LIBRARY
Michigan State .
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.

TO AVOID FINES return on or baton dd. due.

  
 

  

DATE DUE DATE DUE DATE DUE

m 3 0 19937]?

   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmative Action/Equal Opportunity Institution
cMmS-m

EVALUATION OF JUVENILE BROWN TROUT AND STEELHEAD
COMPETITION IN GREAT LAKES TRIBUTARIES

by

John Francis Kocik

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Fisheries and Wildlife

1992

pat/“’7
c ;j

./ ff» 0

ABSTRACT

EVALUATION OF JUVENILE BROWN TROUT AND STEELHEAD
COMPETITION IN GREAT LAKES TRIBUTARIES

by

John Francis Kocik

The importance of competition between steelhead
(gaggrhygghus mygigg) and brown trout (Salmo trutta) was
determined in a three part project. Field and laboratory
experiments assessed the effects of introducing steelhead
fry on brown trout growth, survival, and habitat use. I
examined brown trout abundance, survival, size, growth, and
condition in allopatry and in sympatry with steelhead in
Gilchrist Creek, Michigan and in an artificial stream. In
the field, I established a test and control section on the
stream. I measured the abundance and vital statistics of
salmonines at these stations during 1989. In 1990 and 1991,
steelhead fry were scatter—stocked in the test section.

From 1990 to 1992, I continued population assessments.
Laboratory trials assessed the impact of age-0 steelhead
upon age-0 brown trout in a completely randomized design.
Four replicate cells contained 14 allopatric brown trout and
4 cells contained 7 sympatric brown trout with 7 steelhead.
In field and laboratory studies, the impact of steelhead was
negligible. Steelhead had no impact on brown trout
abundance or survival. Steelhead had a minor impact on
brown trout growth in Gilchrist Creek. However, the impact

had little effect on brown trout size, relative to

intraspecific and abiotic factors. In the artificial
stream, I observed no impact on growth. The earlier
emergence times of brown trout gives them a size advantage
over steelhead. This size advantage decreases over time
since steelhead grow faster. At the post-emergence stage,
these species appear to interact to the detriment of
steelhead. However, if emergence times become closer, the
advantage of brown trout over steelhead may be lost. These
studies indicated that steelhead superimposition of brown
trout redds and factors narrowing the emergence gap between
these species may adversely impact brown trout. Habitat use
of these species was similar. Water depth, cover, and
substrate used during the age-O growing season was not
significantly different. Steelhead did use slower mean
column velocities. Steelhead were suspended in the water
column whereas brown trout were benthic. Despite overlap in
habitat use, no difference in brown trout habitat occurred
between sympatry or allopatry. Vertical stratification of
these species may reduce interactions during the age-0

growth period.

To Linda

iv

ACKNOWLEDGEMENTS

I would like to thank my major professor, William
Taylor, for his unending encouragement, faith and support
throughout my graduate career. Most importantly I thank him
for his friendship. Great appreciation is also extended to
my committee members, Thomas Burton, Thomas Coon, Carl
Latta, and Richard Merritt, for their assistance and advice
and for sharing their experience, knowledge, and insights.

I also thank Scott Winterstein; his statistical assistance
was extremely helpful. His aid contributed greatly to this
dissertation and to my understanding of statistical
procedures.

I extend special thanks to my wife, Linda, for her
constant love, encouragement, and support as well as her
invaluable assistance as a technician, data entry
specialist, and editor. Lin, thanks for always being there
for me.

Thanks to my fellow graduate students for their help in
the field and in the laboratory: Russell Brown, Paola
Ferreri, Michael Hansen, Daniel Hayes, Andre Kabre, Steven
Marod, Paul Padding, Melissa Treml, Susan Walker, and

Michaela Zint. A special thanks to Russ, Dan, and Steve.

Our scientific, theoretical, and philosophical conversations
are much appreciated and have provided insightful views and
approaches. Thanks also to the many dedicated and
industrious undergraduates who worked on this project:
Mandy Dunlop, John Lott, William McMillian, Steven Mero,
Joseph Mion, Kurt Newman, Timothy Nuttle, Sam Noffke, and
Heather Sysak. A special thanks to all those who
volunteered to assist in winter sampling trips. Thanks to
Gaylord Alexander, Andrew Neufer, and Jack Rodgers of the
Michigan Department of Natural Resources Fisheries Division
(MDNR) trout research station for their assistance.

This research was a result of work funded by the MDNR
through the Federal Aid in Fish Restoration Project F-35-R
Number 641. Additional financial support for this research
was provided by the Michigan Agricultural Experiment
station, Michigan Council of Trout Unlimited, West Michigan
Chapter of Trout Unlimited, Oakbrook Chapter of Trout
Unlimited, Challenge Chapter of Trout Unlimited, the
Michigan Polar Equator Club, and Michigan Salmon and

Steelhead Fishing Association.

vi

TABLE OF CONTENTS

LIST OF TABLES . . . . . . . . . . . . . . . . . . . . x
LIST OF FIGURES . . . . . . . . . . . . . . . . . . . xii
LIST OF APPENDICES . . . . . . . . . . . . . . . . . xiv
CHAPTER 1 . . . . . . . . . . . . . . . . . . . . . . . 1
ABSTRACT . . . . . . . . . . . . . . . . . . . . . 1
INTRODUCTION . . . . . . . . . . . . . . . . . . . 2
METHODS . . . . . . . . . . . . . . . . . . . . . 8
Study Site . . . . . . . . . . . . . . . . . . . . 8

Study Design . . . . . . . . . . . . . . . . . . . 10

Steelhead Introduction . . . . . . . . . . . . . . 10

Salmonine Population Assessments . . . . . . . . . 12
Statistical Assessment Methods . . . . . . . . . . 14
Effects of Salmonine Density on Brown Trout Size . 16

RESULTS . . . . . . . . . . . . . . . . . . . . . 18
Steelhead Introduction and Population Assessment . 18
Brown Trout Abundance and Mortality . . . . . . . 20
Brown Trout Size and Growth .-. . . . . . . . . . 24
Brown Trout Condition . . . . . . . . . . . . . . 28
Multiple Regression Models . . . . . . . . . . . . 29
DISCUSSION . . . . . . . . . . . . . . . . . . . . 30

Competition Theory and Interpretation of Results . 30

vii

Steelhead and Brown Trout Abundance . . .
Steelhead and Brown Trout Size and Growth

Combined Effects of Size and Abundance . .

Implications . . . . . . . . . . . . . . .
CHAPTER 2 . . . . . . . . . . . . . . . . . . .
INTRODUCTION . . . . . . . . . . . . . . .
METHODS . . . . . . . . . . . . . . . . .
Stream Design and Macrohabitat . . . . . .

Stream Cells and Microhabitat . . . .
Fish Procurement and Holding . . . . . .
Experimental Design . . . . . . . . .
Feeding and Maintenance Regimes . . . . .

Rate Calculations and Statistical Analysis

RESULTS . . . . . . . . . . . . . . . . .
Brown Trout Mortality . . . . . . . . . .
Brown Trout Size and Growth . . . . . . .
Steelhead Population Parameters . . . . .
Density Effects on Brown Trout . . . . . .
DISCUSSION . . . . . . . . . . . . . . . .

Steelhead Effects on Brown Trout Survival

Steelhead Effects on Brown Trout Growth .

Implications for Natural Systems . . . . .
CHAPTER 3 . . . . . . . . . . . . . . . . . . .
ABSTRACT . . . . . . . . . . . . . . . . .

31

33

35

43

45

46

49

49

51

52

54

57

58

62

62

64

66

67

68

68

68

7O

74

74

INTRODUCTION

METHODS .

Study Site and Design

Salmonine Habitat Use Observations . . . . .

Comparisons of Habitat Use

RESULTS .

Brown Trout Habitat Use

Steelhead Habitat Use and Impact

DISCUSSION

Brown Trout Habitat Use

Steelhead Habitat Use and Impact

APPENDICES . .
Appendix 1
Appendix 2
Appendix 3
Appendix 4

LITERATURE CITED

ix

75

78

78

80

82

83

83

94

101

101

103

107

107

108

109

110

111

Table

Table

Table

Table

Table

Table,

Table

Table

Table

LIST OF TABLES

Salmonines that use riverine environments and
their predominant life—history patterns in the

Great Lakes Region. . .

Selected sets of single degree-of—freedom
linear contrasts used to test for the effect
of steelhead upon juvenile brown trout

population parameters .

Steelhead abundance (S.E.- standard error) and
corresponding instantaneous mortality rates
(Z) for 1990 and 1991 cohorts in the test

section of Gilchrist Creek.

Steelhead total length (TL) and weight (WT)

and corresponding instantaneous growth rates
for 1990 and 1991 cohorts in test section of

Gilchrist Creek . . . .

Average abundance of age-O brown trout per
hectare and standard error (S.E.) in control
(site 1) and test (site 2) sections of
Gilchrist Creek for 1989-1991 cohorts

Instantaneous mortality (Z) rates for age-O
brown trout cohorts 1989-1991 in control
(site 1) and test (site 2) sections of

Gilchrist Creek . . . .

Average abundance of age-1 brown trout per

hectare and standard error (S.E.) in control
(site 1) and test (site 2) sections of
Gilchrist Creek for 1988-1990 cohorts

Instantaneous mortality (2) rates for age-1
brown trout cohorts 1988-1990 in control
(site 1) and test (site 2) sections of

Gilchrist Creek . . . .

Average total length (TL) and weight (WT) of
age-O brown trout and standard errors (S.E.)

for these measurements.

3

17

19

20

21

22

23

24

25

Table 10. Seasonal instantaneous growth rates for age-0
brown trout cohorts 1989-1991 in control
(site 1) and test (site 2) sections of
Gilchrist Creek . . . . . . . . . . . . . . .

Table 11. Average total length (TL) and weight (WT) of
age-l brown trout and standard errors (S.E.)
for these measurements. . . . . . . . . . . .

Table 12. Seasonal instantaneous growth rates for age-1
brown trout cohorts 1988-1990 in control
(site 1) and test (site 2) sections of
Gilchrist Creek. Summer represents June-
October and winter represents October-
February . . . . . . . . . . . . .. . . . . .

Table 13. The timing of fish introductions, collections,
and subsequent time period references . . . .

Table 14. Diving dates for habitat use surveys. . . . .

xi

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

10.

11.

LIST OF FIGURES

Location of Gilchrist Creek and test and
contr01 areas 0 O O O O O O O O O O O O O D O O O 9

Conceptual model of interactions between

juvenile brown trout and steelhead in

riverine environments illustrating areas of
spatial and temporal overlap. . . . . . . . . . 36

Conceptual model of interactions between brown
trout and steelhead in riverine environments
at all life history stages. . . . . . . . . . . 4O

ArtifiCial stream design and experimental
Structure 0 O O O O O O O O O O O O O O O O O O O 5 0

Artificial stream cell illustrating habitat
size and placement. . . . . . . . . . . . . . . 53

Number of brown trout in allopatry and
sympatry surviving in individual artificial
stream cells. . . . . . . . . . . . . . . . . . 63

Mean total length (mm) over experimental time
for brown trout in allopatry and sympatry and
for steelhead in sympatry.. . . . . . . . . . . 65

Map of Gilchrist Creek and study sections
where dives were conducted.. . . . . . . . . . .79

Comparison of water depths utilized by age-O
brown trout between spring and the summer-
autumn time periods.. . . . . . . . . . . . . . 85

Comparison of mean column velocities
utilized by age-0 brown trout between
spring and the summer-autumn time periods. 86

Comparison of substrate types utilized by
age-0 brown trout between spring and the
summer-autumn time periods.. . . . . . . . 87

Comparison of cover types utilized by
age-0 brown trout between spring and the
summer-autumn time periods.. . . . . . . . 88

xii

Figure 13. Comparison of water depths utilized by
age-0 and age-1 brown trout. . . . . . . .

Figure 14. Comparison of mean column velocities
utilized by age-0 and age-1 brown trout. .

Figure 15. Comparison of substrate types utilized by
age-O and age-1 brown trout. . . . . . . .
Figure 16. Comparison of cover types utilized by
age-0 and age-1 brown trout. . . . . . . .
Figure 17. 'Comparison of water depth utilized by
age-O brown trout and steelhead. . . . . .
Figure 18. Comparison of mean column velocities
utilized by age—O brown trout and
steelhead. . . . . . . . . . . . . . . . .
Figure 19. Comparison of substrates utilized by
age-0 brown trout and steelhead. . . . . .
Figure 20. Comparison of cover utilized by age-0
brown trout and steelhead. . . . . . . . .

xiii

Appendix

Appendix

Appendix

Appendix

LIST OF APPENDICES

P values for ANOVA models and linear
contrasts for population estimates and
instantaneous mortality rates for

juvenile brown trout. . . . . . . . . . . 116

P values for ANOVA models and linear
contrasts for total length and

instantaneous growth rates of age-0 and

age-1 brown trout. . . . . . . . . . . . 117

Generalized substrate classes (From Bovee
1986) I O O O O O O O O O O O O O O O O O I 118

Descriptions of cover types. Adapted
from Bovee (1986).. . . . . . . . . . . . 119

xiv

CHAPTER 1

COMPETITION BETWEEN BROWN TROUT AND STEELHEAD
IN A GREAT LAKES TRIBUTARY

ABSTRACT

I examined competitive interactions between juvenile
(age-0 and age-1) steelhead (Oncorhvnchus mykigg) and brown
trout (Sglmg trutta) in Gilchrist Creek, Michigan. I
assessed the abundance and vital statistics of allopatric
juvenile brown trout in two stream sections (test and
control) in 1989. In 1990 and 1991, steelhead fry were
scatter stocked in the downstream test section. After these
introductions, I continued population assessments in both
sections. Steelhead populations established were generally
low but within the range found in the region. There was no
measurable impact of steelhead on juvenile brown trout
abundance, survival, or size. There was an effect of
steelhead on age-O brown trout growth, but it was
insufficient to counter year-to-year size variations. Brown
trout were larger at emergence, which was probably a major
factor in their successful interactions with juvenile
steelhead. However, higher steelhead growth rates
diminished this difference over the first growing season.
If factors such as spawner-alevin interactions, differing
emergence times, and climactic variability alter the initial
size advantage of brown trout, the result of juvenile
competition may be altered. Future research should be
focused on quantification of these other parameters.

1

INTRODUCTION

The indigenous salmonid fauna of Michigan rivers
consisted of grayling (Thvmallius tricolor) and brook trout
(Salvelinus fontinalis). Habitat destruction, overfishing,
and the introduction of exotic species led to the extinction
of Michigan grayling and decimation of brook trout stocks
(Kruger 1985; Kruger et a1. 1985). In response to the
decline of these fish, fishery biologists have introduced
numerous salmonines to Michigan streams since the late
1800's (Parsons 1973). Currently, a multispecies salmonine
complex exists in Michigan tributaries of the Great Lakes
and the region as a whole, consisting of the brook trout and
six exotic species (Table 1). Thus, the result of stocking,
colonization, and range expansion has been that most streams
contain exotic species or reestablished stocks of native
species (Parsons 1973; Kruger et a1. 1985).

The introductions have resulted in salmonine species
combinations unique to the Great Lakes region since several
species do not co-occur in their native distributions.

These species have similar ecological requirements that
could potentially foster interspecific competition (Jones

1991). Hearn (1987) and Fausch (1988) both conducted

3

Table 1. Salmonines that use riverine environments and
their predominant life-history patterns in the
Great Lakes Region. Riverine fish complete their
life-cycle in river systems. Migratory fish
exhibit predictable movement from riverine to
Great Lakes’ lacustrine ecosystems. References
cited in table: a)MacCrimmon and Marshall 1968;
b)Biette et al. 1981; c)Kruger et a1. 1985;
d)Kocik and Taylor 1987; e)Jones 1991.

 

 

 

 

 

 

 

Common Name Scientific Name Life History
pink salmon Oncorhvnchus qorbuscha migratoryd
coho salmon Oncorhvnchus kisutch migratoryd
steelhead Oncorhvnchus mvkiss migratoryb
Chinook salmon Oncorhynchus tshawvtscha migratoryd
brook trout Salvelinus fontinalis riverine‘”c
Atlantic salmon Salmo salar migratorye
brown trout Salmo trutta riverinem°

 

 

extensive reviews of competition between stream salmonines.
Both authors stress the importance of coevolution in the
successful resource partitioning of sympatric salmonine
populations. ‘Given these facts, tributaries with salmonine
species from different geographic origins are prime
candidates for competitive interactions that may be
detrimental to the success of one or both species.

The effects of salmonine competition in Great Lakes
tributaries has received limited attention. Carl (1983)
examined relative abundance of chinook salmon, coho salmon,
and steelhead. He found no evidence of negative

interactions between these species. He attributed limited

4

competition to spatial and temporal segregation. Fausch
(1984) and Fausch and White (1986) examined competition
between brook trout, brown trout, and coho salmon. Both
studies reported that coho salmon exhibited competitive
superiority for food and space over brook trout and brown
trout. Competition between brown trout and brook trout was
dependent upon age, size, and temperature (Fausch and White
1986). However, brown trout were typically dominant over
brook trout (Fausch and White 1981).

Knowledge of competition between brown trout and
steelhead would enhance our understanding of stream
salmonines in Great Lakes tributaries. These salmonines
have been established in the Great Lakes region the longest
(over 100 years), and many populations are naturally
reproducing (MacCrimmon and Marshall 1968; Biette et al.
1981; Seelbach 1987). Successful colonization has resulted
in both species becoming well distributed throughout the
Great Lakes region (MacCrimmon and Marshall 1968; Seelbach
1987). Life-history differences make knowledge of their
interrelationships important since each species represents
distinctive ecological and fishery resources.

Brown trout in the Great Lakes typically utilize
streams for their entire life-history (MacCrimmon and
Marshall 1968). A few migratory populations exist, but
these are rare in Great Lakes populations (Scholl et a1.

1984). Brown trout provide a unique resource, spending

5
their entire life in the riverine environment. With the
demise of indigenous salmonines, brown trout represent the
only riverine-resident salmonine in many systems (MacCrimmon
and Marshall 1968). As the largest fish in many of these
coldwater systems, they may influence ecosystem structure
and function (Alexander 1979). Brown trout also sustain
year-round fisheries in many Great Lakes tributaries (Kruger
et al. 1985).

Steelhead in the Great Lakes are migratory; they use
the riverine environment for spawning and juvenile nursery
areas until age-2 (Biette et al. 1981; Seelbach 1987).

Their distribution overlaps with brown trout in streams that
flow unobstructed into the Great Lakes. Steelhead also
provide an important and distinct fishery resource (Kruger
et al. 1985). A factor contributing to their importance as
a fishery resource is their extended adult residence in
stream environments prior to spawning (Biette et al. 1981).
As such, the loss of either of these species would represent
a change in the structure of these ecosystems and their
fishery resources from past decades.

In the early 1980's, anglers indicated that they felt
there had been a decline in the quality of the Pere
Marquette River brown trout fishery (Kruger et al. 1985).
Kruger (1985) observed an inverse relationship between brown
trout and steelhead abundance in the Pere Marquette River

from 1970-1983. In addition, brown trout growth in size

6
classes equivalent to steelhead parr (< 200 mm) was below
the state average. However, once brown trout exceed the
size at which steelhead outmigrate, their growth rates were
above the state average. Kruger (1985) suggested that
competition between these species might be occurring at the
juvenile stage.

Ziegler (1988) examined age-O brown trout and steelhead
food habits and habitat use in one sympatric and two
allopatric populations. She concluded that the species used
similar habitats and food but found no differences in either
parameter in sympatry or allopatry. She also noted no
difference in relative abundance of age-0 fish in allopatry
or sympatry (Ziegler 1988). However, age-1 brown trout were
more abundant in allopatric than sympatric populations. She
suggested that competition between these fish might occur
during their first winter to the detriment of brown trout
growth and survival.

Understanding the potential interactions between
steelhead and brown trout is important. Both species have
many naturally reproducing populations that provide distinct
and highly valuable sport fisheries (Kruger 1985). To
manage streams for optimal populations of salmonines,
biologists need to know if competitive interactions occur.
In addition, ascertaining the life-history stage and mode of
competition will contribute to a clearer understanding of

sympatric salmonine ecology.

7

The goal of this research is to determine the extent
and mode of competition between juvenile (age-0 and age-1)
steelhead and brown trout. I define competition as the
negative impact of the presence of one species on the
population dynamics of another species. I hypothesize that
juvenile steelhead compete for resources with brown trout to
the detriment of juvenile brown trout. Analyses of the
brown trout population parameters of density, survival,
growth, and condition should provide evidence of the impact
of juvenile steelhead.

I measured brown trout population parameters in
Gilchrist Creek in the northern lower peninsula of Michigan
for three years. In the first year of the study (1989), I
examined the stream in its current state: exclusively brown
trout. In the second and third years of the study, I
introduced steelhead to the downstream section of Gilchrist
Creek. In this chapter, I describe the results of this

introduction.

METHODS

Stu—dxflg

Gilchrist Creek is a second order coldwater stream
located in Montmorency County, Michigan (Figure 1). The
creek flows northeast through forested land into the Thunder
Bay River and Lake Huron. The soils in this region are
primarily glacial sands, and infiltration into deep aquifers
maintains highly stable flow regimes (P. Seelbach, MDNR
pers. comm.). Primary substrates are sand with long
expanses of gravel. This stream type is typical of many
Michigan Great Lakes tributaries harboring brown trout and
steelhead.

I selected the stream based upon its high brown trout
abundance and isolation from migratory Great Lakes
salmonines due to downstream dams on the river system.

Brown trout have been the predominant salmonine species in
the stream since at least the 1960’s (Gowing, MDNR,
Unpublished data). Within Gilchrist Creek, I established
two study sections. Section 1 was the upstream control
section; section 2 was the downstream test section (Figure
1). The two study sections were 2 km apart. Within each
study section, I established three 100 m study reaches

separated by 25 m buffer zones.

 

 

 

 

North .2 g
o E
5’ i
o +
$
3
Lower Michigan 6
Steelhead Stocking Area 7 4
“”83: Section 2
3% 20Test Area

Sectlon 1 $35
COI‘ItI'OI Area

 

 

A:

o

g Greasy Creek
.‘2‘

l-

.1:

2 .2 i
O 1 Kllometer

Figure 1 Location of Gilchrist Creek and test and control

areas .

10
Study Design:

The structure of this experiment quantifies competitive
impacts of steelhead upon brown trout population dynamics.
This was accomplished by comparing juvenile brown trout
(below age-2) abundance, survival, growth, and condition in
allopatry and in sympatry with introduced steelhead. I
collected baseline population data from sections 1 and 2
starting in June 1989. At this point, both sections
contained allopatric brown trout populations (pre-
treatment). In the spring Of 1990, I introduced steelhead
fry into section 2 of Gilchrist Creek, making it a sympatric
(test) treatment. Section 1 remained an allopatric
(control) brown trout population. I continued post-
treatment population assessments through June 1992 to
measure steelhead impacts upon brown trout. I tested the
impact of steelhead upon brown trout using pre- versus post-
treatment comparisons within sections and between-section

comparisons (test versus control) within years.

Steelhead Introduction:

To simulate natural steelhead colonization, I stocked
steelhead swim-up fry in 1990 and 1991. Since it was
impractical and undesirable to impede within-stream fish
movements, I used knowledge of steelhead behavior to isolate
them in only one of the two study sections. Steelhead fry

dispersion is reported to be minimal and predominantly

11
downstream (Hume and Parkinson 1987, 1988). By stocking
steelhead only in the downstream area surrounding section 2
(test area), I successfully isolated allopatric and
sympatric populations. We stocked a total of 700 m of
stream length inclusive of section 2 (Figure 1).

I could not locate any post-emergence population
estimates of steelhead fry for Great Lakes tributaries.
Therefore, I used fall population densities to calculate
stocking rates. The fall densities of age-0 fry in the
Great Lakes region range from 681 fish/ha to 43,771 fish/ha
(Seelbach 1986). Reviewing these data, I determined that a
fall density of 3,500 steelhead/ha to be a realistic target
for Gilchrist Creek. I based this value on steelhead
densities in nearby streams with similar habitat
availability and stream productivity (P. Seelbach, MDNR
pers. comm.).

With the fall target density established, I examined
reports of fry stocking survival. Hume and Parkinson (1987)
summarized data from fry stocking for steelhead, Atlantic
salmon, and brown trout from 10 studies. Using these data,
I predicted steelhead survival from stocking to fall would
be about 10%. I then used this survival rate to calculate
spring stocking rates. Actual 1990 stocking densities were
35,534 fry / ha. In 1991, I used survival data from 1990
(minimum 2.5 %) to calculate stocking densities. In 1991, I

stocked 242,415 fry /ha.

12

I obtained steelhead fry from the Michigan Department
of Natural Resources (MDNR) Wolf Lake Hatchery. These fry
were progeny of wild steelhead collected at the Little
Manistee River. We transported fish to Gilchrist Creek and
released them in stream margins in the center of each 100 m2
surface area (unit) of stream within the stocking section
(Wentworth and La Bar 1984). On 30 May 1990, steelhead fry
averaging 25.5 mm (i 0.199 S.E.) and 0.122 g (i 0.003 S.E.)
were stocked at a rate of 355 fry per unit. In 1991,
average fry size was 27.1 mm (: 0.212 S.E.) and 0.143 g (i
0.004 S.E.). Fry were stocked at a rate of 2424 fry/unit on

23 May.

Salmonine Population Assessments

I collected population data from June 1989 to 1992.
Procedures were the same before and after steelhead
introduction. Assessments were made three times a year:
post salmonine emergence (June), post growing season
(October), and late winter (February). I conducted
assessments within each 100 m study reach.

Abundance estimates were made using the Petersen mark-
recapture method with a 250 volt D.C. electrofishing unit
(Ricker 1975; Peterson and Cederholm 1984). A 24-hour
redistribution period separated the marking and recapture
runs. I marked all salmonines using a site-specific partial

fin clip. Salmonines collected on the marking run were

13
measured for total length (mm) and weight (g). On the
recapture run, all fish were measured for total length and
only unmarked fish were weighed. Using this protocol, I
could calculate cohort population estimates, and sample
sizes for vital statistics were maximized. I collected
scales from all fish during the October collection and from
all fish over 150 mm during other collections. Fish were
aged using length-frequency analysis and scale reading
(MacDonald 1987). This procedure allowed separation of fish
into age groups.

I calculated salmonine abundance by age class for each
species in each study reach using Chapman’s adjusted
Petersen estimate (Ricker 1975). Using these data, the
instantaneous mortality rate (Z) was calculated for each age
class and study reach. The following equation was used

(Ricker 1975):

2: -(loge NM -loge Nt)

Where:
Z = instantaneous mortality rate
Nt+1 = population estimate of study reach at time t+1
Nt = population estimate of study reach at time t

Z was calculated for age-0 and age-1 fish on a seasonal

basis. Growing season mortality was estimated using June
and October abundance. Winter mortality reflected losses
from October to the following February. Spring mortality

quantified losses from February to June. Since the number

14
of steelhead stocked was also known, I calculated their
stocking mortality from initial stocking to June sampling.
Mean lengths were used to calculate instantaneous daily
growth rate (a) for each stream reach. These estimates were
also conducted on a seasonal basis. I calculated u using

the following equation (Ricker 1975):

)1: (loge TLt - loge TLM) / T

Where:
u = instantaneous daily growth rate
TL = mean total length (mm)
t = time at the end of growth period
P1 = time at beginning of growth period
T = time interval in days

To assess differences in the condition of salmonines
between treatments, ordinary least squares regression of the
weight (W) on total length (TL) was used (loge W = loge a +
b loge'FL) as recommended by Cone (1989). This method is
unbiased since the slope of an ordinary least squares
regression can be used as the condition index (Cone 1989).
Condition indices were calculated for each study reach by

age classes.

Statistical Assessment Methods
The design of this experiment allowed comparisons of

juvenile brown trout abundance, survival, growth, and

15
condition to assess the impact of steelhead. Analyses of
these parameters was conducted by cohort. To clarify
cohorts and relate changes to life-history periods (ie.
post-emergence, post-growing season, and late winter), I
assigned the birth date of these fish to be April 15. This
date closely corresponds to brown trout emergence from
redds. Using this method, age-0 fish were sampled in June,
October, and February of their first year of life. By
sampling in their second June they would be considered age-
1. Using this aging scheme, I conducted analyses separately
for age-0 and age-1 cohorts.

I partitioned statistical analysis by month for point
estimates (density and size) and by season for rate
variables (Z and u). However, Z and p are ratio data that
are not typically normally distributed. To account for non-
normality, I transformed these data using the RT-l procedure
(Conover and Iman 1981). To examine the differences in
brown trout between test and control sections in Gilchrist
Creek, I used analysis of variance (ANOVA) and selected sets
of single degree-of-freedom contrasts (Steel and Torrie
1980). I designed contrasts to test biologically meaningful
hypotheses (Table 2). Three comparisons examined
differences between sites within years (test versus
control). The final two compared pre- versus post-treatment
effects within a section. I set the significance levels for

ANOVA's at a = 0.05 and used the Bonferroni routine to

16
adjust for non-orthogonal contrasts (Ott 1988). Using this
routine, an error rate of a = 0.01 was used for these
contrasts.

I tested for differences in condition between test and
control sections within sampling periods using analysis of
covariance (ANCOVA). Using this design, ANCOVA compared
slopes (condition index) for significant (a = 0.05)
differences (Snedecor and Cochran 1982). This test allowed
effective comparisons of fish condition between reaches and
sections. To determine which reaches were significantly
different, I used a Tukey-Kramer multiple comparison test

(Miller 1986).

Effects of Salmonine Density on Brown Trout Size

To further discern differences between the impact of
steelhead on brown trout, I constructed a multiple
regreSsion model. This model examined the relationship
between brown trout total length (Y) and brown trout density
(X1) and steelhead density (X2). I constructed separate
models for age-0 and age-1 brown trout and steelhead from

site 2.

Table 2.

17

Selected sets of single degree-of—freedom linear

contrasts used to test for the effect of steelhead
upon juvenile brown trout population parameters.
The numbers correspond to data comparisons made.

Across rows sum equals 0.

Data values with

different signs are the comparisons being made.

Comparison Site 1

Site 2

Site 1

Site 2 Site 1

Site 2

 

Age-0

1989 Cohort

1990 Cohort

1991 Cohort

 

Between Sites by Year

 

 

 

1989 -1 1 0 O O
1990 0 0 -l
1991 0 O 0 O -1 1
Pre- vs Post-Treatment
Site 1 2 0 -l 0 -l 0
Site 2 0 2 0 -1 O -l
Age-1 1988 Cohort 1989 Cohort 1990 Cohort

 

Between Sites by Year

 

 

1989 -1 l O 0
_1990 0 0 -1 l
1991 O O O 0

Pre- vs Post-Treatment
Site 1 -1 0 -1 0
Site 2 0 -1 0 -1

0 0
0
-1

0

18

RESULTS

Steelhead Introduction and Population Assessment:

Steelhead stocking proceeded well in both years.
Transport mortality was low (< 0.03 %). Temperature
conditions were excellent with the average stream
temperature and average bag temperature varying by no more
than 0.3° C. Fry showed no stress and swam immediately to
cover upon release. Stocking of steelhead swim-up fry
established steelhead cohorts in the test section of
Gilchrist Creek in both years. Steelhead were not found in
the control section.

The population densities and mortality patterns of age-
0 steelhead were similar between 1990 and 1991 (Table 3).
AnalySis of steelhead density using ANOVA found no
significant differences in population density ( P > 0.05) in
June, October, or February samples between years. Age-0
steelhead density averaged 4620 fish in June, 1781 fish/ha
in October and 368 fish/ha in February. No significant
differences were noted in summer or winter instantaneous
mortality rates between the two years (P > 0.05).
Instantaneous mortality in the growing season averaged 1.06
while in the winter it averaged 1.57. However, stocking to

June mortality was significantly higher in 1991 (4.31) than

19

Table 3. Steelhead abundance (S.E.- standard error) and
corresponding instantaneous mortality rates (Z)
for 1990 and 1991 cohorts in the test section of
Gilchrist Creek. P represents the significance
level of ANOVA comparisons of cohort Z values.

 

 

 

 

1990 Cohort 1991 Cohort

Sampling Abundance Abundance

Period #/ha (S.E.) Z #/ha (S.E.) Z P
Stocking 35534 242415
June 4982 (1288) 2.03 4259 (1656) 4.31 0.03
October 2366 (1564) 1.11 1196 (171) 1.01 0.94
February 250 (157) 2.24 486 (79) 0.90 0.35
June 260 (66) -0.42
October 197 (26) 0.26
February 92 (26) 0.78

it was in 1990 (2.03)( P < 0.05).

Steelhead fry size and instantaneous growth rates were
similar between 1990 and 1991 cohorts (Table 4). Fry were
stocked at a significantly larger size in 1991 (27.1 mm)
than in 1990 (25.5 mm). This initial size difference (P <
0.05) was maintained in June where age-0 steelhead averaged
37.8 mm in 1990 and 41.9 mm in 1991. However, by October
and February, age-0 steelhead were not significantly
different between years as summer growth rates were
significantly (P < 0.05) higher in 1990 (0.0070) than in
1991 (0.0055). This growth difference was apparent only in
summer growth. No significant differences in growth

occurred during the winter period (P > 0.05).

20

Table 4. Steelhead total length (TL) and weight (WT) and
corresponding instantaneous growth rates for 1990
and 1991 cohorts in test section of Gilchrist
Creek.

 

 

1990 Cohort 1991 Cohort
Sampling TL mm WT g Inst. TL mm WT g Inst.
Period S.E. S.E. Growth S.E. S.E. Growth
Stocking 25.5 0.12 27.1 0.14
0.20 0.003 0.21 0.004
June 37.6 0.46 0.0158 41.1 0.67 0.0135
0.32 0.025 0.50 0.029
October 77.7 4.53 0.0070 80.4 4.98 0.0055
0.84 0.150 1.20 0.268
February 83.6 5.16 0.0008 84.8 5.35 0.0005
2.08 0.415 1.45 0.291
June 138.3 25.04 0.0034
2.07 1.254
October 164.3 38.17 0.0015

3.56 2.805

February' 168.3 41.43 0.0001
5.36 4.073

firgyn Trout Abundance and Mortality

Age-0 brown trout abundance was not significantly
different (P > 0.05) in June or October samples for any
contrast (Table 5; Appendix 1). Differences did exist in
February samples. Abundance of the 1989 cohort was higher
in site 2 than in site 1 (P < 0.005). Brown trout density
in site 1 averaged 1,656/ha, while site 2 densities averaged
2,892/ha. The high abundance in site 2 also contributed to

a significant difference in pre-treatment versus post-

21

Table 5. Average abundance of age-0 brown trout per hectare
and standard error (S.E.) in control (site 1) and
test (site 2) sections of Gilchrist Creek for
1989-1991 cohorts. The * denotes presence of same

'age steelhead cohort.

 

 

1989 Cohort 1990 Cohort 1991 Cohort

Site 1 Site 2 Site 1 Site 2* Site 1 Site 2*
Sampling #/ha #/ha #/ha #/ha #/ha #/ha
Period S.E. S.E. S.E. S.E. S.E. S.E.
June 23110 118710 4115 5627 6481 4825
14014 104183 1196 1039 1223 1867
October 5627 1998 1919 3116 2472 1814
3326. 329 263 657 500 421
February 1656 2892 1196 1643 1670 1117
224 316 237 329 263 79

treatment contrasts (P < 0.001). These post-treatment
densities averaged 1,643/ha and 1,117/ha for 1990 and 1991
cohorts.

Analysis of instantaneous mortality rates supported
observations of density analysis. Summer mortality was not
significantly different (P > 0.05) for any contrasts (Table
6; Appendix 1). Age-0 winter mortality was significantly
different (P = 0.002) in the between-site comparison for
1989. In February 1989, instantaneous mortality rates were
0.38 in section 2 as compared to 0.90 in section 1. Pre-
versus post-treatment comparisons for section 2 were also
significant (P = 0.002). These differences were caused by
high brown trout abundance in the test section during

February 1990. Two of the three reaches in site 2 exhibited
this trend.

22

Table 6. Instantaneous mortality (Z) rates for age-0 brown
trout cohorts 1989-1991 in control (site 1) and
test (site 2) sections of Gilchrist Creek. Summer
represents June-October. Winter represents
October-February. Spring represents February-
June. The * denotes presence of same age
steelhead cohort.

 

 

1989 Cohort 1990 Cohort 1991 Cohort
Season Site 1 Site 2 Site 1 Site 2* Site 1 Site 2*
Summer 1.33 2.93 0.69 0.60 0.97 0.88
Winter 0.90 0.38 0.48 0.63 0.38 0.44
Spring 0.11 0.83 0.16 0.92

Spring mortality was significantly (P < 0.001) higher
in the test section than in the control section for both
1989 and 1990 cohorts. However, I found no significant
differences in comparisons between pre- and post-treatment
periods for the section 1 (P = 0.738) or section 2 (P =
0.443).

For age-1 brown trout, densities were not significantly
different (P > 0.05) for any between site comparisons within
years (Table 7; Appendix 1). However, pre- and post-
treatment comparisons of cohorts for section 1 were
significant. The 1988 and 1989 age-1 brown trout cohorts
were more abundant in June, October, and February samples
than was the 1990 cohort (P < 0.010). Since these
differences occurred in the control section, factors other
than steelhead were responsible for the change in overall

age-1 abundance.

23

Table 7. Average abundance of age-1 brown trout per hectare
and standard error (S.E.) in control (site 1) and
test (site 2) sections of Gilchrist Creek for
1988-1990 cohorts. The * denotes presence of same
age steelhead cohort.

 

 

1988 Cohort 1989 Cohort 1990 Cohort

Sampling Site 1 Site 2 Site 1 Site 2 Site 1 Site 2*
Period #/ha #/ha #/ha #/ha #/ha #/ha
S.E. S.E. S.E. S.E. S.E. S.E.
June 3563 841 1499 1302 999 1643
1328 234 250 276 105 329
October' 1209 592 999 776 933 670
158 53 39 105 184 157
February 868 171 762 460 868 644
184 53 158 53 105 184

Analysis of instantaneous mortality rates yielded
somewhat different results (Table 8; Appendix 1). Summer
and spring mortality rates were not significantly different
(P >0.01) for any contrast. The only instantaneous
mortality rate that was significant was the section 2
comparison pre- versus post-treatment (P = 0.005). This
difference was driven by the extremely low instantaneous
mortality rate of the 1990 cohort in the winter of 1991-1992
(0.045). As a result, survival was higher in sympatry with
age-0 steelhead in section 2. The effect of steelhead on

age-1 abundance appears to be insignificant.

24

Table 8. Instantaneous mortality (Z) rates for age-1 brown
trout cohorts 1988-1990 in control (site 1) and
test (site 2) sections of Gilchrist Creek. Summer
represents June-October. Winter represents
October-February. Spring represents February to
June of the next year. The * denotes presence of
same age steelhead cohort.

—
1988 Cohort 1989 Cohort 1990 Cohort

 

Season Site 1 Site 2 Site 1 Site 2 Site 1 Site 2*
Summer 0.960 0.303 0.379 0.487 0.108 0.146
Winter 0.373 1.359 0.304 0.511 0.387 0.045
Spring 0.849 0.202 0.347 0.577

 

 

Brown Trout Size and Growth

The size and growth of age-0 brown trout was variable
throughout the course of the study (Table 9 and Table 10).
Since total length and weight were highly correlated (:3 =
0.97), statistical comparisons of size yielded comparable
results using either variable. As such, only length data
are presented.

In June, between-site comparisons within years found no
significant difference (P > 0.05) in size of age-0 brown
trout (Table 9; Appendix 2). However, significant size
differences existed in February (1990, 1991) and October
(1990). In all three cases, age—0 brown trout were larger
in the control section than the test section (P < 0.002).
Only the February 1990 data were from a pre-treatment
sample.

Pre- versus post-treatment comparisons of total length

were also complex (Table 9). Results were significant

25

Table 9. Average total length (TL) and weight (WT) of age-0
brown trout and standard errors (S.E.) for these
measurements. Separate data are given for the
control (site 1) and test (site 2) sections of
Gilchrist Creek for 1989-1991 cohorts. The *
denotes presence of same age steelhead cohort.

 

 

 

 

1989 Cohort 1990 Cohort 1991 Cohort
Date Site 1 Site 2 Site 1 Site 2* Site 1 Site 2*
TL (mm) S.E.
June 47.2 45.1 54.5 52.5 58.4 58.6
0.31 0.26 0.42 0.39 0.67 0.56
October 86.9 82.8 95.6 82.4 91.1 89.0
0.62 0.65 0.72 0.61 0.86 1.09
February 94.9 84.8 99.2 89.9 97.6 93.1
1.04 0.87 0.82 0.78 0.82 1.01
WT (g) S.E.
June 1.11 1.10 1.97 1.51 2.15 2.02
0.036 0.032 0.201 0.055 0.073 0.059
October 7.26 6.88 8.32 5.49 8.37 7.34
0.282 0.297 0.225 0.137 0.288 0.327
February 7.89 5.83 9.06 6.79 8.12 7.19

0.256 0.191 0.234 0.202 0.207 0.227

Table 10. Seasonal instantaneous growth rates for age-0
brown trout cohorts 1989-1991 in control (site 1)
and test (site 2) sections of Gilchrist Creek.
Summer represents June-October. Winter represents
October-February. Spring represents February to
June of the next year. The * denotes presence of
same age steelhead cohort.

”1989 Cohort 1990 Cohort 1991 Cohort
Season Site 1 Site 2 Site 1 Site 2* Site 1 Site 2*
Summer 0.0049 0.0048 0.0054 0.0043 0.0038 0.0036
Winter 0.0006 0.0002 0.0003 0.0007 0.0006 0.0004
Spring 0.0033 0.0037 0.0028 0.0035

26
in 4 out of 6 contrasts. June comparisons found the 1989
cohort to be significantly smaller than the 1990 or 1991
cohorts in both test and control sites (P = 0.001). The
treatment effects were also significant in section 1 in
October (P = 0.006) and in section 2 in February (P =
0.005). However, in both these cases, brown trout were
larger post-treatment than pre-treatment. As such, size
differences between years were of greater magnitude.

Growth rate comparisons provide a clearer examination
of size data since the confounding effect of initial sizes
[is downplayed (Table 10). Only one significant between site
comparison was found (Appendix 2). Age-0 summer growth
rates were significantly higher (p = 0.001) in section 1
than in section 2 for the 1990 cohort. A single significant
within-site comparison was also found. The summer growth of
age-0 brown trout was significantly (P = 0.001) slower post-
treatment than pre-treatment in section 2. The same
comparison in the control section revealed no such effect
(P= 0.183).

Age-1 brown trout size and growth followed a similar
pattern to age-0 fish (Table 11 and Table 12). Two
significant differences were found in between-site
comparisons (Appendix 2). In October 1989, age-1 fish were
significantly larger (P =0.007) in section 2. In June 1990,
age-1 fish were again larger in section 2. All other

between-site comparisons were nonsignificant. Pre-treatment

Table 11.

27

Average total length (TL) and weight (WT) of age-1
brown trout and standard errors (S.E.) for these
measurements. Separate data are given for the
control (site 1) and test (site 2) sections of
Gilchrist Creek for 1988-1990 cohorts. The *
denotes presence of same age steelhead cohort.

1988 Cohort 1989 Cohort 1990 Cohort

 

 

 

 

Date Site 1 Site 2 Site 1 Site 2 Site 1 Site 2*
TL (mm) S.E.
June 124.9 132.3 141.4 132.2 145.4 143.3
1.01 1.20 1.32 1.61 1.41 1.42
October 154.5 165.8 161.7 156.9 171.8 167.0
1.29 1.64 1.41 1.69 1.80 2.15
February 163.4 167.3 170.8 163.9 180.9 180.7
1.97 4.02 1.81 2.60 4.53 3.13
m (g) S.E.
June 20.51 24.68 32.51 26.13 31.85 28.9
0.514 0.678 0.928 0.993 0.906 0.832
October 34.17 41.84 39.32 36.40 47.84 42.35
0.873 1.348 1.034 1.170 1.551 1.695
February 39.86 41.73 47.70 40.42 54.32 53.97
1.522 3.093 1.549 1.937 4.019 2.869

Table 12. Seasonal instantaneous growth rates for age-1

brown trout cohorts 1988-1990 in control (site 1)
and test (site 2) sections of Gilchrist Creek.
Summer represents June-October. Winter represents
October-February. Spring represents February to
June of the next year. The * denotes presence of
same age steelhead cohort. '

Season
Summer
Winter

Spring

1988 Cohort 1989 Cohort 1990 Cohort

 

Site 1 Site 2 Site 1 Site 2 Site 1 Site 2*
0.0017 0.0018 0.0012 0.0016 0.0014 0.0013
0.0004 0.0004 0.0005 0.0004 0.0005 0.0003
0.0027 0.0021 0.0020 0.0027

28
versus post-treatment differences were significant in
section 1 in June and October and in section 2 in June
samples. In all three cases, age-1 brown trout were larger
post-treatment than pre-treatment. These data seem to
indicate no negative effect of steelhead on age-1 size.
Analysis of instantaneous growth rates supports size data
(Table 12; Appendix 2). No significant differences in
growth rates were found in between-site or within-site

comparisons.

BIO!D.I£22§.§QDQ1E12D

Analysis of length-weight regressions yielded excellent
models for age-0 and age-1 brown trout. All slopes were
significant (P < 0.01) and the mean coefficient of
variations were 0.91 and 0.96 for age-0 and age-1 brown
trout. The mean slope was 2.93 for age-0 brown trout and
2.51 for age-1 fish. Comparisons of differences between
stream reaches during each sampling period (blocked by month
and year) were performed by ANCOVA. All ANCOVA were
significant ( P < 0.05). Tukey-Kramer multiple comparison
tests were used to determine which stream reaches were
significantly different. No specific pattern of fish
condition comparisons between reaches in section 1 and
section 2 was found. Reaches that had the greatest
condition factor at one time period had the lowest in the

next. As such, this analysis indicated that condition was

29

independent of the presence of steelhead.

Multiple Regression Models

To discern the role of interspecific competition and
intraspecific competition, multiple regression models were
constructed for age-0 and age-1 brown trout. These models
used brown trout length as the dependent variable and brown
trout and steelhead abundance as the independent variables.
Each model contained only data from fish of the same age
group. The model for age-0 brown trout was significant (P <
0.01) and the r2 for the model was 0.38. The number of
steelhead and the number of brown trout were both inversely
related to age-0 size (P < 0.01). Parameter estimates were
-0.002 for brown trout abundance and -0.049 for steelhead
abundance. As such, density effects of both species are
inversely related to brown trout size and account for 38% of
the observed size variation.

The model for age-1 brown trout was significant (P <
0.01) and the r2 was 0.46. An inverse relationship between
density and size was apparent but only the estimate for
brown trout abundance was significant (P < 0.01). The
parameter estimate for this variable was -0.41, indicating
that at age-1, intraspecific competition was more important
than was interspecific competition. It is important to
remember that steelhead abundance was low by age-1 and that

this could affect the outcome of interactions.

30

DISCUSSION

Wtation of Resulitjs

The outcome of interactions between two species of fish
is a complex process involving habitat segregation (innate
temporal or spatial differences), interactive segregation
(competition related behavioral interactions), and the
varying effects of abiotic factors (Hearn 1987). The
importance of each of these factors in the outcome of
competition depends on the species involved. One central
theme of salmonine competition studies has been the
importance of size (Chapman 1966; Fausch and White 1986;
Hearn 1987). Larger fish tend to dominate in intraspecific
and interspecific conflicts. In addition to size, the
comparative abundance of the two species can have an impact
on the outcome of competition. It is the relationship of
size to competition and the effects of competition upon
abundance that work in concert to form a relationship
between the population dynamics of two species.

In interpreting the results of this experiment, I
assessed the impact of steelhead upon brown trout population
dynamics. In addition, understanding the relative success
of steelhead stocking and their survival can provide

insights into interactions between these species. As such,

31
I will examine the abundance and size relationships of both
species to gain a better understanding of the mechanisms

that may regulate their populations.

Steelhead and Brown Trout Abundance

Steelhead introductions established average fall
densities of 1791 age-0 and 197 age-l steelhead/ha.
Comparable densities were achieved in both years despite a
7-fold increase in stocking in 1991. Densities of age-0
steelhead were within the range (1,000-23,000/ha) reported
in the Great Lakes region while age-1 densities were below
(210-3700/ha) reported values (Seelbach 1986). Overall,
densities obtained were low compared to other Great Lakes
populations. Populations were also low compared to those
obtained by a fry stocking project in Lake Superior
tributaries (Close and Anderson 1992); they achieved late
summer densities averaging 19,250 fish/ha. A major
difference between the two introductions was that the Lake
Superior streams contained few resident salmonines (Close
and Anderson 1992).

During the course of this study, brown trout densities
remained high. The average autumn density of brown trout
was 3389 age-0 and 863 age-1 fish/ha. The only significant
population level response noted was the decrease in winter
abundance of brown trout in section 2 of Gilchrist Creek.

This section contained the highest winter density of age-0

32
brown trout in the 1989-1990 winter when brown trout were in
allopatry. Brown trout form over-winter aggregations as
colder temperatures trigger reductions in territorial
behavior and dominance hierarchies (Bachman 1984; Cunjak and
Power 1986). It is possible that this section contained
favorable age-0 wintering habitat, which concentrated brown
trout from nearby stream sections. Since no immigration was
apparent after steelhead introduction, this could be
interpreted as evidence of competition for wintering
habitat. However, combined brown trout and steelhead
abundance in this section was lower in 1991 and 1992 winters
than brown trout abundance in 1990. Given the lower overall
salmonine abundance, I feel that the presence of steelhead
was not a deterrent to any potential age-0 brown trout
immigration.

Any impact of age-1 steelhead on age-1 brown trout
abundance was not apparent during the course of this study.
Although differences in abundance occurred in both sites,
none could be ascribed to the presence of steelhead. This
was not surprising given the low numbers of steelhead
present by age-1.

Data indicated no impact of steelhead upon brown trout
abundance. This could be a result of the low numbers of
steelhead resulting from stocking. Given high salmonine
productivity in this system, I anticipated the establishment

of a larger steelhead population. Instantaneous mortality

33
rates for steelhead were highest between stocking and June
assessments. High initial mortality was also observed in
other steelhead fry stocking experiments (Wentworth and
LaBar 1984; Hume and Parkinson 1987; Close and Anderson
1992). They implicated competition and spring flooding as
probable causes of mortality. Gilchrist Creek was highly
stable in discharge and no major flooding was seen in the
spring of 1990 or 1991. This would diminish the mortality
due to floods. I suspect that interactions with juvenile
brown trout (competition) and larger brown trout (predation)
were major causes of steelhead mortality in this study

(Alexander 1977).

Steelhead and Brown Trout Size and Growth

Average size of steelhead fry compared favorably with
other studies in the Great Lakes. In this study, age-0 fry
averaged 79 mm and age-1 fish averaged 164 mm in autumn.
Seelbach (MDNR, pers. comm.) reported autumn size ranges of
50-80 mm and 112-196 mm for age-0 and age-1 steelhead in the
Great Lakes region. As such, steelhead were large relative
to other Great Lakes populations.

Brown trout were also large compared to populations
found elsewhere in Michigan (Merna et al. 1981; Alexander
1985). Age-0 brown trout averaged 89 mm and age-1 averaged
163 mm. Size of brown trout varied between years. None of

these differences was attributable to steelhead and occurred

34
in both sites. In between-site comparisons, size was not
different in June. Differences did occur in some autumn and
winter samples. These results show that steelhead played a
minor role in the size obtained by brown trout. Data from
laboratory studies reinforces this observation (Chapter 2).
Other factors influencing brown trout size and abundance
were probably more important. Year-to-year and between-site
differences indicate that factors such as intraspecific
population dynamics or abiotic factors played an important
role (Elliot 1984a, 1984b: Newman and Waters 1989). I feel
these factors were the major ones causing observed size
differences.

Despite the lack of impact of steelhead on the overall
size of brown trout, an impact on growth rate was indicated.
Tests of treatment effects on summer growth were highly
significant in the test section and not in the control
section. The slower growth of brown trout in sympatry with
steelhead indicates that steelhead interaction may inhibit
brown trout growth. Rose (1986) found that steelhead in a
Lake Superior tributary inhibited brook trout growth. This
is similar to the results of my experiment since brook trout
also emerge earlier than steelhead and were at a size
advantage. However, Rose (1986) concluded that steelhead
adversely impacted brook trout growth to the extent that
their wintering mortality was increased. The large response

of brook trout may be due to inherent differences between

35
brook trout and brown trout. However, the brook trout also
did not have as great a size advantage as the brown trout in
this study. The results of both studies indicate that
despite a size disadvantage, steelhead may be able to impact
the growth of larger salmonine species of the same cohort.
This could be important since climactic factors could change
the relative size structures of the two species and, hence,

the outcome of interspecific interactions.

Combined Effects of Siee and Abundance

The impact of juvenile steelhead on juvenile brown
trout population dynamics in this study was insignificant.
Analysis of these results in the context of present
knowledge of salmonine competition ecology, variations in
life-history, and climactic conditions allows a better
understanding of interactions between these species. I
present a conceptual model of these species juvenile
interactions (Figure 2). This model allows interpretation
of the results of this experiment as well as emphasizing

other areas of competition and future research needs.

EESIQ§B2§_§QP

The difference in emergence times leads to a difference
in size of juvenile salmonines. At hatch and emergence,
steelhead and brown trout are roughly the same size (Auer

1982). However, different spawning and incubation times

36

.emaum>o Hmuomemu one Howuoem no woman ocfiumuumSHHH mucmscoufl>co mcfiuu>fiu
cw poocaooum one usouu csoun mawco>9n cowsuon mcowuowumucfl mo Hooos Hmsumoocoo .m ouzmwm

885665 I so... Esem

 

">3.

 

 

_ “mm _ on: .56 L 22. _ >52. .22 _ 5:: mm“. _ zS. _ own _ >oz_ Bo

cocoa 526.0 388.5

 

 

.1. IV.

in.

 

totem monomeeEmLmoe
8.6.88 an. e
_ 1 I coron— oocomeoEm

I
now eczema—cm

 

 

 

 

 

37
create a temporally regulated size difference. The earlier
spawning and emergence of brown trout gives them a distinct
size advantage. In Gilchrist Creek, emergence is usually
complete by the end of April (Chapter 3). Reported
emergence dates for some Lake Michigan tributaries are as
late as mid-May (Fausch and White 1986). Steelhead stocking
did not occur until late May, corresponding to peak natural
emergence (Seelbach 1986). The length of the emergence gap
gave brown trout at least a 30 day head start in exogenous

growth (Figure 2).

Post-Emergent Stage

The effect of the emergence gap was manifested during
the post-emergence stage (Figure 2). In Gilchrist Creek,
age—0 brown trout were 30% larger than steelhead in June.
This size advantage is important in competitive interactions
between riverine salmonines (Fausch and White 1986; Hearn
1987). Direct competition is highly probable since both
species heavily utilize stream margins until mid-summer
(Sheppard and Johnson 1985; Chapter 3). Agonistic behavior
between salmonines starts at both an intraspecific and
interspecific level shortly after emergence. Salmonines
exhibit confrontational behavior, and dominance hierarchies
or territories are formed (Jenkins 1969; Fausch and White
1986; Hearn 1987). Fish gaining the most profitable

positions are most likely to dominate (Fausch 1984).

38

Since diet, feeding strategy, and habitat use of these
species overlaps substantially, interactions between the
species are likely (Elliot 1973; Cada et al. 1987; Ziegler
1988; Chapter 3). Vertical segregation of fry has been
proposed as an isolating mechanism between these species
(Chapter 3). However, the shallow nature of marginal
habitats might negate this isolation mechanism. Thus,
direct interactions between emergent steelhead and
established brown trout probably occurred. Given the
results of this experiment and the relative size of these
species, brown trout should dominate at this post-emergence
stage. Subordinate fish typically exhibit lower growth,
higher mortality, and many emigrate from the area (Chapman
1962; Hearn 1987). In my study, these factors gave brown
trout juveniles a numerical advantage over steelhead that

they maintained.

W

The summer growth stage is characterized by the habitat
shift of age-0 fish to slightly deeper and faster water
(Sheppard and Johnson 1985; Chapter 3)}Figure 3). From this
point onward, habitat use by the two cohorts will be very
similar until steelhead outmigration (Chapter 1; Raleigh et
al. 1984, 1986; Ziegler 1988). Another important element,
the faster growth rate of steelhead, allows them to reduce

size difference with brown trout. This higher growth rate

39
for steelhead has also been seen in other Great Lakes
studies (Kruger 1985; Ziegler 1988). As such, the size
difference between species becomes neutralized as a factor
in competitive interactions. However, with the shift to
somewhat deeper water, vertical stratification could act as
a habitat segregation mechanism (Chapter 3). The influence
of segregation might be to reduce or eliminate potential
competition (Allee 1982; Hearn 1987).

In my study, no impact of steelhead on brown trout
abundance or growth was seen during summer growth periods or
thereafter. This may have been the result of low steelhead
abundance and/or habitat segregation. Had the gap in
emergence not given brown trout the initial competitive
advantage at the post-emergence period, the impact of

steelhead might increase at this time.

Other Life-History Stages

Due to differing life-histories, interactions between
these two species in riverine environments are limited to
only parts of each species’ life history (Figure 2 and
Figure 3). This study only examined the relationship
between juvenile brown trout and steelhead. This is the
longest period of interaction, and previous research
indicated that competition might be occurring here (Kruger
1985; Ziegler 1988). However, it is not the only time

period where these fish interact.

40

.mwmmum known“: mafia Ham um mucuficouw>cm
mcfium>wu Cw pounamoum one Moon» csoun cmuzuun mcofluomumucw no H0605 Hosummocoo .m gunman

862.565 I so; Esem

 

">02

 

 

1 mum _ 0:: 3.. _ 22. _><_2_ E< _ 5:2. mm“. _ z<_. _ 83 >02_ 53

cores 5320 358.5

 

 

 

E...

‘Ili'
UO...On_ Oocomhmsmlumom

_1 l cocon— oocomeoEm

 

 

 

 

 

Esmamuzm Esoamuhm

 

 

 

41

Interactions at the spawning and egg-alevin stage may
also be important (Figure 3). In the Great Lakes region,
brown trout typically spawn from October to November, and
emergence from redds occurs from March to May (Elliot 1984b;
Chapter 3). Steelhead spawning occurs from December to June
(Seelbach 1986). Given the overlap in dates, steelhead redd
superimposition could physically damage or dislodge brown
trout prematurely from redds (Figure 3). During this
overlap in gravel use, steelhead could potentially diminish
the abundance of a brown trout cohort. The temporal
relationship between steelhead spawning and brown trout
emergence would determine the extent and impact of these
interactions. It has been suggested that available gravel
spawning areas are limited in Great Lakes streams, leading
to superimposition and egg overseeding (Seelbach 1986; Kocik
and Taylor 1987, 1991). If spawning steelhead were
overabundant, all spawning gravel in a stream reach might be
excavated during redd construction. If this were the case,
the impact of temporal overlap between these life-history
stages could be substantial. The relationship between these
species at these life-history stages warrants a thorough

investigation and should be a direction for future research.

Impacts of Differing Climape

 

Under environmental conditions observed in this

experiment, juvenile brown trout were not adversely impacted

42
by steelhead. However, annual variations in climate create
an opportunity for differing outcomes of brown trout-
steelhead interactions. Two stages of interactions could be
influenced: the steelhead superimposition period and the
emergence gap (Figure 3). Since brown trout are autumn
spawners and steelhead are spring spawners, atypical or
changing conditions during these period could affect
interactions between these species. For example, a cold
winter followed by a warm spring could cause a greater
overlap in steelhead superimposition of brown trout redds
due to delayed brown trout alevin development. This would
decrease brown trout abundance. In addition, the emergence
gap could be shortened. Elliot (1984a) noted up to a 50 day
difference in brown trout emergence times with temperature
being the predominant factor in these differences. The
impact of a shift of this magnitude would be the reduction
or 1035 of the brown trout size advantage at the post-
emergence stage. In concert, the result could be a stronger
steelhead year class at the expense of brown trout.

This can be illustrated in comparisons of brown trout
size in June between years. In June of 1990 and 1991, brown
trout were 30% larger than steelhead. However, if steelhead
were introduced in 1989 they would have been only 13%
smaller. This could be enough to make a significant
difference in the outcome of competition. This seems

especially important given the results of studies of brook

43
trout and steelhead interactions (Rose 1986). Thus, results
of this study should be used with care in regions with

different or more variable winter and spring climates.

W

In this study, interactions between juvenile brown
trout and steelhead appear to favor brown trout. The major
arena of competition of juveniles appears to be the post-
emergent stage. This stage is critical in determining
future juvenile abundance. After this stage, the juveniles
appear to be compatible (Chapter 2; Chapter 3). As such,
this study provides new directions for assessment of
competition between these species: spawner-alevin
interactions and climatic influences. Culmination of these
studies will complete the assessment of life-history stage
interactions between these species.

Given the results of juvenile competition experiments,
it appears that with adequate temporal segregation, the two
species can coexist. It should be noted that steelhead are
very similar in stream ecology to Atlantic salmon, which
coexist with brown trout in many systems (Gibson and Cunjak
1986; Seelbach 1987; Hearn 1987). This evidence suggests
that coexistence of these species is possible. However, the
effects of human perturbations on Atlantic salmon-brown
trout systems (Hearn 1987) has caused a loss of balance in

some populations to the detriment of one of the species.

44
The long history of natural reproduction of brown trout and
steelhead in the Great Lakes indicates their success in
these systems. It is possible that continued changes in the
Great Lakes lacustrine ecosystems and climate changes are
responsible for changes in the balance between brown trout

and steelhead (Scavia 1991).

CHAPTER 2

IMPACT OF AGE-0 STEELHEAD UPON AGE-0 BROWN TROUT IN AN
ARTIFICIAL STREAM EXPERIMENT DURING THE FIRST SUMMER

ABSTRACT

I examined the impact of age-0 steelhead (Qpppppypgppe
mykiss) on age-0 brown trout (gelmp pppppe) during their
first summer in an artificial stream experiment. I ran a
completely randomized design experiment with 4 replicates
and two treatments. The control treatment was allopatric
brown trout, and the test treatment was sympatric brown
trout with steelhead. A total of 14 fish were placed in
each experimental cell with control cells containing 14
brown trout and test cells containing 7 brown trout and 7
steelhead. The experiment ran for 99 days. Water
temperatures and habitat types were similar to those
encountered by wild trout. Fish were fed invertebrates
twice daily during the experiment. No effect of steelhead
on brown trout survival or growth was detected. Density-
dependent interactions appeared to impact both survival and
growth of brown trout. However, intraspecific interactions
within brown trout populations were more important than
interspecific interactions with steelhead. The results of
this experiment indicate that steelhead have a minimal
impact on brown trout and the species can coexist during

their first summer of life.

45

46

INTRODUCTION

Interspecific competition between stream dwelling
salmonines has been implicated as an important factor in
their population dynamics (Hearn 1987; Fausch 1988).

Concern over competition in riverine salmonines in the Great
Lakes region is high since many of the species are exotics,
habitat use is similar, and most species combinations_have
not coevolved (Carl 1983; Fausch 1984; Fausch and White
1986). Of particular interest has been competition between
brown trout and steelhead (Kruger et al. 1985; Kruger 1985;
Ziegler 1988; Chapter 1).

Michigan anglers, concerned over the apparent decline
of resident brown trout populations, felt that oncorhynchids
were adversely affecting the brown trout fishery. Kruger
(1985) concluded that brown trout abundance had declined and
that juvenile growth was relatively slow in the Pere
Marquette River. He also observed an inverse relationship
between brown trout abundance and steelhead abundance.

These results suggested that steelhead may interact
detrimentally with juvenile brown trout. In a study of 3
Michigan rivers, Ziegler (1988) found that age-0 fish of
both species use similar habitat and food. Given the

results of Kruger (1986) and Ziegler (1988), the need for a

47
clearer understanding of competition between these
salmonines became apparent.

I assessed brown trout population dynamics and habitat
use in a brown trout stream in northern Michigan (Chapter 1;
Chapter 3). In these investigations, I found the effects of
steelhead juveniles on brown trout to be minimal. To
further assess the potential for brown trout-steelhead
juvenile competition, an artificial stream experiment was
conducted. Artificial stream experiments provide valuable
supplements to field studies since abiotic variability can
be minimized (Kalleberg 1958; Li and Brocksen 1977). In
natural ecosystems, food and space can limit the abundance
and growth of stream salmonines (Chapman 1966; Fausch and
White 1986; Cada et al. 1987). These parameters can also
vary between systems and within longitudinal sections of the
a stream (Elliot 1984a; Cada et al. 1987). Artificial
stream experiments allow measured and replicated food
availability, habitat volume, and fish density. As such,
artificial stream experiments allow for true replication for
statistical analysis.

Artificial stream experiments have been successfully
used to analyze competition in stream dwelling salmonines
(Kalleberg 1958; Li and Brocksen 1977; Fausch 1984; Fausch
and White 1986). Previous artificial stream experiments
have illustrated that salmonines form dominance hierarchies

in artificial streams (Li and Brocksen 1977; Fausch and

48
White 1986). These hierarchies allow dominant fish to use
the most energetically profitable stream positions. As a
result, the dominant fish species grew better and exhibited
higher survival (Kalleberg 1958; Li and Brocksen 1977:
Fausch 1984; Fausch and White 1986). Thus, the results of
behavioral interactions and the formation of dominance
hierarchies can be measured by assessing growth and
mortality rates.

The goal of this research was to assess the impact of
age-0 steelhead upon age-0 brown trout survival and growth
in an artificial stream. Competition was assessed through
comparisons of brown trout in allopatry and in sympatry with
steelhead. The experimental design removed abiotic and
biotic variability found in natural systems. By regulating
densities, diet quantity and quality, and habitat, the
effects of steelhead competition upon juvenile brown trout

population dynamics were evaluated.

49

METHODS

Stpeam Design and Macrohabitat

I conducted the competition experiment in a
recirculating artificial stream channel with the inside
dimensions of 60 cm x 60 cm. The channel was partitioned
into 8 experimental cells (Figure 4). Experimental cells
were 1 m in length. Cells were divided by 5 mm mesh nylon
screens held in place by plexiglass frames with edges made
fish-proof using nylon tubing. The screens had a negligible
affect on overall velocity while preventing fish passage
between cells.

Flow was maintained through using of two 1 hp water
pumps adapted to generate 4 jets of water. The filtration
system pump supplemented the main pumps while running an
above ground pool filter. I placed sand, filter floss, and
charcoal into the pool filter to create a biofilter. The
entire volume of water was filtered two times an hour.

Mercury-vapor and fluorescent lights were used to light
the stream. The mercury vapor lights took approximately 30
minutes to reach full intensity. As such, sunrise and
sunset effects were created by these lights. Light cycles
followed ambient norms and were controlled by an electronic

timer. Analysis of variance was used to compare differences

Cell 4

Cell 3

Cell 2

Cell 1

 

 

 

 

H ill ii iii

 

 

 

/

50

O
O

 

Legend

 

 

Cell 5

 

Cell 6

 

Cell 7

 

Cell 8

 

H H} H ill

 

 

 

1 meter

1
i

Brown Trout
«

Steelhead
Qfl

Cooling Unit

Water Filter

 

Water Pump

 

l
l

 

Figure 4. Artificial stream design and experimental

structure.

 

51
in maximum light intensity between cells. No significant
differences were found (P > 0.05), and the mean light
intensity at the water surface was 950 lux.

Two 1 hp chiller units were used to maintain stream
temperatures. I used temperature data from my previous
field studies in Gilchrist Creek as a guide to regulate
temperature in the artificial stream (Kocik 1990,
unpublished data). The mean monthly artificial stream
temperatures were 15.7° C, 16.6° C, 13.2° C, and 12.9° C for
June, July, August, and September. Daily temperature
fluctuations as high as 1.5° C were observed, due to
increased heating by lights and changes in building airflow.
These fluctuations were similar to natural fluctuations with
the highest temperatures in the middle of the day. The
velocity profile of the stream kept the water mixed and at a

uniform temperature at all stream cell locations.

Stream Cells and Microhabitat

Each cell was a replicate in regards to microhabitat
variables of depth, velocity, substrate, and cover. The
water depth in all cells was maintained at 30.5 cm for the
duration of the experiment. Water added at feeding
countered evaporation and excess would drain off through a
standpipe.

Mean column velocity was measured at 20 replicate

locations in each cell. Eight of these represented

52

positions in stream cover and 12 were in open water. Their
was no significant difference in water velocity at either
location set between cells (ANOVA, P > 0.05). Average open
water velocities were 13 cm/s. In cover locations the mean
was 11.4 cm/s. The overall average was 12.4 cm/s. Water
depth and water velocities were low but within ranges
utilized by both species in field experiments (Chapter 3).

The substrate was identical in all cells: clean gravel
ranging in size from 4 to 8 mm comprised the substrate. All
cover types were artificial to ensure uniformity for
replication. Each cell had 5 individual cover structures
placed in identical locations in relationship to the cell
ends and sides (Figure 5). Three artificial plants were
placed in each cell: two corkscrew vallisneria (Vallineria
spiralis) and one ambulia (Limnophila heterophyilia). The
vallisneria were 30 cm in length and the ambulia were 25 cm
long. These species were chosen due to their similar
appearance and size to aquatic plants utilized by these fish
in field studies (Chapter 3). Artificial logs were made
from gray 15 cm pvc pipe cut in half laterally and cut into
21 cm lengths. These logs were then placed in the gravel

with two thirds of the pipe exposed.

's r cu e t oldin
Wild brown trout and hatchery incubated progeny of wild

steelhead were used in this experiment. Brown trout were

53

 

 

 

 

 

 

UPSTREAM
/\
g
@A
KO 2%
A A A
Q A A
DOWNSTREAM A
Corkscrew Vallisneria (Vallineria spiralis) @
Ambulia (Limnophila heterophyllia) mg:

Artificial Log '

 

Figure 5. Artificial stream cell illustrating habitat size
and placement.

54
collected by D.C. electrofishing from Gilchrist Creek,
Montmorency County Michigan. They were transported back to
the laboratory in a 190 l cooler with aeration units and
ice. A total of 130 brown trout were available at the start
of the experiment. I received 200 steelhead swim-up fry
from the Michigan Department of Natural Resources Wolf Lake
Hatchery. The progenitors of these fish were wild steelhead
collected at the Little Manistee River Weir and Egg
Collection Station.

Until introduced into the experimental stream, brown
trout and steelhead were kept isolated in a separate living
stream unit. Fish were fed live and frozen brine shrimp
(Aptemia ep.) twice daily while in the holding tank. Brown
trout were kept in the holding tank for 15 days prior to
introduction into the artificial stream. Steelhead were
kept in the holding tank 32 days. The longer holding period
allowed steelhead to reach a size where they would not go
through the cell partitions (30 mm). In addition, brown
trout were able to establish a hierarchy in the artificial

stream prior to steelhead introduction.

W

To test the impact of steelhead on brown trout survival
and growth, I used a completely randomized design with two
treatments. Control cells would have initial populations of

14 brown trout. Test cells initially contained 7 brown

55
trout. After the 17 day acclimation period for the brown
trout, 7 steelhead were introduced to test cells.

Initial brown trout introductions were made on 21 June
1991. I randomly selected brown trout from the holding tank
using a 15 x 20 cm hand net. To facilitate handling and
minimize stress, fish were anesthetized using tricaine
methanesulfonate at a concentration of 50 mg/l. I measured
all brown trout for total length (nearest mm) and weight
(nearest 0.001 g). I then placed fish in separate
recovery/holding buckets that represented a test or control
cell. Placement of fish in test or control cells was
alternated to minimize any selective effect of the netting.
Fish ranged in length from 48 to 74 mm. I compared fish
total length and weights using an ANOVA and LSD comparison
to check for differences between holding buckets. No
significant differences were detected (P > 0.05). Test and
control buckets were randomly assigned to test and control
cells in the stream and fish were released. This yielded an
experimental design with 14 brown trout in control cells 1,
'3, 7, and 9 (allopatric) and 7 brown trout in test cells 2,
4, 6, and 8 (sympatric). During the first week of the brown
trout acclimation period, any brown trout that died were
replaced with brown trout of similar (within 4 mm) size from
the holding tank. The replacement was conducted to minimize
the effects of transfer mortality from the holding tank to

the artificial stream.

56

On 8 July 1991, I placed 7 steelhead in cells 2, 4, 6,
and 8. Steelhead were randomly selected in the same manner
as brown trout. Their sizes ranged from 35 to 42 mm. I
compared steelhead total lengths and weights using an ANOVA
and LSD comparison to check for differences between cells.
No significant difference between cells was detected (P >
0.05). Steelhead mortalities were replaced with identical
sized fish for one week following introduction. Brown
trout were not replaced at this point. After the 7 day
steelhead introduction period, no mortalities were replaced.

After initial introductions and subsequent
replacements, fish were removed for two periodic vital
statistic assessments (15 August and 4 September 1991) and
at the end of the experiment (27 September 1991). The
collection and measurement protocol was the same for all
three collections. During these checks, I collected all
fish in a cell using a 10 x 15 cm hand net. Fish were
anesthetized using tricaine methanesulfonate at a
concentration of 50 mg/l. Fish were identified to species
and measured for total length and weight. They were then
allowed to recover from anesthesia in an oxygenated recovery
bucket. Once all fish had regained their equilibrium, they
were returned to the stream cell. Fish not accounted for at
this point were considered mortalities. Carcasses were
usually located during fish removal from a cell. All

carcasses were accounted for at culmination of the

57

experiment. All cells were handled individually.

Feeding and Maintenance Regimes

I fed fish twice daily in experimental cells using
live, frozen, or freeze-dried invertebrates. Diet was
varied throughout the experiment but was always identical
for all cells. Using the temperature specific relationship
between maximum food ration eaten per day and fish weight
(Elliot 1975), I calculated feeding biomass. A total of 10%
of fish body weight per day provided ample food. I
recalculated food amounts every time fish were removed for
measurement. Fish were fed at approximately 07:00 and at
16:00 each day. Foods were introduced through a funnel and
tube that released food at the gravel level in the center of
each cell at the upstream cell divider.

A mixture of food types was used to ensure that
nutrient requirements of the fish were met (Elliot 1975).
Predominant food types were frozen brine shrimp, frozen
bloodworms, and frozen and freeze-dried krill. Amounts of
each food were alternated each feeding period between frozen
foods, with one food making up 2/3rd of the frozen ration
and a second 1/3rd of the frozen ration. Frozen foods
comprised 92 % of the biomass of the daily diet. Freeze-
dried krill or live foods were used to supplement the diet.
Freeze-dried krill was fed about every third feeding period.

I also maintained live cultures of mosquitos, brine shrimp,

58

and thponomus riparius. These foods were introduced as
they became abundant enough to use as food.

Stream cells were vacuumed with the filtration system
intake hose each time fish were removed from the cells.
This helped remove excrements and food wastes. The filter
was back flushed as needed (when filter back-pressure
doubled), approximately every two weeks. Water removed from
the tank during back-flushing was replaced with clean water
from a water storage tank that was cooled to the same

temperature as the artificial stream.

Rate Calculations and Statistical Analysis

To identify critical time periods, I calculated
mortality and growth rates over specific time periods. The
pre-steelhead period ranged from initial brown trout
introduction to steelhead introduction (Table 13). After
steelhead introduction, the experiment was divided into
three phases.“ The early phase ran from steelhead
introduction until experiment check 1 (37 days). The middle
phase ran from experimental check 1 until experimental check
2 (20 days). The late phase ran from experimental check 2
until the end of the experiment (23 days). This periodic
analysis of the post-steelhead introduction phases was
designed to find differences as the size ratio of steelhead
to brown trout changed (ie. do steelhead have a greater

effect as they approach the size of brown trout).

59

Table 13. The timing of fish introductions, collections, and
subsequent time period references.

Date Experimental Operation Time Period
(Julian) Reference

 

21 Jun (172) Brown Trout Introduction

8 Jul (189) Brown Trout Measurements Pre-Steelhead
9 Jul (190) Steelhead Introduction

15 Aug (227) Experimental Check 1 Early Phase

4 Sep (247) Experimental Check 2 Middle Phase
27 Sep (270) Experiment End Late Phase

At each time period, the number of each species and
total salmonine density in each cell was known. Given this
abundance data, I calculated the instantaneous mortality
rate (Z) for each cell and species for each period and for
the duration of the experiment. I also calculated the
overall survival rate (S) of each species in each cell. The
following equations were used (Ricker 1975):

Z= - (loge NM-loge Nt) and S= (NM/ Nt)* 100
Where: .
instantaneous mortality rate
overall survival rate

”1 population of cell at time t+1
t population estimate of cell at time t

ZZMN

In order to assess the impact of changing salmonine
densities in cells as differential survival changed
abundance, I analyzed the size of brown trout at each

collection compared to overall salmonine, brown trout, and

60

steelhead abundances. These models were least squares
regressions of fish length to each abundance. If models
were significant, the slopes of the models were compared
using analysis of covariance (ANCOVA). Using this design,
ANCOVA compared slopes for significant (a = 0.05)
differences (Snedecor and Cochran 1982). By assessing the
effects of fish density on fish size, any confounding effect
on fish growth could be explained.

Mean lengths were used to calculate instantaneous daily
growth (u) for each cell using the following equation
(Ricker 1975):

u= (loge TLt--loge TLM)/T

Where:
u = instantaneous daily growth rate
TLt = mean total length (mm) at start of growth period
TLP1 = mean total length (mm) at end of growth period
T = time interval in days

Instantaneous growth rates were calculated for each time
period and for the overall experiment. Total length and
weight data were averaged for each cell to allow direct '
comparisons of fish size at specific points in time.

To assess differences in the condition of salmonines
between treatments, ordinary least squares regression of the
weight W on total length TL was used (loge W = loge a + b
loge'FL) as recommended by Cone (1989). Using this method,
the calculation of condition is unbiased since the slope of

an ordinary least squares regression can be used as the

61
condition index (Cone 1989). This model was used to
calculate this index for brown trout prior to steelhead
introduction and for both species at the end of the
experiment.

The structure of this experiment was designed to test
the impact of YOY steelhead upon YOY brown trout abundance,
survival, growth, and condition under replicate experimental
conditions. Statistical analysis was partitioned by
sampling date for point estimates and by periods for rate
variables. To examine the differences in brown trout
between test and control sections, I used analysis of
variance (ANOVA) (a = 0.05) (Steel and Torrie 1980). This
method was used to compare abundance, survival, and growth.
Analysis of instantaneous growth and mortality rates and
overall survival can be confounded, since ratio data are
usually not normally distributed (Steel and Torrie 1980).
Since small sample sizes precluded effective tests of
normality, I transformed the data using the RT-l
transformation (Conover and Iman 1981). Individual
observations were ranked according to their relative value,
and statistical tests were performed on the ranked data.
This transformation provides a more rigorous substitute to
the non-parametric Kruskal-Wallis test through a standard

ANOVA (Conover and Iman 1981).

62

RESULTS

Bpown Trout Mortality

The number of brown trout in all cells decreased over
time (Figure 6). In control cells, brown trout density at
the conclusion of the experiment fell to a low of 6 in cell
1 and a high of 9 in cells 8 and 3. In the test cells,
abundance fell to a low of 4 in cell 2 and a high of 5 in
cell 4. These data yielded an average instantaneous
mortality rate of 0.6516 in the control sections and 0.3467
in the test sections for the duration of the experiment.
This difference was non-significant (P > 0.05). Comparisons
of instantaneous mortality rates during specific time
periods were only significant during the early phase of the
experimental period. During this time, instantaneous
mortality was higher in the control section (0.4186) than in
the test section (0.1613).

Comparisons of overall survival rates (S) also showed
this trend. Overall survival was higher in test sections
(71.3%) than in control sections (53.5%) (P < 0.10). The
early phase of the experimental period was different with
survival averaging 85.8% in the test sections and 65.8 % in

the control sections (P < 0.05).

63

 

    
 

 

 

 

15

‘5
O Cell 8 8: Cell 3
g: 10—

Cell 1 \
c:
3
2
m .......... ‘ _____ Gel-L5 ________ Cell 6
u—
O ‘ ........ Cell 4
h ‘ ‘ ~ ..........................
0)
.CI 5 _ ‘~
§ ‘ ....... 9 91.2.-.
2

Control Cells
.......... Test Cells
0 1 l 1 1 l

 

 

 

160 180 200 220 240 260 280

Julian Date

Figure 6. Number of brown trout in allopatry and sympatry
surviving in individual artificial stream cells.

64

W

The total length of brown trout did not significantly
differ between test and control cells at the onset of the
experiment (P = 0.18) (Figure 7). Fish in the control cells
averaged 60.0 mm and 2.16 g. In the test section, brown
trout averaged 57.6 mm and 1.91 g. By 15 August, fish in
the test cells (72.3 mm, 3.84 g) were slightly larger than
fish in control cells (74.5 mm, 4.18 g) but differences were
not significant (P = 0.39). Brown trout in the test
sections continued to widen the gap with their counterparts
in the control sections. However, the difference was still
nonsignificant at the end of the experiment (P = 0.15).

Analysis of growth rates supports the observations of
the analysis of fish length and weight differences.
Instantaneous growth rates from the start of the experiment
to the end show that brown trout in the test section grew
faster than those in the control section (P < 0.05). The
mean instantaneous growth rate in the control section was
0.0039; in the test section it averaged 0.0050. Comparisons
during different time periods were only significant during
the early phase of the experimental period. During this
time period, the mean instantaneous growth rate in the
control section was 0.0009, while in the test section it
averaged 0.0013.

Fish condition in cells was measured using the slope of

the loge W-loge TL least squares regression. At the

65

100

 

l

90

00
O
l

N
O
l

O'i
O
l

at
O
l

  

Total Length (mm)

0 Brown Trout- Test

40 A Brown Trout- Control

i

E) Steelhead

 

 

30 1 l l l l l
160 180 200 220 240 260 280 300

 

Julian Date

Figure 7. Mean total length (mm) over experimental time for
brown trout in allopatry and sympatry and for
steelhead in sympatry.

66
point of steelhead introduction into the stream, no
significant differences in the condition of the fish existed
(ANCOVA, P > 0.05). At this point, the average condition
(slope) was 2.94. At the end of the experiment, the
condition of fish had improved to an average of 3.04, and
the difference between test and control sections was still

nonsignificant ( P > 0.05).

W

Steelhead mortality was generally low. For all cells
containing steelhead, the mean instantaneous mortality rate
over the duration of the.experiment was 0.1501 i 0.0526 S.E.
This translates to an overall average survival of 71.5 %.

Growth of steelhead was quite rapid (Figure 7). At the
start of the experiment, steelhead averaged 38.7 mm and 0.45
g. By the end of the trials they grew to an average of 82.5
mm and 6.14 g. The instantaneous growth rates of steelhead
over the course of the experiment averaged 0.0093 1 0.0001.
This growth rate was significantly higher than brown trout
growth rates (P < 0.05). This is illustrated in an
examination of the ratio of mean steelhead total length to
mean brown trout total length. At the start of the
experiment, steelhead were 0.6575 times smaller than brown
trout. By the end of the experiment, this ratio increased

to 0.8979 the size of brown trout. Steelhead growth was

good in all cells, and ANOVA comparisons of the mean total

67
length or weight of steelhead between cells were
nonsignificant at all time periods (P > 0.05).

The condition factor of steelhead averaged 3.39 at
their introduction to the artificial stream. No significant
differences in condition were noted in between-cell
comparisons (ANCOVA; P > 0.05). By the end of the
experiment, average condition had decreased to 2.87.
However, there was no significant difference between the

test cells (P > 0.05).

Density Effegts op Brown Trout

To test for effects of overall salmonine, brown trout,
and steelhead densities on size, I modeled the relationship
of brown trout mean total length to abundance. Overall
salmonine abundance (total number of brown trout and
steelhead in a cell) was inversely related to brown trout
size for control sections (P < 0.01). In test sections, it
had no effect on brown trout size (P = 0.73). Brown trout
abundance was inversely related to brown trout size in both
test and control sections ( P <0.01). ANCOVA of the slope
of these two models was not significant (P > 0.05).
Steelhead density had no effect upon brown trout size (P =

0.22).

68

DISCUSSION

Steelhead Effects on Brpyn Trout Survival

Age-0 steelhead had no apparent effect upon age-0 brown
trout survival in artificial stream experiments. In fact,
survival rates of brown trout in sympatry with steelhead.
were greater than those of allopatric brown trout. These
results support field studies that indicate no effect of
age-0 steelhead upon age-0 brown trout (Chapter 1). Higher
mortality in allopatric cells indicates that intraspecific
competition may be a more important factor in regulating
brown trout abundance in this experiment. Intraspecific
competition of brown trout has been noted in other
artificial stream studies (Hartman 1963; Fausch and White

1986) and in natural systems (Elliot 1984a).

Steelhead Effects on Brown Trout Growth

Brown trout growth rates during the course of this
experiment were comparable to those observed in wild fish
(Chapter 1). Steelhead growth rates were somewhat higher
than those in the companion field study (Chapter 1).
However, they were comparable to fry stocked in a British
Columbia stream (Hume and Parkinson 1988).

No effect of steelhead on brown trout growth was

69
observed. In fact, brown trout growth rates were higher in
sympatry. Analysis of total length illustrated this point
and indicated that in a longer experimental period, test
section brown trout may have grown significantly larger.
This potential "positive effect" of steelhead was probably a
result of the effects of salmonine density on growth.

Analysis of models of brown trout size to overall
salmonine, brown trout, and steelhead abundances clearly
showed the relationship of density to growth. Since fish in
all cells were not significantly different at the beginning
of the experiment, size was an unbiased measure of growth.
The total number of salmonines or steelhead in test cells
had an insignificant effect on brown trout size. However,
the number of brown trout had a negative impact on brown
trout size in both test and control treatments. Thus, at
this life history stage and size distribution, intraspecific
competition between brown trout (density dependence) has a
greater effect than interspecific competition with
steelhead.

I feel that mortality was higher in the control cells
since brown trout carrying capacity was much lower than
initial populations. In test cells, the lower initial
densities resulted in a lower brown trout mortality to reach
carrying capacity. Initial fish densities in this
experiment were also similar to those in comparable

artificial stream experiments (Fausch 1984; Fausch and White

70
1986). However, since my studies were run much longer, the
carrying capacity of the cells decreased as territory size
increased (Chapman 1962; Hearn 1987).

All mortality was assumed to be due to behavioral
interactions since no signs of diseases were observed. The
level of aggressiveness of brown trout can be severe enough
to cause mortality due to physical damage (Kalleberg 1958).
I observed agonistic behavior between and within species in
all cells, although no quantitative assessment was
conducted. Frequently, dominant fish would chase
subordinates to the point where they would jump out of the
stream (This accounted for 60% of total mortality).
Although an artificial response, this could be considered
comparable to emigration from a stream reach in a natural
system. In natural systems, the result of emigration from
inferiority in the dominance hierarchy is typically
downstream displacement and subsequently death (Chapman

1962).

Implications for Natural Systems

The results of this study illustrate that, in their
first growing season, steelhead are likely to have little or
no impact on survival or growth of age-0 brown trout.
Previous research had indicated that the age-0 or age-1
stages were likely arenas of competition (Kruger et al.

1985; Ziegler 1988). The results of this study reject the

71
hypothesis that the age-0 summer growth period is a critical
one for interaction between these two species. These
conclusions have been supported by my research in a natural
stream system (Chapter 1). In that study, I found that the
presence of steelhead had negligible effects on brown trout
size and no effects on abundance. For both size and
abundance, factors related to year-to-year differences in
abiotic conditions and/or intraspecies factors had a more
important effect on brown trout population dynamics. In
this artificial stream study with environmental variables
controlled and equal size fish at the start of the
experiment, results were similar. I conclude that
interactions between these fish are minimal during the
summer growth period.

I feel that one of the factors that allows coexistence
is vertical stratification between the species (Ziegler
1988; Chapter 3). Vertical stratification was consistently
observed in artificial stream experiments. Steelhead were
suspended in the water column and brown trout were closely
associated with the bottom. This may be the mechanism that
allows coexistence between these species. However, prior to
movement to summer habitats, both species utilize shallower
margin habitat (Sheppard and Johnson 1985; Raleigh et al.
1984, 1986; Chapter 3). In these shallow habitats, vertical
stratification could break down and competition could occur.

I feel that this happened in my field studies and that the

72

larger size of brown trout at this time allowed for a
competitive advantage (Fausch and White 1986; Ziegler 1988:
Chapter 1). Field research also showed that the impact of
steelhead in the first winter and yearling period were
minimal under low steelhead densities.

The results of this study indicate that during the age-
0 summer growth period, brown trout and steelhead can
coexist in a riverine environment. While evidence for
interspecies compatibility is substantial at this life-
history stage, other life-history stages must be taken into
account. Of particular interest is the period shortly after
steelhead emergence when both species utilize stream
margins. Assessment of competition at this stage would give
further insights into juvenile competition. Also of
interest is the steelhead spawning stage where interactions
with brown trout alevins my occur (Chapter 1). The effect
of climate variations on the temporal segregation (or lack
of segregation) of alevin and post-emergent stages is also
important. The extent of overlap in spawning as well as the
size and time of each species’ emergence would influence the
outcome of these experiments (Fausch and White 1986; Chapter
1; Chapter 3).

Managers should observe caution in creating sympatric
populations of these two species where a healthy, naturally
reproducing allopatric population of one of these species

exists. The effect of such an introduction on the

73
established population would be uncertain. My results only
assessed the impact of steelhead on brown trout at the
summer growth stage. Interactions to the detriment of brown
trout were minimal, but the impact upon steelhead was not
examined. In addition, until further information is
available on other overlapping life-history stages (ie.
spawner-alevin interactions and climatic effects on temporal
segregation) of these species, the full impact will not be
known. Until this information is quantified, I urge
managers to consider the potential impact these species may
have on each other at all life-history stages before

combining these species in a riverine environment.

CHAPTER 3
HABITAT USE OF JUVENILE BROWN TROUT IN A GREAT LAKES
TRIBUTARY AND THE IMPACT OF JUVENILE STEELHEAD
ABSTRACT
Habitat utilization is described for juvenile brown

trout (Sgipp trutta) in allopatry and in sympatry with
steelhead (Oncorhynchus mykiss). I conducted this study in
Gilchrist Creek, where I had introduced steelhead fry to a
portion of the river. Underwater habitat use observations
were made seasonally from 1989 to 1992. Few winter
observations were made as both species occupied complex
timber and could not be readily observed. In spring, age-0
brown trout occupied margins soon after emergence. By July
they had moved into deeper water. Age-0 brown trout cover
was predominately aquatic vegetation associated with fine
sediments. No seasonal shift was noted in age-1 brown
trout. Compared to age-0 brown trout, they occupied
positions in deeper, faster water and used downed timber and
gravel more frequently. Comparison of age-0 brown trout and
steelhead showed that they use similar water depths,
substrates, and cover during summer and autumn. Steelhead
used slower velocities. Comparisons of age-0 brown trout
habitat use in sympatry and allopatry found no measurable
habitat shift of brown trOut. Vertical stratification
appeared to be an effective isolation mechanism between

these species during the majority of the growing season.

74

75

INTRODUCTION

Research concerning competition between riverine
salmonines has been conducted for over 40 years (Hearn
1987). Currently, this topic has seen renewed interest due
to extensive reviews by Hearn (1987) and Fausch (1988). In
addition, concern over biodiversity has kindled new interest
in the impacts of exotic species on indigenous species and
established fauna (Rinne and Minckley 1985; Ross 1991).

Investigations of competition have often focused on
measuring niche shifts or changes in population dynamics of
one species in the presence (or absence) of another species
(Fausch 1986). Habitat use and segregation have been used
to define niches and the role of competition in niche
development and maintenance. As such, this research
direction has frequently focused on the differing habitat
use of a species when it co-occurs with a potential
competitor (Chapman 1966; Cunjak and Power 1986; Baltz and
Moyle 1984; Gibson and Cunjak 1986; Rose 1986). A trend
seen in many of these studies is that coevolved species have
less overlap in habitat use and are segregated temporally as
well as spatially (Allee 1982; Cunjak and Green 1983; Gibson
and Cunjak 1986; Hearn 1987). In interactions between

exotic and indigenous species, coevolved adaptations do not

76
always act to segregate species in their use of habitat.
This frequently results in the decline in abundance of one
of the competing species. The outcome of competition
between exotic and indigenous forms seems dependent upon the
number and types of species involved and the quality of
stream habitat (compared to historic ecosystems). Ross
(1991) found that exotics were typically more successful in
regions with low numbers of native species and/or degraded
habitats. Given pristine conditions and multiple indigenous
fauna, native species were more likely to flourish.

The scenario described by Ross (1991) appears realistic
for Great Lakes tributaries in the northern portion of
Michigan’s lower peninsula. The alteration of stream
habitat and the introduction of exotic species has
dramatically changed the salmonine fauna in these river
systems over the past century. The current state of most
Great Lakes tributaries are salmonine populations dominated
by exotic fish in altered watersheds (Kruger 1985). This
creates a rather unique situation, since the riverine
salmonine fauna is comprised of predominately introduced
species (Carl 1983; Fausch 1984; Fausch and White 1986).
Hence, these streams contain exotics in species combinations
that have not coevolved.

Brown trout and steelhead are two of these species.
These two species have both been very successful in

colonizing Great Lakes ecosystems and many rivers harbor

77
wild, self-sustaining populations (MacCrimmon and Marshall
1968; Biette et al. 1981) Recent changes in brown trout
growth and abundance have led to concern about negative
impacts of steelhead juveniles upon juvenile brown trout
(Kruger 1985; Kruger et al. 1985; Ziegler 1988; Chapter 1).
Given the past success of natural reproduction of both
species and current concern over adverse impacts of
steelhead upon brown trout, knowledge of the habitat use of
these species is critical to understanding interactions
between these species.

This study was undertaken as part of a project to
determine the effects of steelhead upon brown trout (Chapter
1). A stream containing an allopatric brown trout
population was stocked with steelhead fry to assess shifts
in juvenile brown trout population parameters and habitat
use. This chapter describes my assessment of the
introduction of steelhead upon brown trout habitat use. The
purpose of the investigation was to determine if steelhead
caused any shifts in juvenile (less than age-2) brown trout

habitat use.

78

METHODS

Stugy Site and Design

Gilchrist Creek is a second order stream located in
Montmorency County, Michigan (Figure 8). It flows northeast
through forested land into the Thunder Bay River and
ultimately Lake Huron. The stream is typical of this region
of Michigan being coldwater and characterized by highly
stable flow regimes (P. Seelbach, MDNR pers. comm.). The
salmonine fauna of this stream was exclusively wild brown
trout (Chapter 1).

Investigation of steelhead impact on brown trout
habitat use was conducted in conjunction with assessment of
their impact on brown trout population dynamics (Chapter 1).
The study introduced steelhead to a downstream section of
Gilchrist Creek in late May 1990 and 1991. Populations of
steelhead were moderate (autumn densities averaged 1781/ha)
in density compared to other Great Lakes tributaries
(Chapter 1). Study sections were established representing a
test (downstream) and control (upstream) section separated
by 2 km (Figure 8). These study sections measured 350
meters in length and averaged 7.5 m in width.

Habitat use of brown trout and steelhead was assessed

by observing fish while diving the entire length of the

79

Nonh

 

 

Lower Michigan

 

 

Section 1

Control Area

Gi'ChfiSt Creek

’St
To Thunder Bay River

4
43.33%? Section 2
%>

oeo‘%o
44:4: Test Area

Greasy Creek

I
I

'1 Kilometer

Figure 8 Map of Gilchrist Creek and study sections where

dives were conducted.

80
study section four times a year. Assessments started in May
1989 and ended in April 1992. The design of the study
allowed one year’s worth of observations for brown trout in
allopatry and two years with one allopatric section and one

sympatric section.

Salmopine Habitat Use Observations

A single diver made underwater observations of
salmonine habitat use seasonally. Observations were
conducted in spring, summer, autumn, and winter each year
(Table 14). I made measurements under similar discharge
conditions and only under conditions of high water clarity.
Differences in sampling dates within seasons were the result
of unseasonably high and/or discolored flows or logistical
difficulties. During dives, discharges ranged from 0.85 to

1.28 nfi/sec and averaged 1.09 nfi/sec.

Table 14. Diving dates for habitat use surveys.

 

Season 1989-90 1990-91 1991-92

Spring 05-06 May 27-28 April 18-19 May
Summer 07-08 August 10-11 July 24-25 August
Autumn 03-04 October 09-10 September -----------
Winter ------------ 19-20 January 02-03 February

Dives started no earlier than 09:00 and ended no later
than 15:00. This was done to ensure maximum light

penetration to increase visibility. The time required to

81
dive an entire 350 m study site varied with fish abundance
and water clarity, ranging from 2.0 hours to 3.5 hours.

The diver would start at the downstream section of a
study reach and slowly crawl and swim across and upstream in
a zigzag pattern (Cunjak and Power 1986; Moore and Gregory
1988). When a fish was sighted, the diver would identify
the fish to species and use underwater landmarks to mark the
fishes position and estimate fish length (Cunjak and Power
1986; Ziegler 1988). For data to be recorded, the fish must
have been in an undisturbed state when first sighted.~ The
diver would then move into the area where the fish was
located and place a lead weight with a numbered fluorescent
flag at the focal position of the fish (Fausch and White
1981). The diver would then relay information to an
assistant located approximately 10 m behind the diver.
Information reported by the diver was the weight number,
species, fish length (nearest cm), the substrate the fish
was over, the cover the fish was relating to, the position
relative to cover, and the position in the water column.
Substrate and cover categories are those of Bovee (1986)
(Appendix 3 and Appendix 4). The proximity of fish to cover
was recorded as greater than 0.5 m and less than 1 m from,
less than 0.5 m from, or in cover. Proximity to cover was
related by collapsing the vertical dimension (ie. a fish
could be "in" overhead cover). The location in the water

column was recorded as benthic or suspended.

82

The diver’s assistant would pick up the flagged weight,
placing the base of a top-set wading rod in this position.
The person would then measure the habitat variables of water
depth and mean column velocity. These were measured using
the top-set wading rod and a portable electromagnetic flow
meter. This type of meter allowed for accurate measurements
of velocity even in the midst of heavy vegetative cover.
These data were then added to data recorded by the diver to

complete a data set for an individual fish.

Comparisons of Habitat Use

Since two habitat variables were quantitative measures
and two were categorical, different statistical methods were
needed to make comparisons between seasons, age groups, and
species. Depth and velocity measurements represented the
only quantitative habitat variables. Preliminary analysis
of these variables data indicated most data were not
normally distributed (Shapiro-Wilk Test, P > 0.05). As
such, I used a nonparametric comparison, the Kruskal-Wallis
(K-W) test, for comparisons of multiple groups of
observations (significant at a < 0.05). Substrate and cover
data are qualitative variables and as such were not suited
to this analysis. To test for differences in the
distribution of fish between these habitat variables, the

Kolmogorov-Smirnov (K-S) two-sample test was used.

83

RESULTS

BLEED_IIQBE_BQQLEQE_Q§§

A total of 477 age-0 and 145 age-1 brown trout were
observed. Differences in sample size occurred between
seasons. Only 6% of the observations of age-0 brown trout
and 13% of age-1 samples came from winter samples. These
winter samples were omitted from statistical analysis due to
the low number of observations. However, the observations
during this period were useful in providing insights into
assessment methods for future studies of overwinter habitat
use.

For the remaining seasons, I compared habitat use
within age groups using the K-W test for depth and velocity
data. Since no multiple comparison statistic is suitable
for qualitative data (substrate and cover), I visually
examined their frequency distributions. No significant
differences were noted for age-1 brown trout (P > 0.05). As
such, age-1 habitat use was similar spring through autumn,
and these data were pooled. Differences were significant
for age-0 brown trout. Statistical and graphical analysis
indicated that the spring sample was likely the outlier. I
reanalyzed the summer and fall data with spring removed.

Tests were nonsignificant (P > 0.10) for all four habitat

84
variables, so data were pooled for summer-autumn.

Comparisons of age-0 brown trout habitat use between
spring and summer-autumn were significant for three of four
variables. In the spring, age-0 brown trout were located in
water averaging 34 cm in depth with a mean column velocity
of 22 cm/sec. In the summer, they shifted to deeper (44 cm)
and faster (33 cm/sec) water (P < 0.01; Figure 9 and Figure
10). In the spring, age-0 brown trout were found over
predominately silt and sand substrates (Figure 11). These
two substrates comprised 86% of spring observations. In the
summer-autumn period, habitat use significantly (P < 0.05)
shifted to a greater variety of substrates. During this
period, use of sand and silt declined to 65% and, more use
was made of gravel substrates. Utilization of cover did not
vary significantly (P > 0.10) between spring and summer-
autumn (Figure 12). During both periods, aquatic vegetation
was the most commonly utilized cover. However, as the
season progressed, fish use of downed timer and overhanging
vegetation increased slightly.

Comparison of summer-autumn habitat use of age-0 brown
trout was also significantly different from age-1 brown
trout. These comparisons were significant for all four
habitat variables. Age-0 fish used water averaging 44 cm in
depth and 33 cm/sec velocity. The age-1 brown trout
utilized significantly (P <0.01) deeper and faster waters

(Figure 13 and Figure 14). The mean depth used by age-1

85

 

30

 

_ [I SUMMER-AUTUMN C} SPRING

 

10

(10)

 

PERCENT OCCURRENCE

 

 

(20)

 

 

 

(30) l I l l l l l l L I l l l l i l L i l
0 10 20 3O 40 50 60 70 80 90
5 15 25 35 45 55 65 75 85

WATER DEPTH (cm)

Figure 9. Comparison of water depths utilized by age-0 brown
trout between spring and the summer-autumn time
periods.

86

 

20

 

I. SUMMER-AUTUMN [1 SPRING I

 

 

(10)

PERCENT OCCURRENCE

 

 

(20)

 

 

(30) l l I l l l l l I i l l l l I l
O 10 20 30 40 50 60 7O
5 15 25 35 45 55 65 75

MEAN COLUMN VELOCITY (cm/ sec)

 

Figure 10. Comparison of mean column velocities utilized
by age-0 brown trout between spring and the
summer-autumn time periods.

87

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

60

40
|.l.l
Q 20
Z
I.I.|
SE
3 0
U
U
0
i- (20)
Z
l.l.|
0
m
I.” (40)
D.

(60) —
“ I SUMMER-AUTUMN SPRING
(80) l l l l l l l l L l l l
Detritus Algae Silt Gravel-7 Gr-9 Gr-11
Plants Clay Sand Gr-8 Gr-to Cobble-12
SUBSTRATE CODE
Figure 11. Comparison of substrate types utilized by

age-0 brown trout between spring and the
summer-autumn time periods. Detailed
descriptions of substrates are in Appendix 3.

88

 

80

 

 

- I SUMMER-AUTUMN SPRING

.................

 

 

 

 

 

 

 

 

 

 

 

 

 

40
DJ
‘2’
0:
5
0

O
O
O
P' (20)
Z _
8
CE (40) -
I.I.I _
o.

(60) --

(80) —

(100) l l l l l l l l L

None Un-3 OH Veg-5 Tim-logs Rock/Boulder
Undercut-z OH Veg-4 Aqu. Veg. Tim-complex
COVER CODE

Figure 12. Comparison of cover types utilized by age-0

brown trout between spring and the summer-
autumn time periods. Detailed descriptions
of substrates are in Appendix 4.

89

 

30

 

 

........

I AGE—O 'tf:}:{:}:}t}:}:}: AGE-1

 

 

 

 

 

 

 

 

 

 

 

 

 

(10) -

PERCENT OCCURRENCE
C3

 

 

 

 

 

 

 

 

(20)-
(30)llllllllllllllllllllL
0 10 20 30 40 50 60 70 80 90 100
5 15 25 35 45 55 65 75 85 95
WATER DEPTH (cm)
Figure 13. Comparison of water depths utilized by age-0

and age-1 brown trout.

20
_ ,- AGE-0 AGE-fl
15 —
ll!
0 1O -
Z
DJ
E
D 5—
0
0
O
|- 0
Z
I.l.l
8
u] (5)
O.
(10) — """
(15)llillllllllllllllLil
0 10 20 3O 40 50 60 7O 80 90 100
515 25 35 45 55 65 75 85 95
MEAN COLUMN VELOCITY (cm/ sec)
Figure 14. Comparison of mean column velocities utilized

9O

 

 

 

 

 

 

 

 

by age-0 and age-1 brown trout.

 

91
brown trout was 54 cm and mean velocity was 41 cm/sec. Use
of substrate and cover also shifted significantly (P < 0.01)
(Figure 15 and Figure 16). Age-0 brown trout utilized
predominately silt and sand (65%). The use of these fine
materials declined to only 36% of age-1 observations. The
older brown trout switched to using gravel more frequently,
60% of observations as compared to 31% for age-0 fish. Use
of cover went through a similar shift (Figure 16). Age-0
brown trout were found predominantly in aquatic vegetation
(56%). As the fish entered their second growing season, use
of downed timber became more important. Age-1 brown trout
were found in logs for 20% of observations and in submerged
branches or roots for 25 %. Use of both cover types by age-
0 fish was only 25%. An increase in the use of overhead
cover from 6% at age-0 to 13% at age-1 also occurred.

The horizontal and vertical relationship of brown trout
to coVer and substrate was consistent in all observations.
Brown trout showed a very tight relationship to cover. For
age-0 brown trout, 85% of observations were made directly in
the cover. Another 12% were within 50 cm of cover when
sighted. The remainder of fish (3%) were greater than 50 cm
but less than 1 m from cover. Age-1 brown trout showed a
slightly different relationship to cover. While they were
also most often found in cover (84%), the second most common

location was greater than 50 cm but less than 1 m from

92

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

60
40 -
|.I.l
O _
2
iii
a; 20 —
D
O
o _
O
I-
Z 0 .......
Ill
0
a: _
l.I.l
O.
(20) —
(40) l l l L l l i l l l l l

Detrltue Algae Silt Gravel-7 Gr-S Gr- 1 1
Plants Clay Sand Gr-S Gr- 1 0 Cobble- 1 2

SUBSTRATE CODE

Figure 15. Comparison of substrate types utilized by
age-0 and age-1 brown trout. Detailed
descriptions of substrates are in Appendix 3.

93

80

 

 

 

........

 

I AGE-0 AGE-1

 

 

 

 

 

PERCENT OCCURRENCE

 

..........................

 

 

(20) -

 

.............
.............
.............
.............
..............
..............
..............
..............

 

 

 

 

 

 

 

 

 

(40) l l l l l l l l 1
None Un-3 OH Veg-5 Tim-loge Rock/Boulder
Undercut-z OH Veg-4 Aqu. Veg. Tim-complex

COVER CODE

 

Figure 16. Comparison of cover types utilized by age-0
and age-1 brown trout. Detailed descriptions
of cover types are in Appendix 4.

94
cover (13%). The remaining 3% were found in the middle of
these two positions. Brown trout of age-0 and age-1 were
also most frequently (> 90%) sighted in very close proximity
to the bottom. Divers noted that it was quite common for
the anal, pelvic, and pectoral fins to be in direct contact

with the substrate.

Steelhead Habitat Use and Impect on Broyp_Tpppp

Age-0 steelhead comprised 57 observations and age-1
steelhead only four observations. Given the low number of
observations for age-1 steelhead, only anecdotal conclusions
can be made regarding competition at this age. However,
evaluations of habitat use similarities and differences
between age-0 brown trout and steelhead were possible.
Comparisons were made between summer-autumn habitat use of
brown trout and observations for the same period for
steelhead. Again, winter observations of steelhead were too
low to statistically assess. Spring observations of
steelhead were not made since steelhead were not stocked
until late May (Note: wild steelhead would emerge at
approximately the same time).

Comparisons of habitat use of steelhead and brown trout
showed significant differences in only one of the 4 habitat
variables (P > 0.05). Age-0 steelhead utilized water
averaging 45 cm deep with a mean column velocity of 23

cm/sec. The difference in water depth was not significantly

95
(P > 0.10) different from that utilized by brown trout
(Figure 17). Water speed was significantly (P < 0.01)
slower than that utilized by brown trout (Figure 18). No
significant (P > 0.10) difference in use of cover or
substrate occurred (Figure 19 and Figure 20). Steelhead
used predominately sand and silt (81%) substrates like their
brown trout cohorts. The predominant cover type for both
species was aquatic vegetation, with steelhead found in this
cover in 65% of observations. Use of other cover types was
in approximately the same proportions.

The horizontal and vertical juxtaposition to cover and
substrate varied somewhat from that of brown trout.
Steelhead were closely related to cover, but not to the same
extent as were brown trout. Of all steelhead observed, 74%
were in cover and 22% were within 50 cm of cover. The
remaining 4% were greater than 50 cm but less than 1 m from
cover. Their vertical relationship to cover was very
different. Over 90% of all steelhead observed were
suspended in the water column. Contact of fins with the
bottom was very rare for steelhead observed in this study.

The great amount of overlap in habitat use between
these species implies the potential for competition and
associated shifts in habitat use. To examine any changes, I
compared habitat use of allopatric brown trout to that of
brown trout in sympatry with steelhead. I found no

significant difference in brown trout habitat use between

96

 

 

 

 

STEELHEAD

 

 

 

 

I BROWN TROUT

 

 

             

‘ ‘ ‘ ‘ ‘
”Peron." "on.”

.6...
FF.

4
4

     
     

wan“?!
mfiWWm
mmmmmmm
4.8888448“

V’PPPPPP,

P.

   

    

      
      

 

  
 

 

   

9'.
p . q
4 0.2

  

     

                                

'4

'9

O

      
   

 
 

4V

.hfifififififig
‘

v

‘

   
  

     

so.

 

v

  
 

D

94
o
%>

D

V

     

Pvt
O O
$5

    

  

  

D'O‘
8%

‘M‘N‘M‘M‘M‘M‘M‘M‘M
oIWWPEWWKKWWWWWWWN

P.

POPODOP’OPODOPODODOD DODOPODOPOPOD.
lduddaannwnnnn

 

OvOiOr .

““““““““.
.3V.V¢¢§WWW3KWW”:

PEPPPPP

P

l l

l

0

 

 

30

20 -

_
nu fly
1

mozmmmaooo FZmOm—mn.

(10) -

(20)

10 20 30 40 50 60 70 80 90

15 25 35 45 55 65 75 85

5

WATER DEPTH (cm)

Comparison of water depth utilized by age-0

brown trout and steelhead.

Figure 17.

 

 

 

 

 

 

97

 

 
           

  

    

 

        

    

       

     

     

  
  

  

      

 

 

             

 

  
 
 

 
 

   
 

   

 

nu ..
A V0090.
F—
H”
L
F.
F.
T. ....
.3
T.
w
uiiflfifiififlfifiififififiifififififlfi.
R" wwflnmmmhmmfimmmmmmmhnmhz
T . «O..¢...i.c.a.«.c...¢ .. .. i.4.«.q.«....
N .0»0»0n0»0n0n0n0n0n0»0>0v0>0n0n0n0n0n0n0...
w 30.0%. ............
.0000
' D R R R
o . ‘ . ‘ .
m ”PMS... 5.0.,
a€€€€€€€€€€€€€€€€€€€€€e
-— fifififififififififififlflfififlfifififlfifix
wyfififififififiwfififiyfififi

 

 

         

 
 
 
 
  

   

.0»0» »0» »0» 0 »0»0»0».
.4040004040064940.
”wwwWanu
_ — — p — _

     

P

 
   
      

         

 

 
 

 

20

15-

_ _
m s o a m
( 1

(

(15) *-

mOmemDOOO szomwn—

20 30 40 50 60 70 80
15 25 35 45 55 65 75

MEAN COLUMN VELOCITY (cm/sec)

10
Comparison of mean column velocities utilized

by age-0 brown trout and steelhead.

l

l

 

(20)

Figure 18.

98

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

60
40 ~
Lu
0 20 —
Z
I.l.l _
E
3 O
O
Q _
O
l— (20) —
Z
I“ _
O
a:
UJ (4o) —
O.
(60) -
* - BROWN TROUT STEELHEAD
(80) l l l l l I l l l l l l
Detritus Algae Silt Gravel-7 Gr-S Gr- 1 1
Plants Clay Sand Ctr-8 Gr-10 Cobble-12
SUBSTRATE CODE
Figure 19. Comparison of substrates utilized by age-0

brown trout and steelhead. Complete
descriptions of substrates are given in
Appendix 3.

80

99

 

60

«40

20

(20)

PERCENT OCCURRENCE

(40)

(60)

 

(80)

 

 

 

I BROWN TROUT

 

STEELHEAD

 

 

 

 

 

 

SKHE .

l I l I l

  
     

 

   
  

r'e'e'e'I
:49:
'A’A.A.A.

    
   

       
    
   
  
   
      
       
 

       
 

  
        
      
   
   

     
      
      

   
     
     
    
 

"C"'.'
0000
:0qa

P'O'Q'Q'
. 44:4
9.9 O ‘ ‘ ‘

.'.'.'.
fvvt
0 4
0000
000«
Rfififl
0000

REM?

O

             
   
    
     
  
    
    
 
   
  

     
  
     
  
   

    
 

00
0

%>
0

3»
$9’

g,0

  
         
  

v

  
  
 

.v
.9

      

5L
0

0
.‘ .

   

      
 

v

%>

O
o

    
  
   
   
     
   
   
  
   
  

.V
:3:
.0

   
 
  

.'
9.:
O

O

V

0
3»
,p

 
  
   
   
 
 
 
 
 
   
  
  

3:;
O

O

.9

IS?

i”’
00

0

0
.:

O

    
  

V

     
    
  
    
   
  
    
   
  

v

0
O

0

O
O
O
A

V

O
O
.0

4?

0
0
,3

v

0
O

O

     
  

O

5%
‘%
0

.‘Q

   
   
    
  

    
  

V

%»

‘%

db

5

      
     
   
    
 
    

0
0

$9’

<fi?
0

49

O
- O
'5

O

     
   
    

‘ v

    

l l l l

 

 

Figure 20.

None Un-3
Undercut-2 OH Veg-4

OH Veg-5

Tim-logs Rock/Boulder
Aqu. Veg. Tim-complex

COVER CODE

Comparison of cover utilized by age-0 brown

trout and steelhead.

Complete descriptions

of cover types are given in Appendix 4.

100
sympatry and allopatry for depth or cover variables (P >
0.10). However, significant differences were observed in
mean column velocity and substrate. In sympatry, brown
trout were found in slower water and utilized gravel to a
greater extent. However, the magnitude Of both of these
shifts was in the Opposite direction of the variables most
frequently used by steelhead. Given the Opposite direction
of habitat shifts (ie. to habitats utilized by steelhead), I
suspect that differences were driven by competition between

brown trout cohorts or some other factor.

101

DISCUSSION

Brown Trout Habitat Use

Habitat use of age-0 and age-1 brown trout generally
follow the habitat use patterns of brown trout populations
in other regions (Cunjak and Power 1986; Raleigh et al.
1986; Ziegler 1988; Chaveroche and Sabaton 1989). Limited
habitat use data are available for brown trout in the Great
Lakes region (Gowan 1984; Ziegler 1988). Habitat
utilization assessed in this study compares favorably to
their results. My examination of habitat use on a seasonal
basis provides some additional insights. Little is known of
the habitat use of brown trout soon after emergence, ie.
fish less than 50 mm (Raleigh et al. 1986). My spring
assessments provided information on the habitat use of these
newly emerged fish. These fish typically stayed in the
stream margins in slow, shallow water. Aquatic vegetation
was heavily used as cover and, as such, substrates were
typically fines. Age-0 brown trout used aquatic vegetation
quite frequently when it was available (Mortensen 1977a,
1977b; Mann et al. 1989). As the season progressed through
summer and autumn, these age-0 fish moved to deeper faster
water and utilized downed timber cover and gravel substrate

more frequently. While not well documented for brown trout,

102
this habitat shift has been Observed in other salmonine
species (Chapman and Bjornn 1969; Symons and Heland 1978;
Moore and Gregory 1988; Dauble et al. 1989).

Habitat use of age-1 brown trout did not vary between
spring and fall. Habitat use Of this age group was again
similar to that reported by other authors (Gowan 1984;
Raleigh et al. 1986; Hanson et al. 1987). These fish used
deeper and faster water than age-0 fish. It is thought that
the presence of these older fish may limit the use Of deeper
water by age-0 brown trout (Kennedy and Strange 1986).

While my study has no direct evidence of this intraspecific
competition between age groups, habitat use was partitioned
by age groups.

Wintering habitat of brown trout was not effectively
assessed through diving in this study; Observations of both
age-0 and age-1 brown trout were low. This is in contrast
to the number of fish residing in the stream sections as
indicated by winter electrofishing (Chapter 1). In the
winter, salmonines are often found buried in substrates or
associated with heavy cover (Rimmer et al. 1983; Cunjak and
Power 1986; Cunjak 1988; Heggenes and Saltveit 1990). Both
large substrates and heavy cover were searched in the manner
described by these authors without success. I suspect that
the observational differences were due to habitat
availability and complexity. I feel that fish in this study

were associated with complex timber and hidden in crevices.

103
Therefore, they were not visible to divers without
frightening the fish. It was fairly common for divers to
scare multiple fish from complex cover without making an
adequate observation. This hypothesis is supported by
observations during electrofishing. As such, I feel that
fish were hidden in this cover in aggregates and that winter
habitat use is similar to that noted by Cunjak and Power

(1986).

Steelhead Habitat Use and Impact on Brown Trout

Steelhead habitat use was assessed only for age-0
steelhead due to low numbers of age-1 Observations. They
were first Observed in summer since they were not present
during spring dives. Habitat use by age-0 steelhead during
the growing season is similar to that noted by other authors
(Baltz and Moyle 1981; Raleigh et al. 1984; Sheppard and
Johnson 1985; Ziegler 1988). Steelhead exhibited no
significant shift in habitat use between summer and fall
samples. By the time that fish were sampled in July and
August, it is likely that they had already shifted habitats
from stream margins to deeper waters. Another study in the
Great Lakes observed steelhead fry to be using margin
habitat as late as June (Sheppard and Johnson 1985). Since
this transition is dependent upon size, and fish in
Gilchrist Creek were large, the speculation of a habitat

shift is probably safe. Steelhead habitat shifts during

104
their first growing season are well documented (Raleigh et
al. 1984; Sheppard and Johnson 1985). Since behavior of
age-0 fish is similar to reported values, their habitat at
age-l is also likely to be similar.

Comparisons of the summer and fall habitat use of
steelhead and brown trout showed that each species used the
same range of habitat variables. Only water velocity was
significantly different, with steelhead utilizing slower
waters. This difference corresponds to published habitat
suitability curves for juveniles of both species (Raleigh et
al. 1984; 1986). Ziegler (1988) found similar habitat use
patterns in her examination of allopatric and sympatric
populations of both species in three rivers. Similarities
in habitat use between these two species has also been noted
in other regions (Jenkins 1969; Raleigh et al. 1984, 1986;
Cada et al. 1987).

Despite the large overlap in habitat use in all four
variables, no shifts in brown trout habitat use could be
attributed to steelhead presence. There are three probable
explanations: 1) brown trout typically predominate in
competition, 2) competition between the two species is
negated by vertical stratification, or 3) the abundance Of
steelhead was not large enough to impact brown trout. Since
brown trout maintain a size advantage through age-2 (Chapter
1), it is possible that they have an advantage in habitat

use. Trout dominance hierarchies and use of preferred

105

feeding stations (optimal habitat) typically go to larger
fish (Chapman 1966; Fausch 1984; Fausch and White 1986).
This scenario seems likely given the larger size of brown
trout juveniles. If this is true, then interspecific
competition between the two species would favor brown trout
at the juvenile life stage.

Vertical stratification in the water column could play
a major role in coexistence Of brown trout and steelhead in
similar habitats. I observed steelhead to be suspended in
the water column almost exclusively while brown trout
remained close to the bottom. Ziegler (1988) also noted
this vertical stratification. This type of stratification
has been implicated as a segregation mechanism allowing the
coexistence of salmonines (Allee 1982; Cunjak and Green
1983). This segregation has been termed habitat segregation
(Hearn 1987). In habitat segregation, two species have
innate differences in habitat use regardless of the other
species’ presence. This is supported by the fact that
Ziegler (1988) observed differences in height in the water
column in allopatric populations of both species as well as
in sympatry. This segregation could be a major isolating
factor between these species, allowing coexistence Of these
salmonines at the juvenile stage.

The abundance Of steelhead in this study was relatively
low by Great Lakes standards (Chapter 1). The ratio of

steelhead to brown trout averaged 0.7:1 in June and 0.4:1 by

106
October. Since steelhead frequently outnumber brown trout
in Great Lakes tributaries (Ziegler 1987), the low ratio Of
steelhead to brown trout may have affected the outcome of
habitat use. The higher the ratio of steelhead, the more
probable that individual steelhead will be larger than
individual brown trout. Given higher steelhead densities,
the steelhead may have been able to impact brown trout
habitat use.

This study has shown that steelhead under relatively
low densities have no impact on juvenile brown trout habitat
use. Analysis of three plausible explanations for the lack
of steelhead impact helps explain the interactions between
these species. Given the results of this study, I feel that
brown trout and steelhead have innate habitat isolation
(vertical stratification) that contributes to their
utilization of similar habitats. In addition, the size
advantage and numerical advantage that brown trout have
would benefit them in any interactive competition that could
occur. A shift in species abundance or relative size at
emergence could alter this relationship. This shift could
be caused by differences in spawner abundance and/or the
time of peak spawning (Chapter 1). Under such conditions,
the balance might be tipped toward greater steelhead success

to the detriment of brown trout.

APPENDICES

Appendix 1.

107

P values for ANOVA models and linear
contrasts for population estimates and
instantaneous mortality rates for juvenile
brown trout. Models significant at P < 0.05
and contrasts significant at P < 0.01. The *
denotes comparisons of allopatric brown trout
and brown trout in sympatry with steelhead.

 

 

 

 

 

 

 

 

 

 

Age-0 Between Sites Pre vs Post
Sampling Model 1989 1990* 1991* Site 1 Site 2*
Period
Population Estimates
June 0.401 0.141 0.981 0.979 0.741 0.052
October 0.434 0.096 0.563 0.750 0.072 0.795
Feb. 0.005 0.005 0.239 0.152 0.496 0.001
Instantaneous Mortality
Summer 0.644 0.227 0.768 0.797 0.684 0.193
Winter 0.019 0.002 0.394 0.709 0.166 0.002
Spring 0.001 0.001 0.001 0.738 0.443
Age-1 Between Sites Pre vs Post
Sampling Model 1988 1989 1990* Site 1 Site 2*
Period .
Population Estimates
June 0.037 0.685 0.051 0.575 0.006 0.804
October 0.015 0.053 0.244 0.455 0.003 0.206
February 0.021 0.243 0.734 0.074 0.003 0.136
Instantaneous Mortality
Summer 0.276 0.150 0.693 0.939 0.084 0.277
Winter 0.021 0.027 0.441 0.812 0.220 0.005
Spring 0.475 0.152 0.653 0.283 0.395

108

Appendix 2. P values for ANOVA models and linear
contrasts for total length and instantaneous
growth rates of age-0 and age-1 brown trout.
Models significant at P < 0.05 and contrasts
significant at P < 0.01. The * denotes
comparisons of allopatric brown trout and
brown trout in sympatry with steelhead.

TOTAL LENGTH COMPARISONS

Between Sites Pre vs Post

 

 

 

Age Class , ,
Model 1989 1990* 1991* Site 1 Site 2*

Age-0

June 0.001 0.103 0.136 0.750 0.001 0.001

October 0.001 0.087 0.001 0.507 0.006 0.144

February 0.001 0.001 0.002 0.110 0.092 0.005

 

Age-1 1988 1989 1990*‘ Site 1 Site 2*
June 0.001 0.030 0.001 0.475 0.001 0.002
October’ 0.001 0.007 0.117 0.201 0.001 0.496

February 0.164 0.145 0.383 0.324 0.042 0.536

INSTANTANEOUS GROWTH RATE COMPARISONS

 

 

 

Sampling

Period Model Site 1 vs. Site 2 Pre vs Post
Age-0 ’ 1989 1990* 1991* Site 1 Site 2*
Summer 0.001 0.888 0.001 0.220 0.183 0.001
Winter 0.096 0.070 0.033 0.294 0.466 0.400
Spring 0.090 0.457 0.027 0.114 0.027
Age-1 1988 1989 1990*’ Site 1 Site 2*
Summer 0.034 0.340 0.081 0.816 0.533 0.027
Winter 0.929 0.956 0.978 0.398 0.750 0.496
Spring 0.682 0.319 0.593 0.582 0.326

Appendix 3.

109

1986).

Generalized substrate classes (From Bovee

 

 

 

# ClaeepName Size Range (mm)
01 Organic Detritus
02 Vascular Plants
03 Attached Algae
04 Clay 0.00024 0.004
05 Silt 0.004 0.062
06 Sand 0.062 2.0
07 Very Fine Gravel 2 4

08 Fine Gravel 4 8

09 Medium Gravel 8 16
10 Coarse Gravel 16 32
11 Very Coarse Gravel 32 64
12 Small Cobbles 64 128
13 Large Cobbles 128 256
14 Small Boulders 256 512
15 Medium Boulders 512 1024
16 Large Boulders 1024 2048
17 Very Large Boulders 2048
18 Bedrock Plain, unfractured
19 Bedrock Plain, jointed
20 Bedrock Tilted, perpendicular
21 Bedrock Tilted, parallel

 

110

 

 

 

Appendix 4. Descriptions of cover types. Adapted from
Bovee (1986).
# Cover Code Class Name
1 No Cover
2 undercut bank < 30 cm
3 undercut bank > 30 cm
4 overhanging vegetation > 30 cm above surface
5 overhanging vegetation < 30 cm above surface
6 emergent or submergent aquatic vegetation
7 down timber- logs
8 down timber- branches or roots
9 rock/ boulder

a/b

to combine 2 habitat codes

LITERATURE CITED

111

LITERATURE CITED

Alexander, G. R. 1985. Comparative growth and survival
potential of brown trout from four wild stocks and one
domestic stock. Fisheries Research Report 1929. MDNR,
Lansing, MI.

Alexander, G. R. 1977. Consumption of small trout by large
predatory brown trout in the North Branch of the Au
Sable River, Michigan. Fisheries Research Report 1855.
MDNR, Lansing, MI.

Alexander, G. R. 1979. Predators of fish in coldwater
streams. Pages 153-170 in H.Clepper, editor.
Predator-prey Systems in Fisheries Management.
Washington, D.C.: Sport Fishing Institute.

Allee, B. A. 1982. The role of interspecific competition
in the distribution of salmonids in streams. Pages.
111-122 ip E.L. Brannon and E. O. Salo, eds.
Proceedings of the salmon and trout migratory behavior
symposium. University of Washington Press. Seattle,
Wa.

Auer, N.A. 1982. Identification of larval fishes of the
Great Lakes Basin with emphasis on the Lake Michigan
drainage. Special Publication 82-3. Great Lakes
Fishery Commission. Ann Arbor, Michigan.

Bachman, R. A. 1984. Foraging behavior of free-ranging
wild and hatchery brown trout in a stream.
Transactions Of the American Fisheries Society 113:
1-32 0

Baltz, D. M. and P. B. Moyle. 1984. Segregation by species
and size classes of rainbow trout, Salmo airdneri, and
Sacramento sucker, Catostomus occidentalis, in three
California streams. Environmental Biology of Fishes
10: 101-110.

Biette, R. M., D. P. Dodge, R. L. Hassinger and T. M.
Stauffer. 1981. Life history and timing of migrations
and spawning behavior of rainbow trout (Salmo
qairdneri) populations Of the Great Lakes. Canadian
Journal of Fisheries and Aquatic Science 38:
1759-1771.

112

Bovee, K.D. 1986. Development and evaluation of habitat
suitability criteria for use in the Instream Flow
Incremental Methodology. Instream Flow Information
paper 21. U.S.D.I. Fish and Wildlife Service, Office of
Biological Services. FWS/OBS-86-7. 233 pp.

Cada, G. F., J. M. Loar and M. J. Sale. 1987. Evidence of
food limitation of rainbow and brown trout in southern
Appalachian soft-water streams. Transactions of the
American Fisheries Society 116: 692-702.

Carl, L. M. 1983. Density, growth, and change in density
of coho salmon and rainbow trout in three Lake Michigan
tributaries. Canadian Journal of Zoology 61:
1120-1127.

Chapman, D. W. 1962. Aggressive behavior in juvenile coho
salmon as a cause of emigration. Journal of the
Fisheries Research Board of Canada 19: 1047-1080.

Chapman, D. W. 1966. Food and space as regulators of
salmonid populations in streams. American Naturalist
100: 345-357.

Chapman, D. W. and T.C. Bjornn. 1966. Distribution of
salmonids in streams with special reference to food and
feeding. pp. 153-176 ip T.G. Northcote ed. Symposium
on salmon and trout in streams. University of British
Columbia. Vancouver.

Chaveroche, P. and C. Sabaton. 1989. An analysis of brown
trout (Salmo trutta fario L.) habitat: the role Of
qualitative data from expert advice in formulating
probability-of-use curves. Regulated Rivers: Research
and Management 3: 305-319.

 

Cone, R. S. 1989. The need to reconsider the use of
condition indices in fishery science. Transactions of
the American Fisheries Society 118: 510-514.

Close, T. L. and C. S. Anderson. 1992. Dispersal, density
dependent growth, and survival of stocked steelhead fry
in Lake Superior Tributaries. North American Journal
of Fisheries Management. in press.

Conover, W.J. and R. L. Iman. 1981. Rank transformations
as a bridge between parametric and nonparametric
statistics. American Statistician 35: 124-129.

113

Cunjak, R. A. 1988. Behaviour and microhabitat of young
Atlantic salmon (Salmo salar) during winter. Canadian
Journal of Fisheries and Aquatic Science 45:
2156-2160.

 

Cunjak, R. A. and J. M. Green. 1983. Habitat utilization
by brook char (Salvelinus fontinalis) and rainbow trout
(Salmo gairdneri) in Newfoundland streams. Canadian
Journal of Zoology 61: 1214-1219.

Cunjak, R. A. and G. Power. 1986. Winter habitat
utilization by stream resident brook trout (Salvelinus
fontinalis) and brown trout (Salmo trutta). Canadian
Journal Of Fisheries and Aquatic Science 43:
1970-1981.

Dauble, D. D., T. L. Page and R. W. and Hanf. 1989.
Spatial distribution of juvenile salmonids in the
Hanford Reach, Columbia River. Fishery Bulletin 87:
775-790.

Elliott, J. M. 1973. The food of brown and rainbow trout
(Salmo trutta and S. qairdneri) in relation to the
abundance Of drifting invertebrates in a mountain
stream. Oecologia 12: 329-347.

Elliott, J. M. 1984a. Numerical changes and population
regulation in young migratory trout Salmo trutta in a
lake district stream, 1966-83. Journal of Animal
Ecology 53: 327- 350.

Elliott, J. M. 1984b. Growth, size, biomass and production
of young migratory trout Salmo trutta in a lake
district-stream, 1966-83. Journal of Animal Ecology
53: 979-994.

Elliott, J. M. 1975. Number of meals in a day, maximum
weight of food consumed in a day and maximum rate of
feeding for brown trout Salmo trutta L. Freshwater
Biology 5: 287-303.

Fausch, K. D. 1984. Profitable stream positions for
salmonids: relating specific growth rate to net energy
gain. Canadian Journal of Zoology 62: 441-451.

Fausch, K. D. 1988. Test of competition between native an
introduced salmonids in streams: what have we learned.
Canadian Journal of Fisheries and Aquatic Sciences 45:
2238-2246.

114

Fausch, K. D. and R. J. White. 1981. Competition between
brook trout (Salvelinus fontinalis) and brown trout
(Salmo trutta) for positions in a Michigan stream.
Canadian Journal of Fisheries and Aquatic Science 38:
1220-1227.

Fausch, K. D. and R. J. White.. 1986. Competition among
juveniles of coho salmon, brook trout, and brown trout
in a laboratory stream, and implications for Great
Lakes tributaries. Transactions of the American
Fisheries Society 115: 363-381.

Gibson, R. J. and R. A. Cunjak. 1986. An investigation of
competitive interactions between brown trout (Salmo
trutta L.) and juvenile Atlantic salmon (Salmo salar
L.) in the rivers of the Avalon Peninsula,
Newfoundland. Canadian Technical Report of Fisheries
and Aquatic Sciences No. 1472 1-82.

 

Gowan, C. 1984. The impacts of irrigation water withdrawals
on brown trout (Salmo trutta) and two species of
benthic macroinvertebrates in a typical southern lower
Michigan trout stream. Michigan State University, East
Lansing, Michigan.

Hanson, D.F. Morhardt, J.E., and T.R. Lambert. 1987.
Development of habitat suitability criteria for trout
in the Kings River Basin, California. Prepared for the
Pacific Gas and Electric Company Research Division and
Demonstration Program. Lafayette, California. 70 pp.

Hearn, W. E. Fisheries. Interspecific competition and
habitat segregation among stream-dwelling trout and
salmon: a review. 1987 12: 24-31.

Heggenes J. and S. J. Saltveit. 1990. Seasonal and spatial
microhabitat selection and segregation in young
Atlantic salmon, Salmo salar L., and brown trout, Salmo
trutta L., in a Norwegian river. Journal of Fish
Biology 36: 707-720.

Hume, J. M. B. and E. A. Parkinson. 1987. Effect of
stocking density on the survival, growth, and dispersal
Of steelhead trout fry (Salmo gairdnepi). Canadian
Journal Of Fisheries and Aquatic Science 44: 271-281.

Hume, J. M. B. and E. A. Parkinson. 1988. Effects of size
at and time Of release on the survival and growth of
steelhead fry stocked in streams. North American
Journal of Fisheries Management 8: 50-57.

115

Jenkins, T. M. ,. J. 1969. Social structure, position
choice and micro-distribution of two trout species
(Salmo trutta and Salmo gairdneri) resident in mountain
streams. Animal Behaviour Monographs 2: 56-12.

Jones, M. L. 1991. Biotic constraints to Atlantic salmon
restoration in Lake Ontario: a report to the Atlantic
salmon steering committee.

Kalleberg, H. 1958. Observations in a stream tank of
territoriality and competition in juvenile salmon and
trout (Salmo salar L. and g. trutta L.). Rept. Inst.
Freshwater Research Drottningholm 39: 55-98.

Kennedy, G. J. A. and C. D. Strange. 1986. The effects Of
intra- and inter-specific competition on the survival
and growth of stocked juvenile Atlantic salmon, Salmo
salar L., and resident trout, Salmo trutta L., in an
upland stream. Journal of Fish Biology 28: 479-489.

Kocik, J. F. and W. W. Taylor. 1987. Diet and movements of
age- 1+ pink salmon in western Lake Huron.
Transactions of the American Fisheries Society 116:
628-633.

Kocik, J. F. and W. W. Taylor. 1991. Abundance, size, and
recruitment Of pink salmon (Oncorhynchus gorbuscha) in
selected Michigan tributaries of the upper Great Lakes,
1984-1988. Journal of Great Lakes Research.
17(2):203-213.

Kruger, K. M. 1985. Pere Marquette River angler survey and
brown trout evaluation. Michigan State University, East
Lansing,<Michigan.

Kruger, K. M., W. W. Taylor and J. R. Ryckman. 1985.
Angler use and harvest in the Pere Marquette River near
Baldwin, Michigan. Michigan Academician 317-330.

MacCrimmon, H. R. and T. L. Marshall. 1968. World
distribution Of brown trout, Salmo trutta. Journal of
the Fisheries Research Board Of Canada 25: 2527-2548.

MacDonald, P.D.M. 1987. Analysis of length-frequency

distributions. pg. 371-384 ip. Age and Growth of Fish.
Ames, Iowa.

Mann, R. H. K., J. H. Blackburn and W. R. C. Beaumont.
1989. The ecology of brown trout Salmo trutta in
English chalk streams. Freshwater Biology 21: 57-70.

116

Miller, R.G. 1986. Beyond ANOVA, basics of applied
statistics, John Wiley and Sons, New York

Moore, K. M. S. and S. V. Gregory. 1988. Summer habitat
utilization and ecology of cutthroat trout fry (Salmo
clarki) in Cascade Mountain streams. Canadian Journal
of Fisheries and Aquatic Science 45: 1921-1930.

 

Mortensen, E. 1977a. Density-dependent mortality of trout
fry (Salmo trutta L.) and its relationship to the
management of small streams. Journal of Fish Biology
11: 613-617.

Mortensen, E. 1977b. Population, survival, growth and
production of trout Salmo trutta in a small Danish
stream. Oikos 28: 9-15.

Newman, R. M. and T. F. Waters. 1989. Differences in brown
trout (Salmo trutta) production among contiguous
sections of an entire stream. Canadian Journal of
Fisheries and Aquatic Science 46: 203-213.

Ott, L. 1988. An introduction to statistical methods and
data analysis. Boston, Massachusetts. 835 pp.

Parsons, J.W. 1973. History of salmon in the Great Lakes,
1850-1970. Technical Paper 68 Bureau of Sport
Fisheries and Wildlife. 81 pp.

Peterson, N. P. and C. J. Cederholm. 1984. A comparison of
the removal and mark-recapture methods of population
estimation for juvenile coho salmon in a small stream.
North American Journal of Fisheries Management 4:
99-102.

Raleigh, R. F., T. Hickman, R. C. Solomon and P. C. Nelson.
1984. Habitat suitability information: rainbow trout.
Biological Report 82(10.60). Fish and Wildlife

Service, U.S. Department Of the Interior, Washington,
D.C.

Raleigh, R. F., L. D. Zuckerman, and P. C. Nelson. 1986.
Habitat suitability index models and instream flow
suitability curves: brown trout. Fish and Wildlife
Service, Biological Report 82(10.124). U.S. Department
Of the Interior, Washington, D.C.

Rimmer, D.M., U. Paim, and R.L. Saunders. Autumnal habitat
shift of juvenile Atlantic salmon (Salmo salar) in a

small river. Canadian Journal of Fisheries and Aquatic
Science 40: 671-680.

117

Rinne, J. N. and W. L. Minckley. 1985. Patterns of
variation and distribution in Apache trout (Salmo
apache) relative to co- occurrence with introduced
salmonids. Copeia 285-292.

Ricker, W.E. 1975. Computation and interpretation of
biological statistics of fish populations. Bulletin of
the Fisheries Research Board of Canada Number 191.
Ottawa

Rose, C. A. 1986. Growth decline in subyearling brook
trout (Salvelinus fontinalis) after emergence of
rainbow trout (Salmo qairdneri). Canadian Journal of
Fisheries and Aquatic Science 43: 187-193.

Ross, 8. T. 1991. Mechanisms structuring stream fish
assemblages: are there lessons from introduced
species? Environmental Biology Of Fishes] 30:
359-368.

Snedecor, G.W. and W.G. Cochran. 1982. Statistical
Methods. The Iowa State University Press. Ames, IA

Scholl, D. K., P. J. Peeters and S. T. Schram. 1984.
Migratory brown trout and rainbow trout populations of
the Brule River, Wisconsin. Fish Management Report NO.
123. Madison, Wisconsin.

Seelbach, P.W. 1986. Population biology of steelhead in
the Little Manistee River, Michigan. Doctoral
Dissertation, The University of Michigan, Ann Arbor.

Seelbach, P. W. 1987. Effect Of winter severity on
steelhead smolt yield in Michigan: an example of the
importance of environmental factors in determining
smolt yield. American Fisheries Symposium 1: 441-450.

Sheppard, J.D. and J. H. Johnson. 1985. Probability-of-use
for depth, velocity, and substrate by sub-yearling coho
salmon and steelhead in Lake Ontario tributary streams.
North American Journal of Fisheries Management 5: 277-
282.

Symons, P. E. K. and M. Heland. 1978. Stream habitats and
behavioral interactions of underyearling and yearling
Atlantic salmon (Salmo salar). Journal of the
Fisheries Research Board of Canada 35: 175-183.

Wentworth, R. S. and G. W. LaBar. 1984. First-year
survival and growth of steelhead stocked as fry in
Lewis Creek, Vermont. North American Journal of
Fisheries Management 4: 103-110.

118

Ziegler, R. L. 1987. Interaction of juvenile brown trout
(Salmo trutta) and steelhead trout (Salmo gairdneri) in
Lake Michigan tributaries. Dingell-Johnson Progress
Report.

 

Ziegler, R. L. 1988. Stream resource utilization of
sympatric and allopatric juvenile brown (Salmo trutta)
and steelhead trout (Salmo gairdneri). Michigan State
University, East Lansing, Michigan.

"11111111111411“