. X“
. H.
N... 32;»
1' «31.33351: 5
‘1'!

.. - . :35.V
. ..
‘-.,‘a!‘.".‘-‘ «J» . .
231:1 :““=.Vk'.“

~ .

u;

*2

«—
~.. _n
«.wi ._
$1.093»
...
\s .

n.- 9.“...

nag)“. q.

..
~. »
""‘.,l..‘=
‘- - x-.-, ,
\~“"-"ll‘:-L ‘3

:4?

, \ 5
'23. 3:.

9‘13: .-

;- 4".“ :‘I‘o‘n‘ I).

I'm

1 _ .L‘k
L3”, 1v
’ ‘ ' - . - ' “ . . _ . magic}; ~33.
. 1. ‘ ‘ ‘ . ' , x V ‘ , ~14 , -‘..~i::"
4 ‘ “ . . ' , V . ' v .«u , , w ‘ .16"
_.‘ x2 .. . . ,

 

 

.- “'

 

~
1
l
1'
u
1
~

.

u.

-u .

 

 

.4 ‘4‘; .

~.-,.‘

 

. | 4
2, ,~ f5}

'2' xi 5

3. v » .
«‘3. '. " fl“- ” A, , n 3. .5” f; _ c-zz 4 7.

.
.
.u. f z.- . ,
. .‘1 “to" n ,, . N ’ 3". ‘ 3mg" , ’
. ~I}C"€'~I-":‘:‘ ‘, . _ 5" I; '“ ‘ r..- I,f¢. , J“ ' IEEQC'?!‘ _a'fi‘fi If -53.?
. In ., g ». an‘A‘ _ u. ‘ .I w , ‘ .y ‘ ‘ V » no: IV I“ a . "f a a". u .%‘é?. .. ‘ I“ - £3“):
. L . u ' , - .- b, 1 . ~ ‘ V a” ' _, . ‘ ‘ .3“ "" '2"
I ' ' 1' fl . ' “ * . x- . . ‘ '53: ”23‘“ ‘~§.*.L ’ m 29:“.34rfigr1 1'13"") 5.13.; ’3‘
' ‘ '4: . ‘ .’ q a ‘ ‘I ._~ ‘ " . _ K‘Q‘i " *1: ‘V 33' flit
3;“th . *’ r . ._ . 3 yr. ,." . 7:1 ,‘ '14
\u‘fii‘d {- .. . Ema-K. 3,; ‘ ’I‘" t. .
.~,I :3}: g. I , {£3.23 V :
5:2: . v

 

 

 

 

‘ gang? . h 54:35
. Jr" “avgfigvnmm
, ,
1‘

$::¥'x§<{§ gfi ‘ it}

     

 

SITY UBRARIE

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 0I

3 1293 01020

 

IIIII

 

 

 

 

 

This is to certify that the

dissertation entitled

Dynamics of the Recovery of Lake Trout
(Salvelinus nanaycush) in U.S. Haters of Lake Superior

presented by

Michael Jay Hansen

has been accepted towards fulfillment
of the requirements for

J! . D. degree in W1“ ife

\ ‘ K
, I / / . I
Major professor

 

Date 71//X),/9:‘:/

MS U it an Affirmative Action/Equal Opportunity Institution 0- 12771

 

 

new“

 

 

 

PLACE IN RETURN BOX to remove We chockoutttom your mead.
TO AVOID FINES Mum on or More date duo.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DATE DUE DATE DUE DATE DUE
Its-(,7; l. I ‘ : ~_ 0 00
W-mm .. -' z m"

 

DYNAMICS OF THE RECOVERY OF LAKE TROUT
(SALVELIN US NAMA Y C USH) IN U.S. WATERS OF LAKE SUPERIOR

By

Michael Jay Hansen

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

1994

ABSTRACT

DYNAMICS OF THE RECOVERY OF LAKE TROUT
(SALVELIN US NAMA Y C USH) IN U.S. WATERS OF LAKE SUPERIOR

By

Michael Jay Hansen

Lake trout (Salvelinus namaycush) were nearly extirpated from Lake Superior
during the 19505 due to fishery exploitation and sea lamprey (Petromyzon marinas)
predation. Reproducing populations of lake trout were reestablished in most areas of
Lake Superior by stocking yearlings, controlling sea lampreys, and regulating fisheries.
This dissertation evaluated relative contributions of stocked and wild adult lake trout
to population recoveries and causes of declining survival of stocked lake trout, and
compared current and historic population densities.

In Michigan and Minnesota, stocked lake trout were more strongly correlated to
wild recruitment than wild lake trout. In Wisconsin, stocked and wild lake trout were
both weakly correlated to wild recruitment. I conclude that stocked lake trout
reproduced in Michigan and Minnesota because spawning grounds are inshore where
inexperienced stocked spawners migrate during spawning. I conclude that stocked
lake trout reproduced less successfully in Wisconsin because spawning shoals are
offshore, and require homing ability not possessed by stocked fish.

In Michigan and Wisconsin, survival of stocked lake trout was strongly
correlated to large-mesh gill-net fishing effort and to stocking. In Minnesota, survival
of stocked lake trout was strongly correlated to density of wild lake trout and to

stocking. I conclude that lake trout survival in Michigan and Wisconsin was limited

by fishing mortality, and may be enhanced if large-mesh gill-net fisheries are better
controlled. I conclude that lake trout survival in Minnesota was limited by predation,
and will be more difficult to enhance.

Historic lake trout densities in Michigan, previously thought to be stable prior
to 1939, were declining as early as 1929 in some areas. Wild lake trout densities
exceeded historic densities in some areas during the 1980s, but fell below historic
densities in all areas during the 19903. I conclude that lake trout restoration targets
should be based on modern carrying capacity rather than historic yields or densities
and that further progress in restoration can only be achieved if wild lake trout stocks

are better protected from sea lamprey predation and fishery exploitation.

DEDICATION

For my father, George William Hansen, who planted the seed in me, as a
young boy, that would later sprout into a desire to know about fish and to seek their
conservation. My greatest honor in life is in knowing that he would be proud of the

path that I chose.

ACKNOWLEDGMENTS

I am indebted to my program chair, William W. Taylor, and my supervisors,
Jon G. Stanley (1991-94) and David W. Walsh (1994), who made my doctoral
program a reality in spite of the difficulties imposed by my full-time job. Ivan L.
Mao, Carl W. Ramm, and Scott R Winterstein, the members of my guidance
committee, contributed immensely to my ability to conduct the research, both in their
classrooms and in their individual instruction. Many past and present members of the
Lake Superior Technical Committee, a technical fisheries advisory committee
organized under the auspices of the Great Lakes Fishery Commission, contributed to
discussions that furthered this research. In particular, Mark P. Ebener, Richard G.
Schorfhaar, Donald R Schreiner, Stephen T. Schram, and James H. Selgeby
enlightened me beyond my years about lake trout in Lake Superior and provided the
data used herein. Richard L. Pycha conducted much of the early work that was the
genesis for this research. Wayne R. MacCallum provided the map of Lake Superior
lake trout management areas. Collection of the data used in this work was supported
in part by funds from the United States Federal Aid in Sport Fish Restoration Act.
Joan E. Bratley and Mary T. Halvorsen provided invaluable support in data
management. Connie R. Hansen, my wife, made it all possible, through her endless

support and patiencee—words cannot do justice to her contribution.

TABLE OF CONTENTS

LIST OF TABLES ............................................ viii
LIST OF FIGURES ............................................. x
INTRODUCTION .............................................. 1
History and Causes of Stock Collapse ........................... 2
History and Causes of Stock Recovery .......................... 5
Hypotheses About Lake Trout Recovery ......................... 8
CHAPTER I:
IMPORTANCE OF STOCKED LAKE TROUT
TO RECRUITMENT IN U.S. WATERS OF LAKE SUPERIOR ............ 11
Introduction ............................................ 12
Methods ............................................... 14
Study Area ........................................ 14 I
Abundance ........................................ 15
Statistical Analysis .................................. 16
Results ................................................ 18'
Abundance ........................................ 18«
Sources of Recruitment ............................... 22
Discussion ............................................. 25’
Management Implications ................................... 28-1
CHAPTER II:
DECLINING SURVIVAL OF LAKE TROUT
STOCKED IN U.S. WATERS OF LAKE SUPERIOR .................... 31
Introduction ............................................ 32
Methods ............................................... 34
Recruitment to Age 7 ................................ 34
Factors Potentially Influencing Recruitment ................. 35
Statistical Analysis .................................. 39
Results ................................................ 41
Michigan ......................................... 41
Minnesota ........................................ 45
Wisconsin ........................................ 48
Discussion ............................................. 50

vi

Fishing Mortality ................................... 51

Predation by Wild Lake Trout ................ I .......... 54
Management Implications ................................... 56
CHAPTER III:
STATUS OF LAKE TROUT
RESTORATION IN U.S. WATERS OF LAKE SUPERIOR ................ 58
Introduction ............................................ 59
Methods ............................................... 61
Study Area ........................................ 61
Data Description .................................... 62
Statistical Analysis .................................. 63
Results ................................................ 64
Discussion ............................................. 69
Management Implications ................................... 71
SUMMARY AND CONCLUSIONS ................................ 74
Sources of Recruitment .................................... 74
Causes of Declining Survival ................................ 76
Status of Restoration ...................................... 78
APPENDIX A - ADDITIONAL TABLES ............................ 82
APPENDIX B - ADDITIONAL FIGURES ........................... 101
LIST OF REFERENCES ........... ' ............................. 107

vii

LIST OF TABLES

Table 1. Thousands of tin-clipped lake trout stocked in the four jurisdictional
areas of Lake Superior (updated and modified from Lawrie 1978; fry

and age-Z-and-older stockings excluded). ......................... 6
Table 2. Stock and recruitment data used for modeling survival of yearling

lake trout stocked in Michigan waters of Lake Superior. ............. 36
Table 3. Stock and recruitment data used for modeling survival of yearling

lake trout stocked in Minnesota waters of Lake Superior. ............ 37
Table 4. Stock and recruitment data used for modeling survival of yearling

lake trout stocked in Wisconsin waters of Lake Superior. ............ 38

Table 5. Results of the multiple regression of stocked lake trout CPE at age 7

per million yearlings stocked (logarithms) on yearlings stocked six years

earlier, gill net effort (millions of meters in Michigan and Wisconsin),

and wild lake trout CPE (Minnesota) in Lake Superior. .............. 43
Table 6. Catch/effort of lake trout in spring gill-net assessment fisheries in

western Keweenaw Peninsula waters (MB) of Lake Superior (mean and

SE across N lifts of loge-transformed values). ..................... 82
Table 7. Catch/effort of lake trout in spring gill-net assessment fisheries in

Keweenaw Bay in Michigan waters (MI4) of Lake Superior (mean and

SE across N lifts of loge-transformed values). ..................... 83
Table 8. Catch/effort of lake trout in spring gill-net assessment fisheries

around Marquette in Michigan waters (MIS) of Lake Superior (mean

and SE across N lifts of log-transformed values) ................... 84
Table 9. Catch/effort of lake trout in spring gill-net assessment fisheries

around Munising in Michigan waters (MI6) of Lake Superior (mean and

SE across N lifts of log-transformed values). ..................... 85
Table 10. Catch/effort of lake trout in spring gill-net assessment fisheries

around Grand Marais in Michigan waters (MI7) of Lake Superior (mean

and SE across N lifts of log-transformed values) ................... 86
Table 11. Catch/effort of lake trout in spring gill-net assessment fisheries in

Whitefish Bay in Michigan waters (MI8) of Lake Superior (mean and

SE across N lifts of log-transformed values). ..................... 87
Table 12. Catch/effort of lake trout in spring gill-net assessment fisheries in

western Minnesota waters (MNl) of Lake Superior (mean and SE across

N lifts of loge-transformed values). . ' ........................... 88
Table 13. Catch/effort of lake trout in spring gill-net assessment fisheries in

central Minnesota waters (MN2) of Lake Superior (mean and SE across

viii

N lifts of log-transformed values). ............................ 89
Table 14. Catch/effort of lake trout in spring gill-net assessment fisheries in

eastern Minnesota waters (MN3) of Lake Superior (mean and SE across

N lifts of log-transformed values). ............................ 90
Table 15. Catch/effort of lake trout in spring gill-net assessment fisheries in

western Wisconsin waters (W11) of Lake Superior (mean and SE across

N lifts of loge-transformed values). ............................ 91
Table 16. Catch/effort of lake trout in spring gill-net assessment fisheries in

eastern Wisconsin waters (W12) of Lake Superior (mean and SE across

N lifts of log-transformed values). ............................ 92
Table 17. Results of the multiple regression of recruit CPE (log,) during

1967-93 on native and stocked spawner CPEs (log) during 1959-85 in

Michigan areas MI4-MI7 of Lake Superior. ...................... 93
Table 18. Results of the multiple regression of recruit CPE (loge) during

1967-93 on native and stocked spawner CPEs (loge) during 1959-85 in

Minnesota areas MN2-MN3 of Lake Superior. .................... 94
Table 19. Results of the multiple regression of recruit CPE (log,) during

1967-93 on native and stocked spawner CPEs (log,) during 1959-85 in

Wisconsin area W12 of Lake Superior. ......................... 95
Table 20. Results of the multiple regression of recruitment rate (log,) of the

1963-82 year classes on numbers of yearlings stocked and large-mesh

gill-net fishing effort in Michigan waters of Lake Superior. ........... 96
Table 21. Results of the multiple regression of recruitment rate (log,) of the

1963-82 year classes on numbers of yearlings stocked and density

(CPE) of wild adult lake trout in Minnesota waters of Lake Superior ..... 97
Table 22. Results of the multiple regression of recruitment rate (log,) of the

1963-82 year classes on numbers of yearlings stocked and large-mesh

gill-net fishing effort in Wisconsin waters of Lake Superior. .......... 98
Table 23. Commercial fishery lake trout catch (pounds) and large-mesh gill-net

effort (1,000 feet) in Michigan statistical districts MS-l through MS-3

of Lake Superior during 1929-61 (compiled from Jensen and Buettner

1976). ................................................ 99
Table 24. Commercial fishery lake trout catch (pounds) and large-mesh gill-net

effort (1,000 feet) in Michigan statistical districts MS-4 through MS-6

of Lake Superior during 1929-61 (compiled from Jensen and Buettner

1976) ................................................ 100

ix

LIST OF FIGURES

Figure 1. Lake trout fishery statistics and sea lampreys caught at electrical

weirs in Michigan and Wisconsin waters of Lake Superior during

1929-70 (from Hile et a]. 1951; Pycha and King 1975). .............. 4
Figure 2. Estimated numbers of spawning sea lampreys in United States

tributaries of Lake Superior during 1958-92 (from Klar and Weise

1994) ................................................. 7
Figure 3. Spring abundance of wild and planted lake trout in U.S. waters of

Lake Superior during 1970-92 (3-year moving averages of geometric

mean number per km of 114-mm stretch-measure gill net) (from Hansen

et al. 1994b). ........................................... 10
Figure 4. Lake Superior lake trout management areas. U.S. management areas

are denoted by state: MI, Michigan; MN, Minnesota; WI, Wisconsin.

Areas marked by numbers only are in Canadian waters. ............. 15
Figure 5. Catch/effort of lake trout in spring gillnet assessment fisheries at the

western Keweenaw Peninsula (MI3), Keweenaw Bay (MI4), and

Marquette (MIS), Michigan waters of Lake Superior during 1959-93. . . . . 19
Figure 6. Catch/effort of lake trout in spring gillnet assessment fisheries at

Munising (MI6), Grand Marais (MI7), and Whitefish Bay (MI8),

Michigan waters of Lake Superior during 1959-93. ................. 20
Figure 7. Catch/effort of lake trout in spring gillnet assessment fisheries in

western (MN2) and eastern Minnesota (MN3), and eastern Wisconsin

(W12) waters of Lake Superior during 1959-93. ................... 21
Figure 8. Multiple regression coefficients (i95% CI.) of wild and stocked lake

trout CPEs on recruit CPE in Michigan (MI3-MI8), Minnesota

(MN2-MN3), and Wisconsin (W12) waters of Lake Superior during

1959-93. ............................................... 23
Figure 9. Coefficients of determination of wild and stocked lake trout CPEs on

recruit CPE in Michigan (MI3-MI8), Minnesota (MNZ-MN3), and

Wisconsin (W12) waters of Lake Superior during 1959-93. ........... 24
Figure 10. Recruitment rate of the 1963-82 year-classes of stocked yearling

lake trout to age 7 compared to the number stocked (left panel) and

large-mesh gill-net fishing effort (right panel) in Michigan waters of

Lake Superior. .......................................... 41
Figure 11. Catch per effort of age-7 stocked lake trout caught in assessment

fisheries (dots) and predicted from yearling stocking and large-mesh

gill-net fishing effort (line) in Michigan waters of Lake Superior. ...... 42

Figure 12. Standardized residuals for stocked lake trout recruitment, predicted

from yearling stocking and large-mesh gill-net fishing effort in

Michigan waters of Lake Superior (:h the 95% t-interval). ............ 44
Figure 13. Recruitment rate of the 1963-82 year-classes of stocked yearling

lake trout to age 7 compared to the number stocked (left panel) and

wild lake trout density (right panel) in Minnesota waters of Lake

Superior. .............................................. 45
Figure 14. Catch per effort of age-7 stocked lake trout caught in assessment

fisheries (dots) and predicted from yearling stocking and wild lake trout

density (line) in Minnesota waters of Lake Superior. ................ 46
Figure 15. Standardized residuals for stocked lake trout recruitment, predicted

from yearling stocking and large-mesh gill-net fishing effort in

Minnesota waters of Lake Superior (:1: the 95% t-interval). ........... 47
Figure 16. Recruitment rate of the 1963-82 year-classes of stocked yearling

lake trout to age 7 compared to the number stocked (left panel) and

large-mesh gill-net fishing effort (right panel) in Wisconsin waters of

Lake Superior. .......................................... 48
Figure 17. Catch per effort of age-7 stocked lake trout caught in assessment

fisheries (dots) and predicted from yearling stocking and large-mesh

gill-net fishing effort (line) in Wisconsin waters of Lake Superior. ...... 49
Figure 18. Standardized residuals for stocked lake trout recruitment, predicted

from yearling stocking and large-mesh gill-net fishing effort in

Wisconsin waters of Lake Superior (:1: the 95% t-interval). ........... 50
Figure 19. Catch/effort of lake trout in spring gill net assessment fisheries

(number per km of net) inshore (non-refuge) and offshore (refuge) in

eastern Wisconsin waters of Lake Superior in 1990. ................ 53
Figure 20. Recruitment of the 1982 lake trout year-class at age 7 (CPE/million

yearlings stocked), compared to the average annual large-mesh gill-net

fishing effort during 1983-88, in inshore Michigan areas of Lake

Superior. .............................................. 54
Figure 21. Abundance of stocked and wild lake trout in Michigan west (MI3)

and east (MI4) of the Keweenaw Peninsula in Lake Superior during

1929-93, compared to the average abundance during 1929-43. ......... 66
Figure 22. Abundance of stocked and wild lake trout in Michigan near

Marquette (MIS) and Munising (MI6) in Lake Superior during 1929-93,

compared to the average abundance during 1929-43. ................ 67
Figure 23. Abundance of stocked and wild lake trout in Michigan near Grand

Marais (MI7) and in Whitefish Bay (MI8) in Lake Superior during

1929-93, compared to the average abundance during 1929-43. ......... 68
Figure 24. Scatter-plot matrix of native and stocked spawner CPEs (103,)

during 1959-85 and recruit CPE (log,) during 1967-93 in Michigan areas

MI4-MI7 of Lake Superior. ................................. 101
Figure 25. Scatter-plot matrix of native and stocked spawner CPEs (log,)

during 1959-85 and recruit CPE (log,) during 1967-93 in Minnesota

areas MN2-MN3 of Lake Superior. ............................ 102

xi

Figure 26. Scatter-plot matrix of native and stocked spawner CPEs (loge)
during 1959-85 and recruit CPE (log,) during 1967-93 in Wisconsin area

W12 of Lake Superior. .....................................

Figure 27. Scatter-plot matrix of stocked lake trout survival, millions of
yearlings stocked, average grams/yearling, wild adult lake trout CPE,
stocked adult lake trout CPE, wild juvenile lake trout CPE, and
large-mesh gill-net fishing effort in Michigan waters of Lake Superior

during 1963-89. . . . . ~ .....................................

Figure 28. Scatter-plot matrix of stocked lake trout survival, millions of
yearlings stocked, average grams/yearling, wild adult lake trout CPE,
stocked adult lake trout CPE, wild juvenile lake trout CPE, and
large-mesh gill-net fishing effort in Minnesota waters of Lake Superior

during 1963-89. .........................................

Figure 29. Scatter-plot matrix of stocked lake trout survival, millions of
yearlings stocked, average grams/yearling, wild adult lake trout CPE,
stocked adult lake trout CPE, wild juvenile lake trout CPE, and
large-mesh gill-net fishing effort in Wisconsin waters of Lake Superior

during 1963-89. .........................................

xii

103

INTRODUCTION

The Laurentian Great Lakes collectively arose from several glacial advances
and retreats during the Pleistocene, and their fish faunas are therefore geologically
young (Lawrie and Rahrer 1973). There are 174 species of fish in 71 genera and 28
families in the Great Lakes (Bailey and Smith 1981). The genus Coregonus
differentiated into the greatest number of species (Todd and Smith 1980; Smith and
Todd 1984), but the lake trout (Salvelinus namaycush) differentiated into the greatest
number of distinct morphological forms (Khan and Qadri 1970; Goodier 1981). The
taxonomic status of these morphological forms of lake trout in Lake Superior has been
debated for decades (Khan and Qadri 1970), but there has been little dispute that lake
trout formed discrete spawning stocks that used many offshore shoals, rocky
shorelines, and tributary streams (Lawrie and Rahrer 1973).

Three forms of lake trout are still present in Lake Superior, including the "lean"
lake trout that inhabits most deep, cold lakes in North America, and the "Siscowet" and
"humper" lake trout that occur only in Lake Superior (Khan and Qadri 1970; Goodier
1981; Bumham-Curtis 1993; Bumham-Curtis and Smith 1994). Lean lake trout are
slender, have a low body fat content and straight, pointed snouts, and inhabit inshore
waters less than 70 m deep. Siscowet lake trout are deep-bodied, have a high body fat

content and blunt snouts, and inhabit offshore waters 50-150 m deep. Humper lake

2

trout are intermediate in body depth, have an intermediate body fat content, and
inhabit offshore shoals that are surrounded by waters greater than 100 m deep. Lake
trout restoration in Lake Superior was limited to the lean form, so I will restrict further
discussion and all of my analyses to lean lake trout.

Lean lake trout spawn in Lake Superior from October through early November
(Eschmeyer 1955; Peck 1986; Ebener 1990). Males first mature at 7-8 years of age
while females mature at 9-11 (Rahrer 1967; Peck 1979; Ebener 1990). Females
produce 1,400-1,500 eggs per kg of body weight (Eschmeyer 1955; Peck 1988;

Schram 1993), and annual recruitment of yearling lake trout was 36-10.] million when
fishery yields averaged 2 million kg annually (Sakagawa and Pycha 1971). Lake trout
generally reside within an 80 km home range (90% of marked fish are returned within
this distance, regardless of size at release and length of time at large), though some
individuals move several hundred km (Eschmeyer et a1. 1953; Loftus 1958; Buettner

1961; Pycha et a1. 1965; Rahrer 1968; Swanson 1973; Ebener 1990).
History and Causes of Stock Collapse

Fisheries developed in Lake Superior in increasingly opportunistic pursuit of
new grounds (Goodier 1982), a process known as fishing up that probably occurred
throughout the 18005 and 19003 (Goodier 1989). The annual harvest of lake trout was
less than 1 million kg in 1879 when lakewide harvest statistics were first available,
peaked at 3 million kg in 1903, and averaged 2 million kg per year (CV=13%) during

1913-50 (Baldwin et a1. 1979). The persistence of lake trout harvest during 1913-50

3

suggested that 2 million kg was a sustainable annual yield, but yield was sustained
during the 19405 in Michigan by increased fishing intensity, in spite of declining
abundance (Hile et a1. 1951) (Figure 1). Efficiency of gill nets doubled during the late
1940s as nets were converted from cotton and linen twines to nylon (Pycha 1962) and
fishermen enhanced their ability to locate fish using depth sounders (Hile et al. 1951).
Hile et al. (1951) warned that lake trout in Michigan waters of Lake Superior
in 1949 were ”fast nearing a dangerously low level and in poor condition to withstand
the impending ravages of a growing population of sea lampreys" (Petmmyzon
marinas). During 1949-52, fishing intensity doubled, and sustained yield in spite of a
50% decline in abundance (Pycha and King 1975) (Figure 1). Fishing intensity, yield,
and abundance then declined during 1953-61, as sea lampreys invaded Lake Superior,
increased in abundance, and preyed on remaining lake trout (Dryer and King 1968).
The combined effects of intensive fishing exploitation and sea lamprey predation were
too much for lake trout to sustain, so stocks had essentially collapsed by 1962 when
sea lampreys were reduced and lake trout fisheries were closed (Pycha and King 1975;
Swanson and Swedberg 1980). Coble et a1. (1990) found that the decline of lake trout
in Michigan waters began in 1939, while Hile et al. (1951), Pycha and King (1975),
and Jensen (1978) found that the decline began in 1945. In any case, lake trout
abundance declined well before sea lampreys reached abundances that could otherwise

have caused the decline.

 

 

 

 

 

 

 

 

 

 

 

 

350 . 4o
ldkfiugan I
300- 1‘ S“
o 1 Intensity ;\ LamPteys
g I I ~30
o 3 I
> . I ~ m
< I 8
m I
v 5
as .
N 'o
g: -'20 53
“a I” E
a) H
E 5
V
g 3" ['10
9"
I
o TTTTIIIIIIIPTIWIIIIFTITTT IITITITTI 0
350 40
j ‘Wfimxnmhi ’
I Sea
300': Lanuxeys I
o .
§0 - ~30
g 250-: g on
< . 8
«a 1 I
(I 200- .I g
N 1 “o
.O—i I "~20 3
C... q t- ‘6
3 150-1 1:
on
a . §
5 - . w
8 100‘ ,. 10 V
. P
d: 3 j
503
I
o [TlIIIIIITIIITIIITIITITITTT IIIIIII 1.0

 

 

 

 

 

 

 

 

 

1930 35 40 45 50 55 60 65 70

Year

Figure 1. Lake trout fishery statistics and sea lampreys caught at electrical weirs in
Michigan and Wisconsin waters of Lake Superior during 1929-70 (from Hile et al.
1951; Pycha and King 1975).

5

History and Causes of Stock Recovery

The primary strategy for restoring lake trout into Lake Superior was to increase
recruitment by stocking, and to reduce mortality by controlling sea lampreys and
regulating fisheries (Lawrie and Rahrer 1972, 1973; Pycha and King 1975; Lawrie
1978; LSLTTC 1986). Stocking of lake trout into Lake Superior was begun during
1950. Yearlings composed 88% of the 94 million stocked through 1992 because they
survived 4-10 times better than fingerlings (Buettner 1961; Pycha and King 1967)
(Table 1). Yearling releases increased during 1950-59, but still ranged much lower
during 1960-92 (1.1-3.7 million/year) than the production of wild yearlings in years
prior to stock collapse (3.6-10.l million/year) (Sakagawa and Pycha 1971). Yearling
plantings nonetheless still produced spawner densities in inshore Michigan waters
similar to lightly exploited, offshore Iean lake trout at Michipicoten Island, Superior
Shoal, and the Caribou Islands prior to the invasion of sea lampreys (Lawrie 1978).

Sea lampreys invaded Lake Superior during the early 19405, and reached an
average abundance in the United States of 296,000 (1 80,000) during 1958-61 (Klar
and Weise 1994) (Figure 2). Control subsequently reduced their numbers to an
average of only 44,000 (2h 22,000) during 1962-92 (Klar and Weise 1994). Sea
lampreys reproduced widely, as larvae have been found in 90 United States and 64
Canadian streams since 1950, but most are produced in 19 United States and 9
Canadian streams that have suitable spawning and larval habitat, and adequate flow
(Smith et a]. 1974; Smith and Tibbles 1980; Klar and Weise 1994). Sea lampreys are

controlled by killing larvae with selective chemicals, 3-trifluoromethyl-4-nitr0phenol

6

Table 1. Thousands of fin-clipped lake trout stocked in the four jurisdictional areas of Lake Superior
(updated and modified from Lawrie 1978; fry and age-2-and-older stockings excluded).

 

 

 

 

 

Year Minnesota Wisconsin Michigan Ongrio Total

Class Age 0 Age 1 Age 0 Age 1 Age 0 Age 1 Age 0 Age 1 Age 0 Age 1
1950 0 0 0 0 0 0 50 0 50 O
1951 0 0 0 102 0 0 0 0 0 102
1952 0 0 145 80 65 69 0 0 210 150
1953 0 0 133 102 139 134 50 0 322 236
1954 0 0 142 103 121 61 0 0 264 164
1955 0 O 0 201 0 0 0 0 201
1956 0 0 0 0 0 0 0 0 0 0
1957 0 0 0 184 O 298 0 538 0 1,020
1958 0 0 0 151 0 l 1 0 473 0 635
1959 0 0 0 161 0 393 0 396 0 950
1960 0 0 50 314 0 393 50 434 100 1,141
1961 0 0 0 256 0 705 60 508 60 1,469
1962 77 0 0 31 1 70 1,186 0 477 147 1,974
1963 175 182 0 745 162 1,196 0 472 337 2,596
1964 38 102 0 447 0 659 0 468 38 1,675
1965 150 108 0 352 0 2,218 0 450 150 3,128
1966 151 227 0 235 0 2,059 0 500 151 3,022
1967 154 223 0 239 0 2,260 0 500 154 3,222
1968 153 216 0 254 0 1,860 0 562 153 2,892
1969 0 226 0 204 0 1,916 0 438 0 2,785
1970 0 280 0 206 0 1,055 0 475 0 2,016
1971 0 290 0 259 0 1,063 0 371 0 1,983
1972 0 284 0 227 » 0 894 121 500 121 1,904
1973 0 304 0 436 0 887 0 465 0 2,092
1974 0 337 0 194 0 774 0 510 0 1,814
1975 0 345 0 551 24 785 0 520 24 2,200
1976 0 350 205 368 0 677 0 677 205 2,072
1977 0 355 183 440 101 731 0 629 284 2,155
1978 0 314 181 297 124 789 0 525 305 1,925
1979 0 351 211 342 228 578 0 548 439 1,818
1980 0 312 180 351 200 561 206 811 586 2,035
1981 0 288 287 242 153 676 203 990 644 2,195
1982 161 392 266 274 218 834 206 980 852 2,480
1983 0 212 175 131 0 472 272 1,290 447 2,105
1984 94 358 1 18 408 779 1,552 292 930 1,282 3,249
1985 45 408 222 312 76 1,045 303 1,337 646 3,102
1986 0 91 40 75 120 863 300 1,567 460 2,596
1987 O 212 160 180 0 394 300 1,685 460 2,471
1988 0 370 0 211 150 523 300 1,854 450 2,957
1989 54 361 0 173 150 O 467 1,252 671 1,786
1990 0 542 0 391 150 592 425 2,079 575 3,605
1991 0 499 0 417 150 682 300 2,096 450 3,694
1992 0 540 0 225 150 477 0 1,960 150 3,203
Subtotal 1,250 9,081 2,699 11,154 3,331 32,320 3,905 30,264 11,186 82,819

Total 10,331 13,853 35,651 34,169 94,005

 

7
(TFM) (Smith 1971), and Bayer-73 (Smith et al. 1974). Low-head barrier dams have

also been built on four United States and six Ontario streams to block spawning sea

lampreys without blocking other species.

 

 

 

 

 

is .
2
i
i. ‘ > .
U 1'1 17:1 7'
f l :7. :23“,
'0 .~ 3 2.2: ;;:~
:1 g‘ . f}? :1
I If
v‘. 1 l... ‘ 5‘
‘5 200 -.
5 i712. ’3
o :3! ’
i ‘ ..1 , '
. I , _-. ~~
33p" .4": «‘7
#0.: :2;
1;»: ‘_
1’??? :
L3,": I;
... .. .
I .-
.‘ji ..
I I j
100 t : 7.
J'" «v I
E's-.2; 2*: :5;
("T1 é: t. .
~,'.'. ‘I . 1'
2,:
g". gag
--~a- . . s,_
3. :1 i s 3:
gig-1 a.» ' 1’ p: a;
I 1 If a; TI
i... x». . . -: -
if"! .."9 i;- » . *I i ‘
r . . , r . : ..
" ‘- '. . , _ x, 0, ._ _
.7 k' :41"- -.. .-‘v' 57." r ' " '- 9“}: It‘s; ' - -
0 I; 1;...1 2‘ LA; . ’t .. ._ . I ~ . I:

 

586062646668707274767880828486889092

Year

Figure 2. Estimated numbers of spawning sea lampreys in United States tributaries of
Lake Superior during 1958-92 (from Klar and Weise 1994).

Sport and commercial fisheries for lake trout were closed in 1962 (Pycha and
King 11975), but management agencies reopened restricted sport and commercial
fisheries as lake trout stocks began to recover. Numerous regulations were intended to
limit fishing mortality on lake trout, but were inconsistently applied in the various
jurisdictions. Regulations that were intended to reduce fishing mortality on lake trout

included: (1) limiting the number of commercial fishing licenses, (2) restricting gill

8

netting for other species to depths that minimized the incidental kill of lake trout, (3)
setting quotas on the commercial harvest and creel limits on the angling harvest, (4)
establishing a refuge around the Gull Island Shoal lean lake trout spawning stock, (5)
closing the lake trout spawning season to fishing, (6) converting to entrapment gear
that allows higher survival of released lake trout, and (7) limiting gill-net effort.

Sea lampreys and fisheries have been reduced in Lake Superior, but continue to
exert excessive mortality on lake trout. Sea lampreys caused mortality of 20-82% on
lake trout (age 7 and older) during 1968-78 (instantaneous rates of 0.21 to 1.70), while
fishing mortality was only 16-34% (instantaneous rates of 0.17 to 0.42) and natural
mortality was only 23% (instantaneous rate of 0.26) (Pycha 1980). Consequently, sea
lampreys consumed more of the available lake'trout production than humans during
1968-78 when sea lampreys were at only 15% of their peak abundance. During
1990-92, sea lampreys consumed 41.6% of the total lake trout yield in United States
waters west of the Keweenaw Peninsula and 17.9% from waters east of the Keweenaw

Peninsula (Hansen et al. 1994b).

Hypotheses About Lake Trout Recovery

Interagency management of lake trout restoration is coordinated under the aegis
of the Great Lakes Fishery Commission (GLFC 1980). Interagency committees of
fishery researchers and managers developed a plan for restoring lake trout in Lake
Superior (LSLTTC 1986), objectives for managing the entire fish community (Busiahn

1990), and reports of progress toward the goals and objectives (Hansen 1990, 1994).

9

These reports raised three questions, based on trends in lake trout abundance since
1970 (Figure 3). First, to what extent did stocked lake trout contribute to recruitment
of wild lake trout? Second, why did abundance of stocked lake trout invariably
decline, often to the point where stocking no longer enhanced abundance of spawning
stocks? Third, how does the abundance of current lake trout stocks compare to that of
historic lake trout stocks?

I will address each of these questions in this dissertation. In Chapter I, I will
model the contribution of stocked lake trout to the recruitment of wild lake trout that
became vulnerable to the assessment fishery in Michigan after 1970 and Minnesota
after 1980 (Figure 3). In Chapter II, I will model potential causes of declining
abundance of stocked lake trout that occurred in Michigan during the 19805,
Wisconsin during the 19705, and Minnesota during the late 19805. In Chapter III, I
will evaluate the current status of lake trout stocks by developing a means to directly
compare the abundance of lake trout during the 19905 with the abundance of lake trout

during the 1929-43 historic reference period.

10

 

100

1 Michigan (MB-M17)

 

 

 

 

 

‘ Wisconsin (W12)

Catch/Effort

 

 

 

 

100
Minnesota (MN 2-MNB)

 

 

 

 

Figure 3. Spring abundance of wild and planted lake trout in U.S. waters of Lake
Superior during 1970-92 (3-year moving averages of geometric mean number per km
of 114-mm stretch-measure gill net) (from Hansen et al. 1994b).

CHAPTER I:
IMPORTANCE OF STOCKED LAKE TROUT

TO RECRUITMENT IN U.S. WATERS OF LAKE SUPERIOR

Abstract—Lake trout (Salvelinus namaycush) sustained an average annual
yield of 2 million kg during 1913-50 in Lake Superior, but collapsed to near-extinction
during 1951-62 because of excessive fishery exploitation and sea lamprey (Petromyzon
marinas) predation. Hatchery-reared, juvenile lake trout were stocked, in conjunction
with controls on sea lampreys and fisheries, to reestablish lake trout in the lake. The
contribution of wild and stocked parents on recruitment were evaluated by regressing
catch per effort (CPE) of the two potential parental stocks on CPE of wild recruits.
Data were from lake trout catches in 114-mm assessment gill nets set each spring
during 1959-93 in United States waters of Lake Superior. Stocked lake trout
explained much of the variation in recruitment in Michigan (66%) and Minnesota
(63%). In contrast, wild lake trout explained little of the variation in recruitment in
either Michigan (9%) or Minnesota (14%). In Wisconsin, stocked and wild lake trout
explained much less of the variation in recruitment (29% and 1%, respectively) than in
either Michigan or Minnesota. I conclude that stocked lake trout reproduced in
Michigan and Minnesota because they could easily locate the inshore spawning
grounds there, and were largely responsible for stock recoveries in both states. I
conclude that stocked lake trout reproduced less effectively in Wisconsin because they
could not easily locate the offshore spawning grounds there, and were less responsible

for stock recovery there.

11

12

Introduction

Lake trout (Salvelinus namaycush) sustained an average annual yield of 2
million kg (CV =13%) during 1913-50 in Lake Superior (Baldwin et a1. 1979), but
collapsed nearly to extinction during 1951-62 because of excessive fishery exploitation
and sea lamprey (Petmmyzon marinas) predation (Hile et a]. 1951; Pycha and King
1975; Jensen 1978; Coble et a]. 1990). Hatchery-reared, juvenile lake trout were
stocked, in conjunction with controls on sea lampreys and fisheries, to restore lake
trout into the lake (Lawrie and Rahrer 1972, 1973; Pycha and King 1975). Stocking
has been nearly continuous since 1951 in Wisconsin, 1952 in Michigan, 1957 in
Ontario, and 1962 in Minnesota (Lawrie and Rahrer 1972, 1973; Pycha and King
1975; Lawrie 1978). Sea lampreys peaked in abundance during 1958-61, and were
reduced 87% from 1961 to 1962 using chemicals, barrier dams, and traps (Smith 1971;
Smith et al. 1974; Smith and Tibbles 1980; Klar and Weise 1994). Lake trout
fisheries were closed lakewide in 1962, and reopened later (Pycha and King 1975).

The contribution of stocked lake trout to population recovery in Lake Superior
has been the subject of considerable debate. Dryer and King (1968) predicted
optimistically that the build-up of spawning stocks during 1958-66 and subsequent
reproduction during 1964-66 (the first since 1959) at Gull Island Shoal, Wisconsin,
would soon replace hatchery stockings. They noted, however, that stocked lake trout
generally attempted to spawn near release sites, rather than on offshore reefs where
spawning historically occurred. Pycha and King (1975) also noted that stocked lake

trout tended to spawn inshore near stocking sites in Wisconsin where reproduction had

13

not occurred historically, and suggested that stocking nearer suitable spawning grounds
was needed to imprint the stocked fish to those sites. Wild spawners produced
significantly more young lake trout than stocked spawners in the Apostle Islands
because stocked fish were less able to locate offshore spawning reefs (Krueger et a1.
1986). Stocked fish were substantially more abundant in Michigan than in Wisconsin,
and aggregated in densities rivaling those at Gull Island Shoal on most historically
important offshore spawning reefs (Peck 1979; Peck and Schorfhaar 1991).

The presence of residual native lake trout in most areas of Lake Superior
confounded determination of the importance of stocked lake trout to recruitment
(Lawrie 1978). Wild lake trout were extremely rare but were nonetheless responsible
for recovery of the Gull Island Shoal stock (Swanson and Swedberg 1980). The
presence of even a few wild lake trout in Lake Superior confounded interpretation of
the importance of stocked fish on recruitment. Eshenroder et a1. (1983) stated that
"the Lake Superior example of success may not be appropriate for the situation in the
other lakes where native stocks are believed to be extinct." The only study that has
quantified the relative contributions of. stocked and wild lake trout to recruitment in
Lake Superior confirmed that stocked fish were reproductively ineffective compared to
wild fish (Krueger et a1. 1986). A conventional wisdom emerged, that stocked lake
trout were impaired in their ability to find suitable spawning grounds, and spawned on
sites that were inappropriate for reproduction (Eshenroder et al. 1983).

The contribution of stocked lake trout to recruitment has only been tested for a
single spawning population in one area of Lake Superior (Sand Cut Reef, Wisconsin)

(Krueger et a1. 1986). A similar analysis across more areas would determine whether

14

the results of the analysis by Krueger et a1. (1986) apply to different spawning habitat
distributions (e.g. inshore in Michigan and Minnesota, versus offshore in Wisconsin;
Coberly and Horrall 1980; Goodyear et a1. 1982; Thibodeau and Kelso 1990) and
spawning stock densities (e.g. high in Michigan, versus low in Wisconsin and
Minnesota; Hansen et al. 1994b). My objective was to determine the relative
importance of stocked and wild adult lake trout to wild recruitment in different areas
of Lake Superior. The null hypothesis for my analysis will be that stocked lake trout

had no association with recruitment of wild lake trout in that area one generation later.

Methods

Study A rea

Stock assessment of lake trout in Lake Superior is carried out in accordance
with an inter-agency rehabilitation plan that specifies management areas for reporting
progress in lake trout stock restoration (LSLTTC 1986) (Figure 4). The lake trout
management areas in Michigan, Minnesota, and Wisconsin were modified from areas
described by Hile (1962) for reporting commercial fishery statistics. Large statistical
districts were divided into smaller management areas because of movement studies that
showed 90% of marked lake trout were generally recaptured within 80 km, regardless
of the size at release or length of time at large (Eschmeyer et a]. 1953; Buettner 1961;
Pycha et a1. 1965; Rahrer 1968; Swanson I973; Ebener 1990; Peck and Schorfhaar
1991). Management areas in Ontario are used for managing lake whitefish commercial

fishery quotas, and bear no resemblance to former statistical districts.

15

 

Figure 4. Lake Superior lake trout management areas. U.S. management areas are
denoted by state: MI, Michigan; MN, Minnesota; WI, Wisconsin. Areas marked by
numbers only are in Canadian waters.

A bundcmce

Trends in relative abundance of lake trout were monitored with assessment
gill-nets fished in each lake trout management area during 1959-93 in Michigan and
Wisconsin and 1963-93 in Minnesota. Nets were of 1114-mm stretched-mesh, 210/2
multifilament nylon twine, 18 meshes deep, and hung on the 1/2 basis. Fishing was
conducted from late April to early June, a period when availability was relatively high
and uniform compared to other seasons (Sakagawa 1967). Nets were of non-uniform
length, so catch per effort (CPE) was defined as the number of fish caught per km of
net. Sets were also of non-uniform duration, particularly during 1959-69, so CPE was
standardized to net-nights using corrections derived from an experiment during 1970:

net-nights=1.00 for sets of one night duration, 1.52 for sets of two nights duration, and

16

1.80 for sets of three or more nights duration (Curtis et al., in press).

Hatchery-reared lake trout were all marked by removal of a fin before stocking
(Bailey 1965), so the CPE of unclipped lake trout was assumed to be of wild fish and
the CPE of clipped lake trout was assumed to be of stocked fish. The mean CPEs of
wild and stocked lake trout were distributed log-normally, with heteroscedastic
variances, so one was added to each CPE to account for zero catches and transformed
to natural logarithms for analysis (Sokal and Rohlf 1981). Means and standard errors
were computed for each area and year, then transformed back into geometric means
(Sokal and Rohlf 1981). Assessment fisheries in MIZ, MN], and W11 were only

begun in the 19805, so CPEs from these areas were not analyzed further.

Statistical A nalysis

Multiple linear regression was used to evaluate the relative contribution of
stocked and wild lake trout to subsequent recruitment. The CPEs of stocked and wild
lake trout in spring assessment fisheries were used as indices of parental and recruited
stock sizes because previous movement studies showed that spawning stocks generally
remained within 80 km of the spawning site. Total catches of stocked and wild lake
trout were used because age-specific catches were not available across all areas and
years. I assumed that the CPE of each parental stock would be related to the CPE of
wild recruits eight years later (one generation) if that parental stock was reproductively
important (Krueger et a]. 1986). A generation time of eight years was used because
age 7 is the modal age in 114-mm stretch-measure gill nets, and growth has not

changed enough during the interval of analysis to alter either the modal age or the age

17
of maturity (Hansen et al. 1994b). The model was:

y = Bo + lel + 1321': + 8- (1)

where y = log, catch per effort of wild recruits (CPER),
x , = logc catch per effort of stocked parents (CPES),
x, = loge catch per effort of wild parents (CPEw), and

e = residual variance unexplained by the regression.

Overall model fit was assessed using adjusted multiple coefficients of determination
(R7). Magnitudes of squared, standardized partial regression coefficients from the
regression model (F) were used to assess the relative importance of stocked and wild
parents (CPEs and CPEW) on recruits (CPER) in each management area. Covariance
analyses were used to determine if area-specific relationships were homogeneous
across management areas (N=9). I assumed that homogeneous slope coefficients
indicated similar relationships for the areas tested. The results of covariance analyses
were used to consolidate homogeneous sets of management areas into models
describing stock-recruitment relationships across larger areas. Models were diagnosed
for collinearity among predictor variables, and residual errors were diagnosed for
normality, homogeneity of variance, independence, and linearity (Draper and Smith

1981; Systat, Inc. 1992; Kirby 1993).

18

Results

A bundance

Stocked lake trout were more abundant during 1959-1993 in western Michigan
(MI3-MIS), than in eastern Michigan (MI6-MI8), Minnesota (MN2 and MN3), or
Wisconsin (W12). In western Michigan, stocked fish increased during the late 19605,
remained abundant during the 19705, declined sharply during the 19805, and remained
scarce after 1988 (Figure 5; Appendix A, Tables 6-8). In eastern Michigan, stocked
fish also increased during the late 19605, but then declined quickly during the early
19705, more slowly during the late 19705 and early 19805, and were scarce after 1985
(Figure 6; Appendix A, Tables 9-11). In Wisconsin, stocked fish abundance followed
a similar pattern as in eastern Michigan, but in Minnesota, stocked fish increased
slowly during the 19705, remained high during the 19805, and declined thereafter
(Figure 7; Appendix A, Tables 12-16). During the 19905, stocked fish were extremely
rare throughout Michigan, and were declining elsewhere.

Wild lake trout were generally more numerous in Michigan than in Minnesota
or Wisconsin during 1959-1993, though wild fish were always present in Wisconsin
and nearly absent in Michigan during the late 19605 and Minnesota during the 19605
and 19705. In western Michigan, wild fish increased steadily during the 19705 and
early 19805, and declined slowly thereafter (Figure 5; Appendix A, Tables 6-8). In
eastern Michigan, wild fish increased steadily after 1970, but declined in Whitefish
Bay (MIS) where lake trout restoration was deferred in favor of gill-net fishing for

lake whitefish (Figure 6; Appendix A, Tables 9-11). Wild fish were rare in Wisconsin

l9

 

RAE}

~ 1
30 i {I Stocked
1 I!
I l
I i l /\
/
, / I, / \ / \
‘j l
j l

I, \" / I \\ Wild

    

\
l

 

‘wh “
0 llTr—ITITTIITIIIITHIIIIIITTII TrII

 

: l
120 1 M” ' A

Catch/Effort

I ,’ \ Wild
‘1

 

. / \ —————
0TT‘IIITTIIIIITTTTTTITIIIIIIlIllIII

140

 

 

3 MIS
120 j

 

 

 

 

0 Tll’ITITIIrTIlllITIIITIllITTTTIIII

1960 65 70 75 80 85 90

Year

Figure 5. Catch/effort of lake trout in spring gillnet assessment fisheries at the
western Keweenaw Peninsula (MI3), Keweenaw Bay (MI4), and Marquette (MI5),
Michigan waters of Lake Superior during 1959-93.

20

 

    

 

 

 

 

 

 

 

 

100
.. f
. MI6 A
- j \
80 _ l \‘ Stocked
. I \
j l
d ' \
60 4 I I
. I, \\
I
I / I
40 a J/ \\
‘ f l r“'\\
I , \ /
4 , \v/ \
20 T [I
- I
. //
I _,/ / \\ _____ ’4
0TTIlIlllllIllllllTllTllllTTllllll
100
MI7
80‘, ,r‘ Stocked
I
‘I I“
t - I, 1
° 60- , ‘ '
a: l l I\
‘ I l I \
I I" -1
O I I l, \
I; 40 T I, l”! \\
o 4 [I I \ /\
I
. /
20 .3 //
‘ /
_/
. /" \~‘
0‘T—I‘IIIIIWTnIIIIIIITTIIIIIWIIIITTTITT
100
I
‘ MI8
80 - I
.. [’1
I ’I
. ’ l
I
60- 1’ I Stocked
-4 l 1‘
4 I ‘
. f l
_ 1
‘ I x ’\ t l
I \ I ‘ l ‘
I \ I \I I
‘ I \l \V ‘
20 - I \I ‘
.1 / l \\
IF \ .
, \ /___\ wad
c: ‘w \\- \/ \
0‘ TIIIIIIIIrTTTI
1960 65 70 75 80 85 90
Year

Figure 6. Catch/effort of lake trout in spring gillnet assessment fisheries at Munising

(MI6), Grand Marais (MI7), and Whitefish Bay (MI8), Michigan waters of Lake
Superior during 1959-93.

21

 

 

 

 

 

 

 

 

80
.3MN2
60-
. Stocked
f /\
I \ / \
. / \ ,/ \
/ \ r \\ 1
20— / \ \l
. / V
/
. /\__‘ ,
\ /\ I ~’ WM
‘1 / \sl \\_, d/
0 IIII77“ITITII‘ITIII IITTI IITII‘IITIII
80
fi
« MN3 ,‘I Stocked
. H
n
60 - I I
-I I l
t I\ A
O I i I \
t: ‘ I‘l‘
\ I . \
40 ‘ / \ ' \
o - / \ i ‘Q
*6 I I \ I x
U ‘ / \I \ A
20*
.J
0 l 111?_7rr1 III rlrr'rTI III ITI Ill rrfi’r
80
. ‘WD
\
0 IIIITIMFTTIIITII’IIrTrITw ITTIITII II

 

 

 

 

1960 65 70 75 80 85 90

Year

Figure 7. Catch/effort of lake trout in spring gillnet assessment fisheries in western
(MN2) and eastern Minnesota (MN3), and eastern Wisconsin (W12) waters of Lake

Superior during 1959-93.

22

during the 1960s and Minnesota during the 1960s and 19703, but increased slowly
thereafter in each area (Figure 7; Appendix A, Tables 12-16). During the 19905, wild

fish outnumbered stocked lake trout in all areas except Minnesota and Whitefish Bay.

Sources of Recruitment

Stocked adult lake trout were significantly correlated to recruitment in all areas
except MI8, whereas wild adult lake trout were significantly correlated to recruitment
only in areas MI4-MI7 and MN3 (Figure 8). Stock-recruitment relationships were
significantly different among management areas for both stocked (F=4.l4; df=9, 196;
P<0.001) and wild (F=2.43; df=9, 196; P=0.012) parents. However, stock-recruitment
relationships were similar among Michigan areas MI4-MI7 for both wild (F=1.6l;
df=3, 95; P=0.19) and stocked (F=l.81; df=3, 95; P=0.15) parents, and average CPEs
were similar among areas (F=1.39; df=3, 101; P=0.25). Stock-recruitment
relationships were also similar among Minnesota areas MN2-MN3 for both wild
(F=1.39; df=1, 40; P=0.25) and stocked parents (F=0.13; df=l, 40; P=0.72), but
average CPEs were significantly different among areas (F=27.33; df=1, 42; P<0.001).

Variation in the CPEs of stocked and wild lake trout parents explained much of
the variation in CPE of recruits eight years later (from R’=0.54 in MI6 to R’=0.94 in
MIS) (Figure 9). Variation in the CPE of stocked parents explained the majority of
variation in the CPE of recruits (/=O.67-1.00), whereas wild parents explained little
variation (I’=O.02-0.19). In Michigan areas MI4 through MI7, the combined CPEs of
stocked and wild lake trout explained the majority of the variation in CPE of progeny

(R’=0.79), of which most was explained by stocked parents (r’=0.66) and little was

23

5‘
o

 

111414

f“
o

.N
o

11111111111

p—a
O

 

P
o
I
I

 

 

 

—I _—

"‘ "l— _h
_—

" __ —

Intercept

-2.0 ‘ , ,
1.0

 

 

 

 

I

_.

O

l
'fil

111 1

 

 

 

—1
—1
_4
—I
—
_
—

 

0.5

 

1 1 1 i 1 1 1 1

 

%—} _-—— .117“

 

 

 

I
l
h

 

 

_‘h—

-0.5

Regression Coefficient

Wild Parents

1 1 1 1 1 1 1 1

 

 

 

 

'1-0 l l I l l l l l

1.0 &-
ns} ‘1‘ {F _ % —} *-

0.0

 

L41 1

 

 

 

Stocked Parents ~L

 

 

 

'1-0 l l l I l l l l l

M13 M14 M15 MI6 MI7 MI8 MN2 MN3 WI2
Management Area
Figure 8. Multiple regression coefficients (i95% C1.) of wild and stocked lake trout

CPEs on recruit CPE in Michigan (MI3-MI8), Minnesota (MN2-MN3), and Wisconsin
(W12) waters of Lake Superior during 1959-93.

24

 

I Wild
I Stocked
I Multiple

MN2

Management Area
.5.

MN3

MI4-MI7

MN2-MN3

 

 

 

0 20 40 60 80 100

Coefficient of Determination (%)

Figure 9. Coefficients of determination of wild and stocked lake trout CPEs on recruit
CPE in Michigan (MB-MI8), Minnesota (MN2-MN3), and Wisconsin (W12) waters of
Lake Superior during 1959-93.

25
explained by wild parents (P=0.09) (Appendix A, Table 17; Appendix B, Figure 24).

In Minnesota areas MN2 and MN3, the CPEs of stocked and wild lake trout also
explained the majority of the variation in CPE of progeny (R’=0.63), of which most
was explained by stocked parents (/=O.62) and little by wild parents (r’=0.15)
(Appendix A, Table 18; Appendix B, Figure 25). In Wisconsin area W12, the CPEs of
stocked and wild lake trout explained little of the variation in CPE of progeny

(R’=0.18) (Appendix A, Table 19; Appendix B, Figure 26).

Discussion

These results suggest that stocked lake trout produced more wild progeny in
Lake Superior than wild lake trout, particularly in Minnesota and Michigan. The
relatively weak contribution of stocked fish to recruitment in Wisconsin (W12) may
reflect the offshore, heterogeneous distribution of spawning grounds among the
Apostle Islands (Coberly and Horrall 1980), compared to Michigan and Minnesota,
where spawning grounds are more inshore and homogeneous (Krueger et al. 1986).
Stocked lake trout were probably better able to locate inshore spawning grounds in
Michigan and Minnesota than offshore shoals in Wisconsin. For example, stocked
lake trout were rarely found on offshore spawning shoals in the Apostle Islands area of
Wisconsin during the spawning season, but large concentrations of stocked lake trout
could sometimes be found attempting to spawn on inshore, unsuitable habitat such as
sand beaches as little as 5 km away (Krueger et al. 1986). In contrast, stocked lake

trout spawned on virtually all spawning grounds in Michigan during the early 19705

26
(Peck 1979), and spawner densities were as high or higher than those on Gull Island

Shoal during the same spawning seasons (Swanson and Swedberg 1980). Different
distributions of spawning habitat may thus explain why stock-recruitment relationships
in Michigan and Minnesota were similar across broad areas. This further implies that
lake trout stocks in Michigan and Minnesota interbreed more freely than in Wisconsin,
and that knowledge of Wisconsin's population dynamics cannot be generalized to
Michigan's or Minnesota's populations.

In contrast to the contribution of stocked fish, wild fish contributed relatively
little to the production of wild progeny in Lake Superior. This was surprising, because
stock-recruitment relationships have been documented for at least two spawning stocks
in Wisconsin (Krueger et a1. 1986; Schram et al., in press). Wild lake trout accounted
for 88% of the wild spawners recruited one generation later at Sand Cut Reef in
Wisconsin during 1967-84, but stocked lake trout accounted for only 21% (Krueger et
a]. 1986). Data analyzed from Sand Cut Reef differed from ours in two respects.
First, their analysis was of autumn CPEs from one spawning reef, whereas I analyzed
spring CPEs from larger areas that likely represented multiple spawning stocks. The
Sand Cut Reef analysis detected a relationship between a single spawning stock and
recruitment, whereas my analysis failed to detect any stock-recruitment relationship,
perhaps because several spawning stocks occur in the area, each with its own
stock-recruitment relationship. Second, stocked fish were much more numerous than
wild fish in both analyses, but wild lake trout were present in all years at Sand Cut
Reef and absent in the early years of my analysis. The Sand Cut Reef data therefore

has less overall contrast in lake trout abundance than mine, and may not reflect the

27

colonizing success of stocked lake trout, in the absence of any wild lake trout.

In spite of widespread reproduction, however, stocked lake trout were unable to
replace themselves (regression coefficients for stocked parents were all less than one,
though both parents and recruits were indexed at the same sizes and ages; Ricker
1975). Recruitment must exceed parental stock sizes at some, usually low, parental
stock sizes in order for a population to persist (Ricker 1975). However, parental stock
sizes in Lake Superior were artificially bolstered through intensive stocking, which led
to spawner densities in Michigan that approached 2.5 times the historic average
(Chapter III). These artificially high spawner densities may have contributed to low
reproductive efficiency through competition for spawning habitat or cannibalism on the
progeny produced. Reproductive inefficiency of stocked lake trout in Lake Superior
was previously noted by Krueger et a1. (1986), who hypothesized that stocked
spawners may have been unable to locate suitable spawning habitat on reefs.

Wild lake trout were even less able to replace themselves than stocked lake
trout in inshore areas. Stock-recruitment regression coefficients for wild lake trout
were even less than for stocked lake trout in all areas. During the 19705, densities of
wild lake trout were much lower than stocked lake trout, so wild lake trout spawners
of the 19705 would likely have spawned with stocked lake trout spawners, so the
contribution of rare, wild lake trout to recruitment may have been masked by the
contribution of abundant, stocked lake trout. In contrast, the offshore Gull Island
Shoal spawning stock reproduced successfully even when no female spawners were
detected during spawning surveys (Swanson and Swedberg 1980). Inshore spawning

stocks during the late 1980s and early 1990s were dominated by wild spawners and

28

lower in density than during the 19708. Consequently, I expect that stock-recruitment
regression coefficients for wild lake trout spawning in Michigan during the 1980s will
exceed one, as Krueger et a1. (1986) found for the Sand Cut Reef spawning stock.
The rate of stock recovery in Michigan and Minnesota should therefore improve
during the late 1990s as wild lake trout spawners dominate recruitment.

My results suggest a robust biological relationship between numbers of parental
lake trout and numbers of recruits, because stock-recruitment relationships for both
stocked and wild parents were similar across much of Michigan and Minnesota in spite
of the different time-frames of the relationships. The lack of a stock-recruitment
relationship in Whitefish Bay (MIB) likely resulted from poor data (small samples,
scattered in time), and from extremely low abundance of wild lake trout in all years
analyzed. In western Michigan (MB) and central Minnesota (MN2), relationships
between wild parents and progeny were not significant, rather than between stocked

parents and progeny, which was different than elsewhere in Michigan and Minnesota.

Management Implications

My results suggest that the availability of inshore spawning substrate is a
critical determinant of successful reproduction by stocked lake trout in the Great
Lakes. Other investigators have concluded that stocked lake trout were reproductively
ineffective in Lake Superior (Krueger et a1. 1986). This conclusion was subsequently
held as the primary reason that lake trout restoration in the other Great Lakes had

largely failed (Eshenroder et a1. 1983). Unfortunately, previous stock-recruitment

29

analyses in Lake Superior were restricted to the Apostle Islands area in Wisconsin,
where spawning substrate is restricted to offshore shoals that demand homing ability
by spawning lake trout. In Minnesota, Michigan, and Ontario, however, spawning
substrate is widely distributed inshore, where little homing ability is required by
inexperienced, stocked lake trout spawners.

Lake trout restoration has been deferred in both northern Lake Huron and
northern Lake Michigan where inshore spawning grounds are found. Instead, these
areas have been reserved for maximum sustained harvest of lake Whitefish—mostly by
gillnets that impose high incidental mortality on lake trout (Rybicki and Keller 1978).
As a consequence, spawning stocks have not been permitted to develop in the areas
where inshore spawning grounds occur and the likelihood of successful reproduction is
greatest. Rather, lake trout restoration has been pursued mostly in the southern
portions of both Lakes Huron and Michigan, where inshore spawning grounds are
largely absent. Only since the mid-19805 has lake trout restoration been moved
offshore to the large offshore reefs of Six-Fathom Bank in Lake Huron, the Beaver
Islands in Lake Michigan, and the Mid-Lake Reefs in Lake Michigan. Stocking in
these areas, in conjunction with protection from fishery exploitation, should provide
for successful stock restoration, provided that the fish remain in these areas to spawn.

Stock restoration in northern Lake Huron and northern Lake Michigan can also
succeed, provided that spawning grounds are still in suitable condition and that lake
trout are afforded protection from fishery exploitation and sea lamprey predation.
Surveys of historic lake trout spawning grounds in northern Lake Huron and Lake

Michigan have shown that substrate quality has not been observably degraded (Edsall

30

et a1. 1992). Excessive fishery exploitation and sea lamprey predation may therefore
explain the lack of successful reproduction by stocked lake trout in these areas. The
success of future attempts to restore lake trout stocks in these areas may depend on the
extent to which fishery managers are able to control total annual mortality resulting
from fishery exploitation and sea lamprey predation. Stocking of hatchery-reared fish
continues to be a viable tool for lake trout restoration in both lakes, provided that

these controls on mortality are effective.

CHAPTER H:
DECLINING SURVIVAL OF LAKE TROUT

STOCKED IN U.S. WATERS OF LAKE SUPERIOR

Abstract—The survival of the 1963-82 year classes of stocked yearling lake
trout (Salvelinus namaycush) declined significantly in Lake Superior. To investigate
causes of these declines, a Ricker model of stock-recruitment was used to describe the
catch per effort (CPE) of age-7 stocked lake trout in Minnesota, Michigan, and
Wisconsin waters of Lake Superior as functions of the numbers of yearlings stocked
six years earlier, the CPE of wild adult lake trout (an index of predation), and
large-mesh gill-net fishing effort (an index of fishing mortality). Declining CPEs of
stocked lake trout in Michigan and Wisconsin were significantly associated with
increasing large-mesh gill-net fishing effort. Declining CPEs of stocked lake trout in
Minnesota were significantly associated with increasing densities of wild lake trout.
Sea lamprey abundance varied during the period, so predation by sea lampreys did not
explain declining survival in any state. I conclude that stocked lake trout survival
declined in Michigan and Wisconsin because of increased mortality in large-mesh gill
fisheries, and can be enhanced by better controlling these fisheries. I conclude that
survival of stocked lake trout declined in Minnesota because of increased predation by
wild lake trout that recently recolonized the area Predation by wild lake trout may
also inhibit survival of stocked lake trout in Michigan and Wisconsin, but this effect

appeared to be less important than large-mesh gill-net fishing mortality.

31

32

Introduction

Lake trout (Salvelinus namaycush) sustained 2 million kg of average annual
yield to commercial fisheries during 1913-1950 in Lake Superior (Baldwin et al.
1979), but were nearly extirpated during the 19503 by fisheries and sea lampreys
(Petmmyzon marinas) (Hile et al. 1951; Pycha and King 1975; Jensen 1978; Coble et
al. 1990), which had colonized the lake during the 19403 and 19503 (Smith et al.
1974). Chemical control of the sea lamprey was begun in 1958 and reduced their
abundance 85% by 1962 (Smith et al. 1974), at which time management authorities
closed commercial lake trout fisheries (Pycha and King 1975). Hatchery-reared,
yearling lake trout have been stocked in United States waters since 1952 (Lawrie and
Rahrer 1972, 1973), and totalled 60 million by 1983 (Hansen et al. 1994b).

Stocked yearling lake trout survival was stable from 1959 through 1961, but
declined after 1961, possibly because of predation by increasing numbers of older lake
trout (Dryer and King 1968). The abundance of stocked lake trout increased rapidly
during 1959-66, because of large plantings, and remained high from 1967 through
1970 even though few fish survived past age 9 because of high sea lamprey-induced
mortality (Pycha and King 1975). The abundance of stocked lake trout in Michigan
was high during the 1970s, but declined during the 19803 for unknown reasons
(MacCallum and Selgeby 1987; Peck and Schorfhaar 1991). It has remained low since
1988 (Hansen et al. 1994b). Stocked lake trout abundance declined for unknown
reasons in Wisconsin during the 19703 and 19803 and in Minnesota during the 19903

(Hansen et al. 1994b).

33

This declining survival may have been caused by increased competition for
food with other salmonid species, wild lake trout, and previously stocked lake trout,
and by predation by other species such as sea lamprey and large salmonids (Hansen et
al. 1994a). In Michigan, the abundance of stocked fish declined coincident with
reductions in stocking and growth rates, which slowed recruitment into sizes that were
vulnerable to the assessment nets (MacCallum and Selgeby 1987). Declining
abundance of stocked lake trout in most areas of Michigan also coincided with
increased tribal commercial fishing effort (Peck and Schorfhaar 1991). In Wisconsin,
the abundance of stocked lake trout coincided with reduced stocking and increased
fishing mortality (MacCallum and Selgeby 1987).

Some of these causes of declining abundance of stocked lake trout are not
easily controlled by fishery management actions. For instance, growth rates are
limited by competition within and among species. In contrast, fishing mortality can be
controlled by constraining the numbers and sizes of lake trout caught by sport and
commercial fisheries. Sea lamprey-induced mortality can also be controlled by
reducing the number of sea lampreys. Survival of stocked lake trout can be enhanced
by increasing their average weight prior to stocking. It is important to determine what
forces are currently influencing the survival of stocked lake trout in Lake Superior, to
determine if that survival can be improved. Herein, abundance indices of stocked lake
trout in United States waters of Lake Superior are modeled as functions of the number

lake trout stocked, and of indices of competition, predation, and fishing.

34
Methods

Based on other investigations of lake trout survival, I developed a priori
hypotheses to explain declining survival of lake trout stocked in U.S. waters of Lake
Superior (Tyler and Crawford 1991). Declining survival of stocked lake trout in U.S.
waters of Lake Superior was previously described as a function of reduced stocking
and an unexplained year effect (Hansen et al. 1994a). Factors that were most likely to
account for the year effect included those that affected survival of stocked yearlings
during their first year in the lake, such as competition for prey between stocked and
wild yearlings, size of yearlings at the time of their release, and predation by adult
lake trout (either wild or stocked or both) on stocked yearlings during the first year
after their release (Hansen et al. 1994a). Fishing mortality was not a likely cause of
declining survival because survival was reduced before stocked lake trout were fully
recruited into sport and commercial fisheries (Hansen et al. 1994a). Sea lamprey
abundance varied without trend after 1961 (Klar and Weise 1994), so predation by sea

lampreys did not explain declining survival.

Recruitment to Age 7

Recruitment was indexed as the relative abundance of age-7 lake trout stocked
as yearlings (Hansen et al. 1994a). Catch per effort (CPE) during 1970-89 was
defined asithe number of fish caught per km of standard gill net in U.S. lake trout
management areas (Figure 4). Ages were determined from a sample of scales removed

from fish caught in the gill nets. Sample ages were validated by matching the fin clip

35

observed on each fish to the year class on which that fin clip was used (all lake trout
were marked by removal of a fin before stocking). Ages from the sample were
expanded to the entire catch using an age-length key. Age—7 fish were used to index
recruitment because catch curves revealed that age-7 fish were the first age'class fully
recruited to the gillnets (Pycha 1980), and fish were subject to higher rates of fishing-
and sea lamprey-induced mortality after age 7 (Pycha and King 1975).

Previous analyses of recruitment to age 7 indicated that movement of fish
among management areas precluded analysis of survival within management areas
(Hansen et al. 1994a), but patterns of abundance were relatively homogenous within
each state (Chapter 1). Consequently, average recruitment was computed for Michigan,
Minnesota, and Wisconsin waters for each of the 1963-86 year classes as the sum of
the area-specific CPEs, weighted by the size of each area (Ricker 1975) (Tables 2-4).
The 73-m contour was used for the size of each area because it approximates the
maximum depth limit of lean lake trout in Lake Superior (Dryer 1966). Recruitment
data was only available throughout 1970-93 for Minnesota areas W2-W3),

Wisconsin area W12, and Michigan areas MI3-MI7.

Factors Potentially Influencing Recruitment

Stock size was indexed as millions of yearlings of the 1963-86 year-classes
stocked during 1964-87 (Tables 2-4). All lake trout included in the recruitment index
were known to have come from plantings of fingerlings 7 years earlier and yearlings 6
years earlier, but yearlings were used as the index of stock size because they were

previously shown to survive 4-10 times better than fingerlings in Lake Superior

36

Table 2. Stock and recruitment data used for modeling survival of yearling lake trout
stocked in Michigan waters of Lake Superior.

 

Age-7 Million Adult Lake Trout Juvenile Gill Net
Year Recruit Yearlings Grams/ Wild Stocked Wild Lake Effort

Class CPE Stocked Yearling CPE CPE Trout CPE (1000 km)

 

1963 49.0 1.2 14.7 1.9 3.9 0.6 0.0
1964 32.3 0.7 16.6 1.2 7.3 1.4 0.0
1965 54.3 2.2 21.6 0.7 10.6 4.2 0.0
1966 34.7 2.1 19.2 0.3 23.2 4.6 0.0
1967 44.6 2.2 23.3 0.4 35.3 3.9 0.0
1968 46.5 1.9 24.2 0.3 38.7 6.4 0.0
1969 76.4 1.9 23.0 0.6 68.4 8.1 0.0
1970 25.9 1.1 21.4 1.4 59.8 11.0 0.0
1971 30.4 1.1 21.7 4.2 81.7 12.7 0.0
1972 39.0 0.9 21.0 4.6 83.1 21.0 0.0
1973 21.5 0.9 23.9 3.9 55.6 26.2 0.0
1974 35.3 0.8 22.4 6.4 75.4 36.4 0.1
1975 15.1 0.8 18.7 8.1 70.9 22.6 0.5
1976 16.4 0.7 21.7 11.0 88.2 19.1 0.8
1977 20.0 0.7 19.1 12.7 59.8 24.2 1.3
1978 11.6 0.8 16.5 21.0 65.1 31.1 1.9
1979 5.6 0.6 16.4 26.2 56.5 37.6 2.7
1980 2.0 0.6 18.4 36.4 57.4 29.3 3.7
1981 2.6 0.7 20.5 22.6 42.9 34.4 4.4
1982 1.5 0.8 19.6 19.1 27.7 30.3 5.4

 

Buettner 1961; Pycha and King 1967). Numbers of lake trout stocked previously
appeared to be unrelated to recruitment at age 7 (Hansen et al. 1994a), but lack of
contrast in numbers planted likely explained the lack of a significant relationship.
Size of stocked yearlings at release was indexed as the average weight, in
grams, at release of each year class (total weight/total number stocked) (Tables 2-4).
Size of yearlings at release was included because the percentage of lake trout returned

from plantings in Lake Superior was more closely associated to the weight of the fish

37

Table 3. Stock and recruitment data used for modeling survival of yearling lake trout
stocked in Minnesota waters of Lake Superior.

 

 

Age-7 Million Adult Lake Trout Juvenile Gill Net
Year Recruit Yearlings Grams/ Wild Stocked Wild Lake Effort
Class CPE Stocked Yearling CPE CPE Trout CPE (1000 km)
1963 2.4 0.2 6.4 11.9 1.8 0.4 0.0
1964 1.6 0.1 14.5 3.7 0.6 0.4 0.0
1965 2.0 0.1 20.4 1.7 1.4 0.9 0.0
1966 6.4 0.2 9.6 1.9 3.1 1.1 0.0
1967 10.1 0.2 19.3 0.5 3.2 1.4 0.0
1968 8.8 0.2 20.6 0.1 2.6 1.2 0.0
1969 10.5 0.2 20.5 0.4 10.4 1.1 0.0
1970 10.3 0.3 21.2 0.4 4.1 0.7 0.0
1971 14.8 0.3 22.4 0.9 5.8 0.8 0.0
1972 18.4 0.3 20.8 1.1 13.3 0.6 0.0
1973 9.4 0.3 22.9 1.4 16.1 0.9 0.0
1974 12.3 0.3 21.7 1.2 18.8 2.3 0.0
1975 23.1 0.3 22.0 1.1 16.5 2.3 0.0
1976 25.8 0.4 23.7 0.7 16.9 3.4 0.0
1977 27.0 0.4 15.5 0.8 19.8 1.9 0.0
1978 29.7 0.3 15.7 0.6 20.3 2.9 0.0
1979 22.1 0.4 17.8 0.9 26.6 7.4 0.0
1980 26.1 0.3 24.3 2.3 37.6 4.7 0.0
1981 9.3 0.3 23.5 2.3 43.0 7.7 0.0
1982 15.1 0.4 18.4 3.4 31.3 l6.1 0.0

 

at release than to differences in stocking locations, seasons, or years, eg sources,
rearing stations, or rearing diets (Pycha and King 1967).

Wild and stocked adult lake trout were indexed as predators as average CPEs
of all sizes and ages caught in the adult lake trout assessment fishery (Chapter I)
during years when stocked yearlings were released (1964-83) (Tables 2-4). Wild and
stocked adult lake trout were included as independent potential predators because wild

adults are distributed deeper than stocked adults (Krueger et al. 1986), and therefore

38

Table 4. Stock and recruitment data used for modeling survival of yearling lake trout
stocked in Wisconsin waters of Lake Superior.

 

 

Age-7 Million Adult Lake Trout Juvenile Gill Net
Year Recruit Yearlings Grams/ Wild Stocked Wild Lake Effort
Class CPE Stocked Yearling CPE CPE Trout CPE (1000 km)
1963 49.9 0.7 20.3 7.1 9.5 4.5 0.9
1964 30.3 0.4 22.3 5.2 10.8 9.3 1.0
1965 16.2 0.3 18.4 5.1 21.5 8.5 1.6
1966 12.3 0.2 20.5 2.2 24.1 10.7 2.7
1967 17.4 0.3 20.7 0.4 15.0 9.0 4.5
1968 8.7 0.3 22.9 1.1 35.8 6.9 6.4
1969 11.0 0.2 26.0 4.5 59.4 9.1 8.5
1970 7.2 0.2 22.6 9.3 53.6 6.6 10.7
1971 10.3 0.3 22.0 8.5 32.7 8.0 12.4
1972 7.1 0.2 20.8 10.7 24.5 12.5 13.6
1973 3.8 0.4 20.7 9.0 28.4 11.3 14.0
1974 4.0 0.5 23.8 6.9 15.9 6.6 14.1
1975 8.7 0.6 28.7 9.1 16.7 3.5 14.4
1976 2.5 0.4 24.4 6.6 17.4 3.4 14.5
1977 14.7 0.4 22.8 8.0 19.6 3.9 14.5
1978 3.1 0.3 19.4 12.5 20.7 5.3 14.1
1979 5.8 0.3 22.3 11.3 15.2 12.2 13.6
1980 3.3 0.2 29.0 6.6 14.1 10.9 12.9
1981 2.6 0.2 23.9 3.5 7.6 9.7 12.9
1982 3.9 0.3 27.9 3.4 7.7 13.5 12.9

 

may have overlapped in space with stocked yearling lake trout differently than stocked
adult lake trout. Wild lake trout were indexed as competitors in the same years when
the stocked fish were indexed at age 7 (1970-89) because their interaction with stocked
yearlings may not have been limited to the first year.

Fishing mortality was indexed as the average of the annual large-mesh gill-net
effort (millions of meters) that was fished during the six years between stocking at age

1 and capture in the assessment nets at age 7 (Tables 2-4). Fishing effort was

39

included because gill nets impose mortality on small (young) fish, even though the
focus of their selectivity is on a relatively narrow range of sizes (ages) of fish (Hamley
1975). Gill nets impose increasingly selective mortality on lake trout as they grow in
size (age) to fully recruited sizes (ages), so the gauntlet of gill-net fishing effort faced
by each year class was treated as a moving average of the annual effort during years
between stocking and recruitment. Fishing mortality was indexed using large-mesh
gill-net effort because small-mesh gill nets were restricted to offshore chub fisheries
and inshore floated lake herring (Coregonus artedi) fisheries that impose little mortality
on lake trout. Trap nets are fished inshore but impose little mortality on lake trout

(Schorfliaar and Peck 1993).

Statistical A naIysis
A form of the Ricker (1975) stock-recruitment model was used to model the
effects of size at release, predation, competition, and fishing on the survival of lake

trout in Lake Superior (Walters et al. 1986; Hilbom and Walters 1992):

R = Se”s"x (1)

The model describes recruitment (R) as a function of the parental stock (S), which is
reduced by background, density-independent mortality (a), density-dependent mortality
due to intraspecific competition or cannibalism (bS), and interspecific competition or
predation (cX). Estimates of the model coefficients (a, b and c) can be found using

multiple regression methods on the linear form of the model:

4o
log(R/S) = a - bS - cX + e (2)

In the linear form of the model, the recruitment rate, log(R/S), is a decreasing function
of the parental stock size and interactions with other species or fisheries. Random or
unexplained influences on survival are described in the model as residual error (a).

Potential explanatory variables were regressed sequentially on the logarithms of
the recruitment rate, R/S (CPE at age 7 per million stocked), starting with the number
of yearlings stocked (S), computing partial correlations with remaining X -variables,
and adding the variable with the highest partial correlation and most biologically
sensible coefficient to the model (Henderson and Velleman 1981). Model building
was terminated when remaining variables accounted for little remaining residual error
or had biologically meaningless partial correlations. Model fit was measured using the
adjusted R2 because the number of years sampled was small (N=20). Models were
diagnosed for collinearity among predictor variables, and residual errors were
diagnosed for normality, homogeneity of variance, independence, and linearity (Draper
and Smith 1981; Systat, Inc. 1992; Kirby 1993).

Model performance was judged by predicting recruitment of the 1983-86
year-classes of stocked fish in each jurisdiction, and comparing the predicted values
with observed values (Draper and Smith 1981). Models were therefore fitted to
observed values of the 1963-82 year classes, and validated by comparing predicted to
observed values of the 1983-86 year-classes. Recruitment rates of the 1986 year class
were the lowest observed in all three states, so they provided good tests of model

performance (Tyler 1992). Standardized residuals were computed ([observed value -

41

predicted value]/model SE) (Draper and Smith 1981) to aid in judging if predicted

values of the 1983-86 year-classes were unusually large, compared to those of the

1963-82 year-classes. Predictions were considered satisfactory if their standardized

residuals fell within the 95% t-interval for a sample of 20 observations.

Michigan

Results

Recruitment rates of stocked fish in Michigan were weakly associated with

numbers released (r=0.26; N=20; P=0.26), but were strongly associated with

large-mesh gill-net fishing effort (r=-0.93; N=20; P3001) (Figure 10). Stocking and

100 ,
B
.M
o
8
m
C
.9
g 10;
E
o
M

1

 

 

 

 

 

 

O
O . .
o. o 6 o o ‘
O
o
O
O O
o . . o
O O
0.0 0.5 1.0 1.5 2.0 2.5 0 1 2 3 4 5
Millions Stocked Millions of Meters

Figure 10. Recruitment rate of the 1963-82 year-classes of stocked yearling lake trout
to age 7 compared to the number stocked (left panel) and large-mesh gill-net fishing
effort (right panel) in Michigan waters of Lake Superior.

42

large-mesh gill-net fishing effort each accounted for significant variation in the
recruitment of the 1963-82 year classes of stocked lake trout in Michigan (Figure 11;
Table 5; Appendix A, Table 20). The coefficient for number stocked was negative,
which suggests density dependent competition or cannibalism between the number
released and their recruitment to age 7. However, large-mesh gill-net fishing exerted
greater influence on recruitment than numbers stocked (the standardized coefficient

was larger than for number stocked).

 

80

Catch/Efl‘ort

 

 

 

 

 

Year Class

Figure 11. Catch per effort of age-7 stocked lake trout caught in assessment fisheries
(dots) and predicted from yearling stocking and large-mesh gill-net fishing effort (line)
in Michigan waters of Lake Superior.

Predicted and observed recruitment for the 1983-86 year classes were quite low

(Figure 11), but predicted values for the latter three year classes were substantially

43

Table 5. Results of the multiple regression of stocked lake trout CPE at age 7 per
million yearlings stocked (logarithms) on yearlings stocked six years earlier, gill net
effort (millions of meters in Michigan and Wisconsin), and wild lake trout CPE
(Minnesota) in Lake Superior.

 

Regression Parameter Standardized
Parameter Coefficient SE Coefficient t P

 

Michigan (N=20; R’=0.90; P<0.01)

Intercept 3.90 0.18 0.00 21.37 0.00
Yearlings Stocked -0.36 0.13 -0.23 -2.75 0.01
Gill Net Effort -0.56 0.04 -1.04 -12.66 0.00

Minnesota (N=20; R’=0.67; P<0.01)

Intercept 2.79 0.29 0.00 9.54 0.00
Yearlings Stocked 4.09 0.95 0.60 4.29 0.00
Wild Lake Trout CPE -0.09 0.03 -0.43 -3.12 0.01

Wisconsin (N=20; R’=0.64; P<0.01)

Intercept 4.82 0.357 0.00 13.08 0.00
Yearlings Stocked -0.94 0.74 -0.17 -1.26 0.23
Gill Net Effort -0.13 0.02 -0.82 -5.92 0.00

 

lower than observed values (Figure l2). This indicates that the model fitted to data
for the 1963-82 year classes did not accurately describe data for the 1983-86 year
classes. Recruitment rates for the 1983-86 year classes were among the lowest
observed during 1963-86, and may have been lower than background rates of
immigration into Michigan from neighboring jurisdictions. Immigration was not
modeled explicitly, but could be as high as 10% of the numbers stocked into adjacent
jurisdictions, based on tagging and marking studies (Eschmeyer et al. 1953; Buettner

1961; Pycha et al. 1965; Rahrer 1968; Swanson 1973; Ebener 1990; Peck and

44

Schorfhaar 1991). Such immigration could easily account for differences in predicted

and observed recruitment in Michigan.

Standardized Residual

 

5.0 ,
4.5 a
4.0 -§
3.5 -§
3.0 —§
2.5 -:
2.0 -§
1.5 4,
1.0 —§

 

l “ _
-1 .. n x -.
- t
0 5 — i i i "r
:7 ‘ c .
O " :r. .l ‘3 r
-1 ‘ ‘L l

"‘ “if" ‘2 :r

O O '1 Ilka iii-F
q
. d
.

-o.5 —§
-1.0 -§
-1.5 —§
-2.o —§

 

 

 

 

-2.5 :

l'l‘lTlTl'l’lTTTl'TTl'lT

63 65 67 69 71 73 75 77 79 81 83 85

Year Class

Figure 12. Standardized residuals for stocked lake trout recruitment, predicted from
yearling stocking and large-mesh gill-net fishing effort in Michigan waters of Lake
Superior (:1: the 95% t-interval).

Predation by wild lake trout may have been an important force of mortality on

stocked lake trout in Michigan, but its effect was less significant than that of fishing

mortality. Large-mesh gill-net fishing effort (r=-0.93; N=20; P5001) and CPE of wild

lake trout (r=-0.77; N=20; P5001) were each correlated with recruitment, but wild

lake trout CPE did not explain significant variation in the recruitment rate after

accounting for gill-net effort. Other potential predictor variables for average size of

stocked yearlings, competition with wild lake trout, and cannibalism by previously

45

stocked lake trout were not well correlated with recruitment of stocked lake trout

(Appendix B, Figure 27).

Minnesota

Recruitment rates of stocked lake trout in Minnesota were strongly related to
the number of yearlings previously released (F073; N=20; P3001) and the density
(CPE) of adult wild lake trout in the year of stocking (r=-0.62; N=20; P3001) (Figure
13). The relative abundance of age-7 stocked lake trout in Minnesota was thus
predicted from the number of yearlings released and the density of wild adult lake

trout in the year when yearlings were released (Figure 14; Table 5; Appendix A, Table

 

 

 

 

 

 

 

100, I
. o
. «I 'o
o o
O
8 '0 o. o 9 co 0
.M . O 0.0
§
([3 .0 O .
.8 o o
5%" 10: :
'53
O
0
M
l vvvvrfrrrrxrrrrrvrfi 'v'r”'I'WVIT'*IY'TI"'I"'
0 0.1 0.2 0.3 0.40 2 4 6 8 10 12 14
Millions Stocked Catch/Effort

Figure 13. Recruitment rate of the 1963-82 year-classes of stocked yearling lake trout
to age 7 compared to the number stocked (left panel) and wild lake trout density (right
panel) in Minnesota waters of Lake Superior.

46

21). The coefficient for number stocked was positive, which suggests depensatory
predation between the number released and their recruitment to age 7. Predation on
stocked lake trout by wild adult lake trout was relatively similar to the effect of
numbers of yearling lake trout released (standardized coefficients were of similar

magnitude).

 

35
‘ Modeled Predicted

Catch/Efl‘ort

 

 

 

 

l ‘ l 1

63 65 67 69 71 73 75 77 79 81 83 85

 

Year Class

Figure 14. Catch per effort of age-7 stocked lake trout caught in assessment fisheries
(dots) and predicted from yearling stocking and wild lake trout density (line) in
Minnesota waters of Lake Superior.

Predicted and observed recruitment for the 1983-86 year classes were highly
variable (Figure 14), but predicted values were remarkably similar to observed values
(Figure 15). The recruitment rate for the 1986 year class was the lowest observed in

Minnesota during 1963-86, yet was accurately predicted by stocking and predation,

47

 

2.5 -

 

2.0 —:
1.5 i

1.0-f

 

 

-1.0

Standardized Residual

-1.5
-2.0

 

 

 

Year Class

Figure 15. Standardized residuals for stocked lake trout recruitment, predicted from
yearling stocking and large-mesh gill-net fishing effort in Minnesota waters of Lake
Superior (:1: the 95% t-interval).

which indicates that the model fitted to data for the 1963-82 year classes accurately
described data for the 1983-86 year classes.

Other potential predictor variables were not well correlated with recruitment of
stocked lake trout (Appendix B, Figure 28). Therefore, the average size of stocked
yearlings and wild juvenile lake trout competition were not implicated as causes of
declining lake trout survival in Minnesota Large-mesh gill-nets were not allowed in

Minnesota, so fishing mortality was not related to survival of stocked lake trout.

48

Wisconsin

The recruitment of stocked lake trout to age 7 in Wisconsin was more strongly
related to large-mesh gill-net fishing effort (r=-0.80; N=20; P3001) than to the number
of yearlings previously released (r=-0.087; N=20; P==0.7l) (Figure 16). Recruitment
rates of the 1963-82 year-classes of stocked lake trout in Wisconsin were thus
predicted from the number of yearlings released and the average amount of large-mesh
gill-net fishing effort in intervening years (Figure 17; Table 5; Appendix A, Table 22).
The coefficient for number stocked was not significant, which suggests density
independence between the number stocked and their recruitment to age 7. However,

mortality on stocked lake trout caused by large-mesh gill nets largely explained trends

 

 

 

 

 

 

100, .
1 C j 0
. o o ‘ o .
co « o o
'9 o o .
3 . o. 9 ‘ o.
O
8 .
(I) g 0
5 ° 0 o O o
E 1 . o 1 ‘
J o < o
.e ‘
8
M 1
1 ' ' ' r ' ' v I Y r r r v r 1 YVTr"'I"'I"'r"'I"'1"'I"
0 0.2 0.4 0.6 0.80 2 4 6 8 10 12 14 16
Millions Stocked Millions of Meters

Figure 16. Recruitment rate of the 1963-82 year-classes of stocked yearling lake trout
to age 7 compared to the number stocked (left panel) and large-mesh gill-net fishing
effort (right panel) in Wisconsin waters of Lake Superior.

49

in stocked lake trout CPE in Wisconsin (large standardized coefficient), and may have

masked the effect of numbers of lake trout stocked (small standardized coefficient).

 

60‘

Predicted

Catch/Effort

 

 

 

 

 

63 65 67 69 71 73 75 77 79 81 83 85

Year Class

Figure 17. Catch per effort of age-7 stocked lake trout caught in assessment fisheries
(dots) and predicted from yearling stocking and large-mesh gill-net fishing effort (line)
in Wisconsin waters of Lake Superior. .

Predicted and observed recruitment for the 1983-86 year classes were low, as in
Michigan (Figure 17), but predicted values were remarkably similar to observed
values, as in Minnesota (Figure 18). The recruitment rate for the 1986 year class was
the lowest observed in Wisconsin during 1963-86, as it was in Minnesota, and yet was
accurately predicted by stocking and predation. This indicates that the model fitted to
data for the 1963-82 year classes accurately described data for the 1983-86 year

classes in Wisconsin.

50

 

2.5

 

2.0

 

 

   

Standardized Residual

 

 

 

-1.0
-l.5
-2.0
’z-s’lllTllllTTlllfrlTTllTWT
1965 70 75 80 85
YearClass

Figure 18. Standardized residuals for stocked lake trout recruitment, predicted from
yearling stocking and large-mesh gill-net fishing effort in Wisconsin waters of Lake
Superior (:1: the 95% t-interval).

Other potential predictor variables were not well correlated with recruitment of
stocked lake trout (Appendix B, Figure 29). Consequently, the average size of stocked
yearlings and wild juvenile lake trout competition were not implicated as causes of

declining lake trout survival in Wisconsin.

Discussion

The present analysis suggests that survival of lake trout stocked in Michigan

and Wisconsin declined primarily because of large-mesh gill-net fishing mortality, and

51

in Minnesota because wild lake trout preyed on newly stocked lake trout. Survival of
lake trout stocked in Michigan may also have been reduced by wild lake trout
predation, but the effect was masked by fishing mortality, and therefore survival may
have been no better even if gill-net fishing effort had been less. The present analysis
did not implicate other factors that were previously suggested as significant sources of
mortality on stocked lake trout, such as competition with wild yearling lake trout
(Purych 1977; MacLean et al. 1981; Powell et al. 1986), predation by previously
stocked lake trout (Elrod et al. 1993), or size of lake trout at the time of their release

(Pycha and King 1967; Plosila 1977; Elrod et al. 1988; Gunn et al. 1987).

Fishing Mortality

The importance of fishing mortality on lake trout survival was surprising
because declining survival had previously been shown to occur prior to ages 2-4,
before the fish were recruited into the sport or commercial fisheries (Hansen et al.
1994a). However, survival declined most after the 1982 year class (Hansen et al.
1994a), and may have been undetectable by the present analysis. Fishing mortality
was the primary cause of declining lake trout abundance in Lake Superior prior to the
colonization of the lake by sea lampreys (Hile et al. 1951; Pycha and King 1975;
Jensen 1978; Coble et al. 1990), and large-mesh (2114 mm stretch-measure) gill nets
were the primary gear used to catch lake trout until 1962 when fisheries were closed
(Pycha and King 1975). Large-mesh gill nets were not allowed in Minnesota after the
fishery was closed in 1962, or in Michigan until tribal fisheries for lake Whitefish

(Coregonus clupeqfonnis) reopened during the late 19703 and 19803 (Peck and

52

Schorfhaar 1991). In contrast, large-mesh gill nets were allowed for harvesting lake
Whitefish in Wisconsin after 1970 (Hansen et al. in press).

Gill-net fishing effort significantly limited survival of lake trout stocked in
Michigan and Wisconsin after 1963 even though these fisheries were targeted on lake
Whitefish. Gill nets are extremely selective for fish of certain sizes (Hamley 1975); for
example, 114-mm stretch-mesh gill nets are highly selective for fish 457-610 mm TL
(age 7-9; Pycha 1980). However, larger and smaller fish are also entangled, and many
of these die, along with those that are gilled. The use of gill nets can therefore lead to
high incidental mortality on a species, such as lake trout, even when the fishery
intends to harvest another species, such as lake Whitefish. Large-mesh gill-net effort
increased earlier and was greater in Wisconsin; effort rose from an average of less
than l-million m during 1963-69 to more than 14-million m during 1973-86.
Large-mesh gill-net effort increased later and was lower in Michigan; effort rose from
nil during 1963-73 to an average of nearly 5-million m during 1982-89. Stocked lake
trout apparently suffered high incidental mortality in large-mesh gill nets in both
states, but were spared in Minnesota, where large-mesh gill nets were not used.

In Wisconsin, the effect of high incidental mortality on lake trout in large-mesh
gill nets was evident in a truncated age distribution (Figure 19). In offshore waters,
lake trout of the 1963-82 year-classes were protected by refuges (Swanson and
Swedberg 1980), but inshore, were subjected to high incidental mortality in large-mesh
gill-net fisheries targeted on lake whitefish. As a consequence, age distributions of
inshore lake trout in spring 1990, the first year after those included in this analysis,

were truncated compared to offshore lake trout. The age distribution of inshore lake

53

 

Age Class

 

Non-Refuge Refuge

i—nu—I—ou—nr—v—I—Ir—t—
WQGthN—O‘OWQQM-h

 

 

 

TTTITI'TTI'TJJTTTUIUUUIIUIIIUIIIUFIIIIIJUUIUITITIIITTTITTTITTIITYTIUII'IIT

2018161412108 6420 2 4 6 810121416
Catch/Effort

Figure 19. Catch/effort of lake trout in spring gill net assessment fisheries (number
per km of net) inshore (non-refuge) and offshore (refuge) in eastern Wisconsin waters
of Lake Superior in 1990.

trout shows that their growth rates had increased, such that the first fully recruited age
had dropped by two years compared to offshore lake trout. I In addition, ages beyond
the first fully recruited age were truncated below age 10, whereas ages of offshore lake
trout declined gradually from age 8 through age 18.

In Michigan, incidental mortality on lake trout in large-mesh gill nets was
inversely related to recruitment, even though no refuges were present in which to
contrast the resulting age structure (Figure 20). Large-mesh gill-net effort varied
inversely to lake trout recruitment among management areas within Michigan waters.
Recruitment rates in Michigan were lower for all year-classes after 1974 than for any

previous year-classes, coincident with the onset of large-mesh gill-net fishing. As a

54

 

Management Area

 

 

 

 

4 3 2 l 0 1 2 3
Recruitment Millions of Meters

Figure 20. Recruitment of the 1982 lake trout year-class at age 7 (CPE/million
yearlings stocked), compared to the average annual large-mesh gill-net fishing effort
during 1983-88, in inshore Michigan areas of Lake Superior.

result, the recruitment of lake trout in Michigan was higher for year classes that were
not subjected to large-mesh gill~net fishing and in areas where that fishing effort was
lowest (MIZ, MIS) than for year classes that were subjected to fishing effort and in

areas where that fishing effort was higher (MI3, MI4, MI6, MI7).

Predation by Wild Lake Trout

Predation on newly stocked lake trout by mature wild lake trout may have been
an important source of mortality in Minnesota, and possibly also in Michigan, waters
of Lake Superior. Lake trout stocking success was inversely associated to abundances
of older stocked lake trout during 1959-66, which suggested that cannibalism on newly

stocked lake trout by previously stocked fish was important at that time (Dryer and

55

King 1968). Survival of lake trout stocked in Lake Ontario was also negatively
associated with the density of large previously-stocked lake trout (2550 mm total
length), which suggested that cannibalism was a significant source of mortality on
newly stocked lake trout (Elrod et a1. 1993). Predation by native lake trout may have
limited survival of stocked lake trout in inland lakes as well (Purych 1977; Martin and
Olver 1980; MacLean et a1. 1981; Powell et al. 1986; Evans and Willox 1991).

In spite of these suggestions that cannibalism was an important source of
mortality, stocked juvenile lake trout have rarely been found in stomachs of wild adult
lake trout (Powell et a1. 1986; Elrod et a1. 1993). In Lake Superior, stocked juvenile
lake trout have rarely been encountered in surveys of lake trout feeding (Dryer et al.
1965; Conner et al. 1993; Gallinat 1993), which suggests that predation may not be an
important source of mortality on newly stocked lake trout. A similar model of
predation by Pacific cod (Gadus macrocephalus) on herring (Clupea harengus pallasi)
in the Hecate Strait, British Columbia, also suggested predation rates that were much
higher than stomach contents had indicated (Walters et al. 1986). Walters et al. (1986)
noted that their estimated rate of predation on herring by Pacific cod may have been
too high if predation was spread across several years, if herring abundance had been
overestimated, or if Pacific cod abundance had been underestimated.

Low occurrence of stocked yearling lake trout in wild adult lake trout stomachs
may not adequately indicate the importance of wild lake trout predation to the overall
survival of stocked lake trout. Stocked adult lake trout are distributed nearer to shore
than wild adult lake trout (Krueger et al. 1986), and yearling lake trout move offshore,

away from stocking sites, soon after their release (Pycha et al. 1965). Consequently,

56

stocked yearling lake trout may only be vulnerable to cannibalism by stocked adult
lake trout for a short time after stocking. Provided that stocked yearlings escape
cannibalism near the stocking site, they would become more vulnerable to predation
by wild adult lake trout during the remainder of their first year in the wild. However,
stomach samples of Lake Superior lake trout have generally been obtained from sport
fisheries that primarily operate inshore, or from spring gill-net assessments that
coincide with yearling stocking (Conner et al. 1993; Gallinat 1993). Such sampling is

unlikely to reflect of feeding on newly stocked lake trout by wild adult lake trout.

Management Implications

It appears that survival of lake trout stocked in Lake Superior declined
significantly because of high incidental mortality in large-mesh gill nets and predation
by wild adult lake trout. It is not clear, however, whether survival would have been
better if gill-net fishing effort had been less, because predation by wild adult lake trout
may have reduced survival in the absence of fishing mortality. Survival of stocked
lake trout needs to be tested under conditions of lower gill-net fishing effort in
Michigan, where densities of wild adult lake trout are highest, to determine whether
predation by wild adult lake trout will compensate for reduced fishing mortality.

Such an adaptive management experiment may be focused on the wrong
problem. Reduced survival of stocked lake trout because of excessive incidental
mortality in large-mesh gill nets also indicates a problem for wild lake trout stocks in

Lake Superior. Abundance of wild lake trout has also declined in Michigan and

57

Wisconsin (Hansen et al. 1994b), and likely indicates the same effects of excessive
fishing mortality. Large-mesh gill-net fishing effort needs to be reduced in Michigan
and Wisconsin to enhance survival of both wild and stocked lake trout. Reductions in
large-mesh gill-net effort were imposed in Wisconsin on the state-licensed fishery in
1991 and the tribal-licensed fishery in 1992, but have yet been imposed in Michigan.

Reductions in incidental fishing mortality may lead to increased abundance of
wild lake trout in both Michigan and Wisconsin. Predation by wild adult lake trout
may subsequently increase, and reduce the survival of stocked lake trout that would
otherwise have increased in the absence of fishing mortality. As a consequence of the
interplay between fishing mortality and predation by wild adult lake trout, stocked lake
trout survival may remain low in the future. Stocking may no longer be a useful stock
enhancement technique in Lake Superior, particularly in areas with high densities of
wild lake trout, regardless of the level of fishing mortality. Increasing density of wild
lake trout in Minnesota may therefore lead to failures of stocked year classes, similar
to those in Michigan and Wisconsin, absent excessive incidental fishing mortality.

The importance of predation by wild lake trout on stocked yearling lake trout
needs to be better defined. Stomach samples should be obtained from wild lake trout
throughout their bathymetric distribution and during the entire growing season. The
bathymetric distribution of stocked yearling lake trout should also be determined to
define their spatial overlap with wild adult lake trout. These studies should be done
under conditions of both high and low large-mesh gill-net fishing effort to determine
whether declining survival of stocked yearling lake trout can be improved in the face

of increasing densities of wild lake trout in Lake Superior.

CHAPTER III:
STATUS OF LAKE TROUT

RESTORATION IN U.S. WATERS OF LAKE SUPERIOR

Abstract—Lake trout (Salvelinus namaycush) populations in Lake Superior
sustained 2 million kg of yield annually for four decades before collapsing during the
19503 due to excessive fishery exploitation and sea lamprey predation. Lake trout
restoration was attempted during the ensuing decades through an interagency program
of intensive stocking, sea lamprey control, and fishery regulation. Self-sustaining
populations of lake trout have returned to most areas in Lake Superior, but progress
toward historic yields has been difficult to measure because of losses to sea lamprey
(Petromyzon marinas) predation and unreported harvest. Because of such inherent
weaknesses in yield as a target for restoration, restoration targets were developed that
are based on abundance during the period when historic yields were sustained. Long
time-series of abundance data (1929-93) were developed from linear relationships
between CPE in commercial and assessment fisheries in Michigan. Progress toward
restoration of lake trout populations is described by comparing lake trout abundance
during modern times (1970-93) with their abundance during a historic reference period
(1929-43). Abundances of inshore stocks of wild lake trout exceeded historic
abundances in some years and areas during the 19803, but fell below historic
abundances in all areas during the 19903. Further progress in restoration can only be
achieved if fishery managers adequately protect existing stocks of wild fish from

predation by sea lampreys and exploitation by sport and commercial fisheries.

58

59

Introduction

Lake trout (Salvelinus namaycush) sustained 2 million kg of average annual
yield during 1913-50 in Lake Superior (Baldwin et al. 1979), but collapsed nearly to
extinction during 1951-62 because of excessive fishery exploitation and sea lamprey
(Petmmyzon marinas) predation (Hile et al. 1951; Pycha and King 1975; Jensen 1978;
Coble et a1. 1990). Hatchery-reared, juvenile lake trout were stocked, in conjunction
with controls on sea lampreys and fisheries, to restore populations into the lake
(Lawrie and Rahrer 1972, 1973; Pycha and King 1975). Stocking has been relatively
continuous since 1951 in Wisconsin, 1952 in Michigan, 1957 in Ontario, and 1962 in
Minnesota (Lawrie and Rahrer 1972, 1973; Pycha and King 1975; Lawrie 1978). Sea
lampreys reached peak abundance during 1958-61, and have been maintained at 15%
of that level since 1962 using chemicals, barrier dams, and traps (Smith 1971; Smith et
al. 1974; Smith and Tibbles 1980; Klar and Weise 1994). Commercial and sport lake
trout fisheries were closed lakewide during 1962, and have been strictly regulated ever
since (Pycha and King 1975; Hansen et al. 1994b).

The goal of lake trout restoration in Lake Superior is to restore self-sustaining
stocks that can provide an average annual yield equal to that during 1929-43 (2 million
kg) (LSLTTC 1986; Busiahn 1990). The reference period for lake trout restoration
was set during 1929-43 because the yield during that period was consistent with the
average annual yield dating back to 1913, and because lake trout stocks were thought
to decline after 1943 (Hile et al. 1951; Pycha and King 1975; Jensen 1978). However,

this goal cannot be attained if sea lampreys kill a portion of the annual production, or

60

if much of the yield from fisheries goes unreported, even if self-sustaining lake trout
stocks are restored throughout the lake. For example, sea lampreys may have
consumed as much lake trout production as humans in some areas during the 19903
(Hansen et al. 1994b)—lake trout stocks in these areas may have been much closer to
historic abundances than was indicated by fishery yields. Alternatively, if the
restoration goal were stated in terms of lake trout abundances that are capable of
yielding 2 million kg annually (rather than actual yield), then progress could be
measured regardless of losses to sea lamprey predation or unreported fishery harvest.

Progress in lake trout restoration could be better measured in terms of
abundance, or an index of abundance such as CPE, than in terms of yield. Lawrie
(1978) acknowledged that records of historical lake trout abundance were not
available, but suggested that targets for stock restoration could be inferred from
observations of stocks that had been lightly exploited and little affected by sea
lamprey predation. He noted that the CPE of lake trout averaged 56.4 fish per km of
multifilament nylon gill net during 1938-44 near Michipicoten Island, and ranged from
51.8 to 71.2 near Caribou Island (Lawrie 1978). In contrast, the CPE of lake trout
averaged 241.1 on Superior Shoal during 1967-70 (Lawrie 1978). However, the lake
trout at Superior Shoal averaged 17% smaller than those at either Michipicoten Island
or Caribou Island, so the estimated CPEs of similar-sized lake trout would have been
201.1 at Michipicoten Island and 133.9 at Caribou Island (Lawrie 1978).

Such direct measures of lake trout abundance would facilitate measurement of
progress in lake trout restoration in Lake Superior. My objective is to develop a

quantitative means of evaluating the current status of lake trout stocks in inshore

61
Michigan waters by standardizing CPE data presented by Hile et al. (1951), Pycha and

King (1975), and Hansen et al. (1994b) into a 65-year data set for each inshore
Michigan management area in Lake Superior. These data series will allow direct
comparison of contemporary lake trout abundance (CPE), monitored by ongoing
gill-net assessment fisheries, with historic lake trout abundance, when populations
yielded target levels of production. Progress toward these target levels of production
can therefore be readily judged by the difference between the CPE in the assessment

fishery and the average CPE during the historic (reference) period.

Methods

Study A rea

Stock assessment of lake trout in Lake Superior is carried out in accordance
with an inter-agency rehabilitation plan that specifies management areas for reporting
progress in lake trout stock restoration (LSLTTC 1986) (Figure 4). These management
areas are modifications of statistical districts previously used for reporting commercial
fishery statistics in Michigan, Minnesota, and Wisconsin (Hile 1962). The difference
between the two systems is that some statistical districts were divided into two smaller
management areas that were closer in size to the range of lake trout movement in Lake
Superior—90% of marked lake trout were recaptured within 80 km, regardless of the
size at release or length of time at large (Eschmeyer et al. 1953; Buettner 1961; Pycha
et al. 1965; Rahrer 1968; Swanson 1973; Ebener 1990; Peck and Schorflraar 1991).
Consequently, the statistical district for all of Wisconsin (WI) was divided into two

management areas (W11 and WIZ), the statistical district surrounding the Keweenaw

62

Peninsula in Michigan (MS-3) was divided into two management areas (MI3 and
MI4), and the statistical district in central Michigan waters (MS-4) was divided into
two management areas (MIS and MI6). Lake trout management areas in Ontario are
the same areas used for management of lake whitefish commercial fishery quotas, and

bear no resemblance to former statistical districts.

Data Description

I constructed 65-year data sets of lake trout abundance in Michigan from
previous analyses of lake trout CPE by Hile et al. (1951), Pycha and King (1975), and
Hansen et al. (1994b). Hile et al. (1951) estimated lake trout abundance during
1929-49 from commercial fisheries in Michigan statistical districts MS-l through MS-6
from the catch per lift (CPE) in large-mesh gill nets (114-mm stretch-measure and
greater), set-hooks, and pound nets, expressed as a percentage of the 1929-43 average,
and averaged over statistical districts according to the yield in each area during
1929-43 (see Hile [1962] for a description of methods, and Jensen and Buettner [1976]
for a tabulation of the data).

Pycha and King (1975) updated Hile et al.'s (1951) analysis through 1970 and
expanded the analyses into Wisconsin, but used only large-mesh gill net CPE because
that gear accounted for 96% of the production in 1956-70. The CPEs in gill nets were
adjusted for increased efficiency of nylon twine (Pycha 1962), which replaced cotton
twine during 1950-52. The commercial fishery was closed in 1962, but a few selected
fishers were granted licenses to conduct assessment fishing thereafter. The CPEs of

these fishers was generally higher than the average CPE of all commercial fishers, so

63
Pycha and King (1975) adjusted their CPEs during 1962-70 by the ratio of their CPEs

to those of the entire fishery during 1959-61. The historic average CPE in adjacent
area MIZ was used as the historic average in Wisconsin (Pycha and King 1975).
Hansen et al. (1994b) reported the CPE in gill net (114-mm stretch measure)
assessment fisheries in Michigan, Minnesota, and Wisconsin during 1970-92, but did
not link their data with those of Hile et al. (1951) and Pycha and King (1975). The
analyses by Hansen et al. (1994b) were for lake trout management areas (LSLTTC
1986) (Figure 4), rather than the statistical districts used by Hile et al. (1951) and
Pycha and King (1975). Consequently, I used data from statistical district MS-3 to
construct the historic data series in management areas M13 and MI4, district MS-4 for
areas M15 and MI6, district MS-S for area MI7, and district MS-6 for area M18.
Hansen et al. (1994b) also computed the CPE as the geometric mean across all lifts in
each management area, rather than the pooled catch over the pooled effort (Hile et al.

1951; Pycha and King 1975), to quantify the variance of catches.

Statistical A nalysis

I analyzed commercial gill net fishery catch and effort data compiled by Jensen
and Buettner (1976) for each of the Michigan statistical districts that were analyzed by
Hile et al. (1951) and Pycha and King (1975). First, 1 reconstructed the data tabulated
in Hile et al. (1951) to ensure that the historic data used in my analysis was the same.
Next, I extended the data summaries through the end of the commercial gill net data
series (1962), which provided statistics that should have been the same as those used

by Pycha and King (1975) (Appendix A, Tables 23-24).

64

I used simple linear regression to relate the data series of Pycha and King
(1975) (Appendix A, Tables 23-24) and Hansen et al. (1994b) (Appendix A, Tables
6-11, back-transformed values) during a period of overlap (1959-61), to extend the
CPE of lake trout from 1929 through 1993. The period of overlap was the same one
used by Pycha and King (1975) because it was a period before major restrictions were
imposed on the fishery (1962), and assessment fishers worked as regular licensees
within the overall commercial fishery. Catch and effort statistics were available for
the 1959-61 overlap period in all management areas except MI8. For MI8, I used the
next 3-year period (1962-64) for which data were available in both data sets. The
linear relationship between the data series was then used to standardize the old data
into the same units as the new data The regression equation therefore provides an
omnibus correction of catch from pounds to numbers, net length from feet to meters,
CPE from the commercial fishery to the assessment fishery, and CPE from weighted
averages to geometric means. Target CPEs in each area were computed as the average
reconstructed CPE in the area during 1929-43. Because of the shortness of the overlap

period (N=3), correlations were only judged to be acceptable if they were near unity.

Results

Average annual CPEs in the commercial and assessment fisheries during
1959-61 corresponded well for management areas MI3 (r=0.99), MI4 (r=0.94), M15
(r=0.99), MI6 (r=0.99), and MIS (r=1.00), but not in MI7 (r=0.72). The linear

relations between commercial and assessment fisheries in 1959-61 were:

65
MI3: CPE, = -1.57 + (1.27 x CPEc)

MI4: CPE, = -1.26 + (2.08 x CPEc)
MIS: CPE, = -4.63 + (3.69 x CPE,)
MI6: CPE, = -2.36 + (2.92 x CPEc)
MI7: CPE, = -14.35 + (3.48 x CPEC)

MI8: CPE, = -2.97 + (0.76 x CPE,)

For each relation, CPE, and CPE, are average CPEs in the assessment and commercial
fisheries during 1959-61. The slope of each linear relationship shows the efficiency of
the assessment fishers in that management area, relative to the overall commercial
fishery, provided that average weights of lake trout in the catch and the relationships
between geometric means and weighted averages are relatively consistent among areas.
The target CPE for lake trout restoration (1929-43 average) was 18.40 in MI3,
31.56 in MI4, 70.98 in MIS, 57.36 in MI6, 103.31 in MI7,and 21.14 in MI8. In MI3
and MI4, the CPEs of wild lake trout exceeded the target in several years during the
19803 (1984-85 and 1989 in MI3; 1980-81 and 1986 in M14), but by 1993, fell to only
35% of the target in MI3 and 57% of the target in MI4 (Figure 21). In MI5, the CPE
of wild lake trout exceeded the target in 1985-86, and fell to 54% of the target in 1993
(Figure 22). In MI6, the CPE of wild lake trout never exceeded the target, but rose to
77% of the target in 1993 (Figure 22). In MI7 and MI8, the CPE of wild lake trout
was lower than elsewhere, never exceeded the target in MI7, and exceeded the target
in MI8 only in 1984 (Figure 23). The wild lake trout CPE in 1993 was only 16% of

the target in MI7 and unknown in MI8 (Figure 23).

66

 

100

 

 

Catch/Effort

 

 

 

 

 

140
1 M14
120 d

100 *

 

 

Catch/Efl‘ort

 

 

0 TlllllllfllllllllllllllTIlTTlllFFFTrlllIITTIIIlIllllTlTTlllll
1930 35 40 45 50 55 60 65 70 75 80 85 90

 

 

 

Year

Figure 21. Abundance of stocked and wild lake trout in Michigan west (MI3) and east
(MI4) of the Keweenaw Peninsula in Lake Superior during 1929-93, compared to the
average abundance during 1929-43.

67

 

160 .
140 - i
120 '2
100 -

805

 

Catch/Efi‘ort

 

605

 

 

 

.l
l
160 '1 Total

 

Catch/Efi‘ort

 

 

 

 

 

 

 

0 TTTTTITTTTTTTTTTTTTIflTTTIITTIFTTTTTTTTWTFFTTTWTIIITTTITTTTTTTTI
1930 35 40 45 50 55 60 65 70 75 80 85 90

Year
Figure 22. Abundance of stocked and wild lake trout in Michigan near Marquette

(MIS) and Munising (MI6) in Lake Superior during 1929- 93, compared to the average
abundance during 1929-43.

68

180

 

: M17
160 j

140 —

120 j

W

 

:v - T...

130-}

Catch/Effort

60$
40—:

‘ “- 73". /.-=\
20 — . \a. I" \Vv’ '2‘

 

0 llllllllllllllrllllllllllllllllfiTllTrrTll

 

100
. MI8

80 - Total

60%

Catch/Effort

 

 

20 -/~’ A v ‘
q \
‘ l

1 :3' Wild r4"

   

 

 

 

A \A"\__‘—‘ J ~-:-'"“"""%‘.\
0 llllllllllTlllllllllllllllllTTWrTlllllTlTllllllllllTTrrllllll

1930 35 40 45 50 55 60 65 70 75 80 85 90
Year
Figure 23. Abundance of stocked and wild lake trout in Michigan near Grand Marais

(MI7) and in Whitefish Bay (MIB) in Lake Superior during 1929-93, compared to the
average abundance during 1929-43.

69

Discussion

The perspective of the 1929-93 period shows that abundance of wild lake trout
in Michigan declined steadily from the 19403 through the late 19603 and then
improved from the 19703 through the 19803. High abundances of stocked fish in the
late 19603 produced increased numbers of wild fish in the 19703 (see Chapter 1).
Throughout the 19703, abundances of stocked fish in many areas were much higher
than during 1929-43, but declined sharply in the late 19703 and 19803, and remained
extremely low after 1988 (Chapter H). In most areas, numbers of wild lake trout
increased steadily in the 19703 and early 19803, but declined slowly in the late 19803
and early 19903. The recent decline in the abundance of wild lake trout was partly
caused by an earlier decline in the abundance of stocked lake trout, but was mitigated
by reproduction by wild fish, the progeny of the first stocked spawners. Abundances
of lake trout in Michigan in 1993 remain below the 1929-#43 average in all areas, even
though wild fish dominate the stocks.

Lake trout restoration was previously evaluated in Lake Superior in qualitative
terms, primarily because stocks were far-removed from a restored condition. Early on,
Dryer and King (1968) stated that "[t]he remarkable recovery of lake trout stocks in
the Apostle Islands region makes the outlook for complete success of lake trout
rehabilitation appear excellent. Natural reproduction, which has already been
demonstrated, should soon replace hatchery plantings." A decade later, Lawrie (1978)
stated that "there are now encouraging signs that natural reproduction of lake trout is

increasing so that it may not be necessary to continue planting that species, at least, in

70
perpetuity.” Almost a decade later, MacCallum and Selgeby (1987) stated that

"[i]ntensive planting of lake trout and increasing natural reproduction have led to a
resurgence in lake trout abundance; annual harvests in the commercial and recreational
fishery are now about one third of the harvest formerly sustained.”

Of these accounts of progress in lake trout restoration, only MacCallum and
* Selgeby (1987) attempted to state progress in quantitative terms (relative to historic
sustained yield). However, yield is greatly confounded by restrictions on fisheries,
losses to sea lampreys, and non-reporting, which alter the maximum sustainable level
of yield from one period to the next. Hansen et al. (1994b) attempted to overcome
some of these problems by quantifying progress in terms of the total kill of lake trout
in fisheries (both reported and unreported) and by sea lampreys. They found that the
average yield in 1990-92 was only 25% of the historic average in Canada and 32% in
the United States (Hansen et al. 1994b). Sea lampreys accounted for another 15% of
the historic average yield in the United States (sea lamprey populations were not
estimated in Canada) (Hansen et al. 1994b). Further, sea lampreys accounted for 42%
of all the lake trout yield from United States waters west of the Keweenaw Peninsula
and 18% from waters east of the peninsula (Hansen et al. 1994b).

My targets for wild lake trout abundance are similar, for some areas, to those
suggested by Lawrie (1978). For areas MIS and MI6, our targets were within the
range of CPEs found on Michipicoten and Caribou Islands, whereas those for areas
MI3, MI4 and MI8 were lower and that for area MI7 was higher. The target for area
MI7 was weak due to poor correspondence between commercial and assessment

fishery CPEs during the overlap period, but the average CPE during 1929-43 in the

71

commercial fishery in area MI7 (statistical district MS-5) was higher than in any other
district (Hile et al. 1951). This suggests that my target may be reasonable.

For many areas, target CPEs were quite similar to those that result from simple
conversion of pounds per 1,000 feet of large-mesh gill nets into geometric mean
number of fish per km of gill net. The yield per km is nearly the same as the number
of fish per km because the average size of lake trout caught in ll4-mm, stretch-mesh
gill nets is approximately 1.1 kg. Also, the conversion from weighted average CPE to
geometric mean CPE, a factor of approximately one-half, directly compensates for the
conversion from commercial fishery CPE to assessment fishery CPE, a factor of
approximately two. However, this similarity was not always true, as our target CPEs
for areas MIS (70.98) and MI6 (57.36) were much higher than the target CPE for
district MS-4 computed by simple conversion of units (27.73). Targets for areas MIS

and MI6 were the closest to those suggested by Lawrie (1978).

Management Implications

Lake trout restoration has progressed substantially in several Michigan
management areas. Wild lake trout stocks have been restored to within 23% of
historic abundances in MI6, and abundances are still increasing. The prognosis for the
future is good in this area so long as fishery managers continue to control commercial
and sport fisheries and sea lampreys. In MI4 and MIS, lake trout stocks have been
restored to within 43% and 46% of historic abundances, but stocks are declining in

both areas. Mortality should be reduced in these areas if lake trout restoration is to

72

move forward. Declining abundances in both areas began after 1985, coincident with
the reopening of large-scale tribal commercial gill net fishing and increased catches in
the sport fishery (Peck and Schorflraar 1991). Regulation of each of these fisheries
should be made more stringent in order to reduce fishing mortality on lake trout.

In remaining areas, lake trout stocks remain well below historic abundances,
and will therefore require mortality to be reduced below current levels. In MI3, wild
lake trout stocks recently declined, as in MI4 and MIS, and should also be targeted for
more stringent fishery regulation. In MI7, the estimated historic CPEs are poor, so
stock status is difficult to judge. Further analyses of modern and historic data should
attempt to discover a more reliable target for abundance than the one derived herein.
In the interim, fishery managers should ensure that fishery regulations are sufficient to
sustain current abundances of wild lake trout. In MI8, lake trout restoration was
deferred in 1985 as part of a negotiated settlement between the State of Michigan and
local indian tribes. Lake trout stocks are unlikely to improve until fishery
management changes substantially.

The reference period for lake trout restoration was set during 1929-43 because
the yield during that period was consistent with the average annual yield dating back
to 1913 (Hile et al. 1951), and because lake trout stocks were thought to decline after
1943 (Hile et al. 1951; Pycha and King 1975; Jensen 1978). However, it is possible
that yield during 1913-43 was sustained at an apparently stable level by sequentially
fishing (and depleting) individual stocks, so that total abundance of lake trout in Lake
Superior was declining over that period. For example, abundance of lake trout in Lake

Superior may have begun to decline after 1939 (Coble et al. 1990), rather than 1945,

73
as was previously thought (Hile et al. 1951; Pycha and King 1975; Jensen 1978).

Also, the 1929-93 data series show that abundance in central Michigan (M15 and MI6)
began to decline after 1934 (Figure 20).

Lake trout abundance in central Michigan may have declined earlier than in
other areas because of the proximity of the major ports of Marquette (MIS) and
Munising (MI6), but the decline in these areas was not evident when abundance was
analyzed over a much broader area (Hile et al. 1951; Pycha and King 1975; Jensen
1978; Coble et al. 1990). For this reason, the target CPEs for these two areas should
be considered minimal estimates of stock sizes that are needed to sustain historical
levels of lake trout production. However, records of commercial fishing catch and
effort in Michigan only go back to 1929, so the data needed to investigate the

sustainability of lake trout stocks prior to that year are not available.

SUMMARY AND CONCLUSIONS

Efforts to restore lake trout in Lake Superior have reestablished reproducing
populations in most areas of the lake. Results of my stock-recruitment analyses
suggest that stocked lake trout played a significant role in reestablishing these
populations. Results of my survival analyses suggest that large-mesh gill-net fisheries
reduced the abundance of stocked lake trout in both Michigan and Wisconsin. In
Minnesota, such fisheries were not allowed to develop, so abundance of stocked lake
trout declined later than in Michigan or Wisconsin, mostly in response to predation by
increasing numbers of wild lake trout. Long-term data suggest that wild lake trout in
Michigan during l990-93, after more than 30 years of attempted stock restoration,

were less abundant than during 1929-43, before stocks collapsed.

Sources of Recruitment

My results suggest that the availability of inshore spawning substrate in the
Great Lakes is a critical determinant of successful reproduction by stocked lake trout.
This is in contrast to the widespread belief that stocked lake trout are reproductively
ineffective. Reproductive ineffectiveness of stocked lake trout was thought to explain

the widespread failure of lake trout restoration in other Great Lakes (Eshenroder et al.

74

75

1983), but was based on stock-recruitment analyses in the Apostle Islands area of Lake
Superior where spawning substrate is restricted to offshore shoals that require homing
ability by spawning lake trout (Krueger et al. 1986). Spawning shoals in Minnesota,
Michigan, and Ontario are widely distributed inshore, where little homing ability is
required by inexperienced, stocked lake trout spawners that tend to wander inshore
during the spawning season. Stocked lake trout reproduced effectively in all such
areas of Lake Superior that had abundant inshore spawning habitat.

Lake trout restoration has been deferred in the northern parts of Lakes Huron
and Michigan where inshore spawning grounds are most abundant; these areas have
been reserved for intensive fisheries for lake Whitefish, mostly by gillnets that impose
high incidental mortality on lake trout (Rybicki and Keller 1978). Consequently, lake
trout spawning stocks have not developed in these areas where inshore spawning
shoals occur and the likelihood of successful reproduction is greatest. Rather, lake
trout restoration has been pursued mostly in the southern parts of Lakes Huron and
Michigan, where inshore spawning grounds are rare and the likelihood of successful
reproduction is poorest. Only since the mid-19803 has lake trout restoration been
pursued on large offshore reefs where stocked fish are likely to remain and spawn,
such as Six-Fathom Bank, Lake Huron, the Beaver Islands, Lake Michigan, and the
Mid-Lake Reefs, Lake Michigan. Stocking in these areas should provide for
successful stock restoration if fish are protected from fisheries.

Stock restoration in northern Lake Huron and northern Lake Michigan can
succeed if spawning grounds are still in suitable condition and lake trout are protected

from fishery exploitation and sea lamprey predation. Surveys of historic lake trout

76

spawning grounds in northern Lake Huron and Lake Michigan have shown that
substrate quality has not been observably degraded (Edsall et al. 1992). Excessive
fishery exploitation and sea lamprey predation more likely explain the lack of
successful reproduction by stocked lake trout in these areas. The future success of
lake trout restoration in these areas depends on the extent to which fishery managers
can control total annual mortality resulting from fishery exploitation and sea lamprey
predation. Stocking of hatchery-reared fish continues to be a viable tool for lake trout

restoration in both lakes, provided that these controls on mortality are effective.

Causes of Declining Survival

My analyses suggest that survival of lake trout stocked in Lake Superior
declined because of incidental mortality in large-mesh gill nets and predation by wild
adult lake trout. It is not clear whether survival would have been better if gill-net
fishing effort had been lower, because predation by wild adult lake trout may have
reduced survival in the absence of fishing mortality. Survival of stocked lake trout
should be tested under conditions of lower gill-net fishing effort to determine whether
predation by wild adult lake trout will compensate for reduced fishing mortality.

Such an adaptive management experiment would be interesting, but may be
focused on the wrong problem. Reduced survival of stocked lake trout due to high
incidental mortality in large-mesh gill nets also poses a serious problem for wild lake
trout in Lake Superior. Abundance of wild lake trout also declined in Michigan after

1988 (Hansen et al. 1994b), in conjunction with the abundance of stocked lake trout,

77

most likely due to excessive fishing mortality on both wild and stocked fish.
Large-mesh gill-net fishing effort should be reduced in Michigan and Wisconsin to
enhance survival of both wild and stocked lake trout. Such reductions were imposed
in Wisconsin on the state-licensed fishery in 1991 and the tribal-licensed fishery in
1992. No such reductions have yet been imposed in Michigan.

Reductions in incidental fishing mortality may lead to better survival and
increased abundance of wild lake trout in Michigan and Wisconsin. However, wild
adult lake trout will also increase, and increase their predation on stocked lake trout.
As a consequence of the interplay between fishing mortality and predation by wild
adult lake trout, stocked lake trout survival may remain low in the future. In the
future, stocking may not be a useful enhancement technique in Lake Superior,
particularly in areas with high densities of wild lake trout, regardless of the intensity
of fishing mortality. Increasing density of wild lake trout in Minnesota may therefore
lead to failures of stocked year classes, similar to those in Michigan and Wisconsin, in
the absence of excessive incidental fishing mortality.

The importance of predation by wild lake trout on stocked yearling lake trout
needs to be defined. Stomach samples should be obtained from wild lake trout
throughout their bathymetric distribution and during the entire growing season. The
bathymetric distribution of stocked yearling lake trout should also be determined to
define their spatial overlap with wild adult lake trout. These studies should be done
under conditions of both high and low wild lake trout density to determine whether
declining survival of stocked yearling lake trout can be improved in the face of high

densities of wild lake trout in Lake Superior.

78

Status of Restoration

Lake trout restoration has progressed substantially in several Michigan
management areas. Wild lake trout stocks have been restored to within 23% of
historic abundances in MI6, and abundances are still increasing. The prognosis for the
future is good in this area so long as fishery managers continue to control fisheries and
sea lampreys. In MI4 and MI5, lake trout stocks have been restored to within 43%
and 46% of historic abundances, but stocks are declining in both areas. Declining
abundances in both areas began after 1985 as large-mesh tribal commercial gill net
fisheries were reopened (Peck and Schorfhaar 1991; Chapter 11), so these fisheries
should be regulated more stringently.

In all remaining areas of Michigan, lake trout stocks remain well below historic
abundances. In MI3, wild lake trout stocks recently declined, as in MI4 and N115, and
should also be targeted for more stringent fishery regulation. In MI7, the estimated
historic CPEs are not as reliable and current stock status therefore remains uncertain.
More reliable targets should be developed through further investigation of historic
data In the interim, fishery managers should ensure that fishery regulations are
sufficient to sustain current abundances of wild lake trout. Lake trout stocks in MI8
are unlikely to improve until fishery management changes substantially, because lake
trout restoration was deferred in 1985 as part of a negotiated settlement between the
State of Michigan and local Indian tribes. Large-mesh gill-net fisheries in MI8 should
be converted to trap-net fisheries to advance lake trout restoration in this area.

The reference period for lake trout restoration was set during 1929-43 because

79

yields during that period were consistent with average annual yields dating back to
1913 (Hile et al. 1951), and because lake trout stocks were thought to have declined
after 1943 (Hile et al. 1951; Pycha and King 1975; Jensen 1978). Yet, yield during
1913-43 may have been maintained at an unsustainable level by sequentially fishing
and depleting individual stocks. Lake trout abundance may actually have declined
over that period. For example, abundance of lake trout in Lake Superior may have
begun to decline after 1934 in central Michigan, based on my 1929-93 data series.
Lake trout abundance in central Michigan may have declined earlier than in
other areas because of the proximity of the major ports of Marquette (MIS) and
Munising (MI6). The decline in these areas, however, was not evident when
abundance was analyzed over a much broader area (Hile et al. 1951; Pycha and King
1975; Jensen 1978; Coble et al. 1990). For this reason, the target CPEs for these two
areas should be considered conservative estimates of stock sizes that are needed to
sustain historical levels of lake trout production. However, records of commercial
fishing catch and effort in Michigan only go back to 1929, so the data needed to

investigate the sustainability of lake trout stocks prior to that year are not available.
Conclusions

Self-sustaining lake trout stocks have been reestablished in much of Lake
Superior, but prudent management is required to allow these stocks to recover to
historic levels of abundance and to permit stocks to develop in the rest of the lake.

Sea lamprey control and fishery regulation were effective enough to allow stocking to

80

rapidly build inshore stocks of lake trout that reproduced in all areas with widely
distributed inshore spawning habitat. However, survival of stocked lake trout declined
sharply after 1970 in Wisconsin and after 1980 in Michigan, possibly due to mortality
in large-mesh gill-net fisheries. Lake trout of hatchery origin are now extremely rare
throughout Michigan, and declining rapidly elsewhere. Wild fish have replaced
stocked fish in most areas, and, as the reproductive stocks of the future, should be
protected from by sea lamprey predation and fishery exploitation. Stocking should be
discontinued wherever wild fish dominate stocks, such as in Michigan and eastern
Wisconsin, to protect wild stocks from hatchery diseases and outbreeding depression
(Evans and Willox 1991; Krueger and May 1991).

State and tribal fishery management agencies, particularly in Michigan and
Wisconsin, failed to control exploitation by commercial and angling fisheries after sea
lamprey control and stocking caused inshore lake trout stocks to increase in
abundance. Excessive fishery exploitation stalled lake trout restoration in the Apostle
Islands area of Wisconsin, in waters surrounding the Keweenaw Peninsula in western
Michigan, and in Grand Marais and Whitefish Bay in eastern Michigan. Virtually all
excessive fishery exploitation in Lake Superior, both historically and presently, is
coincident with the use of unregulated amounts of large-mesh gill nets, which impose
incidental mortality on lake trout even when they are set for other species (usually lake
Whitefish). Large-mesh gill-net effort was recently reduced in eastern Wisconsin, by
imposing limits on the total amount of net that can be set in a year by each fisher.
Similar measures should be imposed on tribal fisheries that operate in waters around

the Keweenaw Peninsula, to reverse the downward trend in abundance of wild lake

81

trout in these waters. State and tribal fishery management agencies should renegotiate
the consent order for eastern Michigan waters around Grand Marais and Whitefish
Bay, where lake trout restoration was foregone in favor of intensive large-mesh gill-net
fisheries for lake Whitefish. Lake trout exploitation in Minnesota and Ontario should

be contained at current levels to sustain progress in these jurisdictions.

APPENDIX A - ADDITIONAL TABLES

82
APPENDIX A - ADDITIONAL TABLES

Table 6. Catch/effort of lake trout in spring gill-net assessment fisheries in western
Keweenaw Peninsula waters (MI3) of Lake Superior (mean and SE across N lifts of
log-transformed values).

 

 

 

 

 

Total Wild Stocked
Year Mean SE Mean SE Mean SE N
1959 1.65815 0.09522 1.55456 0.09524 0.41082 0.05743 23
1960 1.31851 0.22253 1.21731 0.21453 0.32797 0.11852 7
1961 1.18169 0.13975 1.05655 0.13114 0.38783 0.06800 25
1962 1.40563 0.15078 1.38454 0.14935 0.09095 0.04137 8
1963 1.13636 0.12167 0.90477 0.11224 0.48904 0.10806 8
1964 1.13037 0.22124 0.56342 0.04007 0.84421 0.26453 3
1965 1.85769 0.11548 1.11303 0.11562 1.43843 0.11734 19
1966 2.17703 0.13515 0.79354 0.09245 2.02967 0.13835 19
1967 2.47818 0.15730 0.51092 0.10327 2.42120 0.15457 7
1968 2.33833 0.13688 0.43418 0.19288 2.27240 0.12856 9
1969 2.42128 0.16948 0.28307 0.07290 2.37318 0.18888 11
1970 2.52076 0.14914 0.07389 0.02944 2.51388 0.14948 15
1971 2.26862 0.24898 0.32466 0.10363 2.23577 0.24696 15
1972 3.88793 0.27468 1.35064 0.24161 3.81507 0.28432 9
1973 3.90670 0.08070 1.14032 0.12014 3.85412 0.08315 16
1974 3.45539 0.10120 1.22993 0.11199 3.36496 0.10417 21
1975 3.68601 0.08362 1.56575 0.10750 3.56921 0.08854 25
1976 4.04048 0.11987 2.11523 0.13660 3.89896 0.11723 15
1977 4.60240 0.21362 2.49987 0.27748 4.47820 0.20548 8
1978 3.52819 0.16756 1.54969 0.16505 3.41366 0.16243 22
1979 3.96728 0.35273 2.25771 0.35426 3.78706 0.35239 6
1980 4.15853 0.27944 2.34882 0.29323 3.99543 0.27154 8
1981 4.26983 0.12340 2.47167 0.19361 4.09267 0.12093 8
1982 4.29120 0.10517 2.41122 0.14789 4.12725 0.11775 8
1983 3.88229 0.06649 2.55119 0.06367 3.59888 0.07683 7
1984 4.41067 0.13726 3.26706 0.16325 4.03952 0.13772 4
1985 4.16164 0.17631 3.13411 0.11819 3.73569 0.21060 5
1986 2.77006 0.14128 1.73353 0.24386 2.34523 0.11709 12
1987 3.08942 0.12718 2.64085 0.15889 2.02757 0.17606 14
1988 3.08170 0.11415 2.87756 0.12753 1.44186 0.11396 24
1989 3.27533 0.10425 3.15884 0.10078 1.21429 0.17653 21
1990 2.99475 0.14164 2.90336 0.14512 0.86970 0.13751 22
1991 2.70692 0.11155 2.56737 0.13136 0.79364 0.10891 31
1992 1.79607 0.14300 1.67280 0.14846 0.42576 0.09163 32
1993 2.09718 0.11753 2.00907 0.11603 0.46735 0.09724 32

 

APPENDIX A

Table 7. Catch/effort of lake trout in spring gill-net assessment fisheries in Keweenaw

Bay in Michigan waters (MI4) of Lake Superior (mean and SE across N lifts of

log-transformed values).

 

 

 

 

 

Total Wild Stocked
Year Mean SE Mean SE Mean SE N
1959 2.22326 0.03069 2.20045 0.03122 0.17397 0.01243 165
1960 1.98110 0.04972 1.93689 0.05057 0.25756 0.01897 97
1961 1.73064 0.05741 1.60769 0.06019 0.44482 0.04035 74
1962 1.70851 0.05026 1.49638 0.05491 0.60826 0.04204 111
1963 2.09071 0.04983 1.61586 0.06667 1.18545 0.05493 111
1964 2.02670 0.05788 1.05103 0.06877 1.68176 0.05544 106
1965 2.36461 0.06154 0.61022 0.04773 2.25977 0.06390 95
1966 2.39127 0.07537 0.22503 0.03417 2.36247 0.07631 104
1967 2.68408 0.06891 0.11161 0.02658 2.67370 0.06909 90
1968 3.77557 0.07614 0.23455 0.05009 3.76754 0.07625 51
1969 3.66895 0.08015 0.20448 0.04092 3.66311 0.08011 44
1970 4.14492 0.12757 0.32234 0.07567 4.13845 0.12732 32
1971 3.65229 0.15241 0.71035 0.07273 3.60918 0.15744 36
1972 4.19486 0.17343 1.40345 0.15131 4.13547 0.17894 20
1973 4.91854 0.15190 1.89237 0.12372 4.87196 0.15657 11
1974 4.41376 0.10466 1.62242 0.10080 4.35637 0.10664 32
1975 4.53550 0.14604 1.69359 0.12123 4.47525 0.15357 15
1976 4.57485 0.11832 1.79799 0.11512 4.50846 0.12162 38
1977 4.90583 0.14447 2.29728 0.10085 4.82770 0.15112 25
1978 4.58469 0.17765 2.37451 0.17998 4.44700 0.18283 29
1979 4.76572 0.17426 3.28558 0.17486 4.51309 0.17514 9
1980 4.84470 0.14393 3.64199 0.16717 4.45386 0.15337 21
1981 4.48602 0.11391 3.62538 0.09876 3.90788 0.13866 25
1982 3.98564 0.11477 2.89007 0.12534 3.51686 0.13057 33
1983 3.84741 0.09799 2.68187 0.12717 3.45842 0.09843 41
1984 3.83152 0.10528 2.94875 0.08676 3.30258 0.12538 24
1985 3.71091 0.14733 3.05067 0.15459 3.02067 0.13952 21
1986 3.91797 0.13934 3.53914 0.13807 2.73090 0.16397 22
1987 3.56890 0.09936 3.31502 0.13054 1.79623 0.14171 27
1988 3.49149 0.11326 3.37657 0.11775 1.34643 0.10066 60
1989 3.29207 0.08213 3.16646 0.08231 1.30139 0.08783 79
1990 3.55046 0.11482 3.45307 0.11983 1.28833 0.09225 48
1991 3.49642 0.09792 3.42301 0.09883 1.09865 0.09553 46
1992 3.38372 0.07818 3.27647 0.07964 1.21306 0.11920 36
1993 3.04780 0.10776 2.93713 0.10746 1.10677 0.10737 40

 

APPENDIX A

Table 8. Catch/effort of lake trout in spring gill-net assessment fisheries around

Marquette in Michigan waters (M15) of Lake Superior (mean and SE across N lifts of
log-transformed values).

 

 

 

 

 

Total Wild Stocked
Xaar Mean SE Mean SE Mean SE N
1959 2.50959 0.06206 2.49714 0.06264 0.12792 0.01078 80
1960 1.83574 0.05176 1.79430 0.05255 0.22066 0.01703 105
1961 1.85016 0.07152 1.78804 0.07044 0.34963 0.03279 57
1962 1.73265 0.06899 1.61678 0.06930 0.49117 0.03553 42
1963 1.95621 0.07726 1.67700 0.07271 0.99697 0.07193 26
1964 1.81756 0.08120 1.05781 0.06281 1.42534 0.08992 44
1965 2.12056 0.06255 0.72011 0.08678 1.95471 0.06816 26
1966 2.82464 0.08095 0.54737 0.06620 2.77375 0.08512 20
1967 2.87104 0.16779 0.19199 0.04391 2.85674 0.16900 20
1968 3.77511 0.21477 0.26603 0.09411 3.76656 0.21569 18
1969 3.93535 0.15682 0.31281 0.07009 3.92782 0.15723 16
1970 4.35491 0.15180 0.48191 0.13238 4.34384 0.15388 13
1971 4.68368 0.10440 1.14275 0.27833 4.65857 0.10205 7
1972 4.51882 0.17657 1.28019 0.23934 4.48873 0.17399 8
1973 4.55978 0.21008 1.81549 0.19979 4.50469 0.20959 7
1974 4.10870 0.13820 1.22137 0.16913 4.06429 0.13645 24
1975 4.95481 0.24507 2.24572 0.30759 4.89227 0.23984 7
1976 4.78553 0.12116 2.46104 0.13169 4.68430 0.12313 24
1977 4.88655 0.12650 2.74029 0.11022 4.76364 0.13216 17
1978 4.80567 0.12265 3.28831 0.10755 4.55923 0.13100 18
1979 5.01752 0.11647 3.66424 0.11959 4.71841 0.11950 18
1980 4.64550 0.16379 3.50660 0.29514 4.21120 0.12848 12
1981 4.80818 0.16129 4.04239 0.15844 4.18470 0.16813 10
1982 4.89686 0.09953 3.96699 0.14458 4.39061 0.07770 10
1983 4.41998 0.08089 3.70556 0.08571 3.75010 0.09284 16
1984 4.39164 0.14784 3.61548 0.13985 3.79247 0.15802 8
1985 4.88179 0.12620 4.31424 0.13547 4.05865 0.12110 5
1986 4.89299 0.26440 4.58382 0.24165 3.57663 0.32298 5
1987 4.32244 0.11719 3.97585 0.24101 2.42378 0.34804 8
1988 4.27323 0.18539 4.22284 0.17940 1.41495 0.26371 14
1989 4.10792 0.11329 3.99451 0.11585 1.83660 0.17052 22
1990 3.99449 0.12067 3.89193 0.12418 1.75126 0.11750 22
1991 4.09348 0.12120 3.97082 0.11786 1.97591 0.17938 18
1992 3.86685 0.11318 3.67614 0.11614 2.04972 0.16552 24
1993 3.85013 0.08291 3.66876 0.08231 2.04311 0144284 A!

 

APPENDIX A

Table 9. Catch/effort of lake trout in spring gill-net assessment fisheries around

Munising in Michigan waters (MI6) of Lake Superior (mean and SE across N lifts of
log,-transformed values).

 

 

 

 

 

Total Wild Stocked
Xaar Mean SE Mean SE Mean SE N
1959 2.40920 0.08179 2.39499 0.08158 0.15281 0.02383 35
1960 1.98054 0.12128 1.93577 0.12323 0.26782 0.04140 25
1961 1.76284 0.05675 1.65217 0.06440 0.40420 0.03880 42
1962 1.98196 0.07845 1.80699 0.08047 0.75354 0.05915 26
1963 2.26695 0.11169 1.66861 0.12755 1.62358 0.12848 13
1964 2.12138 0.08635 1.30878 0.08460 1.69841 0.10029 16
1965 2.40671 0.12086 1.03195 0.12557 2.20302 ' 0.12549 14
1966 2.62936 0.13193 0.70342 0.15359 2.53724 0.12839 13
1967 3.59159 0.14771 0.44839 0.09855 3.57568 0.14658 12
1968 3.61735 0.15744 0.28243 0.09418 3.60648 0.15752 13
1969 3.94331 0.11118 0.29011 0.08675 3.93316 0.11278 19
1970 4.53598 0.13407 0.45132 0.11962 4.52814 0.13429 16
1971 4.43790 0.16956 0.91879 0.20357 4.41790 0.16911 7
1972 5.15415 0.13642 2.42154 0.10768 5.09090 0.15266 2
1973 3.22126 0.23429 1.27070 0.22765 3.09869 0.23149 11
1974 3.67528 0.15370 1.71492 0.13695 3.54800 0.15713 14
1975 3.79169 0.19829 2.11652 0.18961 3.56845 0.21974 16
1976 3.77443 0.09250 2.25177 0.11328 3.51798 0.10312 24
1977 3.55868 0.15712 2.32995 0.16212 3.20705 0.16848 26
1978 3.42437 0.13321 2.45831 0.13743 2.92368 0.14672 30
1979 3.05305 0.11590 2.23523 0.11652 2.51169 0.12540 25
1980 3.24519 0.15158 2.43972 0.18607 2.61589 0.16733 20
1981 4.79360 0.40552 3.69966 0.59724 4.36540 0.28645 3
1982 3.81498 0.26811 2.84831 0.30333 3.32361 0.27976 10
1983 3.05491 0.11223 2.51278 0.08566 2.22440 0.18690 12
1984 3.57591 0.13391 3.01025 0.18501 2.74160 0.11819 8
1985 3.53315 0.22845 3.20521 0.22336 2.27413 0.27641 8
1986 3.81769 0.64868 2.95744 0.89925 2.94694 0.70860 3
1987 3.44753 0.21208 3.34787 0.19952 1.35948 0.28344 8
1988 3.63784 0.25458 3.55666 0.24573 1.33128 0.30382 12
1989 3.48294 0.17826 3.38756 0.17570 1.27317 0.21862 14
1990 3.31388 0.12057 3.20597 0.12324 1.12574 0.15206 32
1 991 3.31830 0.14170 3.26303 0.13402 0.84191 0.18704 28
l 992 3.47285 0.13680 3.35931 0.13826 1.23088 0.18729 26
1993 3.99368 0.2474; 3.49311 02833 250282 023555 16

 

 

APPENDIX A

Table 10 Catch/effort of lake trout in spring gill-net assessment fisheries around

86

Grand Marais in Michigan waters (MI7) of Lake Superior (mean and SE across N lifts
of log-transformed values).

 

 

 

 

 

Total Wild Stogred
Yea; Mean SE Mean_ SE Mean SE N
1959 3.02396 0.08323 3.00193 0.08027 0.35036 0.09787 15
1960 2.49080 0.20953 2.45609 0.19675 0.40729 0.18560 7
1961 1.77336 0.06848 1.71144 0.06748 0.30116 0.03961 17
1962 2.06121 0.09878 1.95634 0.10099 0.57689 0.05305 7
1963 1.68686 0.22576 1.38601 0.21609 0.92617 0.14905 9
1964 2.21341 0.10595 1.20670 0.10551 1.85873 0.17761 7
1965 2.30910 0.24365 0.74003 0.08447 2.19350 0.25534 4
1966 2.27985 0.17124 0.48683 0.10918 2.18891 0.19730 10
1967 4.21703 0.40966 4.20954 1
1968 3.34110 0.21314 0.79242 0.22453 3.27251 0.22648 8
1969 3.48781 0.66473 0.00000 0.00000 3.48781 0.66473 2
1970 4.65451 0.10009 1.25723 0.09808 4.63032 0.09919 2
1971 4.41104 0.38244 1.11705 0.26274 4.38277 0.38582 6
1972 3.53635 0.21583 0.94822 0.15229 3.47254 0.23736 5
1973 4.22605 0.19046 2.03058 0.22885 4.12368 0.18439 6
1974 4.04188 0.14076 2.07900 0.24574 3.89005 0.14550 7
1975 4.10662 0.19030 2.51368 0.29279 3.88898 0.17078 4
1976 3.84157 0.12214 2.75550 0.13020 3.39827 0.13275 28
1977 3.76532 0.23018 2.77035 0243 89 3.28624 0.22936 21
1978 3.44020 0.10767 2.61066 0.11914 2.86051 0.12240 33
1979 3.02287 0.23872 2.28836 0.27168 2.45421 0.18496 17
1980 3.77898 0.19053 3.20248 0.16932 2.98670 0.21180 10
1981 4.04852 0.14325 3.10829 0.16198 3.54300 0.15932 10
1982 3.46228 0.13044 2.62005 0.13708 2.92498 0.13662 18
1983 3.73363 0.17833 3.17746 0.18660 2.92763 0.16885 7
1984 4.02964 0.10866 3.48514 0.09527 3.20171 0.12612 4
1985 2.97874 0.20064 2.61455 0.19071 1.90616 0.22439 6
1986 2.81679 0.16125 2.55924 0.20475 1.34199 0.14062 12
1987 2.97889 0.17775 2.86579 0.17551 1.09015 0.18360 11
1988 2.77748 0.18510 2.67280 0.18936 0.89043 0.12138 26
1989 3.35019 0.13016 3.28331 0.13051 0.94719 0.14679 19
1990 2.82840 0.16567 2.79805 0.16290 0.38024 0.13138 16
1991 3.02187 0.12789 2.96418 0.13604 0.54945 0.10423 24
1992 0
1993 2.96119 0.10529 2.87519 0.11670 0.81137 0.10388 16

 

APPENDIX A

Table 11. Catch/effort of lake trout in spring gill-net assessment fisheries in Whitefish

Bay in Michigan waters (MI8) of Lake Superior (mean and SE across N lifis of

loge-transformed values).

 

 

 

 

 

Total Wild Stocked
Mr Mean SE Mean SE Mean SE N
1959 3.13542 3.13542 0.00000 1
1960 0
1961 0
1962 2.06108 0.13671 1.72890 0.09632 1.17462 0.16224 8
1963 2.04469 0.23042 1.49016 0.39512 0.85570 0.28516 7
1964 2.69995 0.16713 0.10663 0.03164 2.69331 0.16691 11
1965 0
1966 2.80662 0.12714 0.79105 0.25867 2.71765 0.11268 5
1967 3.01989 0.36914 1.07910 0.17403 2.91758 0.38297 4
1968 3.60980 0.40050 0.33660 0.05952 3.59809 0.40321 3
1969 3.55050 0.32337 0.39473 0.14385 3.53968 0.32143 5
1970 2.88287 0.28134 0.26766 0.11008 2.86484 0.28123 4
1971 3.72519 0.04700 0.62873 0.03535 3.70364 0.04928 3
1972 3.08924 0.23343 0.40225 0.17040 3.07116 0.23002 9
1973 4.02848 0.24489 0.21971 0.13817 4.02467 0.24358 3
1974 4.41691 0.18981 0.36368 0.19803 4.40534 0.19633 6
1975 3.33476 0.74234 1.45768 0.85609 3.22101 0.68342 4
1976 2.05392 0.23869 0.89890 0.19483 1.81252 0.24328 28
1977 1.21782 0.23835 0.56276 0.15617 1.01461 0.21689 26
1978 1.31471 0.20913 0.68749 0.15799 0.99412 0.20187 21
1979 1.27147 0.36880 0.93912 0.34320 0.78728 0.30047 8
1980 2.06516 0.52983 1.41968 0.503 79 1.69442 0.45669 8
1981 1.21848 0.20439 0.44529 0.15361 1.05159 0.17509 7
1982 2.35852 0.29992 0.93690 0.17091 2.20057 0.30408 13
1983
1984 3.87324 0.15725 3.29820 0.06730 3.08273 0.26155
1985 2.54371 0.20600 1.29711 0.24820 2.14766 0.25839 1
1986
1987
1988
1989
1990
1991
1992
1993

OOOOOOOONNO

 

APPENDIX A

Table 12. Catch/effort of lake trout in spring gill-net assessment fisheries in western

Minnesota waters (MNl) of Lake Superior (mean and SE across N lifts of
log-transformed values).

 

 

 

 

 

Total Wild Stocked
Year Mean SE Mean: JE Mean SE N
1959 0
1960 0
1961 0
1962 0
1963 2.13421 1.99149 1.22153 1
1964 1.28594 1.14950 0.84696 1
1965 1.85637 0.12922 1.38629 0.09185 1.55121 0.11183 3
1966 1.22079 0.23829 0.70235 0.17891 1.10840 0.21485 9
1967 1.73369 0.38639 0.69175 0.17648 1.69037 0.37513 8
1968 1.53096 0.16685 0.10965 0.10965 1.51752 0.16205 12
1969 0
1970 0
1971 0
1972 0
1973 0
1974 0
1975 0
1976 0
1977 0
1978 0
1979 0
1980 0
1981 0
1982 3.24487 0.11753 0.60100 0.26306 3.22396 0.13714 7
1983 3.39494 0.11918 0.57962 0.26724 3.35330 0.11890 8
1984 3.24952 0.16855 0.78099 0.29723 3.18529 0.14957 9
1985 3.22732 0.15487 0.63093 0.24228 3.18838 0.14859 9
1986 3.34878 0.30876 1.08136 0.25184 3.28135 0.29635 10
1987 4.52471 0.17492 1.67257 0.06156 4.44341 0.18175 5
1988 3.37286 0.19643 1.19889 0.34657 3.27561 0.20960 4
1989 4.01863 0.25227 1.76029 0.31187 3.78070 0.23809 16
1990 3.36930 0.15078 0.99519 0.25043 3.22331 0.14677 20
1991 3.75891 0.14852 1.76814 0.26554 3.40212 0.15993 24
1992 4.26398 0.24542 1.87918 0.33530 3.99179 0.23391 16
$3 4.00479 0.184298 1.91219 0.23035 3.7658L ’ 0.18348: 17

 

 

APPENDIX A

Table 13. Catch/effort of lake trout in spring gill-net assessment fisheries in central

Minnesota waters (MN2) of Lake Superior (mean and SE across N lifts of
log-transformed values).

 

 

 

 

 

Total Wild Stoeked
Xaau Mean SE Mean SE Mean SE N
1959 0
1960 0
1961 0
1962 0
1963 1.48827 0.23398 1.44966 0.21019 0.17486 0.17486 4
1964 2.80293 2.45379 1.76824 1
1965 1.50206 0.03382 1.31749 0.00000 0.56126 0.08654 2
1966 1.71933 1.02712 1.12974 0.99008 1.44236 0.82448 3
1967 2.16000 0.44499 0.87846 0.11804 1.97778 0.49220 2
1968 1.79962 0.27827 0.47764 0.13822 1.69702 0.27519 15
1969 1.39464 0.10525 0.18959 0.04785 1.33314 0.10421 55
1970 2.37400 0.16207 0.18728 0.12806 2.35249 0.15793 9
1971 1.24194 0.13706 0.22363 0.10400 1.12647 0.15385 14
1972 1.49825 0.12547 0.19383 0.09355 1.36632 0.15729 20
1973 2.64273 0.08686 0.36363 0.08919 2.58096 0.09278 42
1974 2.47972 0.10740 0.64646 0.13808 2.36813 0.10524 30
1975 2.39594 0.10228 0.28559 0.06330 2.35153 0.10292 54
1976 2.37053 0.12507 0.19645 0.05996 2.35032 0.12385 45
1977 2.16994 0.12670 0.07140 0.04006 2.15960 0.12690 31
1978 2.68245 0.09809 0.18174 0.05288 2.66296 0.09839 44
1979 2.88629 0.13482 0.13102 0.05559— 2.87265 0.13547 39
1980 3.29792 0.13294 0.25084 0.07494 3.28164 0.13433 45
1981 3.58677 0.14389 0.37030 0.09900 3.57448 0.14255 45
1982 3.82338 0.10000 0.62340 0.10142 3.79131 0.10218 45
1983 3.39298 0.10204 0.46311 0.09109 3.35341 0.10434 52
1984 3.02769 0.11742 0.38761 0.08430 2.98589 0.11858 53
1985 3.40769 0.11622 0.70357 0.11974 3.35866 0.11637 46
1986 3.46693 0.15103 0.83734 0.15083 3.39450 0.15197 51
1987 3.75557 0.11752 1.42009 0.15246 3.63111 0.11951 57
1988 3.88449 0.11930 1.19780 0.18054 3.79270 0.11961 41
1989 3.77279 0.11126 1.64107 0.15915 3.59473 0.11520 54
1990 3.93366 0.10126 1.72628 0.18536 3.74475 0.10914 49
1991 3.62288 0.15389 1.62375 0.18735 3.45724 0.15449 42
1992 2.99145 0.09427 1.00801 0.11264 2.83871 0.09592 98
1993 3.49893 0.06444 1.26402 0.13954 3.35600 0.06370 66

 

APPENDIX A

Table 14. Catch/effort of lake trout in spring gill-net assessment fisheries in eastern
Minnesota waters (MN3) of Lake Superior (mean and SE across N lifts of
log-transformed values).

 

 

 

 

 

Total Wild Stocked
Xaar MeaL SE Mean SE Mean SE N
1959 0
1960 0
1961 0
1962 0
1963 2.26912 0.21733 2.22047 0.21387 0.25358 0.14025 16
1964 2.62719 0.15511 2.61223 0.15525 0.16147 0.06441 15
1965 1.74612 0.16791 1.65369 0.16658 0.45463 0.07676 29
1966 1.04989 0.09504 0.92571 0.08531 0.31542 0.06578 62
1967 1.52372 0.10939 1.13604 0.09935 0.85478 0.11657 30
1968 1.41537 0.12856 0.32367 0.07881 1.25337 0.13959 35
1969 1.26519 0.08384 0.07459 0.02140 1.23878 0.08334 152
1970 2.52752 0.14925 0.38416 0.11242 2.48028 0.14822 22
1971 1.93037 0.11696 0.43014 0.08221 1.81474 0.12438 39
1972 2.35543 0.22772 0.81397 0.10305 2.12993 0.29422 14
1973 2.84794 0.10538 0.91471 0.10348 2.70553 0.11153 65
1974 3.14425 0.12065 0.97829 0.10529 3.02619 0.12883 56
1975 3.30157 0.09300 0.98915 0.09005 3.21436 0.09371 109
1976 3.14335 0.10542 0.94090 0.09185 3.05873 0.10587 74
1977 3.20042 0.10167 0.72600 0.09681 3.13514 0.10361 68
1978 3.24837 0.08731 0.77049 0.08849 3.19270 0.08608 59
1979 3.22692 0.12387 0.60118 0.09705 3.14539 0.13730 51
1980 3.40749 0.10350 0.81728 0.11549 3.33709 0.10416 63
1981 3.80714 0.08511 1.46819 0.14268 3.69229 0.08705 45
1982 3.87316 0.14906 1.41995 0.19544 3.77981 0.15060 34
1983 3.72832 0.12475 1.78974 0.15135 3.53500 0.13967 41
1984 3.40059 0.10659 1.28998 0.11566 3.25596 0.11886 56
1985 4.44092 0.13794 1.60737 0.50313 4.30845 0.12380 10
1986 4.10813 0.13655 2.45976 0.24713 3.80233 0.12848 29
1987 4.26080 0.12646 1.88910 0.33185 4.03527 0.14013 19
1988 3.99709 0.08124 2.45250 0.18722 3.47547 0.12939 53
1989 4.20329 0.09354 3.16337 0.13024 3.64855 0.10648 45
1990 3.74262 0.09749 2.50042 0.15914 3.12733 0.12833 56
1991 3.74691 0.10550 2.55249 0.15177 3.21853 0.14095 50
1992 3.92961 0.12460 2.79414 0.17784 3.33047 0.16008 47
B93 3.0239; 0.17808 2.25513 0.18257 2.20830 0.18113 66

 

APPENDIX A

Table 15. Catch/effort of lake trout in spring gill-net assessment fisheries in western

Wisconsin waters (WIl) of Lake Superior (mean and SE across N lifts of
log-transformed values).

 

Yaar

1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993

Total

MeaL

2.54968
3.71692
2.59677
2.89253

1.62860
1.34175
1.22386
1.28287
1.35549
1.25868
1.56063

SE

0.93087
0.26702
0.15427
0.34385

0.1 1003
0.14223
0.17788
0.13002
0.12284
0.08207
0.10081

Wild

Mean

0.20281
0.94969
0.34784
0.69492

0.80364
0.48900
0.37683
0.62118
0.69563
0.65917
0.87594

SE

0.00000
0.19955
0.12847
0.29479

0.06825
0.08756
0.08583
0.08988
0.10711
0.12017
0.1m9

Stocked

Mean

2.52342
3.67816
2.56087
2.84773

1.50353
1.26320
1.16796
1.16590
1.25422
1.12180
1.44948

SE

0.95018
0.26605
0.16002
0.32658

0.13749
0.14698
0.18070
0.14706
0.12137
0.07592

0.1055;

OOOOOOOOOOOOOOUI-b-ANOOOOOOOOCOZ

APPENDIX A

Table 16. Catch/effort of lake trout in spring gill-net assessment fisheries in eastern

Wisconsin waters (W12) of Lake Superior (mean and SE across N lifts of
log-transformed values).

 

 

 

 

 

Total Wild Stocked
Year Mean SE Mean SE Mean SE N
1959 1.56379 0.11514 1.18054 0.11124 0.86882 0.10159 25
1960 2.22424 0.21331 1.73151 0.21328 1.56687 0.18961 40
1961 1.66293 0.14312 1.43376 0.14424 0.75453 0.09285 22
1962 2.48650 0.11003 2.26394 0.10555 1.21761 0.10374 24
1963 2.37126 0.15680 1.71750 0.14020 1.81157 0.14729 24
1964 2.95438 0.08891 2.08849 0.12321 2.34984 0.10930 32
1965 2.88788 0.15906 1.81951 0.10340 2.47144 0.18757 27
1966 3.35376 0.10317 1.80454 0.13770 3.11488 0.11955 14
1967 3.31669 0.14354 1.16256 0.10973 3.22265 0.14975 15
1968 2.80103 0.07271 0.36041 0.03872 2.76996 0.07349 82
1969 3.63678 0.08310 0.73358 0.07882 . 3.60421 0.08356 31
1970 4.17979 0.10140 1.69638 0.08977 4.10020 0.10432 39
1971 4.17873 0.11905 2.32817 0.12021 3.99952 0.12363 38
1972 3.76564 0.08295 2.25532 0.10366 3.51710 0.08292 38
1973 3.62363 0.10444 2.45782 0.11711 3.23917 0.10981 38
1974 3.70943 0.06833 2.30612 0.11772 3.38082 0.08599 36
1975 3.23261 0.08073 2.06895 0.13309 2.82559 0.07588 36
1976 3.33201 0.11083 2.31336 0.10930 2.87530 0.12932 36
1977 3.39590 0.08860 2.03258 0.15422 2.91034 0.12227 36
1978 3.46160 0.12417 2.19468 0.14573 3.02389 0.14647 36
1979 3.67895 0.07212 2.59976 0.12611 3.07706 0.12493 36
1980 3.42029 0.08798 2.50642 0.10018 2.78316 0.14791 28
1981 3.22153 0.13797 2.02474 0.18693 2.71305 0.19835 18
1982 2.52910 0.13323 1.49430 0.13442 2.15015 0.13726 30
1983 2.62689 0.13398 1.47825 0.18133 2.16257 0.17187 28
1984 2.97113 0.14812 1.59465 0.20089 2.60183 0.15362 30
1985 2.87342 0.14581 1.83430 0.16239 2.44888 0.14668 36
1986 3.36504 0.13622 2.58348 0.15390 2.74063 0.13017 40
1987 3.21625 0.15891 2.47915 0.20658 2.49202 0.13116 31
1988 3.01955 0.13036 2.36810 0.15199 2.24600 0.13026 31
1989 3.29216 0.16499 2.67114 0.18220 2.42473 0.16646 31
1990 3.06239 0.15467 2.49005 0.19969 2.10282 0.13182 31
1991 3.15796 0.15966 2.60588 0.21012 2.16677 0.11351 31
1992 3.08006 0.14477 2.61669 0.18521 1.93280 0.11039 31
1993 3.10047 0.10533 #72104 0.14013 1.84273 0.08831 31

 

 

93
APPENDIX A

Table 17. Results of the multiple regression of recruit CPE (log,) during 1967-93 on
native and stocked spawner CPEs (log,) during 1959-85 in Michigan areas MI4-MI7 of

Lake Superior.

 

EIGENVALUES OF UNIT SCALED X'X

1 2 3
2.71745 .19829 0.08426
CONDITION INDICES
1 2 3
1.00000 .70191 5.67897
VARIANCE PROPORTIONS
1 2 3
CONSTANT 0.01643 .00946 0.97412
NATIVE 0.02850 .70965 0.26185
STOCKED 0.02277 .37743 0.59980

DEP VAR:RECRUITS N:107 MULTIPLE R: 0.890
ADJUSTED SQUARED MULTIPLE R: 0.789

SQUARED MULTIPLE R: 0.793
STANDARD ERROR OF ESTIMATE: 0.54461

VARIABLE COEFFICIENT STD ERROR STD COEF TOLERANCE T P(2 TAIL)
CONSTANT -0.13959 0.14682 0.00000 . -0.95076 0.34393
NATIVE 0.32294 0.04949 0.29251 0.99083 6.52552 0.00000
STOCKED 0.66752 0.03678 0.81352 0.99083 .18E+02 0.00000
CORRELATION MATRIX OF REGRESSION COEFFICIENTS
CONSTANT NATIVE STOCKED
CONSTANT 1.00000
NATIVE -0.56533 1.00000
STOCKED -0.68529 -0.09575 1.00000
ANALYSIS OF VARIANCE

SOURCE SUM-OF-SQUARES DF MEAN-SQUARE F-RATIO P
REGRESSION 118.13219 2 59.06609 199.14349 0.00000
RESIDUAL 30.84647 104 0.29660
WARNING: CASE 95 IS AN OUTLIER (STUDENTIZED RESIDUAL = -3.108)
WARNING: CASE 108 IS AN OUTLIER (STUDENTIZED RESIDUAL = 2.662)
DURBIN-WATSON D STATISTIC 1.654
FIRST ORDER AUTOCORRELATION .155

 

94
APPENDIX A

Table 18. Results of the multiple regression of recruit CPE (log,) during 1967-93 on
native and stocked spawner CPEs (log,) during 1959-85 in Minnesota areas MNZ-MN3
of Lake Superior.

 

EIGENVALUES OF UNIT SCALED X'X

1 2 3
2.57711 0.34908 0.07380
CONDITION INDICES
1 2 3
1.00000 2.71708 5.90923
VARIANCE PROPORTIONS
1 2 3
CONSTANT 0.01688 0.00878 0.97435
NATIVE 0.04084 0.63215 0.32701
STOCKED 0.02342 0.17889 0.79769
DEP VAR:RECRUITS N: 46 MULTIPLE R: 0.802 SQUARED MULTIPLE R: 0.644

ADJUSTED SQUARED MULTIPLE R: 0.627 STANDARD ERROR OF ESTIMATE: 0.49353

VARIABLE COEFFICIENT STD ERROR STD COEF TOLERANCE T P(2 TAIL)

CONSTANT -0.58930 0.21141 0.00000 ‘ . -2.78739 0.00788
NATIVE 0.49141 0.11839 0.38600 0.95782 4.15062 0.00015
STOCKED 0.56539 0.06680 0.78715 0.95782 8.46426 0.00000

CORRELATION MATRIX OF REGRESSION COEFFICIENTS

CONSTANT NATIVE STOCKED
CONSTANT 1.00000
NATIVE -0.61269 1.00000
STOCKED -0.82210 0.20538 1.00000

ANALYSIS OF VARIANCE

SOURCE SUM-OF-SQUARES DF MEAN-SQUARE F-RATIO P
REGRESSION 18.93019 2 9.46510 38.85943 0.00000
RESIDUAL 10.47363 43 0.24357

WARNING: CASE 155 IS AN OUTLIER (STUDENTIZED RESIDUAL = -3.563)
WARNING: CASE 178 HAS LARGE LEVERAGE (LEVERAGE = 0.238)

DURBIN-WATSON D STATISTIC 1.150
FIRST ORDER AUTOCORRELATION .418

 

APPENDIX A

95

Table 19. Results of the multiple regression of recruit CPE (log,) during 1967-93 on
native and stocked spawner CPEs (loge) during 1959-85 in Wisconsin area W12 of

Lake Superior.

 

EIGENVALUES OF UNIT SCALED X'X

1 2 3
2.89864 0.06698 0.03438
CONDITION INDICES
1 2 3
1.00000 6.57850 9.18194
VARIANCE PROPORTIONS
1 2 3
CONSTANT 0.00604 0.00784 0.98612
NATIVE 0.00846 0.46551 0.52603
STOCKED 0.00988 0.73325 0.25687

DEP VAR:RECRUITS N: 27 MULTIPLE
ADJUSTED SQUARED MULTIPLE R: 0.184

R: 0.497 SQUARED MULTIPLE R: 0.247
STANDARD ERROR OF ESTIMATE: 0.54779

VARIABLE COEFFICIENT STD ERROR STD COEF TOLERANCE

P(2 TAIL)

CONSTANT 1.13280 0.46262 0.00000 . 2.44868 0.02202
NATIVE 0.00246 0.20344 0.00219 0.95694 0.01211 0.99044
STOCKED 0.35313 0.12893 0.49609 0.95694 2.73897 0.01144

CORRELATION MATRIX OF REGRESSION COEFFICIENTS

CONSTANT NATIVE STOCKED
CONSTANT 1.00000
NATIVE -0.65266 1.00000
STOCKED -0.57140 -0.20751 1.00000
ANALYSIS OF VARIANCE

SOURCE SUM-OF-SQUARES DF MEAN-SQUARE F-RATIO P
REGRESSION 2.35683 2 1.17842 3.92704 0.03346
RESIDUAL 7.20186 24 0.30008

WARNING: CASE 201 IS AN OUTLIER (STUDENTIZED RESIDUAL = -2.937)

DURBIN-WATSON D STATISTIC .642
FIRST ORDER AUTOCORRELATION .637

 

APPENDIX A

96

Table 20. Results of the multiple regression of recruitment rate (log,) of the 1963-82
year classes on numbers of yearlings stocked and large-mesh gill-net fishing effort in

Michigan waters of Lake Superior.

 

EIGENVALUES OF UNIT SCALED X'X

1
2.18925
CONDITION INDICES
1
1.00000
VARIANCE PROPORTIONS
1
CONSTANT 0.02408
YEARLING 0.02614
GILLNETS 0.05086

DEP VAR:SURVIVAL

N: 20 MULTIPLE

2 3
0.74232 0.06842

2 3
1.71732 5.65647

2 3
0.00389 0.97203
0.04690 0.92696
0.56388 0.38526

R: 0.954 SQUARED MULTIPLE R: 0.911

ADJUSTED SQUARED MULTIPLE R: 0.900 STANDARD ERROR OF ESTIMATE: 0.28529
VARIABLE COEFFICIENT STD ERROR STD COEF TOLERANCE T P(2 TAIL)
CONSTANT 3.90143 0.18255 0.00000 . .21E+02 0.00000
YEARLING -0.35604 0.12942 -0.22594 0.77774 -2.75098 0.01364
GILLNETS -0.55799 0.04407 -1.03985 0.77774 -.13E+02 0.00000

CORRELATION MATRIX OF REGRESSION COEFFICIENTS

CONSTANT YEARLING GILLNETS

CONSTANT 1.00000

YEARLING -0.91064 1.00000

GILLNETS -0.62376 0.47145 1.00000

ANALYSIS OF VARIANCE

SOURCE SUM-OF-SQUARES DF MEAN-SQUARE F-RATIO P
REGRESSION 14.13014 2 7.06507 86.80651 0.00000
RESIDUAL 1.38361 17 0.08139
DURBIN-WATSON D STATISTIC 2.447
FIRST ORDER AUTOCORRELATION -.244

 

APPENDIX A

97

Table 21. Results of the multiple regression of recruitment rate (log,) of the 1963-82
year classes on numbers of yearlings stocked and density (CPE) of wild adult lake
trout in Minnesota waters of Lake Superior.

 

EIGENVALUES OF UNIT SCALED X'X

1 2 3
2.40207 0.56615 0.03178
CONDITION INDICES
l 2 3
1.00000 2.05980 8.69427
VARIANCE PROPORTIONS
1 2 3
CONSTANT 0.00953 0.00754 0.98293
YEARLING 0.01038 0.02096 0.96866
WILDPRED 0.05756 0.76723 0.17521
DEP VAR:SURVIVAL N: 20 MULTIPLE R: 0.839 SQUARED MULTIPLE R: 0.703
ADJUSTED SQUARED MULTIPLE R: 0.668 STANDARD ERROR OF ESTIMATE: 0.31606
VARIABLE COEFFICIENT STD ERROR STD COEF TOLERANCE T P(2 TAIL)
CONSTANT 2.79147 0.29246 0.00000 . 9.54490 0.00000
YEARLING 4.09200 0.95323 0.59640 0.90415 4.29277 0.00049
WILDPRED -0.09290 0.02979 -0.43320 0.90415 -3.11812 0.00626
CORRELATION MATRIX OF REGRESSION COEFFICIENTS
CONSTANT YEARLING WILDPRED
CONSTANT 1.00000
YEARLING -0.95325 1.00000
WILDPRED -0.46765 0.30959 1.00000
ANALYSIS OF VARIANCE
SOURCE SUM-OF-SQUARES DF MEAN-SQUARE F-RATIO P
REGRESSION 4.02587 2 2.01294 20.15074 0.00003
RESIDUAL 1.69820 17 0.09989

WARNING: CASE 1 HAS LARGE LEVERAGE (LEVERAGE = 0.862)

DURBIN-WATSON D STATISTIC
FIRST ORDER AUTOCORRELATION

1.800
.040

 

98
APPENDIX A

Table 22. Results of the multiple regression of recruitment rate (log) of the 1963-82
year classes on numbers of yearlings stocked and large-mesh gill-net fishing effort in
Wisconsin waters of Lake Superior.

 

EIGENVALUES OF UNIT SCALED X'X

1 2 3
2.74717 0.19692 0.05591
CONDITION INDICES
1 2 3
1.00000 3.73509 7.00940
VARIANCE PROPORTIONS
1 2 3
CONSTANT 0.01094 0.00291 0.98615
YEARLING 0.01808 0.33191 0.65000
GILLNETS 0.02311 0.56335 0.41354

DEP VAR:SURVIVAL N: 20 MULTIPLE

R: 0.822 SQUARED MULTIPLE R: 0.676

ADJUSTED SQUARED MULTIPLE R:0.638 STANDARD ERROR OF ESTIMATE: 0.48235
VARIABLE COEFFICIENT STD ERROR STD COEF TOLERANCE T P(2 TAIL)
CONSTANT 4.81613 0.36827 0.00000 . .13E+02 0.00000
YEARLING -0.93529 0.74312 -0.17471 0.98866 -1.25860 0.22519
GILLNETS -0.12848 0.02169 -0.82233 0.98866 -5.92419 0.00002

CORRELATION MATRIX OF REGRESSION COEFFICIENTS

CONSTANT YEARLING GILLNETS

CONSTANT 1.00000

YEARLING -0.75546 1.00000

GILLNETS -0.66321 0.10649 1.00000

ANALYSIS OF VARIANCE

SOURCE SUM-OF-SQUARES DF MEAN-SQUARE F-RATIO P
REGRESSION 8.25806 2 4.12903 17.74731 0.00007
RESIDUAL 3.95516 17 0.23266

WARNING: CASE 1 HAS LARGE LEVERAGE (LEVERAGE = 0.547)

DURBIN-WATSON D STATISTIC
FIRST ORDER.AUTOCORRELATION

2.244
-.135

 

APPENDIX A

99

Table 23. Commercial fishery lake trout catch (pounds) and large-mesh gill-net effort

(1,000 feet) in Michigan statistical districts MS-l through MS-3 of Lake Superior

during 1929-61 (compiled from Jensen and Buettner 1976).

 

 

 

 

MS-l MS-2 MS-3

Year Catch Effort C/f Catch Effort C/f Catch Effort C/f

1929 306,596 13,993 219 61,700 4,139 149 319,275 15,853 201
1930 263,783 12,988 203 93,793 5,938 158 323,978 16,429 197
1931 188,821 10,586 178 65,818 4,808 137 438,577 23,304 188
1932 203,689 8,892 229 38,896 2,723 143 361,739 24,245 149
1933 220,566 10,259 215 47,993 2,324 207 328,570 19,454 169
1934 256,388 10,473 245 47,411 2,563 185 360,858 18,402 196
1935 302,833 11,183 271 44,013 2,870 153 325,075 20,040 162
1936 262,293 9,970 263 112,652 6,366 177 325,946 18,271 178
1937 222,586 10,094 221 108,642 7,560 144 252,051 17,601 143
1938 234,901 10,553 223 67,001 5,091 132 225,592 18,311 123
1939 218,379 11,117 196 91,232 5,351 170 221,538 18,919 117
1940 213,902 10,368 206 122,597 7,798 157 240,754 20,973 115
1941 205,653 9,386 219 102,214 7,284 140 322,943 25,877 125
1942 238,812 9,639 248 81,099 5,253 154 444,437 29,728 150
1943 257,240 9,381 274 96,842 5,646 172 496,905 33,149 150
1944 339,394 12,570 270 143,397 6,450 222 590,641 35,474 166
1945 271,017 10,372 261 159,697 7,973 200 611,669 39,642 154
1946 263,141 12,717 207 147,082 10,200 144 659,628 50,146 132
1947 242,180 10,729 226 127,391 9,516 134 583,512 45,339 129
1948 286,004 12,156 235 142,452 11,295 126 496,398 49,197 101
1949 236,469 12,755 185 125,680 9,108 138 558,009 54,818 102
1950 293,943 11,399 196 112,120 7,237 118 704,146 54,489 98
1951 298,186 13,151 140 103,475 7,055 90 659,034 45,568 89
1952 343,724 15,154 101 75,085 5,637 59 649,508 55,071 52
1953 288,873 15,644 82 62,304 4,430 63 572,158 52,413 49
1954 249,482 15,241 73 37,758 3,196 53 543,795 52,088 46
1955 211,828 11,847 79 57,113 3,229 79 491,937 49,140 44
1956 214,202 16,858 56 89,038 4,494 88 384,114 39,701 43
1957 80,525 6,976 51 35,668 2,601 61 323,711 31,331 46
1958 57,940 4,445 58 30,071 1,842 73 311,422 30,036 46
1959 21,165 2,390 39 12,640 952 59 253,463 24,306 46
1960 11,063 1,542 32 5,112 683 33 108,743 14,448 33
1961 7,980 1,190 30 4,108 520 35 88,591 12,650 31

 

APPENDIX A

100

Table 24. Commercial fishery lake trout catch (pounds) and large-mesh gill-net effort

(1,000 feet) in Michigan statistical districts MS-4 through MS-6 of Lake Superior

during 1929-61 (compiled from Jensen and Buettner 1976).

 

 

 

 

 

MS-4 MS-S MS-6

Year Catch Effort C/f Catch Effort C/f Catch Effort C/f

1929 159,686 6,610 242 377,323 10,321 366 12,231 587 208
1930 249,042 9,612 259 292,085 9,590 305 18,065 783 231
1931 271,700 11,701 232 367,330 10,150 362 38,646 1,721 225
1932 231,366 10,955 211 520,108 11,718 444 62,322 2,568 243
1933 182,149 8,035 227 293,138 5,650 519 98,856 3,031 326
1934 210,721 6,839 308 516,675 9,955 519 86,819 2,461 353
1935 210,81 1 9,792 215 493,884 12,739 388 57,652 1,996 289
1936 196,174 8,937 220 335,200 10,716 313 100,533 3,916 257
1937 169,728 9,792 173 389,099 13,099 297 109,258 4,862 225
1938 177,1 18 1 1,343 156 360,967 17,747 203 73,276 3,220 228
1939 216,422 13,244 163 372,646 18,092 206 21,984 1,542 143
1940 213,459 12,710 168 340,563 14,020 243 63,764 3,347 191
1941 252,311 15,911 159 369,341 14,222 260 119,411 5,705 209
1942 282,828 16,803 168 422,361 14,567 290 98,190 4,522 217
1943 256,610 15,015 171 390,692 10,865 360 107,001 4,190 255
1944 291,167 17,988 162 491,733 14,015 351 127,770 5,850 218
1945 345,763 19,612 176 323,181 11,280 287 135,941 6,519 209
1946 349,819 25,198 139 335,623 17,048 197 192,135 8,843 217
1947 259,940 16,241 160 272,073 13,669 199 114,707 5,192 221
1948 245,241 20,913 1 17 299,473 15,896 188 140,196 7,661 183
1949 283,864 24,501 1 16 341,650 21,315 160 125,009 6,306 198
1950 382,602 26,861 109 311,622 15,480 153 164,400 6,482 193
1951 464,708 31,022 92 279,019 17,009 101 138,104 7,299 116
1952 455,988 32,617 62 214,249 11,119 86 144,227 6,168 104
1953 388,830 32,032 54 177,932 7,769 102 151,604 6,560 103
1954 331,568 29,909 49 222,217 1 1,214 88 151,681 7,027 96
1955 293,287 26,171 50 189,428 8,562 98 85,641 5,015 76
1956 260,844 23,81 1 49 163,153 7,671 95 73,180 4,748 69
1957 208,638 22,577 41 136,220 6,307 96 54,847 4,144 59
1958 188,162 18,515 45 135,155 7,215 83 35,557 2,154 73
1959 208,260 21,413 43 136,644 6,601 92 30,748 2,068 66
1960 68,065 10,702 28 67,094 4,821 62 5,798 444 58
1961 53,126 9,234 26 48,498 3,006 72 14,365 1,131 56

 

APPENDIX B - ADDITIONAL FIGURES

101
APPENDIX B - ADDITIONAL FIGURES

 

 

 

NATIVE
”I, , STOCKED
03 O. 0 ° 50
g 0° 0. . “’0
Q .9 O Q ‘0
0 O o
3 0 ~ 6”
0 ° 9.
we 0 °
. 0.0 9 0
o
.2»,
V1,; .
. °, ,°. RECRUITS
o 0 o 00 . ::§:.
0 o o
.o.? ..°".w : O... ’0‘.
5900$Yԡ:..: 90 ӣ309.
- o
1 a... ‘ O O
‘ o. 9” °
8 o
O. .9. 9 v0

 

 

 

 

 

Figure 24. Scatter-plot matrix of native and stocked spawner CPEs (log,) during
1959-85 and recruit CPE (log,) during 1967-93 in Michigan areas MI4-MI7 of Lake
Superior.

102

 

 

 

 

 

 

 

 

 

 

 

 

 

APPENDIX B
NATIVE
T
1
In [1
° STOCKED
0 Q 9
0:9? 0
13$ 4»
00 o
o $ 0. o .—
9:0 9
o
. 0°. . ° Fi—l—I
° . ° RECRUITS
o
O 0 . ”.9
o o
«:9 . . 990° _
.9 9: .30 _
O 9 o o o
.0... .9. 0 ° 0 o. :. o 3.90
‘6. . 9° 0. :9 0:..20 ll

 

 

 

 

 

 

Figure 25. Scatter-plot matrix of native and stocked spawner CPEs (log,) during
1959-85 and recruit CPE (log,) during 1967-93 in Minnesota areas MN2-MN3 of Lake

Superior.

103

 

 

 

 

 

 

 

 

 

APPENDIXB
NATIVE
I_I_I - ,
° 0 STOCKED
o 9 ’.° 0.
o . '—
o
O I I | I I I —I_I_‘I
:6 °°oo . 33° 3’ o o RECRUITS
o ‘6: 0 ° 6 ,0 8 o o
o 0 o t o
.3 ° : .
° ° m r‘I_-I"I—

 

 

 

 

 

 

 

Figure 26. Scatter-plot matrix of native and stocked spawner CPEs (log,) during
1959-85 and recnrit CPE (log,) during 1967-93 in Wisconsin area W12 of Lake
Superior.

104

APPENDIX B

 

 

O
.0

0
O
Q
.0

 

 

 

. o
o

S 0 .°

0’ o ,

8 o ,
0.60 0 .° °

‘0’ 9 Q ...
o
u 03 o ’ am
0 O

o o . ’ . ° ’ . o
. ' ' ’. . o .0 ° 0 o

9 o

. O o. O

O

0 o

o. . ° ' .' . ° . 005m;
0 . .0 fl . O ..

.. o o o ,
o o o
O O. . .0.

 

o
0 o
o
o o
o . . .
o
o . o .
o. . ° 0 o
o . 9 o .
O o O .

 

 

 

 

 

 

 

 

 

 

WAL
o o ‘ o o . . rrl ”Ah
Figure 27. Scatter-plot matrix of stocked lake trout survival, millions of yearlings

stocked, average grams/yearling, wild adult lake trout CPE, stocked adult lake trout
CPE, wild juvenile lake trout CPE, and large-mesh gill-net fishing effort in Michigan

waters of Lake Superior during 1963-89.

105

 

 

 

 

 

 

 

 

 

 

APPENDIXB
m
nJIIIII
”.0 m
o 2 9 '
. rn IIIII-I
° ° m
. 0. ;.:~ . ..°’.: n
0. °. ; am
a.” ' . 3
o 0‘ O . . .’. IF. 0
° ° ° ° camera
I 9‘ “0‘ o ’ 0. °.... & 0 0 *O”. 1|”
m
o 0. 00.00. 00.0- ... 0.0000000. .00. o
8 ° 0 ‘o ° '. WM.
. e s .r. ca . I
’ o 0 0. {.0 0 o. o 0 ’0 ° 1
o o .0. 0 ° . 8 .
. . . ~ . .o. : : deIIII

 

 

 

 

 

 

 

Figure 28. Scatter-plot matrix of stocked lake trout survival, millions of yearlings
stocked, average grams/yearling, wild adult lake trout CPE, stocked adult lake trout
CPE, wild juvenile lake trout CPE, and large-mesh gill-net fishing effort in Minnesota
waters of Lake Superior during 1963-89.

106

APPENDIX B

 

 

 

0 ° 0’
o o o o 0 O
. o .0
o 0 ° ’ ° *0 o
o o o o

 

 

0 o 00 0. o 0
.0 ° 3 o . o 0 o .9
0.... . .0. o o .0: .00. o
o o o o

 

4'.

’o

.

o

o

o

O

O ..Q
”o o

o

 

o o o o
O 0 o , o. o o 9 . o .9

O O 0 O O
0. o .0 o o o .3. o o o 00
O. o O O o O o o 0 0‘

o. o o . o . o . o
O O O O O 0

Figure 29. Scatter-plot matrix of stocked lake trout survival, millions of yearlings
stocked, average grams/yearling, wild adult lake trout CPE, stocked adult lake trout
CPE, wild juvenile lake trout CPE, and large-mesh gill-net fishing effort in Wisconsin
waters of Lake Superior during 1963-89.

 

 

 

 

 

 

 

 

 

LIST OF REFERENCES

LIST OF REFERENCES

Bailey, M. M. 1965. Lake trout fin-clipping rates at two national fish hatcheries.
Progressive Fish-Culturist 27:169-170.

Bailey, R M., and G. R Smith. 1981. Origin and geography of the fish fauna of the
Laurentian Great Lakes Basin. Canadian Journal of Fisheries and Aquatic
Sciences 38:1539-1561.

Baldwin, N. S., R. W. Saalfeld, M. A. Ross, and H. J. Buettner. 1979. Commercial
fish production in the Great Lakes 1867-1977. Great Lakes Fishery
Commission Technical Report 3.

Buettner, H. J. 1961. Recoveries of tagged, hatchery-reared lake trout from Lake
Superior. Transactions of the American Fisheries Society 90:404-412.

Bumham-Curtis, M. K. 1993. Intralacustrine speciation of Salvelinus namaycush in
Lake Superior: an investigation of genetic and morphological variation and
evolution of lake trout in the Laurentian Great Lakes. Doctoral dissertation.
University of Michigan, Ann Arbor.

Bumham-Curtis, M. K., and G. R. Smith. 1994. Osteological evidence of genetic
divergence of lake trout (Salvelinus namaycush) in Lake Superior. Copeia
1994:843-850.

Busiahn, T. R, ed. 1990. Fish community objectives for Lake Superior. Great Lakes
Fishery Commission Special Publication 90-1.

Coberly, C. E., and R M. Horrall. 1980. Fish spawning grounds in Wisconsin waters
of the Great Lakes. University of Wisconsin Sea Grant Institute,
WIS-SG-80-235, Madison.

Coble, D. W., R. E. Bruesewitz, T. W. Pratt, and J. W. Scheirer. 1990. Lake trout,
sea lampreys, and overfishing in the upper Great Lakes: a review and
reanalysis. Transactions of the American Fisheries Society 1192985-995.

Conner, D. J., C. R Bronte, J. H. Selgeby, and H. L. Collins. 1993. Food of
salmonine predators in Lake Superior, 1981-87. Great Lakes Fishery

107

108

Commission Technical Report 59.

Curtis, G. L., J. H. Selgeby, and R. G. Schorfhaar. in press. Decline and recovery of
lake trout populations near Isle Royale, Lake Superior, 1929-1990.
Transactions of the American Fisheries Society.

Draper, N. R., and H. Smith. 1981. Applied regression analysis, second edition.
John Wiley & Sons, New York.

Dryer, W. R 1966. Bathymetric distribution of fish in the Apostle Islands region,
Lake Superior. Transactions of the American Fisheries Society 95:248-259.

Dryer, W. R, L. F. Erkkila. and C. L. Tetzloff. 1965. Food of the lake trout in Lake
Superior. Transactions of the American Fisheries Society 94:169-176.

Dryer, W. R, and G. R King. 1968. Rehabilitation of lake trout in the Apostle
Islands region of Lake Superior. Journal of the Fisheries Board of Canada
25:1377-1403.

Ebener, M. P. 1990. Assessment and mark-recapture of lake trout spawning stocks
around the Keweenaw Peninsula area of Lake Superior. Great Lakes Indian
Fish and Wildlife Commission Biological Services Division Administrative
Report 90-8, Odanah, Wisconsin.

Edsall, T. A., C.L. Brown, G. W. Kennedy, and J. R. P. French, III. 1992. Swfzcial
Substrates and Bathym etry of Five Historical Lake Trout Spawning Reefs in
Near-shore Waters of the Great Lakes. Great Lakes Fishery Commission
Technical Report 58.

Elrod, J. H., D. E. Ostergaard, and C. P. Schneider. 1988. Comparison of
hatchery-reared lake trout stocked as fall fingerlings and as spring yearlings in
Lake Ontario. North American Journal of Fisheries Management 8:455-462.

Elrod, J. H., C. P. Schneider, and D. E. Ostergaard. 1993. Survival of lake trout
stocked in U.S. Waters of Lake Ontario. North American Journal of Fisheries
Management 13:775-781.

Eschmeyer, P. H. 1955. The reproduction of lake trout in southern Lake Superior.
Transactions of the American Fisheries Society 84:47-74.

Eschmeyer, P. H., R. Daly, and L. F. Erkkila. 1953. The movement of tagged lake
trout in Lake Superior, 1950-1952. Transactions of the American Fisheries
Society 82:68-77.

Eshenroder, R L., T. P. Poe, and C. H. Olver. 1983. Strategies for rehabilitation of

109

lake trout in the Great Lakes: proceedings of a conference on lake trout
research. Great Lakes Fishery Commission Technical Report 40.

Evans, D. 0., and C. C. Willox. 1991. Loss of exploited indigenous populations of
lake trout, Salvelinus namaycush, by stocking of non-native stock. Canadian
Journal of Fisheries and Aquatic Sciences 43(Supplement 1):134-147.

Gallinat, M. P. 1993. Food habits of lake trout in management units WI-2 and MI-2
of Lake Superior, 1992. Red Cliff Fisheries Department Assessment Report
93-2, Bayfield, Wisconsin.

Goodier, J. L. 1981. Native lake trout (Salvelinus namaycush) stocks in the Canadian
waters of Lake Superior prior to 1955. Canadian Journal of Fisheries and
Aquatic Sciences 38:1724-1737.

Goodier, J. L. 1982. The fish and fisheries of Canadian Lake Superior. University of
Toronto, Institute of Environmental Studies, Toronto.

Goodier, J. L. 1989. Fishermen and their trade on Canadian Lake Superior: one
hundred years. Inland Seas 45:284-306.

Goodyear, C. S., T. A. Edsall, D. M. Orrnsby Dempsey, G. D. Moss, and P. E.
Polanski. 1982. Atlas of spawning and nursery areas of Great Lakes fishes,
volume 11: Lake Superior. U.S. Fish and Wildlife Service, FWS/OBS-82-52,
Washington, DC.

GLFC (Great Lakes Fishery Commission). 1980. A joint strategic plan for
management of Great Lakes fisheries. Great Lakes Fishery Commission, Ann
Arbor, Michigan.

Gunn, J. M., M. J. McMurtry, J. N. Bowlby, J. M. Casselman, and V. A. Liimatainen.
1987. Survival and growth of stocked lake trout in relation to body size,
stocked season, lake acidity, and biomass of competitors. Transactions of the
American Fisheries Society 116:618-627.

Hamley, J. M. 1975. Review of gillnet selectivity. Journal of the Fisheries Research
Board of Canada 32:1943-1969.

Hansen, M. J., editor. 1990. Lake Superior: the state of the lake in 1989. Great
Lakes Fishery Commission Special Publication 90-3.

Hansen, M. J., editor. 1994. The state of Lake Superior in 1992. Great Lakes
Fishery Commission Special Publication 94-1.

Hansen, M. J., M. P. Ebener, R. G. Schorfhaar, S. T. Schram, D. R Schreiner, and J.

110

H. Selgeby. 1994a. Declining survival of lake trout stocked during 1963-1986
in U.S. waters of Lake Superior. North American Journal of Fisheries
Management 14:000-000.

Hansen, M. J., M. P. Ebener, J. D. Shively, and B. L. Swanson. 1994b. Lake trout.
Pages 10-30 in M. J. Hansen, editor. The state of Lake Superior in 1992.
Great Lakes Fishery Commission Special Publication 94-1.

Hansen, M. J., and eleven coauthors. in press. Lake trout (Salvelinus namaycush)
populations in Lake Superior and their restoration during 1959-1993. Journal
of Great Lakes Research.

Henderson, H. V., and P. F. Velleman. 1981. Building multiple regression models
interactively. Biometrics 37:391-411.

Hilbom, R, and C. J. Walters. 1992. Quantitative fisheries stock assessment: choice,
dynamics and uncertainty. Chapman and Hall, New York.

Hile, R 1962. Collection and analysis of commercial fishery statistics in the Great
Lakes. Great Lakes Fishery Commission Technical Report 5.

Hile, R, P. H. Eschmeyer, and G. F. Lunger. 1951. Status of the lake trout fishery in
Lake Superior. Transactions of the American Fisheries Society 80:278-312.

Jensen, A. L. 1978. Assessment of the lake trout fishery in Lake Superior,
1929-1950. Transactions of the American Fisheries Society 107:543-549.

Jensen, A. L., and H. J. Buettner. 1976. Lake trout, whitefish, chubs, and lake
herring yield and effort data for state of Michigan waters of the upper Great
Lakes: 1929-1973. Michigan Sea Grant Technical Report 52.

Khan, N. Y., and S. U. Qadri. 1970. Morphological differences in Lake Superior lake
char. Journal of the Fisheries Research Board of Canada 27:161-167.

Kirby, K. N. 1993. Advanced data analysis with SYSTAT. Van Nostrand Reinhold,
New York.

Klar, G. T., and J. G. Weise. 1994. Sea lamprey. Pages 71-80 in M. J. Hansen,
editor. The state of Lake Superior in 1992. Great Lakes Fishery Commission,
Special Publication 94-1.

Krueger, C. C., and B. May. 1991. Ecological and genetic effects of salmonid
introductions in North America. Canadian Journal of Fisheries and Aquatic
Sciences 43(Supplement 1):66-77.

111

Krueger, C. C., B. L. Swanson, and J. H. Selgeby. 1986. Evaluation of
hatchery-reared lake trout for reestablishment of populations in the Apostle
Islands region of Lake Superior, 1960-84. Pages 93-107 in R H. Stroud,
editor. Fish culture in fisheries management. American Fisheries Society, Fish
Culture Section and Fisheries Management Section, Bethesda, Maryland.

Lawrie, A. H. 1978. The fish community of Lake Superior. Journal of Great Lakes
Research 4:513-549.

Lawrie, A. H., and J. F. Rahrer. 1972. Lake Superior: effects of exploitation and
introductions on the salmonid community. Journal of the Fisheries Research
Board of Canada 29:765-776.

Lawrie, A. H., and J. F. Rahrer. 1973. Lake Superior: a case history of the lake and
its fisheries. Great Lakes Fishery Commission Technical Report 19.

Loftus, K. H. 1958. Studies on river-spawning populations of lake trout in eastern
Lake Superior. Transactions of the American Fisheries Society 87:259-277.

LSLTTC (Lake Superior Lake Trout Technical Committee). 1986. A lake trout
rehabilitation plan for Lake Superior. Pages 47-62 in S. D. Morse, editor.
Lake Superior Committee 1986 annual meeting minutes. Great Lakes Fishery
Commission, Ann Arbor, Michigan.

MacCallum, W. R, and J. H. Selgeby. 1987. Lake Superior revisited 1984.
Canadian Journal of Fisheries and Aquatic Sciences 44(Supplement 2):23-36.

MacLean, J. A., D. 0. Evans, N. V. Martin, and R. L. DesJardine. 1981. Survival,
growth, spawning distribution, and movements of introduced and native lake
trout (Salvelinus namaycush) in two inland Ontario lakes. Canadian Journal of
Fisheries and Aquatic Sciences 38:1685-1700.

Martin, N. V., and C. H. Olver. 1980. The lake charr, Salvelinus namaycush. Pages
205-227 in E. K. Balon, editor. Charrs: salmonid fishes of the genus
Salvelinus. Junk Publications, The Hague, Netherlands.

Peck, J. W. 1979. Utilization of traditional spawning reefs by hatchery lake trout in
the upper Great Lakes. Michigan Department of Natural Resources Fisheries
Research Report 1871.

Peck, J. W. 1986. Dynamics of reproduction by hatchery lake trout on a man-made
spawning reef. Journal of Great Lakes Research 12:293-303.

Peck, J. W. 1988. Fecundity of hatchery and wild lake trout in Lake Superior.
Journal of Great Lakes Research 14:9-13.

112

Peck, J. W., and R G. Schorfhaar. 1991. Assessment and management of lake trout
stocks in Michigan waters of Lake Superior. Michigan Department of Natural
Resources Fisheries Research Report 1956.

Plosila, D. S. 1977. Relationship of strain and size at stocking to survival of lake
trout in Adirondack lakes. New York Fish and Game Journal 24:1-24.

Powell, M. J., and six coauthors. 1986. Returns of hatchery-reared lake trout from
eight lakes in northeastern Ontario. Ontario Fisheries Technical Report 22.

Purych, P. R 1977. Poor returns of hatchery-reared lake trout to the sport fishery of
Flack Lake, Ontario, 1968-74. Progressive Fish-Culturist 39:185-186.

Pycha, R L. 1962. The relative efficiency of nylon and cotton gill nets for taking
lake trout in Lake Superior. Journal of the Fisheries Board of Canada
19:1085-1094.

Pycha, R L. 1980. Changes in mortality of lake trout (Salvelinus namaycush) in
Michigan waters of Lake Superior in relation to sea lamprey (Petromyzon
marinas) predation, 1968-78. Canadian Journal of Fisheries and Aquatic
Sciences 37:2063-2073.

Pycha, R L., W. R. Dryer, and G. R King. 1965. Movements of hatchery-reared
lake trout in Lake Superior. Journal of the Fisheries Research Board of Canada
22:999-1024.

Pycha, R. L., and G. R King. 1967. Returns of hatchery-reared lake trout in southern
Lake Superior, 1955-62. Journal of the Fisheries Research Board of Canada
24:281-298.

Pycha, R L., and G. R King. 1975. Changes in the lake trout population of southern
Lake Superior in relation to the fishery, the sea lamprey, and stocking,
1950-70. Great Lakes Fishery Commission Technical Report 28.

Rahrer, J. F. 1967. Growth of lake trout in Lake Superior before the maximum
abundance of sea lampreys. Transactions of the American Fisheries Society
96:268-277.

Rahrer, J. F. 1968. Movements of adult lake trout in Lake Superior. Transactions of
the American Fisheries Society 97:481-484.

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

Rybicki, R. W., and M. Keller. 1978. The lake trout resource in Michigan waters of

113

Lake Michigan, 1970-1976. Michigan Department of Natural Resources
Fisheries Research Report 1863.

Sakagawa, G. T. 1967. An analysis of catch per effort as used to determine

abundance, with particular reference to the lake trout, Salvelinus namaycush, of
Lake Superior, 1939-45. M.S. thesis. University of Michigan, Ann Arbor.

Sakagawa, G. T., and R L. Pycha. 1971. Population biology of lake trout (Salvelinus
namaycush) of Lake Superior before 1950. Journal of the Fisheries Board of
Canada 28:65-71.

Schorfhaar, R. G. and J. W. Peck. 1993. Catch and mortality of non-target species in
lake whitefish trap nets in Michigan waters of Lake Superior. Michigan
Department of Natural Resources Fisheries Research Report 1974.

Schram, S. T. 1993. Fecundity and egg deposition of a wild Lake Superior lake trout
stock. Wisconsin Department of Natural Resources Fisheries Management
Report No. 149.

Schram, S. T., J. H. Selgeby, B. L. Swanson, and C. R Bronte. in press. Stock size,
egg deposition, and recruitment of the Gull Island Shoal lake trout population,
1972-1992. Journal of Great Lakes Research.

Smith, B. R. 1971. Sea lampreys in the Great Lakes of North America. Pages
207-247 in M. W. Hardisty, and LC. Potter, editors. The biology of lampreys,
volume 1. Academic Press, New York.

Smith, 18. R, and J. J. Tibbles. 1980. Sea lamprey (Petmmyzon marinas) in lakes
Huron, Michigan, and Superior: history of invasion and control, 1936-78.
Canadian Journal of Fisheries and Aquatic Sciences 37:1780-1801.

Smith, B. R, J. J. Tibbles, and B. G. H. Johnson. 1974. Control of the sea lamprey
(Petromyzon marinas) in Lake Superior, 1953-70. Great Lakes Fishery
Commission Technical Report 26.

Smith, G. R, and T. N. Todd. 1984. Evolution of species flocks of fishes in north
temperate lakes. Pages 45-68 in A. A. Echelle and I. Komfield, editors.
Evolution of fish species flocks. University of Maine Press, Orono.

Sokal, RR, and F. J. Rohlf. 1981. Biometry: the principles and practice of statistics
in biological research, second edition. W. H. Freeman, New York.

Swanson, B. L. 1973. Lake trout homing, migration and mortality studies, Lake
Superior. Wisconsin Department of Natural Resources Fish Management
Report 65.

114

Swanson, B. L., and D. V. Swedberg. 1980. Decline and recovery of the Lake
Superior Gull Island Reef lake trout (Salvelinus namaycush) population and the
role of sea lamprey (Petromyzon marinas) predation. Canadian Journal of
Fisheries and Aquatic Sciences 37:2074-2080.

Systat, Inc. 1992. SYSTAT for Windows, Version 5 edition. Systat, Inc. Evanston,
Illinois.

Thibodeau, M. L., and J. R. M. Kelso. 1990. An evaluation of putative lake trout
(Salvelinus namaycush) spawning sites in the Great Lakes. Department of
Fisheries and Oceans Canada Technical Report of Fisheries and Aquatic
Sciences 1739.

Todd, T. N. and G. R. Smith. 1980. Differentiation in Coregonus zenithicus in Lake
Superior. Canadian Journal of Fisheries and Aquatic Sciences 37:2228-2235.

Tyler, A. V. 1992. A context for recruitment correlations: why marine fisheries
biologists should still look for them. Fisheries Oceanography 1:97-107.

Tyler, A. V., and W. R. Crawford. 1991. Modeling of recruitment patterns in Pacific
cod (Gadus macmcephalus) in Hecate Strait, British Columbia Canadian
Journal of Fisheries and Aquatic Sciences 48:2240-2249.

Walters, C. J., M. Stocker, A. V. Tyler, and S. J. Westrheim. 1986. Interaction
between Pacific cod (Gadus macrocephalus) and herring (Clupea harengus
paIIasi) in the Hecate Strait, British Columbia. Canadian Journal of Fisheries
and Aquatic Sciences 43:830-837.

"IIIIIIIIIIIIIIIIIIIIII