-
Title (dcterms:title)
-
Eng
Winter Feeding Ecology of Coho Salmon (Oncorhynchus kisutch), Steelhead (O. mykiss), and Cutthroat Trout (O. clarkii) in the Skokomish River, Washington
-
Date (dcterms:date)
-
2010
-
Creator (dcterms:creator)
-
Eng
Wright, Lindsy A
-
Subject (dcterms:subject)
-
Eng
Environmental Studies
-
extracted text (extracttext:extracted_text)
-
Winter Feeding Ecology of Coho Salmon (Oncorhynchus kisutch),
Steelhead (O. mykiss), and Cutthroat Trout (O. clarkii)
in the Skokomish River, Washington
by
Lindsy A. Wright
A Thesis Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
March 2010
© 2010 by Lindsy A. Wright. All rights reserved.
ii
This Thesis for the Master of Environmental Study Degree
by
Lindsy A. Wright
has been approved for
The Evergreen State College
by
_____________________________
Gerardo Chin-Leo
Member of the Faculty
_____________________________
Alison Styring
Member of the Faculty
_____________________________
Roger Tabor
Supervisory Fish and Wildlife Biologist, USFWS
_____________________________
Roger Peters
Supervisory Fish and Wildlife Biologist, USFWS
_____________________________
Date
ABSTRACT
iii
Winter Feeding Ecology of Coho Salmon (Oncorhynchus kisutch),
Steelhead (O. mykiss), and Cutthroat Trout (O. clarkii)
in the Skokomish River, Washington
Lindsy A. Wright
The Skokomish River, Washington state, has frequent flood events which
combined with other factors have caused severe ecological disruption to juvenile
salmonid (Oncorhynchus spp.) habitat. Although most coho salmon populations in this
area are depressed, the habitat is still utilized. The Army Corps of Engineers is
performing a General Investigation to plan ecosystem restoration and flood risk
management; this diet study characterizes the winter diets of juvenile salmonids in this
system in order to inform these investigations and their consequential decisions.
To accomplish this, the diets of juvenile salmonids from four habitat types were
assessed; mainstem, tributaries, backwaters, and off-channel ponds. The diets of 223
coho salmon (O. kisutch), 31 rainbow trout (O. mykiss), and 9 cutthroat trout (O. clarkii)
were assessed to characterize winter feeding habits. The diets of juvenile salmonids in
these different habitat types were compared for prey abundance, relative importance of
prey items, stomach fullness, diet breadth, and diet overlap. In addition, the condition
factor, fish weights, and fork lengths of fish from the different habitats were also
compared.
Prey weights of the salmonids diets indicated that the majority was comprised of
benthic macroinvertebrates; Chironomidae, Ephemeroptera, Plecoptera, Trichoptera,
other Diptera, Megaloptera, and Oligochaetes. Ephemeroptera and Chironomidae had
highest index of relative importance and proportions by weight for coho salmon. Mean
stomach fullness for coho salmon was highest in backwaters and tributaries, and did not
vary between habitat types. Mean stomach fullness for coho salmon did vary
significantly among sites within habitat types: tributaries Swift and Vance Creek were
significantly higher than Hunter Creek (P = 0.004 and P = 0.0001, respectively);
backwater site South Fork downstream of Vance Creek confluence was significantly
higher than North Fork site 2-26 (upstream of X) (P = 0.002). Mean diet breadth was
highest in the mainstem and lowest in backwaters and there were no significant
iv
differences habitat types. Coho salmon diets in the mainstem overlapped significantly
with tributaries (0.62) and ponds (0.78) (Horn’s Index).
Mean condition factor values for coho salmon were not significantly different
between habitat types. However, there were significant differences in mean condition
factor for coho salmon within mainstem sites, tributary sites, and pond sites: the North
Fork mainstem site 2-26 values were significantly higher than South Fork mainstem (at
the North Fork confluence) (P = 0.005); Hunter Creek and Vance Creek values were
significantly higher than Swift Creek (P = 0.002 and P = 0.012, respectively); Skokomish
pond 6-22 was mean values were significantly higher than Skokomish pond 6-14 (P =
0.021).
Mean weights and FL for coho salmon were not significantly different between
habitat types. However, there were significant differences for coho salmon mean weights
and mean FL within mainstem sites, backwater sites, and pond sites: mainstem site 2-31
values were significantly higher than the mainstem site at the South Fork and North Fork
confluence (P = 0.002 weight, P = 0.008 FL); the values at the South Fork backwater site
downstream of the Vance Creek confluence were significantly higher than the North Fork
backwater site 2-26 (P = 0.013 weight, P = 0.041 FL); Skokomish pond site 6-22 values
were significantly higher than Skokomish pond 14 (P < 0.0001 weight, P = 0.0001 FL);
Skokomish pond site 6-21 values were significantly higher than Skokomish pond 14 (P =
0.005 weight, P = 0.024 FL) .
The overall mean weights, fork lengths, and stomach fullness of coho salmon in
this system were lower than typical means for these species in Washington State during
the same time of year, suggesting that Skokomish River fish are relatively small and their
growth may be food limited. Although diets in the different habitats varied, the lack of
differences in fish size suggests that the overall response is similar. The habitats
examined in this diet study served different functions for the fish and each is essential for
their over-wintering diets. All the habitats assessed in this system are supporting a
necessary food base for juvenile salmonids and need to be preserved and further restored.
Before any flood remediation work commences in this area, special emphasis needs to be
placed on protecting current backwater areas, maintaining access into tributaries, and
establishing a system of beaded channels which have demonstrated they are excellent
sources of forage food and refuge.
v
Table of Contents
(a)
List of Tables .................................................................................................... viii
(b)
List of Figures ..................................................................................................... ix
(c)
Acknowledgements ............................................................................................. xi
(d)
Introduction ..........................................................................................................1
(e)
Background...........................................................................................................5
(f)
Study area .............................................................................................................8
(g)
Methods .............................................................................................................. 12
Fish collections ......................................................................................... 12
Fish processing and stomach content sampling .......................................... 15
Laboratory analysis ................................................................................... 16
Data Analysis ............................................................................................ 17
(h)
Results ................................................................................................................ 23
Stomach collections and site characteristics ............................................... 23
Sample size ............................................................................................... 27
Diet composition - Prey Abundance and Importance ................................. 29
(i)
Discussion and Conclusions ................................................................................ 44
(j)
Management Implications and Habitat Recommendations ................................... 51
(k)
Literature Cited ................................................................................................... 57
(l)
Appendix A. Prey codes/categories ..................................................................... 65
(m)
Appendix B. Length frequency (5-mm FL increments) distributions for coho
salmon in all habitat types, steelhead (10-mm FL increments) in backwaters,
mainstem, and tributaries, and cutthroat trout (10-mm FL increments) in
tributaries. Graphs include all fish collected, including those greater than
100-mm FL, which were excluded from all analyses. .................................. 67
(n)
Appendix C. Results of Pielou’s method on individual sample sets for the
determination of adequate sample size for coho salmon. ............................. 68
(o)
Appendix D. Results of Pielou’s method on individual sample sets for the
determination of adequate sample size for steelhead and cutthroat trout. ..... 71
vi
(p)
Appendix E. Weights of all prey items for coho salmon, steelhead and cutthroat
trout in all habitat types. .............................................................................. 72
(q)
Appendix F. Summary of the major prey categories; percent proportion by weight
(%Wi), percent proportion by number (%Ni), and frequency of occurrence
(%Oi) by habitat type for coho salmon. ....................................................... 74
(r)
Appendix G. Summary of the major prey categories; percent proportions by
weight (%Wi), percent proportions by number (%Ni), and percent frequency
of occurrence (%Oi) by habitat type for steelhead and cutthroat trout. ......... 75
(s)
Appendix H. Percent index of relative importance (%IRI) for major prey
categories for coho salmon, steelhead, and cutthroat trout ........................... 75
vii
(a) List of Tables
Table 1.-Habitat types and sample locations of each sampling effort. The numbers in
parentheses are site identification numbers. The centers of the site transect
reaches are labeled “X”; reaches were ten times the bank full width, or 150 m,
whichever was greater. Skokomish pond site SP-6/22 was one very large
pond, but contained two designated sites, SP-6 and SP-22. ............................ 24
Table 2.-Water quality and habitat data from the sites where fish were collected for diet
analysis. The numbers in parentheses after sample locations are site
identification numbers. The centers of the site transects are labeled “X”;
transects were ten times the bank full width, or 150 m, whichever was greater.
Values collected were turbidity (Turb., ntu), dissolved oxygen (D.O., mg/L),
water temperature (°C), pH, GPS coordinates (NAD 83) (X LAT, X LONG),
GPS accuracy (Acc., +/- m), bank full widths (BFW, meters), and reach lengths
(meters). Dashes indicate values which were not collected. ........................... 26
Table 3.-Percent proportion by weight (%Wi), and by number (%Ni) of aquatic and
terrestrial insects. The “Other” categories are comprised of varying
combinations of Arachnida, Amphipoda, Collembola, Copepoda, fish eggs,
Hemiptera, Homoptera, Hydracarina, Hymenoptera, Lepidoptera, Mollusca,
plant material, and unidentified organic matter............................................... 29
Table 4.-Results of Horn’s index of overlap for coho salmon, steelhead, and cutthroat
trout, in the Skokomish River, January through March 2009. Values greater
than 0.60 are considered to have diet overlap, and are totally overlapped with
values equal to 1. Each habitat type represents all the sample sites combined.
...................................................................................................................... 38
viii
(b) List of Figures
Figure 1.-Water is diverted from the North Fork Skokomish River, Washington by two
Cushman Dams and is piped and released near Potlatch, Washington (arrow
points to location where the water is discharged). ...........................................6
Figure 2.-Map of the study area and all sample site locations for diet study, lower
Skokomish River, January through March 2009. Each marker indicates one
site except for the backwater sites. Site identification numbers are used to
identify site locations, see Table 2 for a detailed list of site location
characteristics. .............................................................................................. 13
Figure 3.-Results of Pielou's method to determine adequate sample size within each of
the habitat types for coho salmon; (A) mainstem, (B) tributaries, (C)
backwaters, and (D) ponds. Sample sizes are indicated at the top of each
graph (n). Values for cumulative diversity (Hk) initially start out low, then
increase and level off when adequate sample size has been reached (Hoffman
1978). ........................................................................................................... 27
Figure 4.-Results of Pielou's method to determine adequate sample size within each of
the habitat types. Steelhead in panel (A) mainstem, (B) tributaries, and (C)
backwaters, and cutthroat trout in (D) tributaries. Sample sizes are indicated
at the top of each graph (n). Values for cumulative diversity (Hk) initially
start out low, then increase and level off when adequate sample size has been
reached (Hoffman 1978). Steelhead in tributaries (B) was not used in
analyses because these results indicate the sample size was too small. .......... 28
Figure 5.-Percent proportion by weight (%Wi) of major prey categories present in coho
salmon diets, lower Skokomish River, January through March 2009. Samples
sizes are listed at the top of the graph. The “Other” category includes a
mixture of Arachnida, Amphipoda, Collembola, Copepoda, fish eggs,
Hemiptera, Hydracarina, Hymenoptera, Lepidoptera, Mollusca, plant material,
and unidentified organic matter. ................................................................... 31
Figure 6.-Percent proportion by weight (%Wi) of major prey categories present in coho
stomachs in tributaries of the lower Skokomish River, January through March
2009. Samples sizes are listed at top of graph. The “Other” category includes
a mixture of Arachnida, Amphipoda, Collembola, Copepoda, fish eggs,
Hemiptera, Hydracarina, Hymenoptera, Lepidoptera, Mollusca, plant material,
and unidentified organic matter. ................................................................... 32
Figure 7.-Percent proportion by weight (%Wi) of the major prey categories present in
cutthroat trout and steelhead, lower Skokomish River, January through March
2009. Samples sizes are listed at the top of the graph. The “Other” category
includes a mixture of Arachnida, Collembola, Copepoda, Homoptera,
Hydracarina, Mollusca, Lepidoptera, plant material, and unidentified organic
matter. .......................................................................................................... 33
ix
Figure 8.-Percent index of relative importance of major prey items for coho salmon, in
the Skokomish River, January through March 2009. ..................................... 35
Figure 9.-Mean percent stomach fullness for coho salmon, steelhead, and cutthroat trout
(+ standard error), in the Skokomish River, January through March 2009.
Each bar represents a different sample site. Within each habitat type, bars
with different letters are significantly different (nested ANOVA and Tukey’s
HSD; P < 0.05). *Bars with no letters were sites not statistically compared. 36
Figure 10.-Results of diet breadth determinations for coho salmon, steelhead, and
cutthroat trout, in the Skokomish River, January through March 2009. Index
values range between 1 (one prey type present, i.e. there is no diet breadth)
and infinity. Values less than two indicate little diet breadth (Tabor et al.
2001). Each bar represents a different sample site. ....................................... 37
Figure 11.-Mean weights for coho salmon (+ standard error) in four habitat types
analyzed in the diet analysis, lower Skokomish River, Washington. Each bar
represents a different sample site. There were no significant differences in
mean weights between habitat types (nested ANOVA). Within each habitat
type, bars with different letters are significantly different (nested ANOVA and
Tukey’s HSD; P < 0.05). *Bars with no letters were sites not statistically
compared. .................................................................................................... 39
Figure 12.-Mean FL for coho salmon (+ standard error) in the four habitat types analyzed
in the diet analysis, lower Skokomish River, Washington. Each bar represents
a different sample site. There were no significant differences in mean FL
between habitat types (nested ANOVA). Within each habitat type, bars with
different letters are significantly different (nested ANOVA and Tukey’s HSD;
P < 0.05). *Bars with no letters were sites not statistically compared. .......... 41
Figure 13.-Mean condition factor for coho salmon, cutthroat trout, and steelhead (+
standard error), lower Skokomish River, January through March 2009. Each
bar represents a different sample site. There were no significant differences in
mean condition factor between habitat types (nested ANOVA). Within each
habitat type, bars with different letters are significantly different (nested
ANOVA and Tukey’s HSD; P < 0.05). *Bars with no letters were sites not
statistically compared. .................................................................................. 42
Figure 14.-Mean weight, FL, and condition factor for cutthroat trout and steelhead (+
standard error) lower Skokomish River, January through March 2009. ......... 43
x
(c)
Acknowledgements
I would like to thank the United States Fish and Wildlife, Fisheries Division for
making this research possible. I need to especially acknowledge the time, patience, and
expertise of Roger Tabor and Roger Peters. They offered their honesty, ideas, and
enthusiasm to this research. I would also like to thank Brad Thompson, with U.S. Fish
and Wildlife, for his support and helping me see the big picture. Thanks to Mark
Celedonia for sharing his experience with me. I want to thank those who supported me
through this process including the Skokomish Crew at US Fish and Wildlife; Dan Lantz,
Jamie Sproul, Nathan Hyde, Jeremy Anhalt, Ed Hughes, Eric Meyers, and Sarah Moffitt.
Thank you to my faculty advisors Alison Styring, Gerardo Chin-Leo, and Larry
Dominguez for your time and guidance through the writing process and for your moral
support. There were several occasions in which a few words from you put me right back
on track and encouraged me to keep going. Thank you all for your help, for giving
encouragement and for pushing me to go forward. Thanks to my daughter Maddison for
being such a great sport, and for doing so well at entertaining herself.
xi
(d) Introduction
Flooding in the Skokomish River Valley is considered chronic and extensive, with
generally 10 to 15 events that are classified as floods annually (classified flood stage is
approximately 241 m3s-1/8,500 cfs) (USACE 2000). The highest peak discharge recorded
on the lower Skokomish River of 1,036 m3 s-1/36,600 cfs occurred in 1990 (USACE and
USFWS 2008). Base flows in the mainstem are approximately 6 m3 s-1/205 cfs (Peters et
al. in prep.). Recorded studies of Skokomish River flooding date back to 1941 by the
United States Army Corps of Engineers (Corps), and several have been performed since.
Most recently, the Corps was charged with addressing this flooding after the completion
of a project management plan for the feasibility study of the Skokomish River basin in
2006 (USACE 2006). The Corps are considering levee and dike removals and/or
setbacks, sediment control structures, re-opening of side channels or oxbows, riparian
planting, and dredging of certain sections of the mainstem which experience sub-surface
flow (USACE and USFWS 2008). Goals of the Corps are ecosystem restoration and
flood risk management; by nature the designs and implementation of these projects may
temporarily reduce the availability of juvenile salmonid overwintering habitat and prey
food base.
Winter is a critical time for these fish because they are shifting their diets to
foraging from mainly drift feeding, and their ability to digest food reduces significantly
when the temperature declines, often making the cost of acquiring food to exceed the
benefit of doing so. Emaciation is common for fish over-wintering in temperate
environments, such as the Skokomish River basin. Fish in these areas experience shorter
growing seasons which precede periods of scarce winter resources; this places time
1
constraints on acquiring sufficient energy reserves to survive the winter (Schultz and
Conover 1999; Post and Parkinson 2001). Over-wintering juvenile salmonids require
access into areas of refuge which can protect them from high flows, predators, and can
offer sources of forage food. The Skokomish River habitat conditions are poor, reflecting
a disturbed system, and any actions taken towards habitat-altering improvements will
likely influence the over-wintering salmonid populations.
The Skokomish River has several factors contributing to its current condition; the
river currently has low-levels of channel connectivity, partially due to the intermittent
levels of remediation. The placement of dikes and levees have narrowed water flow
areas, concentrated and accelerated flows, enhanced flooding, and cut off access for fish
into side-channels. Other contributors to the river’s condition are heavy logging in the
lower watershed combined with reduced flows from the installation of two dams on the
North Fork of the river. The diminished flow has caused reduced sediment transport and
increased deposition; the aggradation has caused the channel bed to rise approximately 4
meters (Jay and Simenstad 1996).
The Skokomish Valley floodplain is the primary point of access for coho salmon
(Oncorhynchus kisutch) into most tributaries in the lower watershed. The nature of the
floodplain environment causes fish to seek refuge during high-flow events; high flows in
winter have caused reductions in juvenile salmonid abundance (Cederholm et al. 1997),
and reduced macroinvertebrate densities proportional to the flood magnitude (Robinson
et al. 2004). With increases in water flows, the fish caught in the system will seek refuge
by moving up to 30 km downstream (Peterson 1982a), into ponds which provide a
thermal refuge (Swales and Levins 1989), or from the mainstem to off-channel ponds,
2
sloughs, and wetlands that may not have been accessible during the summer (Quinn
2005). Access into these areas is critical during high flows for avoiding predators,
preventing displacement downstream, providing refuge, and to find alternative sources of
food through the winter season.
Aquatic food sources appear to be more available and important than terrestrial
food sources for juvenile salmonids in winter (Nakano and Murakami 2001). Juvenile
coho salmon (Oncorhynchus kisutch) diets during the winter in Oregon tributaries
showed that aquatic invertebrates accounted for 75% of the total mass ingested, and were
primarily comprised of aquatic chironomid larvae (Diptera), baetid mayfly nymphs
(Ephemeroptera), limnephilid caddisfly larvae (Trichoptera), and winter stonefly nymphs
(primarily Capniid) (Olegario 2006). Recently disturbed habitat is usually in poor
condition and contains low complexity, making them less capable of sustaining adequate
levels of aquatic invertebrates. Skokomish River fish are likely to more susceptible to
emaciation because of the combined effects of the river’s flood state, the effects of
winter, and the continually disturbed, and poor habitat conditions.
Sections containing adequate winter habitat lose fewer fish during freshets and
maintain higher numbers of coho salmon in winter than sections without these
characteristics (Robinson et al. 2004). Acceptable winter habitat for juvenile salmonids
often includes access into stream sections containing deep pools, log jams, and undercut
banks with tree roots and debris (Tschaplinski and Hartman 1983), or areas which allow
access into intermittent headwater streams (Olegario 2006), tributaries, backwaters, and
ponds. During fall and early winter, migrations into riverine ponds and runoff streams
appear to coincide with freshets (Cederholm and Scarlett 1982). Channels that are
3
associated with ponds or swamps form highly productive habitat for overwintering fish
(Peterson and Reid 1984). A fish’s ability to actively access alternative habitats increases
their likelihood of survival.
Habitat containing healthy benthic macroinvertebrate populations like ponds,
backwaters, and small tributaries, tend to offer substantial alternative food sources.
These types of alternative habitats are especially critical for juvenile salmonids that are
transitioning into winter feeding habits. Connected refuge habitats such as tributaries,
side-channels (oxbows), and ponds offer an alternative source of feeding for the fish.
The degree to which an aquatic environment is able to support fish populations is, in part,
directly related to the relative abundance of certain aquatic insects (McCafferty 1981).
The abundance of flying and terrestrial insect’s declines with temperature,
causing the fish to shift their patterns of acquiring food from drift feeding to foraging for
benthic macroinvertebrates available during winter. Although it is known that juvenile
salmonids primarily consume food organisms that are drifting aquatic insects and the
larval stages of terrestrial insects (Quinn 2005), the specifics of their winter diet is not
well understood. Knowledge of which types of winter refuge habitat may be sustaining
populations of Skokomish River juvenile salmonids would be valuable information to the
Corp’s remediation actions.
This research evaluates the diets of juvenile coho salmon, steelhead (O. mykiss),
and cutthroat trout (O. clarkii) in four habitat types for foraging habits during the winter
months. Although more data is available on coho salmon than other salmonid species,
site specific diet information on over-wintering populations is lacking. Diet analysis of
juvenile salmonids in this area may provide benefits such as (1) baseline diet
4
characterization, (2) abundance and importance of prey types, (3) quantifying diet
overlap for fish among habitat types, (4) quantifying diet breadth and stomach fullness
associated with habitat types, and (5) evaluation of which types of habitat may provide
best foraging opportunities.
(e) Background
The first European and American settlers began arriving in the Skokomish River
valley around 1850, and the first logging camps were established in 1887 (Amato 1995).
By the early 1900’s most of the Skokomish floodplain had been converted into pasture
(Canning et al. 1988), and by the 1920’s heavy logging of the area had commenced. In
1930 the Cushman Hydroelectric project was completed and consisted of two dams and
two powerhouses, which diverted approximately 40% of the Skokomish River delta’s
annual mean runoff from the North Fork for power production (Jay and Simenstad 1996).
This also eliminated approximately 19.3 km (12 miles) of prime Chinook salmon (O.
tshawytscha) spawning grounds (Skokomish Tribe and WDFW 2007). The diverted
water flows down a pipe from Lake Kokanee, the reservoir below Lake Cushman, and
out into Hood Canal about 5 km north of the mouth of the Skokomish River, near
Potlatch, Washington.
5
Figure 1.-Water is diverted from the North Fork Skokomish River, Washington by two Cushman Dams
and is piped and released near Potlatch, Washington (arrow points to location where the water is
discharged).
Diversion of the water flow from the North Fork and several other factors
combine and contribute to the current condition of the river basin habitat. The loss of
gravel recruitment into the estuary has diminished eelgrass bed production, and has
caused a reduction in the estuary biotic zone (USACE 2000). Aggradation has caused
heightened water levels resulting in a loss of deltaic surface area, decreased mesohaline
mixing zones, loss of low intertidal habitat and eelgrass beds (Jay and Simenstad 1996),
and has caused frequent and substantial flooding. Eelgrass is a critical nursery
environment for salmonids, and the loss of intertidal marsh combined with the loss of
subtidal estuary area reduces available rearing habitat and refuge areas for juvenile
6
salmonids (USACE 2006); approximately one third of the original marsh areas have been
lost to agricultural activities (Bortelson et al. 1980).
In addition to the physical burden, the flooding incurs a heavy financial burden on
the Skokomish Indian Tribe, Skokomish Valley residents, other local residents, and
taxpayers. The Skokomish Indian Reservation residents are more frequently and severely
affected by the flooding than those in the Skokomish Valley above the U.S. 101 Bridge
(Canning et al. 1988). A feasibility study, the General Investigation (GI), was initiated
using the Corps Puget Sound and Adjacent Water study authority, and funds were
provided by the House of Representatives to study the flooding problems in the
Skokomish River Basin (USACE 2006).
The Corps is responsible for taking actions to alleviate the flooding, and they
intend on restoring proper natural function to the Skokomish River basin while reducing
flood damages to valley residents including the Skokomish Indian Tribe (USACE and
USFWS 2008). The Corps is currently in the feasibility phase of the GI to address
ecosystem restoration and flood risk management. The U.S. Fish and Wildlife Service, in
Lacey, Washington, are currently performing a survey of the distribution, abundance, and
out-migration of juvenile salmonids and resident fish in the Skokomish River Basin. This
diet research has been performed in conjunction with fish sampling and assessments for
the GI. One goal of performing this diet study as a part of the GI was to provide the
Corps of Engineers with more information on habitat types that are providing food
sources to over-wintering populations of juvenile salmonids.
7
(f) Study area
The Skokomish River basin is situated at the southeast corner of Hood Canal, a
fjord tributary to Admiralty Inlet and the Strait of Juan de Fuca (Figure 1) (Peters et al. in
prep.). It drains approximately 622 km2 of the northern part of Mason County (SRBLIT
1994). The basin has the largest estuary and intertidal delta in the Hood Canal basin
(HCCC 2005); tidal influence in the mainstem of the Skokomish extends to near the
confluence of the South Fork and North Fork (Canning et al. 1988). The mean and
diurnal tidal ranges at Union, Washington (on the outer edge of the delta) are 2.4 m (7.8
ft) and 3.6 m (11.81 ft), respectively.
The climate in the Skokomish River basin is generally a temperate, marine
climate with wet winters and dry summers (Peters et al. in prep.). The upper portions of
the watershed receive approximately 304 cm (120 inches) of rain annually, while the
lower portions, near Hood Canal receive approximately 152 cm (60 inches) of rain
annually (Peters et al. in prep.). Almost 90% of this annual rain falls from September
through April (Canning et al. 1988), and stream flows are fed the rest of the year by snow
melt (USDA 1995). Historical daily peaks in precipitation of 15 - 17 cm (6-7 inches)
occur between November and February (Phillips 1968).
The Skokomish River originates in a steep 640 km2 drainage on the southeast side
of the Olympic Mountains (Jay and Simenstad 1996), and enters the southern-most point
of Hood Canal. The Skokomish River consists of three distinct sections, the North Fork,
South Fork, and the mainstem where the North and South Fork converge. The entire
river is approximately 128.7 km (HCCC 2005) comprised of 14.5 km of converged
mainstem, 53.6 km of North Fork and 44.3 km of South Fork (WDOE 1985). There are
8
three main tributary subbasins, the South Fork (269 km2), the North Fork (305 km2), and
Vance Creek (64 km2) (Jay and Simenstad 1996), with tributary streams totaling
approximately 434.5 km, with Vance Creek (17.8 km) being the largest (WDOE 1985).
Both the North and South Forks originate in the mountainous areas of Olympic
National Park and Olympic National Forest. The uppermost watershed is primarily
heavily timbered, while the downstream areas have been extensively logged and suffer
from heavy siltation (WDOE 1985). Approximately 80% of the South Fork subbasin has
been clear-cut since 1947 (Canning et al. 1988), and an mean of 2.8 km km-2 of logging
roads have been constructed in the areas of the South Fork subject to timber cutting (Jay
and Simenstad 1996).
The South Fork of the Skokomish River originates in the Capitol Peak region of
the southern Olympic Range and generally drains southeast for greater than 32 km, with
the upper 7.3 km cutting through very narrow, steep-sloped valleys (SRBLIT 1994). The
river bottom is primarily rubble and boulders, with some bedrock, low occurrence of
gravel riffles (WDF 1975) and is predominately poorly sorted gravels and cobbles (Jay
and Simenstad 1996). Most South Fork tributaries exhibit steep mountain stream
characteristics; narrow and confined channels, cascades and rapids, rubble and boulder
bottoms (SRBLIT 1994). Near river mile 7.0 the South Fork flows through a narrow,
deep, steep-walled canyon and abruptly opens into the broad lower river valley (SRBLIT
1994). The gradient in the valley is moderate and contains excellent gravel substrate,
however, it is unstable and influenced by erosion, channel changes and shifting gravel
bars (SRBLIT 1994).
9
The North Fork originates in the Mount Skokomish-Mount Stone region and at
river mile 28.0 meets and then flows through Lake Cushman and Lake Kokanee. The
diverted water is piped through a spillway to the City of Tacoma Power Generating
Facility on Hood Canal about 5 km north of the mouth of the river. The remaining mean
annual flow in the North Fork is approximately 3 m3s-1, which prior to diversion was near
27 m3s-1 (Canning et al. 1988). This 96% (USACE 2006) reduction of flow causes an
estimated 70% loss of sediment transport (Jay and Simenstad 1996). The upper drainage
(upstream of the dams) has precipitous gradients, numerous cascades and falls, and
contains predominantly rubble and boulder stream bottom (SRBLIT 1994). Downstream
of the dams, the North Fork is characterized by low gradients, eroded banks and heavy
siltation at the confluence, and is predominately poorly sorted gravels and cobbles (Jay
and Simenstad 1996).
The mainstem Skokomish River is approximately 16 km long and extends to a
relatively large estuary at Hood Canal (Peters et al. in prep.). The tidal influence extends
approximately 14.5 km (9 miles) upstream to the confluence of the North and South forks
(Jay and Simenstad 1996). This section of the river is in a floodplain, is relatively low
gradient, and contains sediments that are largely sand and gravel (Jay and Simenstad
1996). Overall, the Skokomish River currently contains approximately 111 river km (69
miles) of anadromous fish habitat (SRBLIT 1994).
Several anadromous and resident salmonid and other non-game fish can be found
in the Skokomish River watershed system. Twenty-three species of fish have been found
in the mainstem and South Fork of the river (Watershed Management Team 1995). Most
of these are salmonids, including Chinook salmon, coho salmon, chum salmon (O. keta),
10
steelhead, cutthroat trout, bull trout (Salvelinus confluentus), and mountain whitefish
(Prosopium williamsoni), which are common in the Skokomish River (Peters et al. in
prep). Sockeye salmon (O. nerka) and pink salmon (O. gorbuscha) were historically
found in the Skokomish River (Peters et al. in prep). Five species of sculpin (Cottus sp.)
are found in the Skokomish River, including prickly sculpin (C. asper), coast range
sculpin (C. alecticus), riffle sculpin (C. gulosus), Reticulate sculpin (C. perplexus), and
shorthead sculpin (C. confusus) (Mongillo and Hallock 1997). River lamprey (Lampetra
ayersii), western brook lamprey (L. richardsoni) and Pacific lamprey (L. tridentata) have
also been observed in the basin (Peters et al. in prep).
Four fish found in the Skokomish River are listed as threatened under the
Endangered Species Act of 1973 or have Evolutionary Significant Units (ESU) there;
Chinook salmon (a Puget Sound ESU), chum salmon (Hood Canal Summer-run),
steelhead (Hood Canal Winter Steelhead), and bull trout (NMFS 1999; USFWS 1988;
USFWS 1999). Salmonid populations within the North Fork watershed above the dams
consist of landlocked steelhead, cutthroat trout, brook trout (Salvelinus fontinalis),
sockeye salmon, mountain whitefish, and a lucastrine stock of Chinook salmon and bull
trout (Brenkman 2001). Pink salmon, spring Chinook, and early chum have been
extirpated from the South Fork of the river (WDNR 1997).
11
(g) Methods
Fish collections
Coho salmon, steelhead, and cutthroat trout were collected from tributaries, offchannel ponds, lateral backwaters, and the mainstem from January 2009 through March
2009 using seining and electrofishing techniques (Table 1 and Figure 2). These sample
sites were randomly selected from the GI study sites used in its overall biological
sampling plan (Peters et al. in prep.); they were selected using Generalized Random
Tessellation Stratified (GRTS) Spatially-Balanced Survey Designs for Aquatic Resources
(Stevens and Olsen 1999). GRTS was used to designate sites for the biological sampling
completed for the Skokomish River GI (Peters et al. in prep.). Because these selected
sites have the same spatial distribution as the stream network from which they were
drawn, measurements made at them can be used to infer conditions within the entire
network for the purposes of the GI (USACE and USFWS 2008). For a complete list of
the coordinates where fish for the diet analysis were obtained please see Table 2.
Stomach samples were collected from habitats within these stream/river reaches
or pond transects (Figure 2). River/stream reaches were centered on the points developed
from the GRTS selection process. The length of the reach where sampling was
completed was equal to ten times the bank-full width or 150 m, whichever was greatest
(Table 2). Pond samples were collected along two 150 m by 3 m transects; one nearshore and one off-shore. These two transects were centered on the GRTS selected point.
For this diet study, if a sufficient sample size was not obtained along these transects the
sample area was increased until an adequate sample size was obtained.
12
Figure 2.-Map of the study area and all sample site locations for diet study, lower Skokomish River,
January through March 2009. Each marker indicates one site except for the backwater sites. Site
identification numbers are used to identify site locations, see Table 2 for a detailed list of site location
characteristics.
Water quality measurements were collected at most sites. Sites in which
measurements were unobtainable due to equipment failure are notated by dashes in Table
2. Water quality values collected include turbidity (ntu), dissolved oxygen (DO),
temperature (°C), pH (Table 2). GPS coordinates (UTM NAD 83) were collected with a
Garmin (model GPS map76S), and bank full widths (BFW) were also collected (Table 2).
Temperature and pH measurements were collected using a Yellow Springs Instrument
(YSI) (model 60), and Dissolved Oxygen (DO) measurements were collected using a YSI
(model 85). All instruments were calibrated prior to use.
13
There are several techniques of classifying habitat. For this study the mainstem is
defined as the section of the river in which flow doesn’t branch off of the main flow of
the river, including braided channels. Braided channels are overlapping islands which
form more than two flow channels, and are generally unstable (Arend 1999a).
Tributaries are considered streams and creeks that actively transported water from the
surrounding drainage basin and converged with the mainstem of the Skokomish River.
Backwater habitat was defined as water collections that occur laterally with the mainstem
of the river, and for this study include locations in tributaries that are near its convergence
with the mainstem. Backwaters generally consist of standing or very slowly moving
water that connected to the mainstem in some manner, and is fed by underground seepage
or river flow. Ponds are considered off-channel habitat as they usually form with higher
flows and remain after flow conditions have stabilized. Ponds are generally connected to
the mainstem by relatively small egress channels that may become intermittent during
low-flow conditions (i.e. summer).
Fish were caught at night, between one to three hours after dusk because research
has shown that nighttime fish counts far exceeded daytime counts in the winter; fish
generally hide and are inactive during the daytime (Roni and Fayram 2000). In less
structurally restrictive areas a 5-mm square size mesh pole seine, with 1.8 m depth, 1.8 m
width, and 9.1 m length was used to collect fish from banks, ponds, and other shallow
and slower moving water. A Smith Root, Inc. (model LR-24) backpack electro-fisher
was used to stun fish found in various log jams and around other in-water structures
where use of the pole seine was ineffective. A Smith Root, Inc. raft Electrofisher (model
14
GPP) on a modified NRS Otter Raft was used in deeper areas (ponds) where foot access
was prohibitive. Fish were removed from the nets and placed in buckets for processing.
Fish processing and stomach content sampling
After fish collections were completed, fish were anaesthetized with MS-222, and
identified. They were measured for fork length (FL, nearest mm) and weight (nearest
0.01 g). The scale (OHaus Scout Pro, model SP402) was calibrated each sampling day
using a standard 200-g weight. A modified pesticide applicator was used to lavage fish
for stomach contents (Foster 1977). The sprayer was fitted with an adapted copper
nozzle of appropriate length and diameter; selected for fish length and girth. Stomach
contents were flushed into a 425 µm mesh sieve, and then rinsed into a whirl-pack bag.
Water-proof data labels containing site number, sample number, date, and fish
information (species, length, and weight) were also placed inside sample bags. Stomach
samples were then placed on dry ice and kept in a cooler. Fish were allowed full
recovery and returned to the location of capture. Stomach samples were transported back
to the laboratory, and placed in a freezer until later analysis.
Only fish greater than 54 mm fork length (FL) and weighing more than 1.67
grams were sampled for stomach contents because fish were anesthetized and the lavage
process needed to be expedited in order to minimize impacts to fish. More time would
have been required to change lavage nozzle heads to a size necessary to perform gastric
lavage on smaller fish. Additionally, due the small size of fish under 54 mm FL and 1.67
g, it was unclear if the lavage process would have harmed them. The nozzle selected was
the appropriate size for the majority of the fish that could be collected and lavaged with
the least amount of time and potential harm.
15
Laboratory analysis
The samples were thawed by placing the frozen whirl-pack sample bags into a
shallow tub of cold water until thawed. Stomach contents were rinsed into a Petri dish,
and then sorted with the aid of a dissecting scope (American Optical Stereo Star Model
A0569 with zoom of 0.7 x 3.0 x). Aquatic insects were identified to family, genus, and
species if possible and only to order if more specific identification was impossible. Some
items were categorized differently than the other prey items because they offered little to
no caloric value to the diet. These items included rocks, plant material, unidentifiable
inorganic matter, and Trichoptera cases, which were categorized as “other”, and were not
included in data analysis. Coleoptera presented a special issue and were classified as
terrestrial insects here, however, they are known to frequently move between aquatic and
terrestrial environments and could have been characterized as aquatic (McCafferty 1981).
Miscellaneous insect parts that were unidentifiable were distributed amongst the
other insect contents in the stomach sample according to their percent proportion of mass.
Stomach contents were assigned prey codes (Appendix A), cataloged, and preserved in
95% ethanol. McCafferty (1981) and Merritt and Cummins (1996) were referenced for
aquatic entomology identification. Ingested fish were classed to genus, and species if
possible. Fish were identified by external characteristics if they are not too degraded, or
by diagnostic bones (i.e., cleithrum and dentaries) if the fish tissue was well-digested.
Sorted contents were dabbed on absorbent towels for at least three seconds before
weighing to the nearest 0.0001 g on a balance (Denver Instrument M-220).
16
Data Analysis
Pielou’s method considers the cumulative diversity and quantity of prey items to
determine if adequate sample sizes have been collected; the purpose of using Pielou’s
method is to perform unbiased data analysis (Hoffman 1978). In Pielou’s method, Hk is
plotted versus k, the number of pooled stomachs, and as the stomachs are pooled, Hk
initially tends to increase, and if the number of pooled stomachs (k) is large enough, Hk
should level off. The following formula was applied to calculate Pielou’s method;
Hk = (1/Nk) log (Nk !/πNki)
where:
Hk = the diversity in k pooled stomachs (k = 1 to n),
Nk = the number or individuals in these stomachs, and
Nki = the number of individuals in the ith species in k pooled stomachs.
Pielou’s method was first calculated on each of the individual sample sets from each
habitat, and then all stomach samples were combined to represent each of the four habitat
types as a whole (separately for each species of predator fish). See Appendix C and D for
results of the determination of adequate sample size for individual sample sets for coho
salmon and trout, respectively.
17
To quantify prey abundance, values of percent proportion by weight (%Wi),
percent proportion by number (%Ni), and percent frequency of occurrence (%Oi) (Liao et
al. 2001; Chipps and Garvey 2007) were calculated;
%Wi =
100Wi
n
W
i 1
%Oi =
n
i 1
%Ni =
n
N
,
i
100 N i
i 1
where:
i
100Oi
O
,
,
i
n = total number of prey categories found in a given sample,
%Wi = percent proportion by weight of prey type,
%Ni = percent proportion by number of prey type,
%Oi = frequency of occurrence*.
* the count of stomachs containing a specific prey item divided by the total
number of stomachs with food in them
Proportion by number (%Ni) can allow small prey items to represent a dominant
component of the diet, while proportion by weight (%Wi) emphasizes the relative
contribution of larger prey, and frequency of occurrence (%Oi) can describe how often a
particular prey item was eaten; however, none of these alone can provide an indication of
the relative importance of prey to the overall diet (Chipps and Garvey 2007).
The four tributaries from which stomach samples were collected varied in their
physical characteristics and were separated and grouped to compare the diets between
streams with similar characteristics. For coho salmon, Vance Creek diet information was
18
compared against Hunter Creek and Swift Creek combined. Vance Creek is a valley
tributary, has its own subwatershed, and has a steep gradient. Hunter Creek and Swift
Creek were grouped because they were different from Vance Creek and were very similar
to each other; lower velocity, and lack a watershed to feed their flow. Vance Creek
would have been grouped with McTaggart Creek and compared to Swift and Hunter
Creeks if coho salmon had been captured there.
It is believed that compound indices like percent index of relative importance
(%IRI) can represent all the unique properties affecting individual measures, and capture
more information than do single, component measures; %IRI can therefore provide a
more balanced view of fish’s diets (Chipps and Garvey 2007). The %IRI values were
calculated according to Bowen (1996);
%IRI =
100 IRI i
n
IRI
i 1
,
where:
IRI = %Oi (%Ni + %Wi).
i
Comparisons here will utilize both the percent proportion by weight and the index of
relative importance; this study is primarily concerned with the overall importance of prey
items in the diet, and both characteristics are valuable. Only prey items which constituted
a significant proportion of the diet according to prey weights (%Wi) and importance
(%IRI) will be represented in the graphs, the remainders are grouped as other.
Additionally, prey items which constituted a significant proportion of the diet according
to the %IRI will be represented in tables; the remainders are grouped as other.
Stomach fullness values can be used to describe the quantity of prey items being
obtained if collected during the appropriate times within a 24 h period, depending on
19
seasonal temperature (Beauchamp et al. 2007). Stomach fullness (K f ) values for each
fish was calculated as;
K f = [wet weight of stomach contents (g)
/
wet weight of stomach contents (g) - wet weight of fish (g) ] * 100.
Diet breadth is commonly used to describe the degree of species present (Levins
1968), and was calculated to compare diet diversity. Diet breadth was calculated for each
species in each habitat type;
B=
1
,
2
pi
where pi = the proportion of the diet that is comprised of food type i.
Index values range between 1 (meaning there is only one prey type present, i.e. there is
no diet breadth) and infinity. Values less than two indicate little diet breadth (Tabor et al.
2001).
Niche overlap indices, or diet overlap indices, are often used to measure the
extent of resource overlap among different species, or to infer competition (Chipps and
Garvey 2007). Diet overlap was determined with Horn’s index of overlap because it is
the method recommended when data are expressed as biomass rather individual prey
counts (Krebs 1989; Chipps and Garvey 2007). The use of prey counts can be misleading
because prey items can be small, have low caloric content, and may require more effort to
obtain than larger prey items of equal mass. High prey counts don’t necessarily indicate
they are acquiring enough caloric sustenance. Horn’s index uses proportions by weight
20
and has been shown to be the least biased when the following are present: changing
quantities of resources, resource distribution, and uneven sample sizes (Horn 1966; Krebs
1989; Chipps and Garvey 2007).
Horn’s index values were tabulated and the most specific information possible for
prey items was used to maximize the robustness of the index values. For several prey
items identification was only possible to order or class due to their small size. Most prey
items were classable to order and family and several were identifiable to species
(Appendix E). Horn’s index values were calculated from the relative proportions by
weight;
s
2 X i * Yi
C
=
i 1
s
s
X i Yi
i 1
2
2
i 1
where:
C
= Horn’s index of diet overlap,
X i = proportion of total diet of species/habitat type X contributed by food category i,
Yi = proportion of total diet of species/habitat type Y contributed by food category i, and
S = food categories.
Overlap index values can range from 0 (no overlap) to 1 (complete overlap). Diet
overlap is commonly considered biologically significant when values exceed 0.60 (Zaret
and Rand 1971). Coho salmon diet overlap was compared between all four habitat types;
backwaters, mainstem, ponds, and tributaries. Steelhead diet overlap was compared
between backwaters and mainstem, and then compared between coho salmon diet in
21
backwaters and mainstem. Cutthroat trout diet overlap was compared between coho
salmon diet in tributaries.
Mean weights and FL’s were used to determine Fulton’s condition factor;
K = (W/L3) * 100,000
where:
K = condition factor (metric),
W = weight of fish (g),
L = length of fish (mm).
Mean weights, FL, condition factor, and stomach fullness were then compared using
nested analysis of variance (P < 0.05) to evaluate whether the effects of habitat on fish
weight, fork length, and condition factor were significant for coho salmon between
habitats types; calculated with statistical software program SAS (SAS Institute Inc.). One
sample site was excluded from mainstem, backwaters, and ponds to produce a balanced
design with equal sample size to allow a nested ANOVA to be completed; tributaries
caused this exclusion because only three tributary sites produced enough coho salmon to
compare. Multiple comparison Tukey’s Honestly-Significant-Difference Test calculated
pair-wise comparisons to determine between which population means significant
differences exist (P < 0.05) (Zar 1984, 1999).
Condition factor values are meant to represent a fish’s overall robustness
(Anderson and Neuman 1996; Cunjak and Power 1987) and to indicate the level of tissue
energy reserves a fish has; based on the assumption that a fish in good condition would
demonstrate faster growth rates, greater reproductive potential, and demonstrate higher
survival than those with lower condition factor levels, given comparable environmental
conditions (Pope and Kruse 2007). There are commonly associated factors with
condition like the season, environment, and spatial variations that can influence fish
22
condition; however, if interpreted correctly, condition factor can be used to characterize
environmental components of fish habitat (Pope and Kruse 2007). The formula applied
to calculate condition factor was not species specific; a more species specific formula
may have produced more accurate condition values. Therefore, coho salmon and trout
condition factors were not compared due to the allopatric growth differences between
species.
(h) Results
Stomach collections and site characteristics
A total of 309 salmonid stomach samples were collected from January to March
2009, which included 226 coho salmon, 64 steelhead, and 18 cutthroat trout samples
(Table 1). The early February attempts to capture coho salmon in the mainstem and
backwaters, 2/3/09 and 2/5/09 respectively, yielded few coho salmon (Table 1); steelhead
were the abundant species captured, and were alternately sampled. It was notable that
cutthroat trout were the only species of salmonid captured at McTaggart Creek. Coho
salmon were consistently captured at all pond sites; only one pond site yielded any
rainbow or cutthroat trout (Table 1).
23
Table 1.-Habitat types and sample locations of each sampling effort. The numbers in parentheses are site
identification numbers. The centers of the site transect reaches are labeled “X”; reaches were ten times the
bank full width, or 150 m, whichever was greater. Skokomish pond site SP-6/22 was one very large pond,
but contained two designated sites, SP-6 and SP-22.
Sampling
date
coho
salmon
steelhead
cutthroat
trout
Mainstem (2-31)
1/29/09
10
11
0
South Fork at Vance confluence
2/03/09
1
14
0
North Fork at confluence
2/12/09
15
0
0
South Fork above and below North Fork
confluence
2/17/09
11
0
0
North Fork mainstem at log jam in braided
channel (2-26)
3/25/09
15
1
0
52
26
0
Habitat type and sample site
MAINSTEM
# Total
TRIBUTARIES
Hunter Creek (2-39)
McTaggart Creek (2-64)
1/28/09
2/26/09
20
0
3
0
0
15
Swift Creek - upper ( 2-72)
3/03/09
15
0
0
Vance Creek - upper (2-41)
3/04/09
14
6
0
49
9
15
# Total
BACKWATERS
South Fork at Vance Creek confluence
2/02/09
5
15
0
South Fork - downstream of Vance Creek
confluence
3/05/09
13
2
0
South Fork - downstream of Vance Creek
confluence
3/10/09
13
4
0
North Fork - downstream of X (2-26)
3/25/09
15
0
0
North Fork - upstream of X (2-26)
3/25/09
15
0
0
61
21
0
# Total
PONDS
Outlet from mainstem to North Fork
Skokomish pond (SP-14)
2/05/09
2/19/09
16
18
0
0
0
0
Skokomish pond (SP-21)
2/23/09
15
8
3
Skokomish pond (SP-6/22)
2/24/09
0
15
0
# Total
64
8
3
Total Fish Collected
% Total
226
73
64
21
18
6
24
Water quality characteristics were similar for all habitats (Table 2). The site
temperatures in the mainstem, tributary, and backwater habitats ranged between 4.0 –
7.0°C and were colder than ponds, which ranged from 5.8 – 9.2 °C. Dissolved oxygen
(DO) levels at the mainstem, tributary, and backwater sites ranged from 10.3 – 15.6 mg/L
and were higher than ponds, which ranged from 7.23 – 9.0 mg/L. The pond site called
“outlet from mainstem to the North Fork” was physically different than the other three
pond sites, because it was smaller, and was closely connected to the mainstem; the other
three pond sites were relatively larger bodies of water, with lower flow velocities and
were more appropriately categorized as off-channel habitat.
The pond outlet from
mainstem to the North Fork had D.O. of 11.03, which was very similar to the other
habitat types with faster moving water.
25
Table 2.-Water quality and habitat data from the sites where fish were collected for diet analysis. The
numbers in parentheses after sample locations are site identification numbers. The centers of the site
transects are labeled “X”; transects were ten times the bank full width, or 150 m, whichever was greater.
Values collected were turbidity (Turb., ntu), dissolved oxygen (D.O., mg/L), water temperature (°C), pH,
GPS coordinates (NAD 83) (X LAT, X LONG), GPS accuracy (Acc., +/- m), bank full widths (BFW,
meters), and reach lengths (meters). Dashes indicate values which were not collected.
Habitat type and
sample site
Turb.
ntu
D.O.
mg/L
°C
pH
X LAT
X LONG
Acc.
+/-m
BFW
Reach
Length
MAINSTEM
Mainstem (2-31)
-
-
7.0
-
485432
5241226
-
-
-
South Fork at Vance Creek
confluence
1.23
15.40
4.0
7.44
-
-
-
-
-
North Fork at mainstem
confluence
3.50
10.66
5.0
6.78
483355
5240897
6.0
17.1
150
South Fork above and below
North Fork confluence
1.40
11.70
4.0
7.50
483442
5240795
6.0
53.1
531
North Fork mainstem at log
jam in braided channel (2-26)
3.40
15.58
5.9
7.60
482343
5244545
9.7
15.0
150
-
-
7.0
-
483916
5239798
11.2
12.3
150
TRIBUTARIES
Hunter Creek (2-39)
McTaggart Creek (2-64)
0.56
10.30
6.3
6.36
481959
5250255
13.5
3.0
150
Swift Creek - upper ( 2-72)
-
11.47
7.1
7.24
482812
5248130
22.0
8.4
150
Vance Creek - upper (2-41)
0.91
11.36
6.0
-
477388
5241632
21.0
30.0
300
South Fork at Vance Creek
confluence
1.23
15.40
4.5
7.44
-
-
-
-
-
South Fork - downstream of
Vance Creek confluence
0.91
11.36
6.9
-
480833
5240387
21.0
30.0
300
South Fork - downstream of
Vance Creek confluence
1.23
15.40
6.4
7.44
-
-
-
-
-
North Fork - downstream of
X (2-26)
3.40
15.58
5.9
7.60
482343
5244545
9.7
15.0
150
North Fork - upstream of X
(2-26)
3.40
15.58
5.9
7.60
482343
5244545
9.7
15.0
150
PONDS
Outlet from mainstem to the
North Fork
Skokomish pond (SP-14)
0.93
11.03
5.8
7.44
481832
5240230
5.5
-
700
-
-
-
-
487715
5239661
-
-
150
Skokomish pond (SP-21)
1.95
7.23
9.2
6.90
488188
5238991
-
-
150
Skokomish pond (SP-6)
-
9.00
8.5
6.25
487627
5240087
-
-
150
Skokomish pond (SP-22)
-
9.00
8.5
6.25
487613
5240130
-
-
150
BACKWATERS
26
Sample size
According to Pielou’s method only coho salmon were collected in sufficient
quantity to claim an adequate sample size was collected across all four of the habitat
types (Figure 3). See Appendices C and D for the results of Pielou’s methods on the
original sample sets of all fish collected, including fish greater than 100 mm FL. After
stomach samples collected from fish greater than 100 mm FL were excluded from the
analysis, Pielou’s method was re-calculated on the remaining samples (Figures 3 and 4).
Figure 3.-Results of Pielou's method to determine adequate sample size within each of the habitat types for
coho salmon; (A) mainstem, (B) tributaries, (C) backwaters, and (D) ponds. Sample sizes are indicated at
the top of each graph (n). Values for cumulative diversity (Hk) initially start out low, then increase and
level off when adequate sample size has been reached (Hoffman 1978).
27
Steelhead and cutthroat trout were also collected, but sample sizes were too small
from each habitat (Figure 4 B) to perform the same analysis performed for coho salmon.
Because the trout sample sizes were too small for each of habitat types in which they
were collected, it was not possible to estimate the diet characteristics for these species in
all habitat types. Adequate sample sizes for steelhead were only collected in mainstem
and backwaters based on Pielou’s method (Figure 4 A and C). An adequate sample size
was collected for cutthroat trout in tributaries (Figure 4 D).
Figure 4.-Results of Pielou's method to determine adequate sample size within each of the habitat types.
Steelhead in panel (A) mainstem, (B) tributaries, and (C) backwaters, and cutthroat trout in (D) tributaries.
Sample sizes are indicated at the top of each graph (n). Values for cumulative diversity (Hk) initially start
out low, then increase and level off when adequate sample size has been reached (Hoffman 1978).
Steelhead in tributaries (B) was not used in analyses because these results indicate the sample size was too
small.
28
Diet composition - Prey Abundance and Importance
Coho salmon fed primarily on benthic macroinvertebrates; aquatic insect nymphs
or larvae which comprised a mean 48% by weight and 94% by number for all habitat
types combined (Table 3). Coho salmon diets contained comparatively low proportions
of exuvia and prey items categorized as “Other”. The “Other” category is comprised of
prey items that had relatively low frequencies of occurrence. Steelhead and cutthroat
trout also fed primarily on aquatic insect nymph and larvae and consumed much higher
proportions (by weight) of prey items categorized as “Other”.
Table 3.-Percent proportion by weight (%Wi), and by number (%Ni) of aquatic and terrestrial insects. The
“Other” categories are comprised of varying combinations of Arachnida, Amphipoda, Collembola,
Copepoda, fish eggs, Hemiptera, Homoptera, Hydracarina, Hymenoptera, Lepidoptera, Mollusca, plant
material, and unidentified organic matter.
Cutthroat
Coho salmon
Steelhead
trout
%
Proportion
mainmainby weight
stem tributaries backwaters tributaries stem backwaters tributaries
Aquatic
58.88
68.56
75.99
73.08
70.22
62.73
73.08
insects
Terrestrial
19.33
26.89
15.19
4.44
0.00
0.10
4.44
insects
Exuvia
Other
%
Proportion
by number
Aquatic
insects
Terrestrial
insects
Other
9.04
12.75
0.04
4.51
3.27
5.55
0.04
22.44
2.24
27.54
1.45
35.72
0.04
22.44
97.11
96.68
97.98
98.31
94.59
99.69
98.31
2.74
5.63
3.09
6.41
1.91
3.93
1.69
3.38
0.00
5.41
0.31
0.00
1.69
3.38
Ephemeroptera comprised over 23% (by weight) of the overall salmonid diet,
with almost 18% of that being the family Baetidae (Appendix E). While Ephemeroptera,
Trichoptera, and Plecoptera were present in comparatively lower proportions (by weight),
the frequency of occurrence for these prey items was consistently high for coho salmon in
29
all habitats. Oligochaeta comprised nearly 15% (by weight) of the overall fish diets,
while Plecoptera was notable at 9% and Trichoptera was nearly 6%.
Prey categories important to coho salmon (by weight) were Ephemeroptera,
Chironomidae, Plecoptera, Trichoptera, Megaloptera, and Oligochaeta (Figure 5 and
Appendix F). Ephemeroptera had the highest proportions (by weight) of all other prey
items, which occurred in backwaters (44%). Ephemeroptera was also the prey item that
was the highest proportions (by weight) in the mainstem (17%). Chironomidae was the
most important prey item in tributaries (39%). For ponds, the prey item in highest
proportions (by weight) was Oligochaeta (32%) and Megaloptera (17%). Megaloptera
and Oligochaeta are comparably large prey items, which contributes to their high
proportions by weight; they have low frequencies of occurrence, but when found
Megaloptera appear to be of great importance (by weight) in ponds, and Oligochaeta (by
weight) in all habitats.
Several prey items had relatively high frequency of occurrence values and for
coho salmon, those prey items were Chironomidae, Diptera larvae, Ephemeroptera,
Plecoptera, Trichoptera, Megaloptera, Coleoptera, and Crustacea (Appendix F).
Chironomidae were a very frequently occurring prey item, with a frequency of
occurrence of 84% for coho salmon in tributaries and backwaters, and over 48% in
mainstem and ponds. Diptera larvae were also highly frequent, with highest frequency of
occurrence values in the mainstem (42%) and lowest in ponds (15%). Ephemeroptera
followed a similar trend being most frequently occurring in the mainstem (76%) and the
least frequent in ponds (47%). Plecoptera and Trichoptera were occurred less frequently
and values were highest in backwaters (51% and 43%) and lowest in ponds (25% and
30
13%), respectively. Coleoptera frequency of occurrence was highest in tributaries (35%)
and lowest in ponds (6%). Crustacea and Mollusca were less frequently occurring, but
most frequently occurred in ponds.
50
5
n
60
100%
n
68
Megaloptera
Proportion by weight %
Plecoptera
80%
Trichoptera
Ephemeroptera
60%
Other Diptera
Chironomidae
40%
Oligochaeta
Coleoptera
20%
Fish
Other
0%
Mainstem
Tributaries
Backwaters
Ponds
Figure 5.-Percent proportion by weight (%Wi) of major prey categories present in coho salmon diets,
lower Skokomish River, January through March 2009. Samples sizes are listed at the top of the graph. The
“Other” category includes a mixture of Arachnida, Amphipoda, Collembola, Copepoda, fish eggs,
Hemiptera, Hydracarina, Hymenoptera, Lepidoptera, Mollusca, plant material, and unidentified organic
matter.
Megaloptera had a mean frequency of occurrence value of 9.4% and it comprised
almost 5% of the coho salmon diet (by weight). Frequency of Megaloptera was highest
in ponds at 17.5 (%Oi) and lowest in the mainstem at 4.8 (%Oi) (Appendix F). Diptera
occurred in every habitat type in this diet study and most frequently occurred in the
mainstem (42%), and least frequently occurred in ponds (14%). Diptera larvae had
consistently higher frequency of occurrence values for coho in all habitats, and for
31
cutthroat trout in tributaries. The Diptera families present in the diets of these fish were
Dixidae, Ceratopogonidae, Simuliidae, and Chironomidae; together they comprise 3.3%
of the overall diets (by weight), of which the family Simuliidae comprised the majority at
2.3% (by weight).
For coho salmon in tributaries, Chironomidae comprised a substantial proportion
(by weight) in Vance Creek (47%). Trichoptera proportions (by weight) were only
slightly higher in Hunter Creek and Swift Creek combined (15%) than Vance Creek
(9%). Oligochaeta comprised substantial proportions (by weight) in Swift Creek (34%).
All other prey categories were fairly similar between creeks.
100%
14
39
23
15
Proportion by weight %
Megaloptera
80%
Plecoptera
Trichoptera
60%
Ephemeroptera
Other Diptera
40%
Chironomidae
20%
Oligochaeta
Coleoptera
0%
Vance Creek
Hunter and Hunter Creek Swift Creek
Swift Creek
Other
Figure 6.-Percent proportion by weight (%Wi) of major prey categories present in coho stomachs in
tributaries of the lower Skokomish River, January through March 2009. Samples sizes are listed at top of
graph. The “Other” category includes a mixture of Arachnida, Amphipoda, Collembola, Copepoda, fish
eggs, Hemiptera, Hydracarina, Hymenoptera, Lepidoptera, Mollusca, plant material, and unidentified
organic matter.
32
For steelhead and cutthroat trout items which were of obvious importance (by
weight) to trout were Ephemeroptera in the backwaters (34%) and Plecoptera (28%) and
Trichoptera (30%) in tributaries (Appendix G). Plecoptera and Trichoptera were
observed in the greatest proportions (by weight) in trout from the mainstem (27% and
30% respectively).
9
19
100%
12
Plecoptera
Proportion by weight %
80%
Trichoptera
Ephemeroptera
60%
Other Diptera
40%
Chironomidae
20%
Coleoptera
Other
0%
Mainstem
Backwaters
Tributaries
Steelhead
Steelhead
Cutthroat trout
Figure 7.-Percent proportion by weight (%Wi) of the major prey categories present in cutthroat trout and
steelhead, lower Skokomish River, January through March 2009. Samples sizes are listed at the top of the
graph. The “Other” category includes a mixture of Arachnida, Collembola, Copepoda, Homoptera,
Hydracarina, Mollusca, Lepidoptera, plant material, and unidentified organic matter.
33
For steelhead and cutthroat trout the prey categories Chironomidae, Diptera
larvae, Ephemeroptera, Plecoptera, Trichoptera, and Coleoptera had the highest
frequency of occurrence values between all habitats (Appendix G). Chironomidae most
frequently occurred in backwaters (75% and 89%) and the least frequently occurred in
the mainstem (20%) (Appendix G). For trout, the frequencies of occurrence values (of
prey items present in appreciable amounts) were highest in tributaries and backwaters,
and lowest in the mainstem (Appendix G). Large amounts of exuvia indicates they are
drift feeding (Tippets and Moyle 1978), and for these trout the proportions of exuvia
were less than 10 percent of any of their diets across all habitats, indicating that they were
mainly foraging in benthic areas (Table 3). For trout in this diet study, the prey items of
most importance (by weight) were Trichoptera, Plecoptera, Ephemeroptera,
Chironomidae, and Diptera larvae (Appendix G and Figure 7).
The prey items with the highest importance (%IRI) values for coho salmon were
Chironomidae and Ephemeroptera (Figure 8 and Appendix H). Chironomidae was the
most important prey item (%IRI ranged 32% – 49%) for all fish in all habitats, except the
mainstem. Ephemeroptera was the prey item of highest importance (%IRI) for coho
salmon in the mainstem (23%). Trichoptera was the prey item of highest importance
(%IRI) in the mainstem for steelhead (29%). Chironomidae %IRI for coho salmon was
highest in tributaries at 49%, and lowest in the mainstem at 18%. The importance of
Chironomidae jumps to near 50% while Ephemeroptera importance drops to 3% for coho
salmon in tributaries (Figure 8).
34
Chironomida e
Ephemeroptera
Diptera la rva e
50
% Index of Relative Importance
Mega loptera
Plecoptera
40
Trichoptera
30
20
10
0
ma instem
tributa ries
ba ckwa ters
ponds
Habitat type
Figure 8.-Percent index of relative importance of major prey items for coho salmon, in the Skokomish
River, January through March 2009.
Stomach Fullness - Diet Breadth – Diet Overlap
Mean stomach fullness values for coho salmon were not significantly different
between habitat types. However, there were significant differences for coho salmon
stomach fullness within tributary and backwater sites (Figure 9). Coho salmon mean
stomach fullness at the tributary site Swift Creek were significantly higher than Hunter
Creek (P = 0.004) (nested ANOVA and multiple comparison Tukey’s HSD; P < 0.05).
Coho salmon mean stomach fullness at the tributary site Vance Creek were also
significantly higher than Hunter Creek (P = 0.0001) (nested ANOVA and multiple
comparison Tukey’s HSD; P < 0.05). Coho salmon mean stomach fullness at the South
Fork backwater site downstream of the Vance Creek confluence were significantly higher
than the North Fork backwater site 2-26, upstream of X (nested ANOVA and multiple
comparison Tukey’s HSD; P < 0.05). Stomach fullness values were highest for coho
salmon in backwaters at 5%, and were the lowest in the mainstem sites (Figure 9).
35
Coho salmon
6
ab
Stomach Fullness %
5
4
3
b
2
b
*
b
*
a
1
*
a
a
a
a
a
a
a
0
Main
6
Tributaries
Backwaters
Ponds
Steelhead and Cutthroat trout
Stomach fullness %
5
4
3
2
1
0
Main
Tributaries
Backwaters
Steelhead
Cutthroat
Steelhead
Figure 9.-Mean percent stomach fullness for coho salmon, steelhead, and cutthroat trout (+ standard error),
in the Skokomish River, January through March 2009. Each bar represents a different sample site. Within
each habitat type, bars with different letters are significantly different (nested ANOVA and Tukey’s HSD;
P < 0.05). *Bars with no letters were sites not statistically compared.
36
Diet breadth were not significantly different between habitat types (ANOVA; P =
0.05). Coho salmon had highest diet breadth in the mainstem and ponds (Figure 10).
Tributaries provided the second highest diet breadth values for coho salmon and cutthroat
trout diets. Diet breadth levels ranged overall from approximately 2 to 10. Backwaters
provided the lowest diet breadth values for coho salmon and trout. The mainstem
appears to be providing the highest overall opportunities for diet diversity.
Coho salmon
14
Diet Breadth
12
10
8
6
4
2
0
Main
Tributaries
Backwaters
Ponds
Steelhead and Cutthroat trout
14
Diet Breadth
12
10
8
6
4
2
0
Steelhead
Cutthroat trout
Steelhead
Main
Tributaries
Backwaters
Figure 10.-Results of diet breadth determinations for coho salmon, steelhead, and cutthroat trout, in the
Skokomish River, January through March 2009. Index values range between 1 (one prey type present, i.e.
there is no diet breadth) and infinity. Values less than two indicate little diet breadth (Tabor et al. 2001).
Each bar represents a different sample site.
37
Approximately 44% of the salmonid diets exhibited overlap (4 out of 9) according
to Horn’s diet overlap index (Table 4). Coho salmon exhibited some diet overlap in
mainstem and ponds and to a lesser degree in mainstem and tributaries. The highest diet
overlap occurred between coho salmon and steelhead in backwaters and they also
overlapped to a lesser extent in the mainstem. No diet overlaps occurred between coho
salmon and cutthroat trout (Table 4). The remaining species and habitat types showed no
presence of diet overlap.
Table 4.-Results of Horn’s index of overlap for coho salmon, steelhead, and cutthroat trout, in the
Skokomish River, January through March 2009. Values greater than 0.60 are considered to have diet
overlap, and are totally overlapped with values equal to 1. Each habitat type represents all the sample sites
combined.
Species
Coho
Coho
Coho
Coho
Coho
Coho
Coho
Coho
Coho
Habitat
backwaters
mainstem
mainstem
mainstem
tributaries
backwaters
tributaries
backwaters
backwaters
Overlap?
Yes
Yes
Yes
Yes
No
No
No
No
No
Species
Steelhead
Coho
Coho
Steelhead
Cutthroat
Coho
Coho
Coho
Coho
Habitat
backwaters
ponds
tributaries
mainstem
tributaries
mainstem
ponds
tributaries
ponds
Horn's Index
0.786
0.775
0.618
0.608
0.581
0.560
0.448
0.335
0.258
Fish size and condition
Mean weights for coho salmon were not significantly different between habitat
types. However, there were significant differences for coho salmon weights within
mainstem sites, backwater sites, and pond sites. Coho salmon weights at mainstem site
2-31 were significantly higher than the South Fork (at the North Fork confluence) (P =
0.002) (nested ANOVA and multiple comparison Tukey’s HSD; P < 0.05) (Figure 11).
Coho salmon mean weights at the backwater site on the South Fork downstream of the
38
Vance Creek confluence were significantly higher than in the North Fork backwater site
2-26 (P = 0.013) (nested ANOVA and multiple comparison Tukey’s HSD; P < 0.05)
(Figure 11). Coho salmon mean weights at Skokomish pond site 6-22 were significantly
higher than those in Skokomish Pond 14 (P < 0.0001) (nested ANOVA and multiple
comparison Tukey’s HSD; P < 0.05) (Figure 11). Coho salmon mean weights at
Skokomish pond site 6-22 were significantly higher than those in Skokomish Pond 21 (P
= 0.005) (nested ANOVA and multiple comparison Tukey’s HSD; P < 0.05) (Figure 11).
8
Coho salmon
a
7
6
*
Weight (g)
5
a
b
ab
b
a
a
a
b
a
a
*
4
*
b
3
2
1
0
Main
Tributaries
Backwaters
Ponds
Figure 11.-Mean weights for coho salmon (+ standard error) in four habitat types analyzed in the diet
analysis, lower Skokomish River, Washington. Each bar represents a different sample site. There were no
significant differences in mean weights between habitat types (nested ANOVA). Within each habitat type,
bars with different letters are significantly different (nested ANOVA and Tukey’s HSD; P < 0.05). *Bars
with no letters were sites not statistically compared.
39
Coho salmon mean FL’s were not significantly different between habitat types.
However, there were significant differences for coho salmon FL within mainstem sites,
backwater sites, and pond sites. Coho salmon mean FL’s at the mainstem site 2-31 were
significantly higher than the South Fork (at the North Fork confluence) (P = 0.008)
(nested ANOVA and multiple comparison Tukey’s HSD; P < 0.05) (Figure 12). Coho
salmon mean FL’s at the backwater site on the South Fork, downstream of the Vance
Creek confluence, were significantly higher than in the North Fork backwater site 2-26
(P = 0.041) (nested ANOVA and multiple comparison Tukey’s HSD; P < 0.05) (Figure
12). Coho salmon mean FL’s on Skokomish pond site 6-22 were significantly higher
than those in Skokomish Pond 14 (P = 0.0001) (nested ANOVA and multiple comparison
Tukey’s HSD; P < 0.05) (Figure 12). Coho salmon mean FL’s on Skokomish pond site
6-22 mean weights for coho salmon were significantly higher than those in Skokomish
Pond 21 (P = 0.024) (nested ANOVA and multiple comparison Tukey’s HSD; P < 0.05)
(Figure 12).
The sites which contained the coho salmon with the greatest mean FL were main
channel #2-31, the South Fork backwater downstream of Vance Creek confluence, and
Skokomish Pond 6-22. Fish greater than 100 - mm FL were eliminated from any
analyses because the frequency distributions indicated there were two size classes present
(Appendix B). There were too few greater than 100 mm FL to analyze a second size
class.
40
Coho salmon
90
80
a
*
ab
b
a
a a
*
70
a
ab
a
b
b
*
b
FL (mm)
60
50
40
30
20
10
0
Main
Tributaries
Backwaters
Ponds
Figure 12.-Mean FL for coho salmon (+ standard error) in the four habitat types analyzed in the diet
analysis, lower Skokomish River, Washington. Each bar represents a different sample site. There were no
significant differences in mean FL between habitat types (nested ANOVA). Within each habitat type, bars
with different letters are significantly different (nested ANOVA and Tukey’s HSD; P < 0.05). *Bars with
no letters were sites not statistically compared.
Mean condition factor values for coho salmon were not significantly different
between habitat types. However, there were significant differences in the mean condition
factor values for coho salmon within mainstem sites, tributary sites, and pond sites. Main
channel site 2-31 mean condition factor values for coho salmon were significantly higher
than values for the South Fork (at the North Fork confluence) (P = 0.005) (nested
ANOVA and multiple comparison Tukey’s HSD; P < 0.05) (Figure 13). Tributary sites,
Hunter Creek and Vance Creek, mean condition factor values for coho salmon were
significantly higher than Swift Creek (P = 0.002 and P = 0.012, respectively) (nested
ANOVA and multiple comparison Tukey’s HSD; P < 0.05) (Figure 13). Skokomish
pond site 6-22 mean condition factor values for coho salmon were significantly higher
than those in Skokomish Pond 14 (P = 0.021) (nested ANOVA and multiple comparison
41
Tukey’s HSD; P < 0.05) (Figure 13). The sites which contained coho salmon with the
highest condition factor values were: the North Fork mainstem site #2-26, which was
mainly comprised of a very large log jam; tributary sites Hunter Creek and Vance Creek
which were very similar in that they did not have their own watersheds and were in the
lower flood plain portion of the watershed; the two backwater sites downstream of Vance
confluence; Skokomish pond sites 6-22 and 21, which were very similar because they
were quite large, relatively shallow ponds, with some presence of deeper pool areas.
1.3
Coho salmon
Condition factor (K)
1.25
1.2
a
b
a
a
*
b
1.15
a
a
a
*
a
ab
*
ab
b
1.1
1.05
1
Main
Tributaries
Backwaters
Ponds
Figure 13.-Mean condition factor for coho salmon, cutthroat trout, and steelhead (+ standard error), lower
Skokomish River, January through March 2009. Each bar represents a different sample site. There were
no significant differences in mean condition factor between habitat types (nested ANOVA). Within each
habitat type, bars with different letters are significantly different (nested ANOVA and Tukey’s HSD; P <
0.05). *Bars with no letters were sites not statistically compared.
42
Steelhead mean weights were slightly higher in the mainstem compared to
backwaters and cutthroat trout in tributaries (Figure 14). The mean FL and condition
factors did not vary much between cutthroat trout and steelhead. Sample sizes for these
collections were too small to perform equal sample size nested ANOVA.
Steelhead and Cutthroat trout
9
8
7
Weight (g)
6
5
4
3
2
1
0
100
90
80
FL (mm)
70
60
50
40
30
20
10
0
1.4
1.2
Condition factor (K)
1
0.8
0.6
0.4
0.2
0
Steelhead
Cutthroat trout
Steelhead
Main
Tributaries
Backwaters
Figure 14-Mean weight, FL, and condition factor for cutthroat trout and steelhead (+ standard error) lower
Skokomish River, January through March 2009.
43
(i) Discussion and Conclusions
The winter diets of coho salmon, steelhead, and cutthroat trout consisted primarily
of benthic macroinvertebrates (aquatic insect nymphs or larvae) and contained very few
terrestrial insects or fish. Aquatic insects in Skokomish River fish were a mean of 65%
by weight. The availability of terrestrial insect’s in forest streams declines to nearly zero
during winter, while aquatic insect biomass generally peaks from December to July
(Nakano and Murakami 2001). This may explain why aquatic invertebrates are more
important than terrestrial invertebrates in the diets of Skokomish River fish, and also for
brook trout (Salvelinus fontinalis) and brown trout (Salmo trutta) in Canadian streams
during the winter (Cunjak and Power 1987). The quantities of aquatic insects ingested by
Skokomish River juvenile fish were low in comparison to other systems. For example,
aquatic insects comprised approximately 75% of juvenile coho salmon diets during early
March in three northern California streams (Gonzales 2006). The diets of Skokomish
River fish are consistent with expected winter food availability; the result of this diet
analysis demonstrates that these fish are eating the most abundant food source, benthic
macroinvertebrates.
Aquatic insects in the Skokomish River fish were 97% (by count). Proportions by
number values are not used to discuss prey importance in this diet study because they are
not reliable indicators of diet characteristics and do not accurately represent the diet. Fish
could consume large quantities of prey items which could comprise relatively low caloric
content (i.e. Chironomidae larvae). If prey importance were based on proportions by
numbers of these types of prey items it would appear the fish’s diets were acquiring
copious amounts of food, which is not the case based on the results of this diet study.
44
The composition of a fish’s diet tends to represent prey availability (Cada et al.
1986). Invertebrate prey has been shown to dominate steelhead and cutthroat trout diets
because it is usually abundant (Milick 1977; Casne 1975) and takes the least amount of
energy to obtain. Because the diets of the fish in this diet study primarily consist of
benthic macroinvertebrates, it is likely that they are the most abundant prey items
available during the winter months in these areas of the Skokomish River. During the
winter many terrestrial (i.e., Trichoptera, Ephemeroptera, Diptera, Plecoptera, Coleoptera
and Megaloptera) insects are in larval or transitional stages and generally occupy a niche
in littoral zones or benthic areas (McCafferty 1981). Ephemeroptera and Chironomidae
are the prey items of most importance to the diets of these juvenile salmonids, with
Plecoptera and Trichoptera following closely behind (Appendix H).
Compound indices like percent index of relative importance (%IRI) provide a
more balanced view of a fish’s diet (Chipps and Garvey 2007). For coho salmon,
Chironomidae had highest importance values in tributaries (49%), and lowest values in
the mainstem (18%), while the reverse was true for Ephemeroptera (Appendix H). The
importance of all other major prey types remained relatively similar across habitat types.
Ephemeroptera and Chironomidae are very similar in the important roles they play in the
diets of many fish species; all stages are important to fish diets (McCafferty 1981). In
tributaries, there was a notable decline in the importance of Ephemeroptera with a
simultaneous increase of Chironomidae importance (Figure 8). This could indicate that
either the environment was more suitable for Chironomidae, or that the fish were
selecting for Chironomidae and not Ephemeroptera.
45
Chironomidae larvae are the most widely adapted family of Diptera and are likely
to be found in almost all inland waters (McCafferty 1981). Chironomidae and Simuliidae
were present in virtually all habitats and fish sampled in this diet analysis. Chironomidae
had the highest importance values in tributaries, backwaters, and ponds. Chironomidae
were more abundant in tributaries for coho salmon (~39 %) and cutthroat trout (~7 %) in
terms of proportions by weight (Figure 5 and 7). Chironomidae may have high
importance in these fish’s diets because they were more available, exist in relatively high
densities, and may be easier to capture compared to other more mobile prey items.
Diet composition and stomach fullness can change dramatically over a 24-h
period, and similar research shows that dusk samples capture peak stomach fullness
values and provide the most representative diet samples (Beauchamp et al. 2007). The
mean stomach fullness of Skokomish River juvenile salmonids was less than 20% for all
fish, and these fish were collected between dusk and midnight. These stomach fullness
values are lower than similar winter studies in which the mean stomach fullness ranged
between 50% - 67% during winter (Cunjak and Power 1987). Winter temperatures slow
evacuation rates and have been shown to limit brook trout to single daily stomach filling
and emptying in southern Ontario, Canada (Cunjak et al. 1987).
Coho salmon in backwaters of the Skokomish River had the highest levels of
stomach fullness at near 5% (Figure 9). In backwaters, high stomach fullness values
could indicate that less effort can be extended while acquiring greater amounts of food
there, however, that is assuming the fish captured were acquiring this food in habitats in
which they were captured. Alternately, the high stomach fullness values in backwaters
could suggest that the fish are utilizing these areas after feeding to rest and maintain low
46
metabolic rates while they digest their food. Coho salmon mean stomach fullness in the
backwater site on the South Fork, downstream of Vance Creek confluence, were
significantly higher than the North Fork backwater site 2-26, upstream of X (P < 0.05)
(nested ANOVA and multiple comparison Tukey’s HSD; P < 0.05).
Coho salmon in the tributaries Vance Creek and Swift Creek had significantly
higher stomach fullness values than Hunter Creek (P = 0.0001, and P = 0.004,
respectively) (nested ANOVA and multiple comparison Tukey’s HSD; P < 0.05). Hunter
Creek and Swift Creek occupy the lowland flood plain area, and do not have their own
steep drainage area and watershed like Vance Creek does. However, Hunter Creek likely
provides lesser opportunity to forage feed due to the invasive species which occupy the
majority of its edge areas. While tributaries, backwaters, and ponds provided the highest
values for stomach fullness they offered the lowest values of diet breadth (Figure 10).
Low diet breadth values imply that these habitat types offer limited levels of prey
diversity or that the fish have selective diets during the winter months, January through
March. Diet breadth values were highest for coho salmon in the mainstem, followed by
tributaries and ponds; backwaters offered the lowest values (Figure 10); suggesting that
mainstem and tributaries may offer more diet diversity. Sample sizes were relatively
similar for coho across all four habitat types; therefore, it is unlikely that diet breadth is
influenced by sample size. However, prey items were in larval/transitional stages, and
were quite small; classification to more specific levels was not always possible and this
may have caused diet breadth to be underestimated. Although backwaters and ponds
lacked diversity compared to the mainstem and tributaries, they may compensate for it by
providing higher levels of prey biomass. Higher abundance of prey biomass may be
47
more valuable because fish growth rate is limited by food supply and consumption rate
(Martin 1983).
Coho salmon diets overlapped between the mainstem and ponds and between the
mainstem and tributaries (Table 4). This overlap suggests that the mainstem and
tributaries, and the mainstem and ponds are providing similar levels of diet diversity.
Significant levels of diet overlap are often attributed to higher abundance levels of
invertebrate prey availability, especially when seasonal peaks in prey availability are
occurring. The fact that little diet overlap occurred between other habitat types may be
indicative of low prey abundance levels in this system, however, many prey items were
too small to identify, and diet overlap values could have been underestimated. The low
occurrence of diet overlap could also be attributed to the combination of winter
conditions and overall poor habitat health because they can reduce the ability of these
habitats to sustain stable forage food.
Overall, the health and well-being of coho salmon in this system appear to be
relatively similar and typical for winter conditions. There were no significant differences
in condition factor between habitat types (nested ANOVA and multiple comparison
Tukey’s HSD; P < 0.05). The condition factor for coho salmon was similar between all
habitat types, with a mean condition of 1.17 (Figure 13). Condition factor values for
cutthroat trout and steelhead were also very similar, and didn’t differ much between
habitat types; however, both trout species had lower condition values than coho salmon.
Similar research found that the condition of brook trout in winter was typically below 1
and that low condition factors in winter were not a function of reduced food availability
or quality, but was instead due to the fish’s inability to digest and assimilate more food
48
(Cunjak et al. 1987). A species specific formula that considers their differing growth
patterns may have been appropriate to compare between coho salmon and trout condition.
Condition factor is a measure of a fish’s health, which is related to a fish’s amount of
energy stores, and small fish with lower lipid stores can exhaust their energy reserves
earlier, experiencing high mortality rates sooner than larger fish with greater lipid content
(Biro et al. 2004).
Although there were no significant differences in coho salmon mean weight and
FL’s between habitat types (nested ANOVA and multiple comparison Tukey’s HSD; P <
0.05), the mean weights and FL’s in the mainstem were larger (Figures 11 and 12,
respectively). Larger coho salmon mean weights and FL’s in mainstem may suggest that
that only the healthier, more fit, fish in the Skokomish River system can sustain activity
in the mainstem, and are more capable of foraging in these areas of greater forage food
diversity, although similar research suggests otherwise. Smolts originating from ponds
have been shown to be generally larger than those from tributaries (Peterson and Reid
1984; Cederholm and Scarlett 1991). Peterson and Reid (1984) also found that small size
may reflect the conditions of their past spring-summer rearing environment, and they
could grow more rapidly when given a suitable environment (Chapman 1962; Mason
1969)
Overall, the mean FL of coho salmon in this diet study was slightly lower than
typical means. Coho smolts in Washington State are usually a mean FL of 89 to 129 mm;
smolting generally occurs in late spring (Wydowski and Whitney 2003). Coho salmon in
winter California streams are generally a mean FL 73 mm in early March (mean 4.5 S.E.)
(Gonzales 2006). The coho salmon in the Skokomish River were pre-smolts and
49
averaged less than 76 mm FL across all habitat types (minimum 54 mm, maximum 100
mm, and median of 72 mm), and these values may have been biased by the fact that fish
smaller than 54 mm were not sampled. These fish may reach mean average smolt sizes
in this area by late spring, but further studies would be necessary to determine if this were
occurring. The relatively small sizes of these juvenile salmonids could indicate poor
habitat health year-round, an insufficient prey food base during winter, or that their
growth may soon be enhanced when spring insects emerge. There were no significant
differences in mean coho salmon FL between habitat types (nested ANOVA and multiple
comparison Tukey’s HSD; P < 0.05).
This study of the winter feeding ecology of juvenile salmonids in the lower
Skokomish River provides small insight into a much larger ecological system.
Substantial habitat degradation has occurred and is exhibited by the small size of the
juvenile salmonids inhabiting it during the winter months. Because freshwater fish diets
tend to shift primarily to benthic foraging in the colder weather, adequate supplies of
aquatic insects and access to areas of refuge will be necessary to sustain them. Exuvia,
discarded exoskeleton, is usually acquired through drift feeding, and relatively large
amounts of it in the diet can indicate that drift feeding is the primary means of acquiring
prey (Tippets and Moyle 1978), especially in the mainstem. These diets exhibited low
amounts of exuvia, indicating that these fish are not primarily drift feeding and are
relying on benthic foraging opportunities in the substrate. Therefore, maintaining healthy
undisturbed habitat which can sustain juvenile salmonids and their prey base is critical to
their survival in winter.
50
More accurate results could be produced if more sites were sampled in future diet
studies. There was a lot of variability within the habitat types sampled in this diet study;
as demonstrated by the figures for coho salmon weight, length, condition factor, and
stomach fullness. Future research of this system should perform prey availability
assessments in addition to diet sampling. Collecting multiple years of diet comparisons
during the winter could also provide better descriptions of how sustainably these habitat
types can provide food for juvenile salmonids over time. Adhering to a consistent
collection time, between dusk and dawn, of stomach samples would prevent the
possibility of obtaining stomach samples contain partially digested contents. And finally,
larger sample sizes would provide better interpretive and comparative ability, ensuring
the analyses were adequately representing the populations’ characteristics.
(j) Management Implications and Habitat Recommendations
The conditions on the Skokomish River have been gradually accumulating since
the 1800’s and are not likely to be reversed in any short length of time. Conditions are
continually accentuated by a combination of factors; extensive logging in the lower
watershed; heavily introduced sediment from logging; reduction in sediment transport
and consequential river bed aggradation; transformation of lowlands to agricultural areas;
regularly frequent and increasing intensity of flooding; and the present installation of
levees, dikes, and setbacks that intensify the effects of the flooding. Much more
restoration is needed to provide better habitat for these fish and their food sources in
these systems.
51
It is well established that healthier habitat is better equipped in sustaining overwintering populations of juvenile salmonids. Pre-settlement characteristics of the
Skokomish Valley describe it as densely covered by shrubs and trees associated with wet
or periodically flooded soils and surveys performed in the 1860’s describe few portions
of the valley as inundated or swampy (Canning et al. 1988). There is evidence that the
lowland reaches of the Skokomish Valley was characterized by multiple channels,
numerous sloughs and old side channels, and an abundance of snags and log jams, with
sluggish stream flow through multiple channels (Canning et al. 1988). This may explain
why it was once prime spawning habitat for salmon. Access to healthy habitat, including
side-channels and multiple area of refuge are critical to their survival.
Although Quinn (2005) and others state that several factors support smolt survival
and growth, the major contributors are water flow, lengthier streams, lower gradients
causing more pools in the streams, and most importantly healthy habitat and food
availability. Scarcity of suitable over-wintering habitat can cause fish to migrate long
distances (Peterson 1982a). Ocean survival of salmon has been shown to be closely
linked with the early–life stages of growth and size during freshwater rearing (Bilton et
al. 1982; Holtby and Healey 1990; Nislow et al. 1999), suggesting that a larger body size
is advantageous (Olegario 2006). Access into areas of refuge appears to determine the
long-term survival of these fish.
Healthy habitat needs are demonstrated by the fact that juvenile salmonids prefer
habitat with certain combinations of depth, velocity, and other physical characteristics
(Quinn 2005). Constructed complex habitat can be created by adding large woody debris
to newly created alcoves and dammed pools; these additions showed a significant
52
increase in over-winter survival of juvenile coho salmon in treatment streams, and
increases in downstream migrant numbers (Solazzi et al. 2000). Coho salmon have been
shown to selectively use deeper, slower water characteristic of pools rather than
shallower, faster moving water (Healy and Lonzarich 2000). Additionally, juvenile
salmonids shift from using both pool and riffle habitat to predominantly deeper water
depths, as found in pool habitat, during winter conditions in small streams (Hartman
1965; Bustard and Narver 1975; Bisson and Nielsen 1983; Murphy et al. 1984).
Survey of Washington state creeks showed that coho salmon were most likely to
be found in pools, while other salmonid species preferred shallower, faster water, or
intermediate conditions (Bisson et al. 1988). Ponds can provide over-wintering fish
benign thermal refuge during the harshest time of year, when their lipid reserves are most
constrained. However due to intense predation by avian and mammalian predators
(Peterson 1982b), installation of ponds will need to be specifically created to prevent
predation, provide ample cover, refuge and depth. Fish will need to be able to safely
forage in the benthos areas while taking refuge in the deeper portions; deeper parts of the
ponds tend to have lower benthos densities (Peterson 1982b). Ponds are capable of
providing ample high-quality detrital base for insect production and rich invertebrate
fauna is directly associated with aquatic macrophytes (Hodkinson 1975). It is possible to
maximize both survival and growth of overwintering fish by combining the productivity
of shallow ponds with the cover of a riverine environment (Peterson 1982b)
Stream-restoration projects that increase the quantity of large woody debris
(LWD) and pools tend to increase coho density but decrease that of other species (Roni
and Quinn 2001). Productive sites for fish tend to possess hard waters with relatively
53
high inorganic nutrient concentrations; moderate temperatures, especially in spring-fed
streams where temperatures are buffered by groundwater inputs year-round; relatively
low vegetative canopy coverage allowing ample sunlight to reach the streams and
abundant macrophytes and mosses, or dense growths of filamentous algae (Bisson and
Bilby 1998). Well developed overhanging vegetation in a riparian area can enhance the
input of terrestrial invertebrates, and the presence of meandering and/or braided stream
channels can be expected to increase the supply of emerging aquatic insects per unit area
of forest (Nakano and Murakami 2001). Additionally, flood plains have been shown as
important in the rearing of juvenile salmonids; salmon increased in size substantially
faster in seasonally inundated agricultural floodplain than in the river, suggesting better
growth rates (Sommer et al. 2001), and larger size is advantageous to ocean survival
(Bilton et al. 1982; Holtby and Healey 1990; Nislow et al. 1999; Olegario 2006). The
Skokomish River has all these characteristics or can possess them to some degree.
The Corps has proposed several options of remediation for the Skokomish River,
including levee and dike removals, setbacks, sediment control structures, re-opening of
side channels which experience sub-surface flow, riparian planting, and dredging portions
of the mainstem (USACE and USFWS 2008). All of these actions could be beneficial if
applied appropriately. However, any actions that disturb salmonid habitat, at any time of
the year will likely have a substantial effect on their prey food base, especially during the
winter. Disturbing the lotic and lentic areas could potentially eliminate the prey food
base during their over-wintering period.
Dredging could have a substantial impact on overwintering juvenile salmonids.
Dredging may isolate critical off-channel habitat or these habitats could potentially be
54
eliminated. If dredging is performed, it must be done such that juvenile salmonids can
still access tributaries and off-channel ponds (Peters pers. comm. 2010). In addition,
dredging depth will influence groundwater levels, which may cause off-channel ponds to
become shallower (and potentially more productive) or completely dewater (Peters pers.
comm. 2010). Backwater areas will likely be completely eliminated if dredging is
completed and will likely require re-creation as part of the dredging process (Peters pers.
comm. 2010).
Flooding is now occurring more frequently on the Skokomish River and access
into areas of refuge is essential during times of high flows. The river already has lowlevels of channel connectivity caused in part by the intermittent level of remediation and,
any additional channel elimination could be devastating to the fish populations. A food
web analysis between birds and freshwater fishes shows that the loss or degradation of
one habitat has a more detrimental effect on neighboring communities that previously
recognized (Nakano and Murakami 2001).
My recommendation is that remediation activities consider the food and habitat
needs of juvenile salmonids during all seasons, but particularly during winter. Peterson
and Reid (1984) suggest that providing additional overwintering habitat within the lower
reaches of stream systems may provide refuge to large numbers of resident and
immigrant coho salmon. Considering that all the habitat types assessed in this diet study
are sustaining these fish to some degree during winter will ensure that remediation efforts
are successful at creating more suitable habitat for them.
55
Several options could alleviate pressure to these populations, including
performing relatively small projects over time which would allow the disturbed habitats
to achieve some form of stability before other areas of refuge are disturbed. Also, if
dredging and other major alterations must occur which further reduce channel
connectivity, prior installation of new areas of refuge should be performed and can be
very beneficial if created properly. Installing beaded channels, as described by
Cederholm and Scarlett (1991), before remediation projects commence is a low-cost
technique which has been shown to successfully enhance winter habitat for coho salmon.
These beaded channels contained a system of ponds which considerably increased the
overwinter survival of juvenile coho salmon. These areas must be allowed enough time
to build some level of ecosystem stability, measured by the presence of indicator species,
before other large remediation efforts commence.
56
(k) Literature Cited
Amato, C. 1996. Historical changes affecting freshwater habitat of coho salmon in the
Hood Canal Basin, pre -1850 to the present. Point No Point Treaty Council,
Kingston, WA.
Anderson, R. O., and R. M. Neumann. 1996. Length, weight, and associated structural
indices. Pages 447-482 in B. R. Murphy and D. W. Willis, editors. Fisheries
techniques, 2nd edition. American Fisheries Society, Bethesda, Maryland.
Arend, K. K. 1999a. Classification of streams and reaches. Pages 57-74 in M. B. Bain
and N. J. Stevenson, editors. 1999. Aquatic habitat assessment: common
methods. American Fisheries Society, Bethesda, Maryland.
Arend, K. K. 1999b. Microhabitat identification. .Pages 75-103 in M. B. Bain and N. J.
Stevenson, editors. 1999. Aquatic habitat assessment: common methods.
American Fisheries Society, Bethesda, Maryland.
Bain, M. B., and N. J. Stevenson, editors. 1999. Aquatic habitat assessment: common
methods. American Fisheries Society, Bethesda, Maryland.
Beauchamp, D. A., D. H. Wahl, and B. M. Johnson. 2007. Predator-Prey Interactions.
Pages 765-842 in C. S. Guy and M. L. Brown, editors. 2007. Analysis and
interpretation of freshwater fisheries data. American Fisheries Society, Bethesda,
Maryland.
Bilton, H. T., R. B. Morley, A. S. Coburn. 1982. Time and size at release experiment:
four releases of three size categories of juvenile Chinook salmon from the
Quinsam Hatchery in the spring of 1982. Nanaimo, B.C. Department of
Fisheries and Oceans, Fisheries Research Branch, Pacific Biological Station,
1983.
Biro, P. A., A. E. Morton, J. R. Post, E. A. Parkinson. 2004. Over-winter lipid depletion
and mortality of age-0 steelhead (Oncorhynchus mykiss). Canadian Journal of
Fisheries and Aquatic Sciences 61:1513-1519.
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 N. B. Armantrout, editor.
Acquisition and utilization of aquatic habitat inventory information. American
Fisheries Society, Bethesda, Maryland.
Bisson, P. A. and J. L. Nielsen. 1983. Winter habitat utilization by salmonids in Pacific
Northwest streams: the role of channel morphology and woody debris.
Weyerhaeuser Company, Technology Center.
57
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.
Transactions of the American Fisheries Society 117:262-273.
Bisson, P. A., and R. E. Bilby. 1998. Organic matter and trophic dynamics. Pages 373398 in R. J. Naiman and R. E. Bilby, editors, River ecology and management.
Springer-Verlag, New York.
Bortelson, G. C., M. J. Chrzastowski, and A. K. Helgerson. 1980. Historical changes of
shoreline and wetland at eleven major deltas.
Bowen, S. H. 1996. Quantitative description of the diet. Pages 513-522 in B. R.
Murphy and D. W. Willis, editors. Fisheries techniques, 2nd edition. American
Fisheries Society, Bethesda, Maryland.
Brannon, E. L. and E. O. Salo, editors. 1981. Salmon and trout migratory behavior
symposium. School of Fisheries, Seattle, Washington.
Brenkman, S. J. 2001. Spawning migration of lucastrine-adfluvial bull trout in a natural
area. Transactions of the American Fisheries Society 130:981-987.
Bustard, D. R. and D. W. Narver. 1975. Aspects of the winter ecology of juvenile coho
salmon (Oncorhynchus kisutch) and steelhead trout (Salmo gairdneri). Journal of
Fisheries Research Board of Canada 32:667-679.
Cada, G. F., J. M. Loar, and D. K. Cox. 1987. Food and feeding preferences of rainbow
and brown trout in southern Appalachian streams. American Midland Naturalist
117:374-385.
Canning, D. J., J. L. Randlette, W. A. Haskins. 1988. Skokomish River comprehensive
flood control management plan. Washington Department of Ecology, Report 8724, Olympia, Washington.
Casne, S. R. 1975. Production and food of salmonid populations in three sections of the
Cedar River, Washington. Master’s thesis, University of Washington, Seattle.
Cederholm, C. J. and W. J. Scarlett. 1982. Seasonal immigrations of juvenile salmonids
into four small tributaries of the Clearwater River, Washington. Pages 1977-1981
in E.L. Brannon and E. O. Salo, editors. 1981. Salmon and trout migratory
behavior symposium.
Cederholm, C. J., R. E. Bilby, P. A. Bisson, T. W. Bumstead, B. R. Fransen, W. J.
Scarlett, and J. W. Ward. 1997. Response of juvenile coho salmon and steelhead
to placement of large woody debris in a coastal Washington stream. North
American Journal of Fisheries Management 17:947-963.
Chapman. D. W. 1962. Aggressive behavior in juvenile coho salmon as a cause of
emigration. Journal of the Fisheries Research Board of Canada 19(6):1047-1080.
58
Chipps, S. R., and J. E. Garvey. 2007. Assessment of diets and feeding patterns. Pages
473-514 in C. S. Guy and M. L. Brown, editors. 2007. Analysis and
interpretation of freshwater fisheries data. American Fisheries Society, Bethesda,
Maryland.
Cunjak, R. A., and G. Power. 1987. The feeding and energetics of stream-resident trout
in winter. Journal of Fish Biology 31:493-511.
Cunjak, R. A., R. Allen Curry, G. Power. 1987. Seasonal energy budgets of brook trout
in streams: implications of a possible deficit in early winter. Transactions of the
American Fisheries Society 116:817-828.
Foster, J. R. 1977. Pulsed gastric lavage: an efficient method of removing the stomach
contents of live fish. Progressive Fish-Culturalist 39:166-169.
Gonzales, E. 2006. Diet and prey consumption of juvenile coho salmon (Oncorhynchus
kisutch) in three northern California streams. Master’s thesis, Humboldt State
University.
Groot, C., and L. Margolis. 1991. Pacific salmon life histories. University of British
Columbia Press, Vancouver, British Columbia.
Guy, C. S., and M. L. Brown, editors. 2007. Analysis and interpretation of freshwater
fisheries data. American Fisheries Society, Bethesda, Maryland.
Hartman, G. F. 1965. The role of behavior in the ecology and interaction of under
yearling coho salmon (Oncorhynchus kisutch) and steelhead trout (Salmo
gairdneri). Journal of Fisheries Research Board of Canada 22:1035-1081.
Healy, B. D., and D. G. Lonzarich. 2000. Microhabitat use and behavior of
overwintering juvenile coho salmon in a Lake Superior tributary. Transactions of
the American Fisheries Society 129:866-872.
Hodkinson, I. D. 1975. Energy flow and organic matter decomposition in an abandoned
beaver pond ecosystem. Oecologia (Berlin) 21:131-139.
Hoffman, M. 1978. The use of Pielou’s method to determine sample size in food
studies. Pages 56–61 in Gutshop ’78 fish food habits studies; Proceedings of the
2nd Pacific Northwest technical workshop, October 10-13, 1978, Lake
Wilderness Conference, Maple Valley, California.
Holtby, B. L., and M. C. Healey. 1990. Sex-specific life history tactics and risk-taking
in coho salmon. Ecology 71(2): 678-690.
HCCC (Hood Canal Coordinating Council). 2005. Salmon habitat recovery strategy for
Hood Canal & eastern Strait of Juan de Fuca. Version 09-2005. Poulsbo,
Washington.
59
Horn, H. S. 1966. Measurement of “overlap” in comparative ecological studies.
American Midland Naturalist 100(914):419-424.
Jay, D. A., and C. A. Simenstad. 1996. Downstream effects of water withdrawal in a
small, high-gradient basin: erosion and deposition on the Skokomish River delta.
Estuaries 19(3):501-517.
Krebs, C. J. 1989. Ecological methodology. Harper & Row, New York.
Levins, R. 1968. Evolution in changing environments: some theoretical explorations.
Princeton University Press, Princeton, New Jersey.
Liao, H., C. L. Pierce, and J. G. Larscheid. 2001. Empirical assessment of indices of
prey importance in the diets of predacious fish. Transactions of the American
Fisheries Society 130:583-591.
Martin, D. J. 1983. Growth, food consumption, and production of cutthroat trout in
relation to food supply and water temperature. Pages 135-144 in J. M. Walton
and D. B. Houston, editors. Proceedings of the Olympic wild fish conference,
March 23-25, 1983. Fisheries Technical Program, Peninsula College, Port
Angeles, Washington.
Martin, D. J. 1985. Production of cutthroat trout (Salmo clarki) in relation to riparian
vegetation in Bear Creek, Washington. Doctoral dissertation, University of
Washington, Seattle.
Mason, J. C. 1976. Response of underyearling coho salmon to supplemental feeding in a
natural stream. Journal of Wildlife Management 40(4):775-778.
McCafferty, P. W. 1981. Aquatic entomology: The fishermen’s and ecologist’s
illustrated guide to insects and their relatives. Science Books International,
Boston, Massachusetts.
McDonald, B. S., G. W. Mullins, and S. Lewis. 1991. Macroinvertebrates as indicators
of stream health. The American Biology Teacher, Volume 53, No. 8.
Merritt, R. W., and K. W. Cummins. 1996. An introduction to the aquatic insects of
North America. 3rd edition. Kendall/Hunt Publishing Company. Dubuque,
Iowa.
Milick, J. G. 1977. Ecology of benthic insects of the Cedar River, Washington.
Doctoral dissertation, University of Washington, Seattle.
Mongillo, P. E. and M. Hallock. 1997. Distribution and habitat of native nongame
stream fishes of the Olympic Peninsula. Technical Report #FRD 97-05,
Washington Department of Fish and Wildlife, Olympia, Washington.
60
Murphy, M. L., J. Heifetz, S. W. Johnson, K. V. Koski, and J. F. Thedinga. 1984. Role
of large organic debris as winter habitat for juvenile salmonids in Alaska streams.
Pages 251-262 in Proceedings of the annual conference western association of
fish and wildlife agencies.
Murphy, B. R. and D. W. Willis, editors. 1996. Fisheries techniques, 2nd edition.
American Fisheries Society, Bethesda, Maryland.
Nakano, S., and M. Murakami. 2001. Reciprocal subsidies; dynamic interdependence
between terrestrial and aquatic food webs. Proceedings of the National Academy
of Science 98:166-170.
NMFS (National Marine Fisheries Service). 1999. Endangered and threatened wildlife
and plants; listing of nine evolutionary significant units of Chinook salmon, chum
salmon, sockeye salmon, and steelhead. Final rule. Federal Register
64(57):41835-39.
Nislow, K. H., C. L. Folt, and D. L. Parrish. 1999. Favorable foraging locations for
young Atlantic salmon: application to habitat and population restoration.
Ecological Applications 9(3):1085-1099.
Olegario, A. O. 2006. Over-wintering diet, growth, and prey availability to juvenile
coho salmon (Oncorhynchus kisutch) in the west fork Smith River, Oregon.
Master’s thesis, Oregon State University.
Peters, R., M. Li, L. Wright, K. Denton, M. Liermann, and J. Duda. In prep. Biological
sampling in the Skokomish River Basin, Washington: Army Corps of Engineers
General Investigation. Prepared for the U. S. Army Corps of Engineers, Seattle,
WA.
Peters, R. 2010. Personal communication. March 20, 2010.
Peterson, N. P. 1982a. Immigration of juvenile coho salmon (Oncorhynchus kisutch)
into riverine ponds. Canadian Journal of Fisheries and Aquatic Sciences 39:13081310.
Peterson, N. P. 1982b. Population characteristics of juvenile coho salmon
(Oncorhynchus kisutch) overwintering in riverine ponds. Canadian Journal of
Fisheries and Aquatic Sciences 39:1303-1307.
Peterson, N. P. and L. M. Reid. 1984. Wall-based channels: their evolution, distribution,
and use by juvenile coho salmon in the Clearwater River, Washington. Pages
215-225 in J. M. Walton and D. B. Houston, editors. Proceedings of the Olympic
Wild Fish Conference, March 23-25, 1983. Fisheries Technical Program,
Peninsula College, Port Angeles, Washington.
61
Phillips, E. L. 1968. Washington climate for these counties: King, Kitsap, Mason, and
Pierce. Publication E.M. 2734. Cooperative Extension Service, College of
Agriculture, Washington State University, Pullman.
Post, J. R., Parkinson, E. A., and Johnston, N. T. 1999. Density dependent processes in
structured fish populations: assessment of interaction strengths in whole-lake
experiments. Ecological Monographs 69:155–175.
Pope, K. L., and C. G. Kruse. 2007. Condition. Pages 423-471 in Guy, C. S. and M. L.
Brown, editors. 2007. Analysis and interpretation of freshwater fisheries data.
American Fisheries Society, Bethesda, Maryland.
Quinn, T. P. 2005. The behavior and ecology of Pacific salmon and trout. University of
Washington Press. Seattle, Washington.
Robinson, C. T., S. Aebischer, and U. Uehlinger. 2004. Immediate and habitat-specific
responses of macroinvertebrates to sequential, experimental floods. Journal of
North American Benthological Society 23(4):853-867.
Roni, P., and A. Fayram. 2000. Estimating winter salmonid abundance in small western
Washington streams: a comparison of three techniques. North American Journal
of Fisheries Management 20:683-692.
Roni, P., and T. P. Quinn. 2001. Density and size of juvenile salmonids in response to
placement of large woody debris in western Oregon and Washington streams.
Canadian Journal of Fisheries and Aquatic Sciences 58:282-292.
SAS Statistical Software. SAS Institute Inc. Cary, North Carolona.
Schultz, E. T., and Conover, D. O. 1999. The allometry of energy reserve depletion: test
of a mechanism for size-dependent winter mortality. Oecologia (Berlin) 119:474–
483.
Skokomish Indian Tribe and WDFW (Washington Department of Fish and Wildlife).
2007. Skokomish River Chinook salmon recovery plan. December 2007.
SRBLIT (Skokomish River Basin Local Interagency Team). 1994. Skokomish River
basin preliminary watershed assessment. Olympia, Washington.
Solazzi, M. F., T. E. Nickelson, S. L. Johnson, and J. D. Rodgers. 2000. Effects of
increasing winter rearing habitat on abundance of salmonids in two coastal
Oregon streams. Canadian Journal of Fisheries and Aquatic Sciences 57:906-914.
Sommer, T. R., M. L. Nobriga, W. C. Harrell, W. Betham, and W. J. Kimmerer. 2001.
Floodplain rearing of juvenile Chinook salmon: evidence of enhanced growth and
survival. Canadian Journal of Fisheries and Aquatic Sciences 58:325-333.
62
Stevens Jr., D. L., and A. R. Olsen. 1999. Spatially restricted surveys over time for
aquatic resources. Journal of Agricultural, Biological, and Environmental
Statistics 4:415-428.
Swales, S., and C. D. Levings. 1989. Role of off-channel ponds in the life cycle of coho
salmon (Oncorhynchus kisutch) and other juvenile salmonids in Coldwater River,
British Columbia. Canadian Journal of Fisheries and Aquatic Sciences 46:232242.
Tabor, R. A., S. Hager, A. Hird, and R. Piaskowski. 2001. Predation of juvenile
salmonids by predatory fishes in the lower Cedar River, 1999. U.S. Fish and
Wildlife Service, Washington Fish and Wildlife Office, Lacey, Washington.
Tippets, W. E. and P. B. Moyle. 1978. Epibenthic feeding by steelhead (Salmo
gairdneri) in the McCloud River, California. Journal of Animal Ecology 47:549559.
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. Canadian
Journal of Fisheries and Aquatic Sciences 40:452-461.
USDA (United States Department of Agriculture). 1995. Forest Service: South Fork
Skokomish watershed analysis. Olympia, Washington, Olympic National Forest.
USACE (United States Army Corps of Engineers). 2000. Project management plan for
feasibility phase study of Skokomish River Basin, Mason County, Washington.
Final report. Seattle, Washington.
USACE (United States Army Corps of Engineers). 2006. Skokomish River General
Investigation reconnaissance study 905(b) analysis. Seattle, Washington.
USACE and USFWS (United States Fish and Wildlife Service). 2008. Distribution and
abundance of juvenile salmonids and resident fish in the Skokomish River Basin,
Washington. Draft statement of work. Seattle District U. S. Army Corps of
Engineers, and Western Washington Fish and Wildlife Office, Lacey Washington.
January 18, 2008.
USFWS (United States Fish and Wildlife Service). 1988. Endangered Species Act of
1973, as amended through the 100th Congress. U.S. Dept. of the Interior,
Washington, D.C.
USFWS. 1999. Endangered and threatened wildlife and plants; determination of
threatened status for bull trout in the coterminous United States. Final rule.
Notice of intent to prepare a proposed special rule pursuant to section 4(d) of the
Endangered Species Act for the bull trout, proposed rule. Federal Register
64(210):58910-33.
63
USGS (United States Geological Survey). 2008. Water-Data Report 2008: 12061500
Skokomish River near Potlatch, WA. U.S. Geological Survey. Available online:
http://wdr.water.usgs.gov/wy2008/pdfs/12061500.2008.pdf
Walton, J. M. and D. B. Houston, editors. 1983. Proceedings of the Olympic Wild Fish
Conference, March 23-25, 1983. Fisheries Technical Program, Peninsula
College, Port Angeles, Washington.
WDF (Washington Department of Fisheries). 1975. Catalog of Washington streams and
salmon utilization. Olympia, Washington. Volume I.
WDOE (Washington Department of Ecology). 1985. Instream Resources Protection
Program. Water resources inventory area 16, Skokomish – Dosewallips Water
resource inventory area. Prepared by the Water Resources Planning and
Management Section.
WDNR (Washington Department of Natural Resources). 1997. Washington State
salmonid stock inventory: bull trout/Dolly Vardon. Olympia, Washington.
Watershed Management Team. 1995. South Fork Skokomish watershed analysis: U.S.
Dept. of Agriculture, Forest Service.
Wydowski, R. S., and R. L. Whitney. 2003. Inland fishes of Washington. 2nd edition.
University of Washington Press, Seattle, Washington.
Zar, J. H. 1984. Biostatistical analysis. 2th Edition. Prentice-Hall, New Jersey.
Zar, J. H. 1999. Biostatistical analysis. 4th Edition. Prentice-Hall, New Jersey.
Zaret, T. M. and A. S. Rand. 1971. Competition in tropical stream fishes: support for the
competitive exclusion principle. Ecology 52:336-342.
64
(l) Appendix A. Prey codes/categories
Code
Prey Item
550 Unknown_Ephemeroptera
10
Unidentified_Salmon_or_Trout
551 Ephemeroptera_Baetidae
50
CHIRONOMIDS_MIDGES
552 Ephemeroptera_Heptageniidae
501
Chironomid_Larvae
553 Ephemeroptera_Caenidae
502
Chironomid_Pupae_and_emergent
554 Ephemeroptera_Siphlonuridae
503
Chironomid_Adults
555 Ephemeroptera_Leptophlebiidae
OTHER_AQUATIC_DIPTERA_LARVAE
556 Ephemeroptera_Ephemerellidae
51
510
Unknown_Aquatic_Diptera_Larvae
511
Diptera_Larvae_Simuliidae
560 Unknown Odonata
512
Diptera_Larvae_Ceratopogonidae
561 Odonata_Coenagrionidae
513
Diptera_Larvae_Dixidae
514
Diptera_Larvae_Chaoboridae
570 Unknown_Coleoptera
515
Diptera_Larvae_Tipulidae
571 Coleoptera_Elmidae
516
Diptera_Larvae_Empididae
572 Coleoptera_Hydrophilidae
517
Diptera_Larvae_Athericidae
573 Coleoptera_Dystisidae
520
COLLEMBOLA_Isotomis_Springtails
574 Coleoptera_Staphylinidae
53
56 ODONATA_Dragonflies
57 COLEOPTERA_Aquatic Beetles
TRICHOPTERA_Caddisflies
575 Coleoptera_Noteridae
530
Unknown_Trichoptera
576 Coleoptera_Hydrochidae
531
Trichoptera_Rhyacophilidae
577 Coleoptera_Circulionidae
532
Trichoptera_Leptoceridae
533
58 OTHER_AQUATIC_INSECTS
Trichoptera_Hydroptilidae
581 Megaloptera_Sialis_Alderfly
534
Trichoptera_Brachycentridae
582 Lepidoptera
535
Trichoptera_Limnephilidae
583 Chrysomelidae_Aquatic_Leaf_Beetle
536
Trichoptera_Psychomyiidae
59 OTHER
537
Trichoptera_Hydropsychidae
590 Unknown
538
Trichoptera_Wood_Leaf_Cases
591 Exuvia_Aquatic_Insect_Exoskeleton
539
Trichoptera_Sand_Gravel_Cases
592 Misc_Insect_Parts
593
594
54
Trichoptera_Glossomatidae
Trichoptera_Polycentropodidae
PLECOPTERA_Stoneflies
60 NEOMYSIDS
600 Neomysid_Neomysis
61 AMPHIPODS
540
Unknown_Plecoptera
610 Unknown_Amphipods
541
Plecoptera_Perlidae
611 Amphipod_Gammaridae_Gammarus spp
542
Plecoptera_Perlodidae
612 Amphipod_Talitridae_Hyalella_azteca
543
Plecoptera_Nemouridae
62 Crayfish_Astacidae
544
Plecoptera_Leuctridae
63 COPEPODS
545 Plecoptera_Capniidae
630 Unknown_Copepods
546 Plecoptera_Chloroperlidae
631 Copepod_Cyclopoid
595 Plecoptera_Sericostomidae
632 Copepod_Calanoid
55 EPHEMEROPTERA_Mayflies
633 Copepod_Harpactacoid
65
64 CLADOCERANS_Water_Fleas
831 Arachnids_Spiders_
65 ISOPODS_Asellidae
832 Homoptera_Aphids_
70 GASTROPODA_Snails_Limpets
833 Hemiptera_Water_Bugs_Arthropod
701 Gastropod_Gyraulus
834 Thrips_Terrestrial_Arthropod
702 Gastropod_Campeloma
85 Terrestrial_Oligochaetes_Earthworms
703 Gastropod_Goniobasis
86 Terrestrial_Mollusks_Slugs
71 PELECYPODA_Clams_Mussels
87 Terrestrial_Cicadellidae_Leafhopper
72 HIRUDINEA_Leeches
90 Detritus
73 AQUATIC_OLIGOCHAETES
91 Plant_Material
741 Aquatic_ Horsehair_Worm
92 Rocks
75 HYDRACARINA_Water_Mites
93 Fish_Eggs
76 OTHER_AQUATIC_INVERTEBRATES
94 Unidentified_Organic_Matter
80 Diptera_Adult_Flies
95 Unidentified_Inorganic_Matter
81 Hymenoptera_Ants_Bees
96 Other
82 Coleoptera
83 OTHER_TERRESTRIAL_ARTHROPODS
835 Centipedes_Terrestrial_Arthropod
84 Terrestrial_Isopods_Sow_Bugs
66
(m) Appendix B. Length frequency (5-mm FL increments) distributions for coho
salmon in all habitat types, steelhead (10-mm FL increments) in backwaters,
mainstem, and tributaries, and cutthroat trout (10-mm FL increments) in
tributaries. Graphs include all fish collected, including those greater than 100mm FL, which were excluded from all analyses.
25
25
salmon Mainstem
CohoCoho
salmon
Mainstem
Coho salmon Tributaries
Coho salmon Tributaries
20
20
15
Count
Count
15
10
10
5
5
0
0
50
55
60
65
70
75
80
85
90
95
100 105 110 115 120
50
55
60
65
70
75
Fork Length mm
80
85
90
95
100
105
110
115
120
Fork Length mm
25
Coho salmon Backwaters
Coho
salmon Backwaters
25
Coho
salmon
Ponds
Coho
salmon
Ponds
20
20
15
Count
Count
15
10
10
5
5
0
0
50
55
60
65
70
75
80
85
90
95
100
105 110
115
120
50
55
60
65
70
Fork Length mm
75
80
85
90
95
100
105
110
115
120
Fork Length mm
10
10
Rainbow
trout Main
Steelhead
Mainstem
Rainbow trout Tributaries
Steelhead Tributaries
8
8
6
Count
Count
6
4
4
2
2
0
0
60
70
80
90
100 110 120 130 140 150 160 170 180 190 200 210
60
70
80
90
100
Fork Length mm
110
120
130
140
150
160
170
180
190
Fork Length mm
10
10
Rainbow trout Backwaters
Cutthroat
trout
Tributaries
Cutthroat
trout
Tributaries
Steelhead Backwaters
8
6
6
Count
Count
8
4
4
2
2
0
0
60
70
80
90
100
110
120
130
140
Fork Length mm
150
160
170
180
190
60
70
80
90
100
110
120
130
140
150
160
170
180
190
Fork Length mm
67
(n) Appendix C. Results of Pielou’s method on individual sample sets for the
determination of adequate sample size for coho salmon.
Backwaters
68
Mainstem
Ponds
69
Tributaries
70
(o) Appendix D. Results of Pielou’s method on individual sample sets for the
determination of adequate sample size for steelhead and cutthroat trout.
Mainstem
Backwaters
Tributaries
71
(p) Appendix E. Weights of all prey items for coho salmon, steelhead and cutthroat
trout in all habitat types.
Mainstem
Tributaries
coho
salmon
coho
steelhead salmon
0.0442
0.0442
0.0232
0.0232
Mollusca
Bivalvia
Gastropoda
Backwaters
cutthroat
coho
trout
salmon
Ponds
coho
steelhead salmon
0.0008
0.0454
0.0008
0.0454
0.0689
0.0211
0.0478
0.6171
0.6822
0.0002
0.0002
0.0000
0.0000
Annelida
Oligochaeta
Nematomorpha
Gordiacea
0.2559
0.4595
0.0023
0.0023
0.0012
0.0012
0.0078
0.0078
Arthropoda
Insects
Aquatic
Collembola
Isotoma
Ephemeroptera
Baetidae
Ephemerellidae
Heptageniidae
Leptophlebiidae
Unknown sp.
Plecoptera
Capniidae
Leuctridae
Nemouridae
Perlidae
Perlodidae
Sericostomidae
Unknown sp.
Megaloptera
Sialidae
Trichoptera
0.0024
0.2462
0.1389
0.0262
0.0671
0.0090
0.0050
0.1257
0.0013
0.0264
0.0279
0.0093
0.0541
0.0194
0.0093
0.0079
0.0006
0.0016
0.0615
0.0003
0.0145
0.0930
0.0248
0.0362
0.0026
0.0258
0.0035
0.2093
0.1079
0.0092
0.0698
0.0051
0.0463
0.0192
0.0067
0.0098
0.0031
0.0720
0.1339
0.0672
0.0718
0.1862
Hydropsychidae
Hydroptilidae
Limnephilidae
Polycentropodidae
Psychomyiidae
Rhyacophilidae
Unknown sp.
Chironomidae
Adults
Larvae
Pupae and emergent
Other Diptera larvae
Ceratopogonidae
Chaoboridae
Dixidae
0.0555
0.0008
0.0005
0.0687
0.0043
0.0064
0.0173
0.1388
0.1339
0.0006
0.0040
0.0003
0.0217
0.0020
0.0176
0.0148
0.0537
0.0003
0.1603
0.0525
0.0074
0.1617
0.0073
0.1261
0.0271
0.0012
0.0882
0.0131
0.0408
0.0021
0.0303
0.0216
0.0010
0.0221
0.0931
0.2158
0.0497
0.3713
0.1238
0.0014
0.0436
0.0011
0.0949
0.0258
0.0672
0.0075
2.5456
2.1566
0.0956
0.2835
0.0020
0.0080
0.7907
0.0079
0.6296
0.0174
0.0025
Brachycentridae
Glossomatidae
0.0033
0.0288
0.0040
0.0127
0.0057
0.1451
0.0048
0.0463
0.0987
0.0023
0.0394
0.0001
0.0391
0.0002
0.0050
0.0252
0.0526
0.0035
0.0405
0.0086
0.0273
0.0026
0.0032
0.0009
0.0040
0.0001
0.1993
0.0018
0.0512
0.1463
0.0502
0.0001
0.0046
0.0010
0.0036
0.0040
0.0001
0.9388
0.0165
0.5858
0.3365
0.0819
0.0034
0.0001
0.0421
0.0348
0.0073
0.0266
0.0129
0.0078
0.0131
0.4340
0.0200
0.3005
0.1135
0.2613
0.0521
72
Empididae
Simuliidae
Tipulidae
Unknown sp.
Odonata
Coenagrionidae
Lepidoptera
0.0057
0.0420
0.0024
0.0070
0.0170
0.0025
0.0040
0.0240
0.0135
0.0139
0.2336
0.0138
0.0050
0.0002
0.0036
0.0005
0.0068
0.0141
0.0111
0.0076
0.0089
Arthropoda
Insects
Terrestrial
Coleoptera
Chrysomelidae
Circulionidae
Elmidae
Hydrochidae
Noteridae
Staphylinidae
Unknown sp.
Other Diptera
adult
Unknown sp.
Homoptera
Cicadellidae
Hymenoptera
Formicidae
0.0210
0.0681
0.0013
0.0110
0.1038
0.0226
0.0026
0.0146
0.0262
0.0225
0.0065
0.0116
0.0110
0.0065
0.0537
0.0223
0.0057
0.0126
0.0087
0.0009
0.0160
0.0002
0.0035
0.0108
0.0965
0.1024
0.0004
0.0611
0.0059
0.0059
0.0669
0.1211
0.1066
0.0145
0.0031
0.0031
0.0048
0.0275
Arthropoda
Insects-Other
Other - Insect Unknown sp.
Exuvia
Insect eggs
Trichoptera Cases
Sand/Gravel
Cases
Wood/Leaf
Cases
0.1347
0.1345
0.0050
0.0050
0.0400
0.0010
0.0021
0.0021
0.0188
0.0474
0.0128
0.0429
0.1953
0.1893
0.0060
0.0042
0.0123
0.0402
0.0123
0.0323
0.0017
0.0631
0.0065
0.0072
0.0005
0.0106
0.0025
0.0038
Crustacea
0.0217
0.0004
0.0213
Copepoda
Amphipoda
0.0455
0.0455
0.0905
0.0009
0.0896
0.0518
0.0001
0.0000
0.0081
0.2492
0.0022
0.0573
0.0573
Arachnida
Unknown sp.
Hydracarina
Fish
Fish Eggs
Unidentified
salmonid fry
0.0014
0.0001
0.0620
0.0006
0.0004
0.0116
0.0116
0.2470
Other
Other
Plant Material
Unidentified
Organic Matter
0.1595
0.0524
0.0141
0.0082
0.0965
0.0403
0.0105
0.0080
0.0737
0.0733
0.0781
0.0080
0.1498
0.0760
0.1071
0.0059
0.0562
0.0025
0.0004
0.0701
0.0738
73
(q) Appendix F. Summary of the major prey categories; percent proportion by
weight (%Wi), percent proportion by number (%Ni), and frequency of
occurrence (%Oi) by habitat type for coho salmon.
Habitat Types
Mainstem
Tributaries
Backwaters
Ponds
Prey Category
%Wi
%Ni
%Oi
%Wi
%Ni
%Oi
%Wi
%Ni
%Oi
%Wi
%Ni
%Oi
Mollusca
Annelida
2.97
0.43
4.00
0.96
0.07
1.96
0.00
0.00
0.00
3.27
2.96
8.82
17.19
0.29
3.92
18.96
0.15
8.16
10.66
0.13
8.20
32.36
0.56
4.76
Oligochaeta
Arthropods
Insects
Aquatic Insects
Collembola
Isotoma
0.16
1.59
15.69
0.60
1.53
30.61
0.13
0.24
18.03
0.35
4.26
12.70
Ephemeroptera
16.54
18.18
76.00
3.84
6.15
60.78
43.97
45.89
60.00
7.67
28.15
47.06
Plecoptera
8.45
6.49
50.00
8.64
2.76
47.06
13.66
6.55
51.67
4.18
5.93
25.00
4.84
0.29
3.92
2.96
0.11
6.12
1.61
0.15
9.84
17.61
2.96
17.46
Megaloptera
Sialidae
Trichoptera
9.00
5.63
36.00
7.68
1.30
33.33
3.73
2.10
43.33
5.87
2.59
13.24
Chironomidae
13.39
58.01
48.00
38.74
76.96
84.31
7.50
38.54
88.33
2.49
35.71
48.53
Other Diptera
larvae
3.37
5.92
42.00
3.38
7.23
39.22
4.51
4.38
36.67
1.29
3.33
14.71
1.41
1.59
20.00
2.81
1.19
35.29
1.79
0.56
30.00
1.07
0.74
5.88
0.72
0.87
7.84
3.98
1.72
32.65
1.77
1.14
47.54
2.90
3.70
12.70
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.90
0.00
0.88
0.00
0.37
0.00
0.34
0.00
11.76
10.20
0.00
0.79
0.59
0.19
0.59
0.09
0.06
0.02
0.06
5.00
1.64
3.28
4.29
0.00
4.25
0.00
8.33
13.24
Amphipoda
Gammaridae
Unknown sp.
0.00
0.00
0.00
0.00
5.19
0.00
14.29
0.00
Copepoda
Cyclopoida
0.00
0.00
0.00
0.02
2.04
0.00
0.00
0.00
0.04
3.15
3.17
0.78
0.00
8.00
0.00
0.00
4.30
0.06
3.33
2.72
0.00
4.41
Terrestrial Insects
Coleoptera
Diptera
Adult
Crustacea
Fish
0.04
0.00
74
(r) Appendix G. Summary of the major prey categories; percent proportions by
weight (%Wi), percent proportions by number (%Ni), and percent frequency of
occurrence (%Oi) by habitat type for steelhead and cutthroat trout.
cutthroat trout
steelhead
Tributaries
Prey Category
Mollusca
Nematomorpha
%Wi
Mainstem
%Ni
%Oi
%Wi
Backwaters
%Ni
%Oi
%Wi
%Ni
%Oi
0.32
0.42
11.11
0.00
0.00
0.00
0.00
0.00
0.00
3.17
2.54
33.33
0.00
0.00
0.00
0.00
0.00
0.00
Isotoma
1.34
1.27
22.22
0.00
0.00
0.00
0.00
0.00
0.00
Ephemeroptera
11.66
9.75
77.78
8.69
22.97
60.00
34.19
28.93
66.67
Plecoptera
7.00
4.24
66.67
27.54
6.76
20.00
5.35
1.89
33.33
Trichoptera
21.76
17.80
100
30.10
55.41
50.00
12.25
18.24
66.67
Chironomidae
17.05
55.08
88.89
2.05
8.11
20.00
9.71
48.11
75.00
Diptera larvae
10.77
7.20
66.67
1.79
1.35
5.00
1.23
2.52
16.67
4.44
1.69
44.44
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.10
0.31
8.33
Paragordius
tricuspidatus
Arthropoda
Insects
Aquatic Insects
Collembola
Terrestrial
Insects
Coleoptera
Diptera
Adult
0.00
0.00
0.00
0.00
(s) Appendix H. Percent index of relative importance (%IRI) for major prey
categories for coho salmon, steelhead, and cutthroat trout
Prey Category
Mollusca
Arthropoda
Mainstem
0.11
coho salmon
Tributary
Backwater
Pond
cutthroat
trout
Tributary
Steelhead
Backwater
Mainstem
0.00
0.00
0.57
0.02
0.00
0.00
Insects
Aquatic
Collembola
0.23
0.29
0.02
0.60
0.17
0.00
0.00
Ephemeroptera
22.84
2.74
23.84
17.55
5.13
19.27
13.32
Plecoptera
Megaloptera
larvae
Trichoptera
6.46
2.42
4.61
2.63
2.30
1.10
4.81
0.17
0.08
0.07
3.74
0.00
0.00
0.00
4.55
1.35
1.11
1.16
12.19
9.30
29.91
Chironomidae
18.19
49.23
32.27
33.9
34.43
32.97
1.63
Diptera Larvae
3.37
1.88
1.44
0.70
3.69
0.28
0.11
0.51
0.00
0.63
0.06
0.31
0.01
0.11
1.74
0.84
0.00
0.00
0.00
0.00
0.00
0.05
0.00
0.06
0.12
0.00
0.00
0.00
Terrestrial
Coleoptera
Crustacea
Fish
75
