
55
18
27
-
14
62
-
X2
June 88
Coho Salmon
Ere -smolt
32
15
48
4
1
289.65
<0.01
"
93
3
0
4
55.35
<0.01
<0.01
47
20
33
60
14
26
0
0
0
0
10
4
41
19
18.39
<0.01
39.65
<0.01
709.38
<0.01
-not sampled; not present; Fcombined with steelhead
26
83
14
0
0
0
0
0
0
129.37
<0.01
128.54
<0.01
178.53
<0.01
1
2
11
10
24
18
Steelhead -86 year class
Coho-87 year class
Steelhead -87 year class
Coho-88 year class
8000
Sept 67 Apr S8 June SS Aug 00 Opt 86 Apr 06
Slope 67 Apr 00 Jun. ea Aug 00 OW 00 Apr 00
Steelhead-88 year class
Coho-89 year class
6000
x
6000 '-'
..5
li, 4000 -'
1 2000Sopit 87 Ap 88 Jurs 88 Aug ea OW 88
Cutthroat >1-I- old
Apr 70
0
Sow 87 Apr 00 June ea Aug 08 Oat ea Apr 80
Total Salmonids
good 87 Apr 66 Juno 06 Aug 88 Oat 06 Apr SO
EI Pool IIIGlide El Riffle El Side Channel 0 Valley Floor Tributary
Fig. 3.
Abundance of each species, age class and total salmonids by habitat type
and date.
19
fluctuations are common. Abundance of coho salmon fry in floodplain habitats
decreased as streamflow decreased from June to August, such that these habitats
were only used in proportion to their availability by August (Table 2).
Large woody debris was instrumental in forming and maintaining pool
habitats and floodplain habitats in Cummins Creek. LWD formed 57-68% of pool
habitats in August 1988 and April 1989, and it was positively correlated with pool
volume in August 1988 and April 1989 (r = 0.38 and 0.27,159-df and 140-df, P <
0.01), maximum pool depth in August 1988 (r = 0.44,155-df, P < 0.01), and slow
surface velocity (< 0.5 m/s) area in pools in April 1989 (r = 0.25,131-df, P < 0.01).
LWD was also positively correlated with pieces of small woody debris (r =
0.40,31-df, P = 0.02) in April 1989 which added habitat complexity and cover.
Pieces of LWD at the head of side channels was responsible for diverting flow into
many of these habitats. In addition, pools within floodplain habitats were often
formed and maintained by pieces of LWD or beaver dams. Beaver dams also
increased pool area and volume in the main channel in late summer.
Longitudinal Distribution
Although each species was distributed throughout the basin, there was
a general trend in most sampling periods where reaches with highest abundance for
steelhead were in the lower basin, cutthroat trout in the upper basin and coho
salmon between the two other species (Table 3). Reaches with greatest fish
abundance varied seasonally and between years but were often located where
transitions between species and age classes occurred.
Longitudinal distribution of the salmonid assemblage did not differ from
habitat distribution seasonally or between years (P > 0.10 Table 3). However,
longitudinal distribution of individual species and age-classes varied from habitat
20
Table 3.
Percent longitudinal habitat availability and use by each species and
associated Chi-square value and significance. Habitat availability in
September 1987 and August and October 1988 is based on volume, and
in April and June 1988 and April 1989 is based on area. It is assumed
that all 0+ fish become 1+ or Pre-smolts in April.
Steelhead
Date
Reach
Habitat
0+
1± Pre-smolt
1
10
13
14
16
13
12
12
10
12
13
8
12
16
21
16
8
Sept 87
2
3
4
5
6
7
8
X2
Prob
April 88
20
23
13
6
9
4
13.38
0.06
6
7
12
14
18
14
12
13
12
8
5
1
2
3
4
5
June 88
1
2
3
4
S
6
7
8
12
14
17
14
13
13
12
5
X2
Prob
August 88
7.18
0.41
29.84
<0.01
15
14
14
27
21
20
8
8
8
8
8
8
8
8
#
33
12
15
18
7
6
3
0
8
8
8
8
8
8
8
14
7
16
4
16
10
0
13
18
19
13
13
12
9
3
55.12
<0.01
3.06
0.88
1
5.66
0.58
37.85
<0.01
56.83
<0.01
57.36
<0.01
12
11
15
12
32
7
9
13
12
0
5
100
0
9
16
20
3
14.59
0.04
15
15
6
12
5
2.55
0.92
4.02
0.78
51.52
<0.01
15
13
21
17
18
17
1
18
18
3
9
7
29
27
19
0
18
7
19
13
19
9
614.29
<0.01
54.45
<0.01
75.87
<0.01
51.80
<0.01
7.60
0.37
8
8
14
12
18
10
15
17
17
10
10
15
15
13
10
13
9
9
5
16.95
0.02
11.74
0.11
2.10
0.95
4
8
9
0
11
7
8
9
6
8
8
8
11
4
2
6.68
0.46
4.52
0.72
11.46
0.12
13.78
0.06
18
8
23
22
19
14
16
10
10
8
5
21
17
2
3
4
5
6
7
8
15
17
15
14
13
15
17
14
18
12
11
7
9
6
9
12
21
8
8
7
8
3
3.41
0.85
11.50
0.12
9
12
19
17
19
11
10
1
10.86
0.15
10
10
13
12
18
16
17
4
6
1
9
6
2
19
5
X2
Prob
18
0
0
33
16
0
0
0
8
17
10
15
12
12
17
9
11
8
8.35
0.3
14
23
24
22
9
7
19
15
3
52.80
<0.01
34
37
15
6
6
4
11
18
12
2
8
13
18
19
17
8
13
4
12
7
12
7
15
12
10
7
3
6
7
3
Salmonid
Assemblage
2
23
15
16
9
6
3
4
X2
Prob
11
11
Cutthroat Trout
10-20 cm >20 cm >10 cm
15
17
15
14
13
1
2
Oct 88
10
24
21
3
23
"
2
14
15
10
14
5
14
16
X2
Prob
Coho Salmon
0+ Pre-smolt,
"
"
11
10
15
17
19
11
5
26
8
5
31
14
4
26.30
<0.01
71.87
<0.01
20
11
5
23
13
24
12
6
35.85
<0.01
4
12
20
16
17
13
9
9
4
2.87
0.90
21
Table 3 continued...
April 89
1
2
3
4
10
16
20
16
14
14
15
15
14
20
23
23
6
5
5
17
11
32
4
4
15
17
17
18
28
14
7
5
11
8
5
4
4
13
6
19
41
5
7
6
4
13
7
10
15
15
10
12
8
3
4
1
3
0
4
12
0
27
16
24
3
3.87
0.8
16.14
0.02
49.97
<0.01
16.32
0.02
61.95
<0.01
73.27
<0.01
54.46
<0.01
5
6
11
X2
Prob
11
22
19
27
11
14
14
20
18
10
11
2
4.82
0.68
not present; ireornblned with steelhead
distribution in some sampling periods. Coho salmon were distributed significantly
different than habitat in each sampling period except August and October 1988 (P
< 0.05; Table 3). Coho salmon fry were concentrated in reaches 5-7 in April of
both years and in September 1987 (Table 3). Reach 5 contains two large debris
torrent deposits that have accumulated the largest concentration of spawning
gravel in the basin and has created relatively open canopies and abundant
floodplain habitats. Similar conditions exist in reach 7, although to a lesser extent,
where three large tributaries enter Cummins Creek.
Age 1+ steelhead were distributed in proportion to the longitudinal
distribution of habitat in all sampling periods (P > 0.10; Table 3), except as pre-
smolts in April and June 1988 and April 1989. Both steelhead pre-smolts and
coho salmon pre-smolts were most abundant in pool habitats in the lower reaches
in spring of both years (Table 2, 3). Mean pool area and volume were significantly
larger in reaches 1-4 than in reaches 5-8 (P < 0.01) in April of both years; and
reaches 1-4 contained 65-72% of the total pool volume (Fig. 4). In April 1989,
amount of pool area with surface velocities less than 0.5 m/s was strongly
correlated with total pool area (r = 0.73; 183-df, P < 0.01), and approximately
70% of the basin wide slow velocity area in pools was located in the reaches 1-4.
22
September 1987
April 1988
4
0
E
2
3
4
5
6
4
8
Reach
5
Reach
August 1988
April 1989
4
4
5
6
Reach
Pool
Glide
7
8
2
4
5
Reach
Riffle N Side Channel 0 Valley Floor Tributary
Fig. 4. Habitat volume by habitat type, reach, and date. Reach 8 was only
partially sampled in April 1988 and April 1989.
e
23
s
Both coho salmon pre-smolts and steelhead pre-smolts were significantly
correlated with the area of slow surface velocity (<0.5 m/s) in pools in April 1989
(r = 0.66 and r = 0.57,45-df and 47-df, P < 0.01), and pool volume in April 1988
and April 1989 (r > 0.67,>24-df P < 0.01).
Age 0+ steelhead were distributed in proportion to longitudinal habitat in
all sampling periods (P > 0.10, Table 3), except in September 1987. At this time,
they were concentrated in reaches 3 and 4, which were just downstream from areas
of high coho salmon abundance.
Cutthroat trout were distributed significantly different than longitudinal
habitat in each sampling period (P < 0.10; Table 3), except in August 1988. In all
sampling periods, except April 1988, cutthroat trout were most abundant in upper
reaches, typically just upstream from areas of high coho salmon abundance (Table
3).
Fish abundance appeared to have a strong influence on longitudinal
distribution. Total salmonid abundance was 50 to 60% greater in 1988-89
compared with 1987-88, primarily because of greater coho salmon numbers (Fig.
3B, 3D, 3H). When abundance was high, fish were distributed more widely than
when abundance was low. The relatively high fish abundance in August and
October 1988 resulted in large increases in fish abundance in downstream reaches
while upstream reaches had similar numbers between years (Fig. 5). A similar, but
opposite, pattern was observed in spring. Higher abundance in April 1989
expanded fish distribution upstream when compared to April 1988. There was a
large increase in fish abundance in upstream reaches while downstream reaches
had similar numbers between years (Fig. 6). The result was that certain reaches
24
had consistent numbers of fish between years while the number of fish in other
reaches varied widely.
2500
2000
N 1500
/
'45
a)
0
/
=
1000
z
/
\
\
(
'/
/
/
/
/
\
\
500
1
2
3
4
5
6
7
8
Reach
September 1 987
Fig. 5.
October 1988
Longitudinal distribution of the salmonid assemblage in September 1987
and October 1988.
25
1
2
3
4
5
6
7
8
Reach
April 1988
Fig. 6.
April 1989
Longitudinal distribution of the salmonid assemblage in April 1988 and
April 1989. Reach 8 was only partially sampled in April 1988 and 1989.
26
Survival
The timing of losses of 0+ steelhead, 1+ steelhead and 0+ coho salmon
varied among species. Fifty-five percent of 0+ steelhead and 73% of 1+ steelhead
lost between August 1988 and April 1989 were lost during low flow conditions
between August and October (Fig. 3C, 3E). However, for 0+ coho salmon, only
18% of the losses occurred between August and October with the remaining losses
occurring after October (Fig. 3D). Fall reductions in fish abundance were most
apparent in riffle habitats in the lower basin for 0+ steelhead, in riffle and pool
habitats basin-wide for 1+ steelhead, and in glide habitats in all reaches except for
reach 2 where a 67% increase was observed for 0+ coho salmon (Fig. 3D, Table
3). Hiding behavior, due to low stream temperatures, is not believed to account
for the fall losses of salmonids since temperature only decreased from 14°C in
August to 12°C in October (Table 1; Fig. 2).
Overwinter survival of 1+ steelhead and 0+ coho salmon was higher when
fish abundance was higher. From August 1988 to April 1989, 44% of 1+
steelhead and 46% of 0+ coho salmon survived, whereas from September 1987 to
April 1988, 39% of 1+ steelhead and 31% of 0+ coho salmon survived.
The abundance of certain species and age-classes was similar between
years in the same season. In April 1988, the estimated number of 0+ steelhead was
2998 and in 1989, it was 3131 (Fig. 3C, 3E). The estimated number of large
cutthroat trout (>20 cm) was consistently higher in late summer (202-267) than in
spring (121-138) of both years. Population estimates for 0+ steelhead and 1+
steelhead in September 1987 were lower than the estimates in August 1988 and
higher, or about equal to that for 1+ steelhead, than the estimates in October 1988
(Fig. 3A, 3C, 3E). This pattern was also observed for large cutthroat trout (>20
27
cm). This suggests a possible decreasing trend with consistent numbers of these
fish in fall in both years.
28
DISCUSSION
A major focus of salmonid ecology is identifying patterns in species
distribution and abundance and relating these patterns to factors and processes that
influence them. Species specific distribution patterns have been observed at most
spatial scales ranging from microhabitats (Everest and Chapman 1972) to
longitudinal reaches (Huet 1959; Cederholm and Scar lett 1982; Newman and
Waters 1989; Schwartz 1991; Decker and Erman 1992; Frissell 1992). In
Cummins Creek, there was a general longitudinal pattern where reaches with
highest abundance of steelhead were located in the lower basin, cutthroat trout in
the upper basin, and coho salmon between the two species of trout. Schwartz
(1991) found a similar pattern in Drift Creek, Oregon where highest abundance of
chinook salmon were in the lower basin, cutthroat trout in the upper basin, coho
salmon between these species, and steelhead were more evenly distributed.
Temporal shifts in distribution patterns are well recognized (Bjornn and Mallet
1964; Saunders and Gee 1964; Lister and Genoe 1970; Skeesick 1970; Bjornn
1971; Everest and Chapman 1972; Elliott and Reed 1974; Cederholm and Scarlett
1982; Peterson 1982; Bilby and Bisson 1987; Hartman and Brown 1987; Hillman
et al. 1987; Meyers et al. 1992). The relative importance of individual factors that
influence distribution often depend on interrelationships among factors (Hawkins
et al. 1983; Wilzbach 1985; Bilby and Bisson 1987), and temporal changes in life
history characteristics and habitat conditions. Factors that appeared to influence
the seasonal distribution and abundance of salmonids in Cummins Creek during the
study period are discussed below. An attempt was made to relate these factors to
current management and research strategies to improve our understanding and
management of salmonids.
29
Salmonid habitat use in Cummins Creek highlights the importance of pool
habitats for juvenile rearing. Pool habitats consistently contained the majority of
salmonids and became increasingly important during low flow conditions, which is
similar to findings of other investigators (Saunders and Gee 1964; Hartman 1965;
Everest and Chapman 1972; Bustard and Narver 1975; Tschaplinski and Hartman
1983; Bisson et al. 1988; Frissell 1992; Nickelson et al. 1992). For the most part,
salmonid abundance was directly related to pool size. Large pools are more likely
to maintain slow velocity areas, suitable for salmonids, during winter fluctuations
in streamflow when compared with small pools. Nickelson et al. (1992) found that
coho salmon in coastal Oregon streams in winter preferentially selected pool
habitats with low current velocity and little turbulence during freshets, which is
similar to findings of other investigators (Bustard and Narver 1975; Tschaplinski
and Hartman 1983; Baltz et al. 1991).
Use of riffle habitats by 0+ steelhead in Cummins Creek appeared to be
directly related to total fish abundance in summer, when riffles were extremely
shallow. At relatively low total fish abundance (i.e., September 1987) riffle
habitats were not used in proportion to their availability, whereas at higher
abundance (i.e., August 1988) riffles were used in greater proportion than their
availability by 0+ steelhead (Table 2). It is assumed that competition for food and
space was more intense in 1988 due to higher fish abundance, and species
interactions stimulated 0+ salmonids to increase their use of riffle habitats when
compared with 1987. Other investigators have found that riffle habitats are
extensively used by 0+ cutthroat trout and 0+ steelhead in summer (Hartman
1965; Bisson et al. 1982; Bisson and Sedell 1984; Schwartz 1991). In Drift Creek,
Oregon, 0+ trout, 1+ cuttroat trout and 1+ steelhead used riffle habitats in greater
30
proportion than their availability in the mainstem of the basin in summer, but they
were more concentrated in pool and glide habitats in tributaries where the riffles
were shallow (Schwartz 1991). The extent of riffle use seems to depend on riffle
depth (Schwartz 1991), species composition (Hartman 1965; Glova 1984, 1986,
1987), season (Saunders and Gee 1964; Hartman 1965; Glova 1986), and total fish
abundance. As stream temperatures decrease in fall, however, salmonids are often
found in pool habitats (Saunders and Gee 1964; Hartman 1965).
Floodplain habitats, such as side channels and valley floor tributaries, were
found to be important habitats for coho salmon fry in spring. These habitats
provide relatively persistent slow velocity areas for small fry during spring when
main channel flow fluctuations are common, and often, they are slightly warmer
than the main channel which can accelerate growth (Murphy et al. 1989). Use of
side channels and intermittent tributaries by coho salmon fry in spring has been
observed in other basins (Cederholm and Scarlett 1982) but has not highlighted,
perhaps, because the number of fry using off channel habitats in spring was less
than one-quarter of the number using them in fall. Nickelson et al. (1992) found
that coho salmon fry densities in spring were highest in backwater pools which
were characterized by slack water even at high flows. Several researchers have
observed juvenile salmonids immigrating into side channels, sloughs, and small
tributaries with the onset of freshets in fall in northern California (Kralik and
Sowerwine 1977), Oregon (Skeesick 1970; Everest 1973, Everest et al. 1987),
Washington (Cederholm and Scarlett 1982), and British Columbia (Bustard and
Narver 1975; Swales et al. 1986; Hartman and Brown 1987). Fish in these
habitats have relatively high overwinter survival rates. It is likely that floodplains
provide important habitats for all species during high streamflow.
31
The importance of pool and floodplain habitats for juvenile salmonids and
the strong influence that LWD has on their creation, maintenance and quality
(Keller and Tally 1979; Beschta 1989; Sedell and Froggatt 1984; Robison 1987;
Stack and Beschta 1989; Bilby and Ward 1991; Fausch and Northcote 1992)
emphasizes that buffers are not only needed to protect stream channels from
management activities but also should be designed to protect floodplains. By
protecting floodplains, management agencies will not only be protecting much of
the critical interface between terrestrial and aquatic systems but will likely be
protecting future main channel habitats in case the main channel shifts position on
the floodplain.
Longitudinal distribution of fish is initially determined by adult selection of
spawning areas. Although spawning areas were not identified in Cummins Creek,
juvenile distribution may indicate where spawning occurs. Since coho salmon fry
were most abundant in reaches 5-7 in spring of both years (Table 3), it is believed
that most coho salmon spawned in these reaches. Age 1+ cutthroat trout were
most abundant in the headwaters above the study site in spring, which is also the
season when large cutthroat trout (>20 cm) in the study site were least abundant.
This suggests that cutthroat trout may migrate upstream and possibly into
tributaries to spawn. Other investigators have noted that cutthroat trout migrate
upstream and into tributaries to spawn (Bjornn and Mallet 1964; Johnston 1982;
Trotter 1989) and their spawning and rearing areas are often located immediately
upstream from coho salmon and steelhead (Johnston 1982). Although 0+
steelhead were evenly distributed when they began to emerge in June 1988 (Table
3), they were most abundant in the lower basin in late summer (Table 3) of both
years which indicates that steelhead may spawn mostly in the lower river.
32
However, 0+ steelhead may have been forced to disperse into these reaches by
larger (Bohlin 1978) or more aggressive fish (Chapman 1962). In larger coastal
Oregon basins (>180 km2), 0+ trout were most abundant in headwater and
tributary sections (Schwartz 1991; Frissell 1992). Hartman and Gill (1968)
speculated that juvenile steelhead and cutthroat trout distributions in several
stream in southwestern British Columbia were directly related to adult migratory
and spawning behavior. They suggest that adult size, drainage size, and stream
velocity influenced the larger steelhead to spawn in larger basins with faster
velocities than cutthroat trout.
Habitat type preferences by fry and the distribution of those habitat types
probably modify the initial distribution established by spawning adults.
Disproportionate use of floodplain habitats by coho salmon fry in spring and the
concentration of those habitats in reaches 4-7 probably limited dispersal of these
fish from spawning areas. Similarly, the wide range of habitat types used by 0+
steelhead suggest that most reaches probably contained suitable habitat for these
fish.
However, the longitudinal segregation of species and the apparent
influence that coho salmon abundance appeared to have on the distribution of all
species suggest that species interactions may also influence distribution. Several
investigators have demonstrated that juvenile salmonids use habitat differently in
the presence of other species and age classes. Young-of-the-year coho salmon are
believed to displace 0+ cutthroat trout and 0+ steelhead from pools (Glova 1984,
1986, 1987; Hicks 1990), and steelhead appear to dominate cutthroat trout in
riffles (Hartman and Gill 1968). In experimental streams with sympatric coho
salmon and steelhead, Hartman (1965) found that steelhead defend territories in
33
riffles but not in pools and coho salmon did the opposite. Perhaps the longitudinal
segregation of juvenile salmonids in Cummins Creek (Table 3) and in other basins
(Cederholm and Scar lett 1982; Schwartz 1991; Decker and Erman 1992; Frissell
1992) is, in part, the cumulative large scale result of many microhabitat
interactions between species.
The fact that each species had a unique longitudinal distribution has
important implications. Areas of high fish abundance are often species and age
class specific (Cederholm and Scarlett 1982; Newman and Waters 1989; Schwartz
1991; Decker and Erman 1992; Frissell 1992), and it is only through a basin-wide
survey that longitudinal differences in species abundance can be determined.
Quality of the habitat and the assemblage of fish present can influence the
distribution of species. Managers and researchers need to consider the entire fish
assemblage, otherwise they risk altering it to favor one species over another. Like
other factors that influence fish distribution, the assemblage of fish that use similar
habitat must be considered when assessing habitat use by a single species.
The observation that the salmonid assemblage was distributed in proportion
to habitat availability in all sampling periods, even though certain species were not,
suggests that longitudinal segregation of species is a mechanism that allows for full
utilization of available habitat. However, the relatively even distribution of high
quality pool habitat throughout Cummins Creek (Fig. 4) probably allowed the
salmonid assemblage to be distributed in proportion to habitat availability in each
sampling period. If habitat quality and quantity had been more variable among
reaches (i.e., several reaches dominated by riffle habitat), it is likely that the
salmonid assemblage might not have been distributed in proportion to longitudinal
habitat as consistently.
34
This study indicates that species distribution cannot be inferred from the
distribution of habitat when fish abundance is low (Table 3). With low fish
abundance, which is characteristic of many Pacific Northwest streams (Nehisen et
al. 1991), particular reaches may contain the majority of a species individuals while
other reaches, even those with abundant, high quality habitat, contain relatively
few fish. Several investigators have noted that fish dispersal is proportional to
population density (Chapman 1962; Bjornn 1971; Solomon 1982). Close and
Anderson (1992) demonstrated that steelhead fry only disperse from stocking areas
as far as needed to find suitable rearing densities. Current emphasis by
management agencies to survey habitat conditions without assessing fish
distribution and abundance may prevent the agencies from identifying seasonally
important reaches for fish. These areas need to be identified to adequately protect
them from degradation and to improve the effectiveness of stream rehabilitation
efforts.
When fish abundance is high, however, habitat availability becomes a better
index of species distribution (Table 3) if habitat is suitable for their survival and
growth. Since habitat availability and pool size typically increase downstream
(Stack and Beschta 1989), these areas would be expected to provide the majority
of habitat if current efforts to increase salmonid spawning escapement are
successful (Northwest Power Planning Council 1987). However, habitat
conditions in the lower portions of many basins are often not suitable for salmonids
due to habitat degradation by humans. Reduced interactions between stream
channels and their floodplains (Sedell and Froggatt 1984), low pool frequency and
quality (Sedell and Everest 1991), high stream temperatures (Frissell 1992) or
other factors may prevent salmonids from utilizing these areas to their historic
35
capacity. This may limit the expected benefits of increased spawning escapement
until habitat conditions in the lower portions of degraded basins improve.
The observation that salmonids in Cummins Creek are more closely
associated with habitat availability at relatively high fish abundance (Table 2, 3) is
consistent with the Theory of Habitat Distribution of a dispersive organism
(Fretwell and Lucas 1970; Fretwell 1972), which states that populations tend to
become more uniformly distributed as population size increases. This theory
predicts that at low population size all individuals should be in the most suitable
habitat. As density increases in the most suitable habitats, their suitability
decreases such that other habitats become more suitable. This process continues
until all suitable habitats are filled.
If this theory is correct, when abundance is low reaches with high
abundance would be expected to offer the most suitable habitat, if these reaches
were located near sources of fish. In September 1987, when fish abundance was
relatively low, reaches 3-5 had the most salmonids and reach 7 had considerably
more fish than the adjacent reaches (Fig. 5). Reach 5 and 7 both have more open
canopies than the other upper basin reaches (M. Hemstrom, USDA Forest Service,
P.O. Box 25127, Lakewood, CO 80225, unpubl. data). Open canopies allow more
light to reach the stream which has been shown to increase primary production
(Bilby and Bisson 1987), invertebrate and salmonid populations (Hawkins et al.
1983; Bisson and Sedell 1984) and foraging efficiency by salmonids (Wilzbach and
Hall 1985) in streams on the west side of the Cascade mountains in summer.
Holtby and Hartman (1982) concluded that food limited survival and growth of
coho salmon in Carnation Creek, British Columbia.
36
In April 1988, when fish abundance was relatively low, reaches 2 and 3
contained most salmonids (Fig. 6). The relatively abundant large pools with slow
surface velocities may provide the best spring habitat available in Cummins Creek,
particularly for pre-smolts of coho salmon and steelhead. Other investigators have
observed that salmonids are most abundant in pool habitats that are relatively deep
with slow velocities and abundant LWD in winter (Bustard and Narver 1975;
Tschaplinski and Hartman 1983; Baltz et al. 1991; Nickelson 1992). Changes in
the longitudinal distribution of coho salmon from August to October seem to
support that coho salmon immigrate out of the upper basin and into the lower
basin in fall (Table 3). In addition, the disproportionate amount of coho salmon
smolts (36%) in reach 2 in April 1988 (Table 3) indicates that this reach may be
the first to fill with coho salmon seeking suitable winter habitat.
The observation that abundant, high quality habitat appeared to be
underutilized in reaches 1-2 in September 1987 (Fig. 4,5) and in reach 4 in April
1988 (Fig. 4,6) suggests that Cummins Creek was probably not rearing as many
juvenile salmonids in 1987-88 as was possible. This indicates that Oregon's
General Fish Management Goal to manage populations of naturally reproducing
fish so that they take full advantage of the productive capacity of natural habitats
(Oregon Department of Fish and Wildlife 1992) was not met in Cummins Creek in
1987-88. Additional escapement may be needed to fully seed available habitat in
Cummins Creek.
This study documents that the relative distribution of fish, both
longitudinally and by habitat type, in Cummins Creek is perpetually changing.
Many researchers have documented changes in distribution of stream-dwelling
salmonids. Most of these studies focused on major changes in distribution with the
37
emergence of fry in spring (Hartman et al. 1982), the establishment of territories
(Chapman 1962), and the onset of large freshets in fall (Skeesick 1970; Cederholm
and Scarlett 1982). This study indicates that gradual, relatively minor, shifts in
distribution occur continuously and can have a major influence on the basin-wide
distribution of fish.
Some researchers highlight the fact that some stream salmonids maintain a
relatively permanent position in summer with little movement (Saunders and Gee
1964; Edmondson et al. 1968). This conclusion is based on recaptures or
observations of marked fish, yet the percentage of fish recaptured is often low and
decreases through time. Such results indicate that a fraction of fish either do not
maintain a permanent station or change stations regularly through time. Pucket
and Dill (1985) found that a proportion of salmonids are non-station holders.
Gibson (1981) found that steelhead are more likely to hold permanent stations than
are coho salmon. One can infer from the constantly changing distributions in this
study and others (e.g., Decker and Erman 1992) that the fraction of fish without
permanent stations can be significant and can have a major effect on the overall
distribution of fish in a basin. It is likely that fish occupy a preferred position only
temporarily, and then they change positions as fish size (Everest and Chapman
1972; Bisson et al. 1988), assemblage composition (Saunders and Gee 1964;
Hartman 1965), density (Chapman 1962; Bilby and Bisson 1987; Close and
Anderson 1992), and habitat conditions (Hartman 1965; Bustard and Narver 1975;
Mason 1976; Cederholm and Scarlett 1982; Hartman et al. 1982; Tschaplinski and
Hartman 1983; Frissell 1992) change.
The changes in salmonid distribution observed in Cummins Creek
emphasize the limited information that only one annual survey can provide. The
38
data from each survey act to artificially "freeze" the position of salmonids in the
basin, even though the fish distribution may be continually changing. Habitat that
contain the majority of salmonids in one season may contain relatively few fish in
another. For example, many of the floodplain habitats important for coho salmon
fry in spring were dry and overgrown with vegetation during summer and would
not have been identified without spring samples. Results of this study indicate that
seasonal surveys are needed to assess how salmonids use a basin through time, and
to identify seasonally important habitat.
The substantial losses of 0+ steelhead and 1+ steelhead during low flow
conditions in fall 1988 indicates that low flow conditions may limit the production
these fish. Predation by river otters may contribute to these losses since they were
observed in the lower basin during this period and fish are more vulnerable during
low flow conditions (Erlinge 1968). In addition, the relatively large size of 1+
steelhead may make them exceptionally vulnerable due to their relatively low
mobility (Erlinge 1968). Other predators such as heron, mergansers and
kingfishers may also have high feeding efficiency during low flow conditions.
However, the mechanism responsible for losses in Cummins Creek is unknown and
merits further investigation. The reduction in numbers may not indicate mortalities
since hiding behavior (Saunders and Gee 1964; Bustard and Narver 1975; Frissell
1992), ocean migrations (Pemberton 1976), unrepresentative sampling or other
reasons may be responsible for the apparent losses. The observation that relatively
few coho salmon in Cummins Creek were lost during low flow conditions in fall
supports observations by Tschaplinski and Hartman (1983) that overwinter losses
of coho salmon occurred mostly in early autumn with the onset of the first fall
freshets. Whether most coho salmon in Cummins Creek were lost with the first
39
fall freshets or at other times during the winter was not determined, yet higher
streamflow appears to be associated with overwinter losses of coho salmon.
The relatively low overwinter survival of 0+ coho salmon and 1+ steelhead
in 1987-88 could be due to persistent low flow conditions in fall 1987. In 1987,
low flow conditions persisted throughout October, whereas in 1988 flow began to
rise in late September (Fig. 2). The result was that average daily streamflow in
October in 1987 was about half that observed in 1988. Considering the decreases
in salmonid abundance during low flow conditions in fall 1988, the persistent, low
flow conditions in fall 1987 may partially account for the lower overwinter
survival. It is also possible that at relatively high fish abundance, a larger fraction
of fish were closer to high quality overwinter habitat (i.e., lower basin; Fig 5) prior
to the first fall freshets, resulting in less migration distance to overwinter habitat
and higher survival.
Although coho salmon spawning escapement throughout the Oregon Coast
was estimated to be higher in 1986 than in 1987 (Cooney and Jacobs 1993), the
abundance of juveniles in Cummins Creek indicated that the opposite may have
been true for this basin. However, differences in streamflow between years may
also account for the differences in coho salmon abundance. Lower coho salmon
abundance in 1987-88 may have been due to lower egg to fry survival since in
1987 the peak flow was higher (25 vs. 17 m3/s in Big Creek) and later (Feb 1 vs.
Jan 15) than in 1988 (Fig. 2). High flows can scour developing embryos from the
streambed and displace small fry (Elwood and Waters 1969). In addition, the
relatively low average streamflow in February and March 1987 (2.6 m3/s in Big
Creek) compared with 1988 (5.0 m3/s in Big Creek; Fig. 2) may have restricted fry
access to side channels and valley floor tributaries which may result in lower fry
40
survival (Brown and Hartman 1988). Solomon (1982) found a direct relationship
between April streamflow and 0+ brown trout abundance in October.
In summary, the location of successful spawning sites initially determines
the distribution of 0+ salmonids. Distribution is modified as fish seek suitable
habitat for rearing. The extent of dispersal depends on the longitudinal distribution
of suitable habitat and food availability. Fish abundance and existing distribution
of fish also has a major influence on subsequent distribution patterns. At low
abundance much of the suitable habitat may not be utilized to its full potential,
particularly areas most distant from sources of fish. Throughout the freshwater life
history of salmonids other factors operate to modify distribution patterns. These
include changes in habitat requirements as fish grow and habitat conditions change,
species and age class interactions, and mortality factors such as predation or
extreme streamflow and temperature.
Results of this study emphasize that the basin-wide distribution of juvenile
salmonids in Cummins Creek varies among species, age classes, seasons, and
years. Reaches and habitat types that in some seasons contained the bulk of a
given species or age class, had relatively few fish at other times. Species specific
patterns in longitudinal distribution and habitat type use suggest that our
understanding of salmonid distribution and abundance could be greatly enhanced
by adopting a basin-wide and community perspective. Seasonal changes in habitat
use by salmonids in Cummins Creek emphasize the "snap shot" nature of data that
single surveys provide, and suggest that more than one survey annually is needed
to understand the dynamics of habitat use by anadromous salmonids. The
modified Hankin and Reeves (1988) method employed in this study offers one way
41
to assess the seasonal changes in the basin-wide distribution and abundance of fish
and their habitats in a relatively quick, inexpensive, and non-destructive manner.
42
LITERATURE CITED
Baltz, D. M., B. Vondracek, L. R. Brown, and P. B. Moyle. 1991. Seasonal
changes in microhabitat selection by rainbow trout in a small stream. Trans.
Amer. Fish. Soc. 120:166-176.
Beschta, R. L. 1989. The effects of large organic debris upon channel
morphology: a flume study. Water Resources Bull. 25:1031-1036.
Bilby, R. E., and P. A. Bisson. 1987. Emigration and production of hatchery
coho salmon (Oncorhynchus kisutch) stocked in streams draining an oldgrowth and a clear-cut watershed. Can. J. Fish. Aquat. Sci. 44:1397-1407.
Bilby, R. E., and J. W. Ward. 1991. Characteristics and function of large woody
debris in streams draining old-growth, clear-cut, and second-growth forests
in southwestern Washington. Can. J. Fish. Aquat. Sci. 48:2499-2508.
Bisson, P. A., J. L. Nielsen, R. A. Palmason, and L. E. Grove. 1982. A system of
naming habitat types in small streams, with examples of habitat utilization by
salmonids during low streamflow. Pages 62-73 in Armantrout, N. B. (ed.)
Acquisition and Utilization of Aquatic Habitat Inventory Information.
Proceedings of a symposium, October 28-30, 1981, Portland, Oregon.
Hagen Publishing Co., Billings, Montana.
Bisson, P. A., and J. R. Sedell. 1984. Salmonid populations in streams in
clearcuts vs. old-growth forests of western Washington. Pages 121-129 in
W. R Meehan, T. R. Merrell, Jr., and T. A. Hanley (ed.) Fish and wildlife
relationships in old-growth forests: Proceedings of a symposium held in
Juneau, Alaska, 12-15 April 1982. Amer. Inst. Fish. Res. Biol. 425 p.
Bisson, P. A., K. Sullivan, and J. L. Nielsen. 1988. Channel hydraulics, habitat
use, and body form of juvenile coho salmon, steelhead, and cutthroat trout in
streams. Trans. Amer. Fish. Soc. 117:262-273.
Bjornn, T. C. 1971. Trout and salmon movements in two Idaho streams as
related to temperature, food, stream flow, cover, and population density.
Trans. Amer. Fish. Soc. 100:423-438.
Bjornn, T. C., and J. Mallet. 1964. Movements of planted and wild trout in an
Idaho river system. Trans. Amer. Fish. Soc. 93(1):70-76.
43
Bohlin, T. 1978. Temporal changes in the spatial distribution of juvenile sea-trout
Salmo trutta in a small stream. Oikos 30:114-120.
Brown, T. G., and G. F. Hartman. 1988. Contribution of seasonally flooded lands
and minor tributaries to the production of coho salmon in Carnation Creek,
British Columbia. Trans. Amer. Fish. Soc. 117:546-551.
Bustard, D. R., and D. W. Narver. 1975. Aspects of the winter ecology of
juvenile coho salmon (Oncorhynchus kisutch) and steelhead trout ( Salmo
gairdneri). J. Fish. Res. Board. Can. 32:667-680.
Cederholm, C. J., and W. J. Scarlett. 1982. Seasonal immigrations of juvenile
salmonids into four small tributaries of the Clearwater River, Washington,
19771981. Pages 98-110 in E. L. Brannon and E. 0. Salo (ed.) Salmon and
Trout Migratory Behavior Symposium, June 1981. College of Fisheries,
Univ. Wash., Seattle.
Chapman, D. W. 1962. Aggressive behavior of juvenile coho salmon as a cause
of emigration. J. Fish. Res. Bd. Can. 19:1047-1080.
Close, T. L., and C. S. Anderson. 1992. Dispersal, density-dependent growth,
and survival of stocked steelhead fry in Lake Superior tributaries. North
Amer. J. Fish. Manage. 12:728-735.
Cooney, C. X., and S. E. Jacobs. 1993. Oregon coastal salmon spawning surveys,
1991. Oregon Department of Fish and Wildlife, Information Reports (Fish)
93-1, Portland.
Culp, J. M., and R. W. Davies. 1982. Analysis of longitudinal zonation and the
river continuum concept in the Oldman-South Saskatchewan River system.
Can. J. Fish. Aquat. Sci. 39:1258-1266.
Decker, L. M., and D. C. Erman. 1992. Short-term seasonal changes in
composition and abundance of fish in Sagehen Creek, California. Trans.
Amer. Fish. Soc. 121: 297-306.
Edmundson, E., F. E. Everest, and D. W. Chapman. 1968. Permanence of station
in juvenile chinook salmon and steelhead trout. J. Fish. Res. Bd. Canada
25:1453-1464.
Elliott, S. T., and R. D. Reed. 1974. A study of land-use activities and their
relationship to the sport fish resources in Alaska. Alaska Dep. Fish. Game,
Study D-1, Job D-1-B, Vol. 15.
44
Elwood, J. W., and T. F. Waters. 1969. Effects of floods on food consumption
and production rates of a stream brook trout population. Trans. Amer. Fish.
Soc. 98:253-262.
Erlinge, S. 1968. Food studies on captive otters Lutra lutra L. Oikos 19:259270.
Evans, J. W., and R. L. Noble. 1979. The longitudinal distribution of fishes in an
east Texas stream. Amer. Mid. Nat. 101:333-343.
Everest, F. H. 1973. Ecology and management of summer steelhead in the Rogue
River. Fishery Research Rep. Num. 7, Oregon State Game Commission,
Corvallis, 48 p.
Everest, F. H., and D. W. Chapman. 1972. Habitat selection and spatial
interaction by juvenile chinook salmon and steelhead trout in two Idaho
streams. J. Fish. Res. Bd. Canada 29:91-100.
Everest, F. H., G. H. Reeves, J. R. Sedell, D. B. Hohler, and T. Cain. 1987. The
effects of habitat enhancement on steelhead trout and coho salmon smolt
production, habitat utilization, and habitat availability in Fish Creek, Oregon,
1983-86. U. S. Dept. of Energy, Bonneville Power Administration, Division
of Fish and Wildlife, Portland, OR. 128 p.
Fausch, K. D., and T. G. Northcote. 1992. Large woody debris and salmonid
habitat in a small coastal British Columbia stream. Can. J. Fish. Aquat. Sci.
49:682-693.
Fretwell, S. D. 1972. Populations in a seasonal environment. Princeton
University Press, Princeton, New Jersey, 217 p.
Fretwell, S. D., and H. L. Lucas, Jr. 1970. On territorial behavior and other
factors influencing habitat distribution in birds. I. Theoretical development.
Acta Biotheoretica XIX:16-36.
Frissell, C. A. 1992. Cumulative effects of land use on salmonid habitat in
southwest Oregon coastal streams. Ph.D. dissertation, Oregon State
University, Corvallis, Oregon.
45
Gibson, R. J. 198L Behavioural interactions between coho salmon
(Oncorhynchus kisutch), Atlantic salmon (Salmo salar), brook trout
(Salvelinus fontinalis), and steelhead trout (Salmo gairdneri) at the juvenile
fluviatile stage. Canadian Technical Report of Fisheries and Aquatic
Sciences 1029 p.
Glova, G. J. 1984. Management implications of the distribution and diet of
sympatric populations of juvenile coho salmon and coastal cutthroat trout in
small streams in British Columbia, Canada. Progressive Fish Culturist
46:269-277.
Glova, G. J. 1986. Interaction for food and space between experimental
populations of juvenile coho salmon (Oncorhynchus kisutch) and coastal
cutthroat trout ( Salmo clarki) in a laboratory stream. Hydrobiologia
132:155-168.
Glova, G. J. 1987. Comparison of allopatric cutthroat trout stocks with those
sympatric with coho salmon and sculpins in small streams. Environmental
Biology of Fishes 20:275-284.
Hankin, D. G., and G. H. Reeves. 1988. Estimating total fish abundance and total
habitat area in small streams based on visual estimation methods. Can. J.
Fish. Aquat. Sci. 45: 834-844.
Hartman, G. F. 1965. The role of behavior in the ecology and interaction of
underyearling coho salmon (Oncorhynchus kisutch) and steelhead trout
(Salmo gairdneri). J. Fish. Res. Bd. Canada 22:1035-1081.
Hartman, G. F., B. C. Anderson, and J. C. Scrivener. 1982. Seaward movement
of coho salmon (Oncorhynchus kisutch) fry in Carnation Creek, an unstable
coastal stream in British Columbia. Can. J. Fish. Aquat. Sci. 39:588-597.
Hartman, G. F., and T. G. Brown. 1987. Use of small, temporary, floodplain
tributaries by juvenile salmonids in a west coast rain-forest drainage basin,
Carnation Creek, British Columbia. Can. J. Fish. Aquat. Sci. 44: 262-270.
Hartman, G. F., and C. A. Gill. 1968. Distribution of juvenile steelhead and
cutthroat trout (Salmo gairdneri and S. clarki clarki) within streams in
southwestern British Columbia. J. Fish. Res. Board Can. 25:33-48.
46
Hawkins, C. P., M. L. Murphy, N. H. Anderson, and M. A. Wilzbach. 1983.
Density of fish and salamanders in relation to riparian canopy and physical
habitat in streams of the northwestern United States. Can. J. Fish. Aquat.
Sci. 40: 1173-1185.
Hawkins, C. P., and J. R. Sedell. 1981. Longitudinal and seasonal changes in
functional organization of macroinvertebrate communities in four Oregon
streams. Ecology 62:387-397.
Hicks, B. J. 1990. The influence of geology and timber harvest on channel
morphology and salmonid populations in Oregon coast range streams. Ph.D.
dissertation, Oregon State University, Corvallis, Oregon.
Hillman, T. W., J. S. Griffith, and W. S. Platts. 1987. Summer and winter habitat
selection by juvenile chinook salmon in a highly sedimented Idaho stream.
Trans. Amer. Fish. Soc. 116:185-195.
Holtby, L. B., and G. F. Hartman. 1982. The population dynamics of coho
salmon (Oncorhynchus kisutch) in a west coast rain forest stream subjected
to logging. Pages 308-347 in Proceedings of the Carnation Creek
Workshop, A 10-yr Review, Pacific Biological Station, Nanaimo, British
Columbia, February 24-26, 1982.
Huet, M. 1959. Profiles and biology in western European streams as related to
fish management. Trans. Amer. Fish. Soc., 88:153-163.
Hughes, R. M., and J. R. Gammon. 1987. Longitudinal changes in fish
assemblages and water quality in the Willamette River, Oregon. Trans.
Amer. Fish. Soc. 116:196-209.
Johnston, J. M. 1982. Life histories of anadromous cutthroat with emphasis on
migratory behavior. Pages 123-127 in E. L. Brannon and E. 0. Salo (ed.)
Salmon and Trout Migratory Behavior Symposium, June 1981. College of
Fisheries, Univ. Wash., Seattle.
Keller, E. A., and T. Tally. 1979. Effects of large organic debris on channel form
and fluvial processes in the coastal redwood environment. Pages 169-197 in
D. D. Rhodes and G. P. Williams, (ed.). Adjustments of the fluvial system.
Kendall/Hunt, Dubuque, Iowa.
Kralik, N. J., and J. E. Sowerwine. 1977. The role of two northern California
intermittent streams in the life history of anadromous salmonids. Joint
Master's thesis, Humboldt State University, Arcata, California.
47
Leidholt-Bruner, K., D. E. Hibbs and W. C. McComb. 1992. Beaver dam
locations and their effects on the distribution and abundance of coho salmon
fry in two coastal Oregon streams. Northwest Science 66:218-223.
Lister, D. B., and H. S. Genoe. 1970. Stream habitat utilization by cohabitating
underyearlings of chinook (Oncorhynchus tshawytscha) and coho (0.
kisutch) salmon in the Big Qualicum River, British Columbia. J. Fish. Res.
Bd. Canada 27:1215-1224.
Mason, J. C. 1976. Response of underyearling coho salmon to supplemental
feeding in a natural stream. J. Wildl. Manage. 40:775-788.
Meyers, L. S., T. F. Thuemler, and G. W. Kornely. 1992. Seasonal movements of
brown trout in northeast Wisconsin. North Amer. J. Fish, Manage. 12:433441.
Murphy, M. L., J. Heifetz, J. F. Thedinga, S. W. Johnson, and K. V. Koski. 1989.
Habitat utilization by juvenile Pacific salmon (Oncorhynchus) in the glacial
Taku River, southeast Alaska. Can. J. Fish. Aquat. Sci. 46:1677-1685.
Nehlsen W., J. E. Williams, and J. A. Lichatowich. 1991. Pacific salmon at the
crossroads: Stocks at risk from California, Oregon, Idaho, and Washington.
In Fisheries (Bethesda) 16(2):4 -21.
Newman, R. M., and T. F. Waters. 1989. Differences in Brown Trout (Salmo
trutta) production among contiguous sections of an entire stream. Can. J.
Aquat. Sci. 46:203-213.
Nickelson, T. E., J. D. Rodgers, S. L. Johnson, and M. F. Solazzi. 1992.
Seasonal changes in habitat use by juvenile coho salmon (Oncorhynchus
kisutch) in Oregon coastal streams. Can. J. Fish. Aquat. Sci. 49:783-789.
Northwest Power Planning Council. 1987. Columbia River Basin fish and wildlife
program. Portland, Oregon.
Oregon Department of Fish and Wildlife. 1992. General fish management goals,
635-07-510(2). Portland, Oregon.
Pemberton, R. 1976. Sea trout in North Argyll sea lochs; population, distribution
and movements. J. Fish. Biol. 9:157-179.
48
Peterson, N. P. 1982. Immigration of juvenile coho salmon (Oncorhynchus
kisutch) into riverine ponds. Can. J. Fish. Aquat. Sci. 39:1308-1310.
Pucket, K. J., and L. M. Dill. 1985. The energetics of feeding territoriality in
juvenile coho salmon (Oncorhynchus kisutch). Behaviour 92:97-111.
Robison, E. G. 1987. Large woody debris and channel morphology of
undisturbed streams in southeast Alaska. Master's thesis, Oregon State
University, Corvallis, Oregon, 136 p.
Saunders, R. L., and J. H. Gee. 1964. Movements of young Atlantic salmon in a
small stream. J. Fish. Res. Bd. Can., 21:27-36.
Schwartz, J. S. 1991. Influence of geomorphology and land use on distribution
and abundance of salmonids in a coastal Oregon basin. Master's Thesis,
Oregon State University, Corvallis, Oregon.
Sedell, J. R. and F. H. Everest. 1991. Historic changes in pool habitat for
Columbia River Basin salmon under study for TES listing; draft report,
December 1990. USDA Forest Service, Pacific Northwest Research
Station, Corvallis, OR. 7 p.
Sedell, J. R., and J. L. Froggatt. 1984. Importance of streamside forests to large
rivers: The isolation of the Willamette River, Oregon, U.S.A., from its
floodplain by snagging and streamside forest removal. Verh. Internat.
Verein. Limnol. 22:1828-1834.
Sheldon, A. L. 1968. Species diversity and longitudinal succession in stream
fishes. Ecology 49:193-198.
Skeesick, D. G. 1970. The fall immigration of juvenile coho salmon into a small
tributary. Oreg. Fish Comm. Res. Rep. 2:90-95.
Solomon, D. J. 1982. Migration and dispersion of juvenile brown and sea trout.
Pages 136-145 in E. L. Brannon and E. 0. Salo (ed.) Salmon and Trout
Migratory Behavior Symposium, June 1981. College of Fisheries, Univ.
Wash., Seattle.
Stack, W. R., and R. L. Beschta. 1989. Factors influencing pool morphology in
Oregon coastal streams. Pages 401-411 in D. F. Potts and W. W. Woessner
(ed.) Headwaters Hydrology. Proceedings of a symposium held in Missoula,
Montana. American Water Resources Association, Bethesda, Maryland.
49
Swales, S., R. B. Lauzier, and C. D. Levings. 1986. Winter habitat preferences of
juvenile salmonids in two interior rivers in British Columbia. Can. J. Zool.
64:1506-1514.
Trotter, P. C. 1989. Coastal cutthroat trout: A life history compendium. Trans.
Amer. Fish. Soc. 118:463-473.
Tschaplinski, P. J., and G. F. Hartman. 1983. Winter distribution of juvenile coho
salmon (Oncorhynchus kisutch) before and after logging in Carnation Creek,
British Columbia, and some implications for overwinter survival. Can. J.
Fish Aquat. Sci. 40:452-461.
Wilzbach, M. A. 1985. Relative roles of food abundance and cover in
determining the habitat distribution of stream-dwelling cutthroat trout
(Salmo clarki). Can. J. Fish. Aquat. Sci 42: 1668-1672.
Wilzbach, M. A., and J. D. Hall. 1985. Prey availability and foraging behavior of
cutthroat trout in an open and forested section of stream. Verh. Int.
Ver. Limnol. 22:2516-2522.
Table A.1. Mean densities (number/m3) of each species and age class sampled by habitat type and reach in Cummins Creek in
September 1987. Density of fish in pools were calculated for each reach while glides and riffles were calculated for
reaches 1-4 and reaches 5-8. Side channels and valley floor tributaries were not sampled for fish in September 1987.
Habitat
Pools
Reach
1
2
3
4
5
6
7
8
Glides
1-4
5-8
Riffles
1-4
5-8
Cutthroat
> 20 cm
Total
Assemblage
1.237
1.596
0.145
0.013
0.045
0.023
0.034
0.157
0.088
0.053
0.000
0.235
0.244
0.207
0.344
0.011
0.010
1.320
1.180
0.085
0.109
0.000
0.000
0.000
0.011
0.956
0.530
Steelhead
0+
0.920
0.695
1.325
1.184
1.026
0.293
0.613
0.036
Steelhead
0.867
0.582
0.871
0.410
1+
0.220
0.267
0.390
0.394
0.438
0.190
0.397
0.090
Coho Salmon
0+
0.085
0.589
0.719
0.329
1.238
0.596
1.201
2.457
1.940
2.859
1.167
2.265
0.272
Table A.2. Mean densities (number/m2) of each species and age class sampled by habitat type and reach in Cummins Creek in
April 1988. Density of fish in pools were calculated for each reach while glides and riffles were calculated for reaches
1-4 and reaches 5-8. Side channels and valley floor tributaries were not sampled for fish in April 1988.
0.051
0.058
0.095
0.044
0.064
0.051
0.023
0.025
Steelhead
Pre-smolt
0.122
0.082
0.066
0.039
0.039
0.045
0.026
0.021
Coho Salmon
0+
0.006
0.022
0.054
0.000
0.058
0.067
0.035
0.017
0.010
0.010
Steelhead
Habitat
Pools
Reach
1
2
3
4
5
6
7
8
Glides
1-4
5-8
Riffles
1-4
5-8
1+
Coho Salmon
Pre-smolt
Cutthroat
> 20 cm
0.031
0.061
0.106
0.096
0.023
0.035
0.023
0.023
0.012
0.000
0.011
0.004
0.000
0.004
0.002
0.010
0.007
0.000
Total
Assemblage
0.222
0.262
0.239
0.122
0.186
0.196
0.129
0.152
0.038
0.021
0.003
0.040
0.025
0.009
0.003
0.000
0.104
0.086
0.012
0.001
0.001
0.018
0.000
0.000
0.000
0.000
0.024
0.029
Table A.3. Mean densities (number/m2) of each species and age class sampled by habitat type and reach in Cummins Creek in
June 1988. Density of fish in pools were calculated for each reach while glides, riffles, side channels, and valley floor
tributaries were calculated for reaches 1-4 and reaches 5-8.
Habitat
Pools
Reach
1
2
3
4
5
6
7
8
Glides
1-4
5-8
Riffles
1-4
5-8
Side Chls
1-4
5-8
V.F. Tribs.
1-4
5-8
Steelhead
0+
0.029
0.035
0.069
0.033
0.019
0.021
0.021
0.006
Steelhead
0.016
0.032
Steelhead Coho Salmon Coho Salmon
Cutthroat
10-20 cm
0.004
0.009
0.004
0.007
Cutthroat
Total
> 20 cm Assemblage
0.007
0.184
0.167
0.000
0.000
0.287
0.001
0.147
0.007
0.349
0.014
0.332
0.009
0.506
0.027
0.275
Pre-smolt
0.008
0+
0.071
0.001
0.000
0.002
0.003
0.008
0.008
0.006
0.076
0.154
0.058
0.163
0.178
0.278
0.067
Pre-smolt
0.000
0.003
0.000
0.000
0.000
0.000
0.000
0.000
0.036
0.095
0.002
0.000
0.034
0.096
0.000
0.000
0.000
0.015
0.000
0.007
0.088
0.245
0.011
0.035
0.021
0.014
0.000
0.000
0.000
0.007
0.000
0.000
0.000
0.000
0.000
0.000
0.032
0.056
0.113
0.004
0.000
0.000
0.000
0.000
0.080
0.192
0.000
0.000
0.000
0.000
0.003
0.000
0.193
0.196
0.084
0.001
0.000
0.000
0.000
0.000
0.339
0.267
0.000
0.000
0.000
0.000
0.000
0.000
0.423
0.268
1+
0.065
0.043
0.060
0.045
0.122
0.093
0.169
0.135
0.035
0.017
0.021
0.033
Table A.4. Mean densities (number/m3) of each species and age class sampled by habitat type and reach in Cummins Creek in
August 1988. Density of fish in pools were calculated for each reach while glides, riffles, and channels were
calculated for reaches 1-4 and reaches 5-8. Valley floor tributaries were not present in August 1988.
Habitat
Pools
Steelhead
8
Steelhead
0+
0.388
0.685
0.195
0.476
0.386
0.134
0.515
0.897
1-4
0.498
5-8
1-4
Reach
1
2
3
4
5
6
7
Glides
Riffles
5-8
Side Chls
1-4
5-8
1+
0.220
0.336
0.319
0.367
0.277
0.188
0.304
0.206
Coho Salmon
0+
0.345
0.400
Cutthroat
10-20 cm
0.007
0.023
0.008
0.024
Cutthroat
> 20 cm
Total
Assemblage
0.975
0.749
1.114
0.147
0.040
0.087
0.076
0.015
0.020
0.024
0.035
0.047
0.065
0.043
0.078
1.193
0.189
0.122
0.353
0.642
0.009
0.027
0.018
0.005
1.067
1.988
1.432
0.421
0.173
0.110
0.009
0.050
0.031
0.000
0.000
0.012
1.645
0.593
0.097
0.005
0.000
0.010
0.775
0.407
0.000
0.015
0.000
0.005
0.872
0.440
1.003
0.833
1.104
0.031
1.465
1.549
1.735
1.845
1.176
2.063
1.403
Table A.S. Mean densities (number/m3) of each species and age class sampled by habitat type and reach in Cummins Creek in
October 1988. Density of fish in pools were calculated for each reach while glides, riffles, and channels were
calculated for reaches 1-4 and reaches 5-8. Valley floor tributaries were not present in October 1988.
Habitat
Pools
Reach
1
2
3
4
5
6
7
8
Glides
1-4
5-8
Riffles
1-4
5-8
Side Chls
1-4
5-8
Steelhead
0+
0.411
0.393
0.308
0.577
0.420
0.100
0.445
0.720
Steelhead
0.362
0.640
1+
0.219
0.242
0.140
0.277
0.106
0.145
0.163
0.056
Coho Salmon
0+
0.395
0.756
0.951
0.639
0.884
0.695
Cutthroat
10-20 cm
0.024
0.009
0.000
Cutthroat
> 20 cm
0.008
0.017
0.010
0.048
0.007
0.086
Total
Assemblage
1.056
1.416
1.410
1.595
1.470
1.105
1.690
0.928
0.972
0.084
0.055
0.053
0.080
0.059
0.053
0.158
0.086
0.067
0.267
0.000
0.089
0.000
0.039
0.587
0.577
0.353
0.033
0.044
0.000
0.000
0.033
0.000
0.000
0.000
0.642
0.397
0.059
0.005
0.000
0.005
0.626
0.276
0.000
0.010
0.000
0.000
0.685
0.305
0.051
0.015
1.121
Table A.6. Mean densities (number/m2) of each species and age class sampled by habitat type and reach in Cummins Creek in
April 1989. Density of fish in pools were calculated for each reach while glides, riffles, side channels, and valley floor
tributaries were calculated for reaches 1-4 and reaches 5-8.
Steelhead
Habitat
Pools
Coho Salmon
0+
0.000
0.010
0.008
0.038
0.002
Coho Salmon
Pre-smolt
0.082
0.092
0.098
0.148
0.109
8
0.041
0.061
0.133
0.071
0.079
0.127
Steelhead
Pre-smolt
0.047
0.030
0.032
0.039
0.038
0.013
0.014
0.015
1-4
5-8
0.055
0.080
0.015
0.018
1-4
0.023
0.044
Reach
1
2
3
4
5
6
7
Glides
Riffles
5-8
Side Chls
1-4
5-8
V.F. Tribs.
1-4
5-8
1+
0.126
0.063
0.047
0.067
Cutthroat
10-20 cm
0.017
0.004
0.004
Cutthroat
> 20 cm
Total
Assemblage
0.274
0.204
0.192
0.308
0.267
0.195
0.277
0.309
0.060
0.000
0.011
0.027
0.032
0.054
0.034
0.001
0.004
0.002
0.006
0.020
0.003
0.009
0.000
0.024
0.005
0.046
0.008
0.001
0.021
0.000
0.142
0.132
0.016
0.000
0.000
0.005
0.001
0.000
0.000
0.001
0.000
0.000
0.040
0.050
0.001
0.000
0.000
0.000
0.168
0.454
0.005
0.040
0.002
0.000
0.000
0.000
0.176
0.495
0.000
0.000
0.000
0.000
0.221
0.426
0.000
0.011
0.000
0.000
0.000
0.000
0.221
0.437
0.071
0.035
0.001