ax..-

WWIWWWWWWWWIWWWWIWWWWWWWIWWIWWW

'WE319310503 7265

A
.L-

 

 

 

W LmRARY W
W Michigan State

1 University

‘ -7 e w w..._r

 

This is to certify that the

thesis entitled

POPULATION DYNAMICS AND STOCK DIFFERENTIATION
OF LAKE NHITEFISH, COREGONUS CLUPEAFORMIS,
IN NORTutASTERN LAKE MICHIGAN

presented by

Paul David Scheerer

has been accepted towards fulfillment
of the requirements for

M.S. Fisheries & Wildlife

degree in

 

 

 

 

DateA ust 30 1982

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

MSU

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from

 

 

 

LIBRARIES
.fizulzgszsl. .your record. FINES will
be charged if book is
returned after the date
stamped below.
in? 2 Ct” !

VI
’ , '9'“ on

4‘ .. f...“ , ,,-,_A
I“? I‘i‘ & :‘li *‘c

 

"o
E}

 

 

POPULATION DYNAMICS AND STOCK DIFFERENTIATION
OF LAKE WHITEFISH, COREGONUS CLUPEAFORMIS,

IN NORTHEASTERN LAKE MICHIGAN

BY

Paul David Scheerer

A THESIS

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

MASTER OF SCIENCE

Department of Fisheries and Wildlife

1982

ABSTRACT
POPULATION DYNAMICS AND STOCK DIFFERENTIATION

OF LAKE WHITEFISH, COREGONUS CLUPEAFORMIS,
IN NORTHEASTERN LAKE MICHIGAN

BY

Paul David Scheerer

Lake whitefish, Coregonus clupeaformis, were tagged and
the commercial trap net catch was sampled in the North
Shore, Leland, Beaver Island (BI), and Grand Traverse Bay
(GTB) areas of Lake Michigan, to determine vital statistics
and differentiate discrete stocks.

Distribution of tag returns and statistical comparisons
of certain population parameters indicate the existence of
three stocks in these areas. The North Shore stock was
considerably larger in numbers and biomass than the Leland
stock.

Average exploitation rate was higher for the North
Shore stock (49.6%) than the Leland stock (24.2%).

The Leland and GT3 catch had broader age compositions,
older fish, and larger fish than the North Shore and BI
samples.

Instantaneous growth rates and mean back-calculated
length were consistently higher for fish in the Leland
stock.

North Shore whitefish mature at a younger age and
smaller size and are recruited at an earlier age than the

Leland fish.

ACKNOWLEDGEMENTS

I would like to acknowledge Michigan State University,
Department of Fisheries and Wildlife, for providing this
research opportunity and Michigan Sea Grant for funding the
study.

Special thanks are in order to Peter Jacobson for his
helpful suggestions and comradeship throughout the study.

I thank Steve Kraus for his computer work and guidance,
and Brian Raber for the fine cartography work. I appreciate
the patience and helpful suggestions of Dr. John Gill.

I thank Dr. William Taylor, Dr. John Gill, Dr. Niles
Kevern, and Peter Jacobson for reviewing the manuscript.

This study would not have been possible without the
cooperation and friendship of the commercial fishermen in
northern Lake Michigan, especially the Frazier brothers,
King's Fisheries, Ross Lang, and Bill Carlson.

I would like to thank P. Jacobson, G. Curtis, G.
Fleischer, D. Kononen, S. Cornelius, M.~ Ultis, and R.
Wehrmeister for their help in the collection of field data.

I also express my most sincere appreciation and love
for my wife, Jennifer Kraus. I am grateful to her for the
construction of figures, typing, and most of all for her

patience.

ii

LIST OF TABLES . . .
LIST OF FIGURES . .
LIST OF APPENDICES .
INTRODUCTION . . . .
METHODS . . . . . .

Sampling of the

TABLE

OF CONTENTS

Commercial Catch

Abundance and Movements . . . . .

Aging . . . . .

Weight-Length Relationships . . .

Growth . . . .

Mortality Estimates .

Sex, Maturity, and Age at Recruitment

RESULTS AND DISCUSSION . .

Distribution of Tag Returns . . .

North Shore Returns

Leland Returns . .

Grand Traverse Bay Returns . .

Tag Loss . . .

Aging . . . . .

Age Compositions

Population and Biomass Estimates

Mortality Estimates .

iii

12
13
13
14
16
l7
l7
17
26
31
33
35
37
47
51

Page

Length and Weight . . . . . . . . . . . . . . . . . 60
Weight-Length Relationships . . . . . . . . . . . . 75

Mean Back-Calculated Growth . . . . . . . . . . . . 83
Instantaneous Growth Rates . . . . . . . . . . . . 89

Age at Recruitment . . . . . . . . . . . . . . . . 93

Sex and Maturity . . . . . . . . . . . . . . . . . 94
SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . 99
APPENDICES . . . . . . . . . . . . . . . . . . . . . . . 102

LI ST OF REFERENCES 0 O O I O O O O O O O O O O O C O O O 123

iv

Table

Table

Table

Table

Table

Table

Table

Table

LIST OF TABLES
Page

Sampling dates, locations, and numbers of lake
whitefiSh' 1980-1982 I O I O O O O O O O O O O O 7

Monthly tag returns of lake whitefish tagged
during November 1980 and November 1981 in the
Leland and North Shore areas. Percentages are
based on the total number of returns from each
tagging during the sampling period . . . . . . 19

Distribution of tag returns by statistical grid
during 1980-1982 from the November 1980 and
November 1981 taggings in the North Shore area.
Percentages are based on the total number of
returns from each tagging date during each

year . . . . . . . . . . . . . . . . . . . . . 23

Distribution of tag returns by statistical grid
during 1980-1982 from the November 1980 and
November 1981 taggings in the Leland area and

from the June 1981 tagging in the Grand Traverse
Bay area. Percentages are based on the total
number of returns from each tagging date during
each year . . . . . . . . . . . . . . . . . . 28

Percent age compostition of commercial trap net
(TN) and gill net (GN) catch in northeastern
Lake Michigan in 1965 through 1982 . . . . . . 38

Percent of the sampled catch (pounds) in each age
class from northeastern Lake Michigan in 1980
through 1982 O O O O I O O O I O O O O O O O O 39

Age composition comparisons among sampling
locations using a Chi square goodness of fit
teSt O O O O O O O O O O O O O O O O O O O O O 40

Mean age of lake whitefish from the sampled catch
in the North Shore, Leland, Beaver Island, and
Grand Traverse Bay areas of Lake Michigan . . 46

Table

Table

Table

Table

Table

Table

Table

Table

Table

Table

10.

11.

12.

13.

14.

15.

16.

17.

18.

Page

Population numbers and biomass (kg) of Lake
whitefish by age class for November 1980 in the
North Shore, Leland, and Leland spawning areas.
All figures were derived from the October 1980
percent age compositions in the respective

areas O O O O O O O O O O O O O O O O O O O O 49

Annual (A), instantaneous total (2), instantaneous
fishing (F), and instantaneous natural (M)
mortality rates of lake whitefish in northeastern
Lake Michigan . . . . . . . . . . . . . . . . 52

Annual (A), instantaneous total (2), instantaneous
fishing (F), and instantaneous natural (M)
mortality rates for lake whitefish from Lake
Michigan and other lakes . . . . . . . . . . . 53

Mean length (mm) and weight (g) of lake whitefish
from the sampled catch in the North Shore, Leland,
Beaver Island, and Grand Traverse Bay areas of

Lake Michigan . . . . . . . . . . . . . . . . . 61

Length composition comparisons among sampling
locations using a Chi square goodness of fit
teSto O O O O O O O O O O O O O O O O O O O O O 64

Length compositions of the sampled catch from
northeastern Lake Michigan in 1980-1982 . . . . 65

Results of comparisons of mean length (mm) at age
for lake whitefish from northeastern Lake

Michigan using Scheffé tests. The hypothesis of
equality of means was rejected if the interval

a + 95% MSD did not include zero . . . . . . . 69

Mean length (L), mean weight (W) and number (N)

of lake whitefish at various ages at each

sampling date in northeastern Lake Michigan.
Length is in millimeters. Weight is in grams. . 71

Mean length at age in millimeters (L) and numbers
(N) of male and female lake whitefish from the
Leland and North Shore areas of Lake Michigan . 74

Predictive weight-length relationships for lake
whitefish from the North Shore, Leland, Beaver
Island, and Grand Traverse Bay areas of Lake
Michigan. Total length was measured in milli-
meters and weight was measured in grams . . . . 76

vi

Table

Table

Table

Table

Table

Table

Table

19.

20.

21.

22.

23.

24.

ZSO

Page

Tests of equality of slopes of weight-length
regressions between sampling areas using an
analysis of covariance (Snedecor and Cochran,

. 1967) O O O O O O O O O O O O O O O O O O O O O 78

Slopes (b) of the predictive weight-length
regression for lake whitefish from northern
Lake MiChigan O O O O O O O O O O O O O O O O O 79

Comparisons of mean back-calculated lengths (mm)
at annulus formation for lake whitefish from
northeastern Lake Michigan . . . . . . . . . . 85

Predictive regressions of total fish length in
millimeters (Lc) on the anterior scale radius in
millimeters (S) for combined 1980-1982 samples

of lake whitefish from northeastern Lake

Michigan. The spring 1981 and spring 1982

Beaver Island samples are presented separately

for comparison . . . . . . . . . . . . . . . . 86

Instantaneous true growth rates (G) for lake
whitefish from northern Lake Michigan. The
predictive length-weight exponenet and back-
calculated lengths were used in the formula:
G = b (loge L2 - loge Ll) (Ricker, 1975) . . . 90

Percent maturity of male and female whitefish
in 10 millimeter length categories from the
North Shore and Leland areas . . . . . . . . . 95

Percentage of lake whitefish that were mature at

each age in the North Shore and Leland areas of
Lake MiChigan O O O O O O O O O O O O O O O O O 97

vii

Figure 1.

Figure 2.

Figure 3.
Figure 4.

Figure 5.

Figure 6.

Figure 7.

Figure 8.

LIST OF FIGURES
Page

Commercial harvest of lake whitefish in
statistical districts MM-3 and MM-S of Lake
MiChigan from 1948-1981O O O O O O O O O O O O 3

Commercial harvest of lake whitefish in
statistical district MM-S of Lake Michigan from
1948-1981 O O O O O O O O O O O O O O O O O O O 4

Map of the study area in northeastern Lake
Michigan and adjacent waters of Lake Huron . . 8

Map of the statistical grids within the study
area in northeastern Lake Michigan . . . . . 20

Distribution of tag returns from the 1980 and
1981 taggings in the North Shore, Leland, and
Grand Traverse Bay areas. Grand Traverse

Bay returns are circled. Returns from the fall
1981 taggings are in parentheses . . . . . . 21

Age compositions of the sampled catch of lake
whitefish, in springs 1981 and 1982, from the
Beaver Island area of Lake Michigan . . . . . 42

Catch curves for stocks of lake whitefish from
the North Shore, Leland, Beaver Island, and
Grand Traverse Bay areas of Lake Michigan, in
spring and fall. a - North Shore, springs of
1981 and 1982 combined b - North Shore, falls
of 1980 and 1981 combined c - Leland, springs
of 1981 and 1982 combined 6 - Leland, falls
of 1980 and 1982 combined e - Beaver Island,
spring 1981 f - Beaver Island, spring 1982

g - Beaver Island, springs of 1981 and 1982
combined h - Grand Traverse Bay, spring

1981 O O O O O O O O O O O O O O O O O O O O 55

Length compositions of the spring 1981 and

spring 1982 sampled catch of lake whitefish

from the North Shore and Beaver Island

areas of Lake Michigan. . . . . . . . . . . . 66

viii

Figure 9.

Figure 10.

Figure 11.

Page

Mean length of the 1977 and 1976 Year classes
in the sampled catch from northeastern Lake
Michigan in October 1980 through May 1982 . . 68

Mean back-calculated length (mm) at age of
lake whitefish at various ages from
northeastern Lake Michigan. . . . . . . . . . 84

Instantaneous growth rates from spring (a)

and fall (b) samples of lake whitefish at
various ages from northeastern Lake Michigan . 91

ix

Appendix

Appendix

Appendix

Appendix

Appendix

Appendix

Appendix

LIST OF APPENDICES
Page

Monthly tag returns (R) and commercial catch
(C) of lake whitefish by statistical grid
during 1981 for the North Shore area. Catch
is in pounds and recaptures are in numbers.

O O O O O O O O O O O O O O O O O O O O O 102

Monthly tag returns (R) and commercial catch
(C) of lake whitefish by statistical grid
during 1981 for the Leland and Grand Traverse
Bay areas. Catch is in pounds and recaptures
are in numbers. Grand Traverse Bay tag
returns are in parentheses . . . . . . . . 104

Estimation of the instantaneous rate of tag
loss (L) for the 1980 Floy dart tags . . . 106

Comparison of ages assigned by the two
principal scale readers, including calculation
of the index of average percent error . . 107

Comparison of ages assigned by scales and tin
rays from lake whitefish in the North Shore
area of Lake Michigan, including calculation
of the index of average percent error . . 108

Comparison of girth-length relationships

between northeastern Lake Michigan and Hammond
Bay, Lake Huron. Girth (G) and length (L) are
in millimeters . . . . . . . . . . . . . . 109

Population estimates of lake whitefish
(2430mm) in Lake Michigan for the North Shore,
Leland, and Leland spawning grounds during
November 1980. The North Shore estimate
includes grids 115-119, 213-220. The Leland
estimate includes grids 615, 714, 812-814 and
911-912. Estimates include only catch and
recaptures from cooperating fishermen . . 110

Appendix

Appendix

Appendix

Appendix

Appendix

Appendix

Appendix

Appendix

10.

11.

12.

13a.

13b.

13c.

Page

Calculations of the adjustment of the 1981
commercial harvest by cooperating fishermen,
of lake whitefish in the North Shore and
Leland areas for the recruitment of those fish
that were sublegal (<430mm) in November 1980,
yet grew into the harvestable portion of the
population during 1981 . . . . . . . . . . 111

Annual exploitation rates, after correction
for tag loss, for lake whitefish (>430mm)
tagged in the Leland and North Shore areas of
Lake Michigan. Values were calculated from
the catch and returns from known cooperating
fishermen, then expanded by the ratio of
their catch to the total catch in each area.

O O O O O O O O O O O O O O O O O O O O O 113

Details of an analysis of variance on mean
length (mm) at age for representatives of the
1976, 1977, and 1978 year classes of lake
whitefish in northeastern Lake Michigan

during 1980-1982. Analyses were conducted on
all samples where n220 fish . . . . . . . 114

Tests of the assumptions of homogeneous
variance and normally distributed residuals
for the analyses of variance on mean length at
age . . . . . . . . . . . . . . . . . . . 115

Results of the quadrant test of the joint
distribution of length and weight for
bivariate normality (Cramer 1946). The
hypothesis was rejected when the calculated
Chi square value exceeded the tabular value at
the 0.05 level of significance . . . . . 117

Mean back-calculated lengths for lake white-
fish from the North Shore area of Lake
Michigan. Standard deviations are in
parentheses . . . . . . . . . . . . . . . 118

Mean back-calculated lengths for lake white-
fish from the Leland area of Lake Michigan.
Standard deviations are in parentheses . 119

Mean back-calculated lengths for lake white-
fish in the Beaver Island area of Lake
Michigan. Standard deviations are in
parentheses . . . . . . . . . . . . . . . 120

xi

Page

Appendix 13d. Mean back-calculated lengths for lake white-

Appendix 14.

fish in the Grand Traverse Bay area of Lake
Michigan. Standard deviations are in
parentheses O O O O O O O O O O O O O O O 121

Graphical estimation of age at recruitment
from the mean back-calculated length (mm) at

‘age of lake whitefish from northeastern Lake

MiChigan O O O O O O O O O O O O O O O O 122

xii

I NTRODUCTI ON

Lake whitefish (Coregonus clupeaformis) have supported
a major commercial fishery in Lake Michigan since the mid
1800's (Baldwin et a1 1979). Historically, this species has
exhibited wide fluctuations in abundance (Smith 1968, Wells
and McLain 1973). Pollution of the spawning grounds,
introduction of exotic species, variable year class
strength, and overharvest have all been cited as probable
causes of these fluctuations (Smith 1968, Wells and McLain
1973, Lawler 1965, Cucin and Regier 1966).

The whitefish is the last on a long list of
commercially valuable coregonids in Lake Michigan (Baldwin
et a1 1979). Several species of large, commercially

valuable ciscoes (Coregonus sp) were fished to extinction by

 

the early fishery (Wells and McLain 1973). The whitefish
has persisted as a result of its resilient conpensatory
dynamics, and is presently managed under stringent
regulatory authority. Healey (1975) characterizes the
resilient nature of the whitefish by its early maturation
and increased growth rates in response to increased
exploitation.

A recent increasing trend in the annual production of
lake whitefish has raised concern regarding the stability of

the whitefish stocks in the northern portion of Lake

2

Michigan (Figures 1 and 2). This increase in harvest in
statistical district MM-3 is due in part to an increase in
effort by native American fishermen and to the presence of
an abundant year class of whitefish in the fishery. The
ecological stability of this species is of concern. Healey
(1975) suggests that the scope of a whitefish population to
compensate for increased mortality can be measured by the
difference between the growth rate of the population and the
maximum growth rate. Whether or not the whitefish can
contiunue to maintain their abundance and support the
commercial fishery depends upon their compensatory reserve
for increasing their growth rate.

The commercial fishery for whitefish is regulated under
a zone management plan implemented by the Michigan
Department of Natural Resources in 1972. Licensing is
limited to those fishermen who were licensed the previous
season. Each license permits the fisherman to fish ten
large-mesh deep water trap nets (4 1/2 inch stretched
measure).

Current management plans include the implementation of
a quota system by which to regulate fishing effort by
geographical lake quadrants. Differentiation of discrete
stocks and knowledge of their relative abundance and growth
capabilities are essential to proper management of
whitefish. A tagging study in combination with statistical
comparisons of certain population parameters was implemented

for this purpose.

.Hmmfi - mama Song cmeeOHz mean no

 

 

mus: can miss mpownpmflw Hmowpmwpmpm E.” :mHMmpacz 33..” .Ho unwind: Hmwopmgoo .H onsmfim
ecu»
coma mama opal mama owns mmma ommz
. _ .. l- A _ A _ rid... o
r oom
r ooo_ u H
U
w w
m n
S
r d 1
com“ m
w
B
w==-::- r coon
m==.||.
. oomN

 

 

 

mmw»

mesa Lo m-22 poahpmne HAOHPmAPaPm ch emflcapfles mama

oom—

mbmu

0mm”

mmma
_

chma
.r - ii

I IO.

‘I'OOOOOO

.Hmmfi - mama aona cam“:0A2

mo pmm>nmz HmwonmEEoo

mmmd ommd

 

Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q

 

1.

T

I

. N mpsmfim

O
(D

D
D

[spunod to spuesnorm
183A HVH

I cod

r ON“

T o:

 

The objectives of this study were to analyze the vital
statistics of lake whitefish in the northeastern portion of
Lake Michigan with special emphasis on mortality rates,
growth rates, age structure, population numbers and biomass,
and movements for the purpose of stock differentiation.

For the purpose of this investigation, a discrete stock
has been defined as a manageable unit of reproductively
isolated mature fish with homogenous characteristics of age
composition, length composition, length at age, growth, and

mortality.

METHODS
Sampling of the Commercial Catch

Commercial trap net catch was sampled for length, weight,
and scale samples during the period from November 1980
through May 1982 (Table l). The other primary fishing gear
in the study area was the large mesh gill net. This gear is
fished exclusively by the treaty fishermen and is prohibited
for use by the state licensed commercial fishermen. The
gill net catch was not sampled due to the inadequate narrow
range of selectivity of this gear for obtaining length,
weight, and age data. The selectivity curve of the large
mesh gill net exhibits a peak of efficiency which diminishes
as the mean length increses (McCombie and Fry 1960).

Sampling was concentrated around the southeast shore of
the upper peninsula of Michigan , hereafter referred to as
the North Shore area, and the western shore of the Leelanau
peninsula from Leland south to Empire, hereafter referred to
as the Leland area (Figure 3).

Additional sampling of the commercial catch was
conducted in the Beaver Islands area and in the northeastern
Grand Traverse Bay area. The Grand Traverse Bay samples
were collected from the catch of an experimental purse seine
funded by the Michigan Sea Grant and Carlson Fisheries in
Leland. The purse seine is the only state licensed gear
operating in that region of the Bay. The purse seine has a

larger minimum size limit (19 inches) than that for the trap

6

Table l.

whitefish, 1980-1982.

Sampling dates, locations, and numbers of lake

 

 

 

Port Grids Date Sample Type N
North Shore Area
Naubinway 115,116 10/23/80 SLW 513
10/29/80
Naubinway 116 11/4/80 Tagged 1683
Epoufette 117 6/29/81 SLWM 107
Epoufette 218 8/24/81 SLW 264
Epoufette 218 8/25/81 SLWF 36
Naubinway 116,117 10/17/81 SLW 331
10/24/81
Naubinway 117 11/3/81 Tagged 1024
Epoufette 216,218 5/17/82 SLW 263
Epoufette 216,218 5/17/82 CL 86
Epoufette 216,218 5/18/82 SLW* 63
Leland Area
Leland 714,814 10/29/80 SLW 114
10/30/80
Leland 714 11/7/80 Tagged 415
Leland 812,814,912 6/15/81 SLW 81
Leland 812,814,912 8/27/81 SLW 94
Leland 812 10/21/81 SLW 134
Leland 812 10/22/81 SLWM 110
Leland 812 11/2/81 Tagged 117
Leland 812,814,912 5/20/82 SLW 111
Leland 812,814,912 5/24/82 SLWG l4
Leland 812,814,912 5/24/82 SLW 107
Leland 812,814,912 5/24/82 LWSG* 11
Leland 812,814,912 5/24/82 LWS* 16
Beaver Island Area
Beaver Island 316 11/5/80 Tagged 19
Beaver Island 316 6/16/81 SLW 219
Charlevoix 317,418 5/18/82 SLW 169
5/19/82
Grand Traverse Bay Area
North Port 615 6/14/81 Tagged 163
North Port 615 6/14/81 SLW 140
S = scale sample, L = length, W = weight, G = girth,
F = fin rays, M = sex and maturity, * = nonrandom sample

 

 

 

 

 

 

 

 

 

 

 

 

9 140‘ 2L 39 ‘2 59 6B 710 N b‘nuudyono E f 7
mm. I 3“ “my. .pou em
1.. has a“
N St. James$a sHelenat t. 'amncMfl’Lnggnn‘ 33:;
13.
Island
’0
“a («43‘
volx
a
N
C3
72
‘2
Y’
Q MICHIGAN 3'9 CW
5
Q.
2
E m Muskegon
(D
Z
C)
C)
2
3 LL)
’5: L
.4 o ansing
Detroit
lllhnb V Irdlana Ohio

 

 

Figure 3. Map of the study area in northeastern Lake
Michigan and adjacent waters of Lake Huron.

9
nets (1? inches), thus rendering these samples uncomparable
for most purposes.

Certain locations were only sampled during certain
seasons. This was a result of the relocation of commercial
fishing effort in response to changes in whitefish abundance
on the various fishing grounds. Whitefish have been noted
to move into deeper water during the summer and return to
shallower water in the fall to spawn (Lawler 1965).

A random sample of fish was selected from the unsorted
catch at dockside. Total fish length was measured to the
nearest 5 millimeters and weight was measured to the nearest
10 grams. Scales for age and growth analysis were removed
from the lateral region of the fish directly below the

dorsal fin.
Abundance and Movements

A total of 3239 fish were tagged during the closed
fishing season (November 1-30), 2098 in 1980 and 1141 in
1981. In June 1981 an additional 163 sublegal fish (<483mm)
were tagged from the purse seine operated in Grand Traverse
Bay.

The fish tagged in November 1980 and June 1981 were
tagged using Floy dart tags. In November 1981 a smaller
diameter Floy anchor tag was used due to several complaints
by the commercial fishermen regarding the unhealed sore

apparent on the recaptured fish.

10

All fish were tagged directly below their dorsal fin so
that the tag lodged under the interneural bones. Total
length was recorded and the fish were returned immediately
to the lake.

A reward system was implemented to encourage
cooperation with the commercial fishermen. A $1 reward was
paid for each returned tag. Additional information that was
requested included the date and location of capture.
Considerable personal contact and frequent mailings of
newsletters were used to encourage cooperation and maintain
interest of the whitefish fishermen. Double-tagging of 158
fish with one dart and one anchor tag was implemented in
November 1981 to estimate tag loss.

Population numbers and biomass were estimated for all
fish larger than 430 millimeters (17 inches) in November

1980 using a Petersen mark-recapture equation,

N = MC/R

where, N is the population estimate,

M is the number of fish tagged in November 1980,

C is the total catch of whitefish during the

1981 season, and

R is the total number of recaptured tags during the

1981 season.

Chapman (1951) states that when the total numbers of

recaptures are few, population size is overestimated, and an

adjusted Petersen should be used. The adjustments amount to

11
the addition of one to the values of M, C, and R in the
above equation. Robson and Regier (1964) indicate that the
bias will be less than two percent if the product of the
marks and the recaptures is greater than the total
population size. Since this product exceeded the estimated
population sizes no adjustments were made.

Due to a presumed non-cooperation by certain fishermen
in the study area, the values for C and R represent only
those portions attributed to fishermen who were known to be
cooperating.

The number of recaptured fish was adjusted for tag loss
by a procedure described by Seber (1973). The total catch
was adjusted to account for the recruitment of those fish
which were sublegal (<430 millimeters) at the time of
tagging, yet grew into the catchable population during the
1981 fishing season. To estimate recruitment the change in
mean length, of the youngest age class present in the fall
1980 catch (age III), was followed through the sampling
period. The seasonal increments of growth for this age
class were used as the best available estimate of the
seasonal growth of the fish that were sublegal in November
1980. That portion of the sampled legal dockside catch
(weight) which was represented by the fish shorter than the
mean length of an average November 1980 sublegal fish, at
the time of each subsequent sample, was subtracted from the
total catch. These fish were not part of the legal-sized

tagged population at the time of tagging.

12

Examination of the distribution of tag returns, in
conjunction with statistical comparisons of age
compositions, length compositions, and mean length of
several representative age classes, were used to

differentiate discrete stocks of lake whitefish.
Aging

The ages of lake whitefish were determined by scale
analysis. Scales were cleaned in the laboratory with a
toothbrush and water then projected with a Bell and Howell
ABR-1020 microfiche reader at a magnification of 22x. Scale
ages were assigned by counting the number of annuli present.
The primary criteria for distinguishing annuli was "cutting
over" along the anterio-lateral ridges and spatial
disruptions of the circuli (van Oosten 1923). One scale
from each fish was selected at random and the distance from
the focus of the scale to each annuli was measured along the
center radius of the anterior field. The assigned ages were
verified by comparisons with ages assigned by pectoral fin
ray sections and a fifteen percent overlap by the principal
scale readers. Pectoral fin rays were aged via examination
of microtome sections immersed in oil under the low power of
a compound microscope. Annuli were discerned according to

procedures described by Ovchynnyk (1962).

12

Examination of the distribution of tag returns, in
conjunction with statistical comparisons of age
compositions, length compositions, and mean length of
several representative age classes, were used to

differentiate discrete stocks of lake whitefish.
Aging

The ages of lake whitefish were determined by scale
analysis. Scales were cleaned in the laboratory with a
toothbrush and water then projected with a Bell and Howell
AER-1020 microfiche reader at a magnification of 22x. Scale
ages were assigned by counting the number of annuli present.
The primary criteria for distinguishing annuli was "cutting
over" along the anterio-lateral ridges and spatial
disruptions of the circuli (van Oosten 1923). One scale
from each fish was selected at random and the distance from
the focus of the scale to each annuli was measured along the
center radius of the anterior field. The assigned ages were
verified by comparisons with ages assigned by pectoral fin
ray sections and a fifteen percent overlap by the principal
scale readers. Pectoral fin rays were aged via examination
of microtome sections immersed in oil under the low power of
a compound microscope. Annuli were discerned according to

procedures described by Ovchynnyk (1962).

Weight-Length Relationships

The weight-length relationship,
W = aLb,
where, W is the weight of the fish in grams,
L is the length of the fish in millimeters, and
a and b are constants,
was transformed to natural logarithms and fit by a least

squares predictive regression. The underlying regression

assumptions were statistically validated.
Growth

A least squares predictive regression of the form,
Lc = a + bS,

where, Lc is the total fish length at capture,

S is the anterior scale radius,

a is the y-intercept of the regression, and

b is the slope of the regression,
was used to predict the length of the fish at the time of
formation of each annulus. A correction factor (f) of the

form,

f = Lc/Lc*,
where, Lc is the observed length at the time of capture, and
Lc* is the length at capture predicted from the body
length-scale radius regression for the observed total
scale radius,

was multiplied by the calculated lengths under the

13

l4

assumption that the proportional deviation of lengths from
the regression is the same at each annulus as at the time of
capture (Carlander 1981). The underlying regression
assumptions of a normal distribution of residuals,
linearity, and homogeneous variance of Le over the range of
S were tested. The mean back-calculated length at each
annulus was computed for all fish in the samples, then
averaged by age class.

The instantaneous true growth rate (G) for individual
fish was estimated from the back-calculated lengths at
annulus formation and the slope (b) of the transformed

weight-length relationship,

G = b (1n (L2) - 1n (Ll)).
where, L2 is the length at formation of the most recent
annulus,
L1 is the length at formation of the next to the last
annulus, and
1n is the natural logarithm.
The instantaneous growth rate was calculated from the last
two annuli on the scales, providing the best estimate for

the most recent year of growth (Ricker 1975).
Mortality Estimates

The annual survival rate (5) was estimated from tag
returns from the equation,

5 ((R12)(M2)/(Ml)(R22)),

15
where, R12 is the total number of recaptures of November
1980 tags returned during the period from January
1982 - June 1982,
R22 is the total number of recaptures of November
1981 tags returned during the period from
January 1982 - June 1982,
M1 is the number of fish tagged in November 1980,
M2 is the number of fish tagged in November 1981.
The annual mortality rate (A) was computed from the
relationship,
A = l - S.
The annual exploitation rate (u) was calculated from

tag returns from the relationship (Ricker 1975),

u = Rll/Ml,
where, R11 is the number of recaptured fish, adjusted for
tag loss, that were caught during the 1981 fishing
season.

The survival rates and the exploitation rates were
calculated soley from those tags returned by cooperating
fishermen. The exploitation rates were expanded to the
entire fishery in each area by multiplying by the ratio of
the total commercial catch to that of the cooperating
fishermen in each area.

The instantaneous total mortality rate (2) was
estimated from the relationship,

2 = -ln (S),

where, ln is the natural logarithm.

16

The instantaneous total mortality (2) was also
estimated from catch curves (Robson and Chapman 1961) to
serve as a check for the tagging estimates, and as the sole
estimate where no tagging was conducted.

The instantaneous rate of fishing mortality (F) was

estimated from the formula (Ricker 1975),

F = uZ/A.

The instantaneous rate of natural mortality (M),
defined as that portion of the instantaneous total mortality
attributable to non-fishing sources, was computed as the
differerce z - F.

'Sex, Maturity, and Age at Recruitment

Length at maturity was estimated from one sample from
the Leland area and one from the North Shore area. The
North Shore sample was acquired on a day when the Michigan
Department of Natural Resources was collecting samples for
the same purpose. The other sample was obtained when fish
were being filleted in Leland, Michigan. Sex ratios of
these samples were estimated. Additional samples of this
nature were not available because whitefish are marketed
primarily in the round (not eviscerated).

Age at recruitment was estimated graphically from a
plot of mean back-calculated length against age. The point
on the abscissa (age) where the growth curve intersected the
length at recruitment (432 millimeters) was the estimated

age at recruitment.

RESULTS AND DISCUSSION

Distribution of Tag Returns

The distribution of tag returns indicates the existence
of at least three discrete stocks of whitefish in the study
area. One stock inhabits the waters south of Naubinway and
north of the Beaver Islands. One is located along the
eastern shore of the Leelanau peninsula from Cat Head Point
south to Empire (Figure 3). The third is located southeast
of the Beaver Islands. No mixing of the North Shore and the
Leland tagged populations occurred. The shallow reef
extending westward from Waugoshance Point through Hog and
Garden Islands appears to act as a barrier to the movement
of the North Shore stock (Figure 3). No tagged fish were
recaptured in the vicinity of the Beaver Islands south of
the reef and north of grids 615-616. In addition, the
distribution of tag returns indicates that these three
stocks are distinct from the stock(s) fished in Muskegon,
the stock(s) in the southern arms of Grand Traverse Bay, and
the stock(s) west of Seul Choix Point.

Nggth Shore Tag Returns. A total of 505 tags,
representing 18.7 percent of the tagged population, were
returned from the North Shore area. This includes all
recaptures from November 1980 through June 1982. During the
1981 fishing season 334 tags, representing 19.9 percent of

the fish tagged in November 1980, were returned.

17

18

Eighty-six percent of the returns from the November
1980 tagging were returned during the 1981 fishing season
(Table 2). Recaptures paralleled the peak fishing periods,
with the largest number being reported in October 1981
(Appendix 1).

Tag returns from the North Shore taggings ranged from
as far east as the the Duck Islands in Lake Huron, as far
west as the mouth of the Menomonee River, Wisconsin, and as
far south as Door County, Wisconsin (Figures 4 and 5).
North Shore tags were primarily recaptured (94%) from grids
115-117 and 215-219 south of Naubinway, Michigan. The
precise location of recapture of the 58 tags assigned to
grid 116 for October 1981 is unknown. The fisherman failed
to record the locations of recapture, yet stated that most
of the tags came from grid 116 and the remainder were caught
inshore, presumably from grids 115 and 117.

The first reported recapture from the North Shore
taggings came from Lake Huron just east of the Mackinac
bridge. This fish was caught less than three weeks after it
was tagged. During the 1981 fishing season, three more
recaptures were reported from Lake Huron. This indicates
that there is some movement of the fish from the North Shore
stock into Lake Huron. The few tag returns reported from
northwestern Lake Huron, compared to the abundant returns
from Lake Michigan grids 216-218, suggest that the stocks
fished in these areas are distinct. No commercial catch in

the waters of northwestern Lake Huron was sampled, thus

19

Table 2. Monthly tag returns of lake whitefish tagged during
November 1980 and November 1981 in the Leland and North Shore

areas. Percentages are based on the total number of returns

from each tagging during the sampling period.

 

 

 

 

North Shore Leland
1980 Tags 1981 Tags 1980 Tags 1981 Tags
Month # % # % # % # %
1980
November 1 0.3
December 38 31.4
1981
January 1 0.8
February 5 1.7
March 1 0.8
April 27 9.0 23 19.0
May 38 12.7 9 7.4
June 40 13.3 10 8.3
July 29 9.7 14 11.6
August 9 3.0 3 2.5
September 13 4.3 3 2.5
October 96 32.0 5 4.1
November 1 0.3 1 0.8
December 1 0.3
1982
January 1 0.9
February 2 1.8
March 1 0.8
April 3 1.0 9 8.3 1 7.7
May 21 7.0 51 46.8 4 30.8
June 16 5.3 46 42.2 12 9.9 8 61.5
Subtotal 300 109 121 13
Unknown 76 20 11 0

 

Total 376 129 132 13

 

20

 

 

MICHIGA N

  

WISCONSIN

Music-yon

To Saul! Sta. Marie

MICHIGAN

 
   
   

To Lansing I

 

  
   
  

 

 

 

LAKE

 
 

 

 

   

Naubinway

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  

i> o .-

O

1&\,NEpoufette
SeuIChoibeint
WA{ St. 1
Helena
211 212 213 214 215 216 217 218 219 ,
A ardcn 8:. Q:
. ‘ 319 320 7°53"
Willi. .323"? -... """'
311 312 313 314 316 317 318 4
avcrls.
L/T CrossVillage
411 412 413 414 415 416 1417 4 8 -
<3
511 513i) 514
611
711
811
911

 

 

Figure 4.

area in northeastern Lake Michigan.

Map of the statistical grids within the study

 

.mmmmcpcmpmm cw mum mwcwmwmw Hmma Hang map Souk menspmm .vmaouwo
ohm manspmn ham mmnm>mna undue .mdmum ham mmnm>mne osmnw can panama .mnonm swnoz
629 cm mmcfimmmw Hmaa can omma one Scum mauspmu map mo soapsnwnwmfin .m madman

 

 

21

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a 952.0.— o... / 1— 9:23.— o... /
53m \ 83...... \
Amt. . W?
ANV Mp
ON
252:2: 54v. . n - 55:10.: 3\ -
MN
8 E
c. “i E
AU .0
h N—
53135 595:...
A AV
7 7
36> 39.0 .3.th wan—u) 880
. . .. m .3
“M...“ .3. ft... . VLNH...“ A3
0 a i 1 0 : AQvON
2...». in. $3
. a a kgrduyLESSm . o a a. 5%328
7L \ sum 56
0:0 :2. C/OL 3.0 :0.— 0.03m
_ m 8356...: .Q 0:20.. _ m g to:

 

 

 

 

22
examination of vital statistic parameters was not a possible
avenue for discerning discrete stocks.

A few discrepancies between the reported recapture
locations and the monthly catch locations were found in the
data. For example, no catch was reported from grid 216
during February 1981 (Appendix 1), yet five recaptures were
reported from grid 216 during that month. Further, a
reported 300 thousand pounds of whitefish were harvested
from grid 218 during 1981. No tags were reported from that
grid, yet many were recorded in the nearby grids 217 and
219. During the first six months of 1982, 47 percent of the
reported recaptures were from grid 218 (Table 3). It is
suspected that the accuracy of the locations of the reported
catch and tag returns is somewhat less than perfect.

In April 1981, after the winter ice cover had
sufficiently melted, fishing was concentrated in the
offshore grids 215-217 and those portions of grids 316-317
north of the Beaver Islands. Consequently, tag returns were
primarily from these areas (Appendix 1). May 1981
recaptures were also concentrated in these grids. A few
tagged fish had moved both east and west, as evidenced by
recaptures near Seul Choix Point (grid 213) and Saint Helena
Island (grid 219).

June and July returns were reported primarily from the
offshore grids 216, 217 and 219. One tagged fish was
recaptured in June from the Wisconsin waters of Lake

Michigan near Door County. Another was caught near Saint

23

Table 3. Distribution of tag returns by statistical grid
during 1980-1982 from the November 1980 and November 1981
taggings in the North Shore area. Percentages are based on
total number of returns from each tagging date during each
year.

 

 

1980 Tagging 1981 Tagging
1980 1981 1982 1980 1982 Grand
Returns Returns Returns Total Returns Total
Grid # % # % # % # % # % # %

 

Lake Michigan

115 10 3.7 2 5.0 12 3.8 2 1.8 14 3.3
116 68 24.9 1 2.5 69 22.0 69 16.3
117 19 7.0 19 6.1 1 0.9 20 4.7
213 6 2.2 6 1.9 6 1.4
215 10 3.7 6 15.0 16 5.1 13 11.9 29 6.9
216 101 37.0 8 20.0 109 34.7 18 16.5 127 30.0
217 29 10.6 7 17.5 36 11.5 14 12.8 50 11.8
218 15 37.5 15 4.8 55 50.5 70 16.6
219 19 7.0 1 0.3 20 6.4 l 0.9 21 5.0
315 1 0.4 1 0.3 l 0.2
316 1 0.4 1 0.3 l 0.2
317 2 0.7 2 0.6 2 0.5
318 4 3.7 4 1.0
408 2 0.7 2 0.6 2 0.5
604 l 0.4 l 0.3 1 0.2
806 1 0.4 l 0.3 l 0.2
Lake Huron
301 1 100 1 0.3 l 0.2
302 2 0.7 2 0.6 2 0.5
303 l 0.4 1 0.3 l 0.2
412 1 0.9 l 0.2

 

Total 1 273 40 314 109 423

 

24
Martin Island, Lake Huron during July.

Harvest dropped off in August and September, thus fewer
tags were returned durings these months. Recaptures were
mostly from grid 216. Two tagged fish were caught in
northwestern Lake Huron (grids 302 and 303).

During October 96 tags were returned. Fishing effort
had shifted inshore, as the fish moved to the shallow areas
to spawn. The recaptures were primarily from grids 115-117.
The apparent inshore movement of spawning fish was evidenced
by the presence of breeding tubercles and the emission of
eggs and milt when the fish were handled during sampling.
This indicates a homing of the stock tagged in November 1980
back to the same spawning grounds'in November 1981.

The catch information for 1982 is not available at the
present. Commercial catch is compiled by the Michigan
Department of Natural Resources during the winter months and
is not available until March of the following year.

The fishing season got a late start in 1982, as a
result of extensive inshore ice in the early spring. Three
tags were reported during the winter months near Waugoshance
Point. These were returned by a treaty fisherman,
presumably fishing through the ice. In April 1982, tags
were all returned from the inshore grids 115-117. The April
1981 tags were caught primarily from the offshore grids
215-217 and 316-317. The colder winter of 1981-82 and the
later melting of the ice cover may have delayed the offshore

spring movements of the whitefish. When catch is available

25
for 1982, it can be determined if the April 1982 catch was
primarily from inshore or deep water areas.

The number of tag returns increased in May. For every
1980 tag returned, two 1981 tags were returned. There were
nearly twice as many recaptures during May 1982 compared to
May 1981. This would be expected since 1024 fish were added
to the tagged population during November 1981. No
difference between the distribution of the 1981 tags and the
distribution of the 1982 tags was apparent in the spring
months of 1982. This suggests that the same population was
tagged during subsequent November taggings in this area.

For the first time during the study, tags were reported
from grid 218. Fifty percent of the tags returned during
the first six months of 1982 were from this grid. The June
1982 recaptures were abundant. Returns were mostly from the
offshore grids 217 and 218.

Occasional long distance movements of whitefish are not
uncommon in tagging studies of this species. An occasional
fish tagged in the Green Bay and North-Moonlight Bay area by
the University of Wisconsin was recaptured near Epoufette
(Ebner 1980).

In summary, North Shore tag returns indicate that this
stock inhabits a large portion of the lake and contributes
to the catch of the fishery from the Mackinac Bridge area
westward to Seul Choix. The reef extending eastward from
Waugoshance Point appears to act as a barrier to whitefish

movements southward, thus limiting the mixing of the North

26

Shore and Beaver Island stocks. For management purposes,
these stocks can be considered independently in regard to
limiting or redistributing fishing effort. Protection of
the spawning population of the North Shore area can be
insured by the regulation of the fall inshore fishery, since
this stock appears to contribute heavily to the catch of
that entire region.

Leland Tag Returns. A total of 145 tags, representing
27.3% of the tagged population, were returned from the
Leland stock (Table 2). Thirty-nine tagged fish were
returned between December 1, 1980 and January 5, 1981 from
the same grid where they were tagged the preceeding
November. During the 1981 fishing season 21.5 percent of
the 376 remaining tags were recaptured. Very few tagged
fish were returned during the first six months of 1982.

The eastern shoreline of Lake Michigan, south of Cat
Head Point, is fished for whitefish by only three state
licensed fishermen. Two of these fishermen are located at
Muskegon and the other fishes between Leland and Empire.

Seasonal relocation of fishing effort in the Leland
area complicates the interpretation of the true movements of
the tagged fish. Tag returns from the North Shore area
suggest that adult whitefish migrate seasonally as a unit.
The patchy distribution of fishing effort in the Leland area
does not permit the movements of the tagged population to be
traced throughout the fishing season. Between January 1 and

June 30, 1981, fifty-six percent of the catch of the Leland

27

fisherman was reported from grids 615, 614, and 715.‘ After
July all of the catch from the Leland area was from grids
812, 814 and 912. Also, the tagging locations and the
numbers of *fish tagged were different between years. The
November 1981 tagging was conducted on the spawning grounds
in grid 714. The November 1982 tagging was conducted in
grid 812. A total of 415 fish were tagged the first fall
and 117 were tagged the following fall.

The Leland fisherman fished primarily in grids 812-814
and 911-912 during the spring of 1982. Nine out of ten of
this fisherman's nets were set in these grids at the time of
the May 1982 sample. The relocation of fishing effort, in
combination with the smaller number of fish tagged in
November 1981, can explain the smaller number of recaptures
during the first six months of 1982, compared to those from
the same period the previous year. Perhaps, when the Leland
fisherman changed the location of his nets, different
stocks, or subpopulation, were being harvested.

Proportionately more recaptures, per pound of catch,
were reported from grids 714 and 814 during the 1981 fishing
season (Appendix 2). Returns ranged from as far south as
Muskegon, as far west as Door County in Wisconsin, and
northeast into Grand Traverse Bay (Figure 5).

Ninety-two percent of all Leland area recaptures were
reported from grids 615, 714, 812-814, and 911-912, which
extend along the shoreline from Cat Head Point south to

Empire (Table 4). Returns from Muskegon suggest a partial

28

Table 4. Distribution of tag returns by statistical grid
during 1980-1982 from the November 1980 and November 1981

taggings in the Leland area and from the June 1981 tagging
in the Grand Traverse Bay area. Percentages are based on
the total number of returns from each tagging during each

year.

 

 

 

 

 

1980 Tagging 1981 Tagging

1980 1981 1982 1980 1982 Grand
Returns Returns Returns Total Returns Total

Grid # % # % # % # % # % # %

Leland
615 8 11.3 4 33.3 12 9.8 12 8.8
616 7 9.9 1 0.8 7 5.2
703 1 7.7 1 0.7
706 1 1.4 1 0.8 1 0.7
714 38 100 9 12.7 47 38.5 1 7.7 48 35.6
715 3 4.2 3 2.5 3 2.2
812 2 2.8 2 1.6 2 1.5
813 l 8.3 l 0.8 1 0.7
814 27 38.0 2 16.7 29 23.8 2 15.4 31 23.0
911 1 8.3 1 0.8 1 0.7
912 9 12.7 4 33.3 13 10.7 9 69.2 22 16.3
1810 6 8.5 6 4.9 6 4.4
Total 38 70 12 122 13 135
Grand Traverse Bay

615 2 40.0 2 33.3 4 36.4 4 36.4
616 1 20.0 3 50.0 4 36.4 4 36.4
715 2 40.0 2 18.2 2 18.2
912 1 16.7 1 9.1 1 9.1

 

Total 0 5 6 ll 0 ll

 

29

mixing of the Leland stock with one or more stocks ’to the
south. It is believed that the whitefish harvested near
Muskegon (grid 1810) originate from a stock separate from
those harvested in Leland. Only six tags were returned from
over 250 thousand pounds of whitefish harvested in that area
in 1981. If the Muskegon and Leland fish originated from
the same stock, one would expect a higher number of
recaptures from the Muskegon area.

Three tagged fish from the Leland area were recaptured
in the Northport Bay area. The movement of fish from the
Leland stock into northeastern Grand Traverse Bay indicates
overlap of the ranges of the fish in these areas. Rybicki
(1980) suggested the existence of three subpopulations of
whitefish in Grand Traverse Bay, one of which resides in the
northern portion of the Bay.

The results of this study do not allow the distinction
of stocks between the Leland area and upper Grand Traverse
Bay. Only one state licensed fishing vessel, the purse
seine, operates in the Bay. The purse seine catch is
restricted under a yearly quota of 50 thousand pounds.
Treaty fishermen harvested 260 thousand pounds from the
upper portion of the Bay (grids 715 and 716), although only
one treaty fisherman is known to be cooperating. Suspected
incomplete reporting of tags, and the patchy distribution of
fishing effort in this area, limits the interpretation of

the available data.

30

April marked the beginning of the 1981 fishing season,
after the winter ice cover on the lake had melted. Twenty
three tagged fish were recaptured during April. Ninety-one
percent of these were reported from grid 814. Two tags were
returned from the Muskegon area. In May, only a few tagged
fish were recaptured and these were from grids 714 and 814
near Leland. During June, the recaptures were distributed
evenly along the shoreline in the Leland study area.
Another tag was returned from Muskegon. July tag returns
were reported primarily from grids 615, 616, and 714. One
fish was recaptured on the purse seine in Northport Bay
(grid 715) and another in Muskegon.

Few tags were returned during the remainder of the 1981
fishing season. Fishing effort was concentrated in grids
615, 616, and 715 near the mouth of Grand Traverse Bay and
in grids 812 and 912 near Empire. The relatively few
recaptures suggests that the tagged stock may have moved
back to their spawning grounds in grid 714. Homing of
whitefish was noted in the North Shore stock and, although
no harvest nor recaptures were reported from grid 714 after
July, a homing tendency of the Leland tagged stock is
suggested.

Recaptures during the first six months of 1982 were
scarce. The returns were concentrated in grid 615 near Cat
Head Point and grid 912 near Empire.

The patchy distribution of the fishing effort, both

geographically and seasonally, limits the utility of tagging

31
operations in the Leland area. The tag returns indicate a
broad geographical distribution of this stock, and suggest a
homing tendency of the tagged fish. These findings are
consistent with those from the North Shore area.

Grand Traverse Bgy Tag Returns. Only eleven of the 140
tagged fish from the June 1981 tagging in Grand Traverse Bay
were returned (Table 4, Appendix 2). The small number of
returns is, in part, due to the later tagging date in this
region (Table 4). These fish were only available for
harvest from the June 1981 until the end of the study. The
tagged fish in the other areas were available for harvest
from November 1980 until the end of the study. Also, the
fish tagged on the purse seine were all shorter than the 19
inch minimum size limit for this gear, thus many of them
were not of legal size for harvest by either trap nets or
the purse seine. Several sublegal tagged fish were reported
to have been caught by the purse seine during July 1981.
These were all returned to the water with no record made of
their numbers. It has been mentioned previously that
incomplete reporting of tagged fish is suspect in this area.
No tags were reported from grid 716 where a substantial
portion of the catch of lake whitefishi in the Bay was
reported.

The tags that were returned from the Grand Traverse Bay
tagging suggest an overlap of the fish from the Leland and
upper Grand Traverse Bay areas. Continued investigation is

required to discern the discreteness of the Leland, Grand

32
Traverse Bay, and Muskegon stocks, if it exists.’ Index
trawling by the Michigan Department of Natural Resources, in
the waters from Leland south to Pentwater, showed few
residual populations of whitefish (Rybicki 1980). The
non-sedentary nature of these whitefish in this area

complicates stock differentiation.

Tag Loss

A total of 158 fish were tagged in November 1980 with
two tags each. This double-tagging procedure was conducted
to estimate the frequency of tag loss in the population
(Seber 1973). The double-tagged fish had one green dart
tag, identical to those used in November 1980, and one
yellow anchor tag, identical to those used in November 1981.
One of the double tagged fish was returned with only the
green tag remaining. Another of the double tagged fish was
returned with only the yellow tag remaining. The former was
recaptured on June 21, 1982. The latter was recaptured on
June 19, 1982. Assuming that the tags were lost at the
midpoint of the interval between the date of tagging and the
date of recapture, an instantaneous rate of tag loss was
calculated (Appendix 3). The value estimated for the
instantaneous rate of tag loss (T) for the November 1980
tags was 0.2184.

In order to adjust the population estimates for tag
loss, the reported tag returns were separated into
intervals, such that the dates of recapture were known
precisely. These intervals spanned several months during
the fishing season since the only information concerning the
date of recapture of several tags was that they were caught
after the last sampling date and prior to the next sampling
date. The midpoint of each interval was used as the time
(t), expressed in years, to estimate the percentage of tags

lost during that interval from the relationship, R = exp

33

34
(-T)(t), where R is the percent of the returned
double-tagged fish which retained both tags. The number of
recaptures during each interval was divided by the percent
tag loss estimated for that interval. The adjusted
recaptures were summed over all intervals.

An instantaneous rate of tag loss of 0.2184 is the same
as an annual tag loss of 19.6 percent. This value is larger
than the 11.1 percent reported by Ebner and Copes (1982),
but is approximately the same as the 19.3 percent estimated
by Humphreys (1978).

My estimate is crude, at best, because it is based on
the return of a single incomplete double-tag, for each tag
type. Continued return of double-tagged lake whitefish in
the remainder of the 1982 season, and in future seasons,

will put more strength in this estimate.

Aging

Scale aging was considered to be a reliable technique
for the purposes of this study. A ninety-four percent
agreement was found between ages assigned by scales and
those assigned by fin rays (Appendix 5). Mills and Beamish
(1980) noted similar agreement (93%) between ages assigned
by scales and fin rays of experimentally reared whitefish.

A fifteen percent overlap by the two principal scale
readers for the first two samples showed an overall 82.3
percent agreement. The difference between ages assigned by
the investigators only varied by one year for any fish
(Appendix 4). This is better agreement than the 78 percent
reported by Christie (1963) and the 60 percent reported by
Healey (1980). Ricker (1975) states that 80-90 percent
agreement is good, and is only attainable in fast growing
populations.

Beamish et a1 (1976) found scales to underestimate the
true ages of the older whitefish in northern Canada.
Humphreys (1978) did not find this to be the case in
northwestern Lake Michigan. Perhaps more distinct seasonal
temperature variation, which is a major factor influencing
the formation of annular marks on scales (Hoagman 1968),
results in the more reliable aging of whitefish in the
relatively warmer waters of Lake Michigan compared to

northern Canada.

35

36

A method of comparing the precision of a set of age
determinations (Beamish and Fournier 1981) was used to
estimate the "index" or average of the mean percent error of
the overlap by the two principal scale readers. This method
was also applied to the scale and fin ray overlap. Since
there was no disagreement in the assignment of ages to
whitefish by move than one year the index was the same for
both scale readers and for both aging techniques. The index
of average percent error was 2.4 percent for the overlap by
the principal scale readers, and 0.8 percent for the
scale-fin ray overlap (Appendices 3 and 4). These measures
imply that the assignment of ages to lake whitefish in this

study was consistent, i.e. precise.

Age Compositions

The percent age composition of the sampled catch from
the North Shore, Leland, and Beaver Island areas indicates
the existence of at least one distinct stock in each area.
The presence of the strong 1977 year-class, which first
appeared as age III fish in the November 1980 catch, is
noted throughout the study area (Table 5). This age class
dominated the catch during all seasons (Table 6).
Statistical analysis of the age compositions between the
sampling locations showed that they were all significantly
different (p<.01)(Table 7).

The apparent absence of fish older than age V in the
June 1981 North Shore sample contrasts sharply with the 30.8
percent and the 10.6 percent of older fish present in the
Leland and Beaver Island areas, respectively. The
increasing representation of the 1977 and 1978 year classes
in the catch, as the seasons progress, suggests that these
fish were not fully vulnerable to the gear until the end of
the 1981 season, or later.

Eschenroder et a1 (1980) conducted an investigation
concerning the selectivity of large mesh trap nets for
whitefish in Hammond Bay, Lake Huron. They noted that
whitefish are not fully vulnerable to the gear until the
reach a length of 489 millimeters in total length. The
girth-length regression for the North Shore stock is nearly
identical to that from Hammond Bay (Appendix 6). This

indicates that only a small proportion of the whitefish in

37

Table 5.

:38

in northeastern Lake Michigan in 1965 through 1982.

Percent age composition of commercial trap net (TN) and gill net (GN) catch

 

 

 

 

 

 

 

Area Age
Gear N Ref.
Date 1 2 3 u 5 6 7 9 1o 11 12 :13

Leland (a)
May 1982 TN 232 1.7 12.9 53.0 7.8 1.6 O “ 3.9 3.9 2.2 1.7 0.9
Oct. 1981 TN 2““ O.“ 6.6 67.2 11.1 7.“ 1.2 0.8 0.8 2.0 1.6 O.“ 0.“
Aug. 1981 TN 9“ 6.“ 67.0 5.3 7.“ 5.3 1 1 1.1 “.3 2.1
June 1981 TN 81 2.5 ““. 22.2 12.3 6.2 2 5 1.2 2.5 “.9 1.2
Oct. 1980 TN 11“ 57.0 28.9 7.0 2.6 1 8 1.8 0.9
Grand Traverse Bay . (a)
June 1981 TN 1“0 “.3 57.1 29.3 3.6 2.1 .“ 0.7 1.“
Grand Traverse Bay (Grid 715) (b)
Fall 1981 TN 223 11.2 “5.3 20.6 1.8 “.9 3.1 “.5 1.3 “.9 2.2
Grand Traverse Bay - North end (d)

1973 ON 53 51 22 21 u 2

1971 GN 28 29 21 39 11

1969 ON 371 1 22 37 18 8 2 1

1968 ON 189 6 6“ 27 2 1
North Shore (a)
May 1982 TN 263 0.4 17.1 79.5 3.0
Oct. 1981 TN 331 7.9 85.8 6.0 0.3
Aug. 1981 TN 300 13.0 83.7 3.0 0.3
June 1981 TN 107 0.9 1“.O 79.“ 5.6
Oct. 1980 TN 513 0.2 77.6 20.9 1.0 0.2 0.2
North Shore (Grids 214,215,216) (b)
Summer 1980

GN 371 36.7 56.3 6.5 0.3 0.3

North Shore (C)
July 1978 TN “07 31.9 36.6 16.5 10.3 2.5 0.7 1.0 0.5
May 1979 TN 198 a. 59.6 29.2 5.6 5.1 0.5 0.5
Epoufette (d)
Oct. 1976 TN 18“ 1.6 83.2 1“.1 1.1
North Shore (d)

1973 TN 1u1 1 1 65 3o 1 1 1

1972 TN 1“1 1 28 68 2 1

1971 TN 296 8 81 10 1

1970 TN 169 ““ 5“ 2
North Shore (e)
Oct. 1966 TN 3“6 0.9 16.5 77.5 5.2
Beaver Island (a)
May 1982 TN 169 31.“ 63.9 “.1
June 1981 TN 219 2.7 68.0 18.7 5.5 2.3 2.3 0.5
Cross Village (Grid “18) (b)
Fall 1981 GN 46 10.9 59.3 17.9 6.5 2.2 4.3 2.2 2.2
Hog IslandZIle aux Galets (c)
June 1979 TN 122 12.3 30.3 16.“ 6.6 “.8 2.5 6.6 “.9 “.1 1.6
Oct. 1978 TN 2“2 1.1 21.8 50.7 15.2 1“.7 2.“ 3.2 1.9 2.“
July 1978 TN 211 0.5 12.8 38.“ 15.2 22.3 3.3 2.“ 2.“ 2.8
Oct. 1977 TN 12“ “.0 36.3 36.3 20.2 0.8 1.6 0.8
Hog Island (e)
Oct. 1965 ON 59 6.8 81.3 8.5 1.7 1.7

 

this study

)

) unpublished data from the
) Rybicki (1980)
) unpublished data from the
) Piehler (1967)

Sault Ste. Marie tribe of Chippewa Indians

Michigan Department of Natural Resources

39

 

 

 

:.: 6.H :.m 5.5 6.: m.mm H.om m.m o:H Hmmfi mean
wan mmmm><me ozamo

m.m 5.m 5.no 5.5m 56H mama has

m.o 5.m m.m H.5 N.Hm m.mo m.H 5H5 Hmmfi mesa
ozaqu mm>¢mm

6.5 m.: 6.: m.m :.5 5.o m.mH 5.5 H.H: m.m 5.0 mmm mama has

N.H :.H 5.: m.m 5.H a.H 5.H H.0H o.HH 0.6m w.n H.o ::N Hmmfi .960

6.: 5.0a m.m 5.H :.oH 0.5 0.6 6.0m ~.: :5 Hmmfi .ms<

:.: m.5 0.6 5.5 :.m m.o o.:H m.mH :.mm m.m Hm Hmmfi mesh

o.m m.m o.m w.m :.HH H.:m :.m: :HH ommfi .poo
ozaqmq

H.: H.Hm m.:H m.o mom «was 562

:.o o.m 5.mm 5.m Hmm Hams .poo

m.o 5.: o.:m n.3H oom Hmmfi .ms<

5.5 m.om N.HH m.o 5HH Hams mash

m.o m.o 6.H o.mm m.m5 H.o mam ommfi .poo
mmoxm mamoz

max NH as ea 5 m 5 o m : n m 2 mean
mop¢

 

.mmmfi :wsousp ommfi cu cmwflnOMS mama

campmmmnpmoc Eopm mmmao mwm comm CH Amocoomv copmo noamemm map Ho psoopwm

.0 manme

“0

Table 7. Age composition comparisons among sampling
locations using a Chi square goodness of fit test.

 

 

Samples Compared Age Categories X df a
Fall 1980
North Shore (115,116) vs. <4,4,5,6+ 50.13 3 <.001
Leland (714,814)
Spring 1981
North Shore (117) vs. <4,4,5,6,7+ 58.49 4 <.001
Leland (812,814,912)
North Shore (117) vs. 3,4,5,6,7+ 35.42 4 <.001
Beaver Island (316)
Leland (812,814,912) vs. 3,4,5,6,7+ 21.95 4 <.001
Beaver Island (316)
Summer 1981
North Shore (218) vs. 3,4,5,6,7+ 65.64 4 <.001

Leland (812,814,912)
Fall 1981

North Shore (116,117) vs. 3,4,5,6,7+ 57.94 4 <.001
Leland (812)

Spring 1982

North Shore (216,218) vs. 3,4,5,6,7+ 86.32 4 <.001
Leland (812,814,912)

North Shore (216,218) vs. 4,5,6+ 13.39 2 <.01
Beaver Island (317,418)

Leland (812,814,912) vs. 3,4,5,6,7+ 61.89 4 <.001

 

41
the North Shore catch are fully vulnerable to the trap nets,
thus skewing the age composition of the sampled catch to the
right, i.e. older age classes.

A slightly higher percentage of fish aged IV and V in
the October 1981 sample, compared to those in the August
1981 sample, indicates a possible movement of older fish
onto the spawning grounds in the fall. Seasonal inshore
movements of older whitefish have been observed in Green Bay
(Ebner 1980, Humphreys 1978, Gunderson 1978). If the tagged
fish are indeed homing to the same spawning grounds in
subsequent years, this magnifies the importance of the
protection of the North Shore spawning stock to the entire
fishery in that area. A depletion of failure of this
spawning stock could mark the collapse of the fishery
between Seul Choix Point and the Mackinac bridge.

The apparent absence of fish older than age VI in the
May 1982 sample from the Beaver Islands area (grids 317 and
418), compared to the June 1981 sample from that vicinity
(grid 316), suggests the existence of either a second stock
or a segregation of the population by age classes in the
region southeast of the Islands (Figure 6). The absence of
tag returns from this area does not permit a distinction
between these possible explanations.

Comparisons of the 1980-1982 age compositions from the
North Shore area, with those reported by the Michigan
Department of Natural Resources, show strong similarities

(Table 5). The 1970-1973 catch sampled in the North Shore

“2

.cmwflnowz oxmq mo wopm UCmHmH po>dmm mSp EoHM .NmmH

 

 

 

pcm mefi mmcasmm SH .5ma5mpflcs oxma ho :oPMo voamemm map 50 mCoHPamoQSoo mw¢ Aw opswwm
nw¢¢w>u was
Q m h m w v m
Wan... E E _l._ W W E o
5 5
5 5
W k .. on .52
m
5 m
\
5 w
5 4..
5s ”nan“ n. 5
5
“t T 8
. E

 

 

 

 

 

Db

43
area was dominated by ages III and IV. The harvest during
that period was dominated by the 1965 and the 1969
year-classes. The narrow age distribution and the
dependence of the fishery on one or two age classes, is a
characteristic of this fishery.

The presence of numerous age classes in the samples
from Grand Traverse Bay is also consistent with the values
reported by the Michigan Department of Natural Resources.
Minor differences between their results and this study are
probably a reflection of the different selectivity
characteristics and minimum size limit of the purse seine.
Gill net catch was sampled for the Michigan DNR's estimates.

Any differences in the age composition between two
areas may be a reflection of differential fishing pressure
exerted on the stocks, segregation of the population by age
class or by sex, variable recruitment (Christie 1963, Lawler
1965), predation (Wells and McLain 1973), or food
availability during the life stages, from egg to adult. The
presence of the strong 1977 year-class in the catch,
throughout the study area, leads me to believe that the
factors affecting recruitment are rather homogenous between
areas.

Cucin and Regier (1966) and Roelofs (1958) showed that
the strong 1943 year classes of whitefish in Lake Huron and

the lake herring Coregonus artedi Leseur in Lake Michigan

 

corresponded with the strong 1943 year class of whitefish in

Lake Michigan. This indicates that the factors favoring the

44
abundance of the whitefish and similar species are indeed
somewhat homogenous within the northern Great Lakes.

A pattern in the annual variation of year class
strength has been observed in Lake Michigan in past
investigations, whereby a strong year class is followed the
next year by a weaker year class (Roelofs 1958, Humphreys
1978). The 1978 year class in the present study follows
this trend, in that it is relatively less abundant than the
1977 year class. This succession of an abundant cohort by a
less abundant cohort is probably due to competitive
intraspecific interactions among the prerecruited whitefish.

Lawler (1965) suggested the existence of some
relationship between the temperature during spawning,
embryonic development, and hatching and the abundance of
year classes of whitefish. The effect of cold winter
temperature on the earlier formation and longer duration of
the ice cover on the lake was discussed as a possible
mechanism by which the effects of the strong November winds,
upon the turbidity of the water on the spawning grounds, may
be reduced. This, in turn, would enhance survival of the
eggs.

In order to determine the true factors limiting year
class success, detailed investigation of egg, larval, and
juvenile mortality are necessary. This information would
aid tremendously by enabling the prediction of the strength
of prerecruited age classes prior to their entry into the

fishery. Ultimately, catch quotas could be tailored to each

45
stock, by incorporating knowledge of the relative abundance
of the prerecruited stock with that of the recruited stock.

The 1977 year-class had a strong influence on the mean
age of the North Shore samples during the study period
(Table 8). The steadily increasing mean age from sample to
sample throughout the sampling period indicates that the
same stock was being sampled at all times.

In Leland this pattern is not as prevalent. The higher
mean ages for this area reflect the presence of numerous
fish aged V and older in the catch. The mean age in the
spring samples are higher than those for the summer and fall
samples. This may be a result of a change in the location
of the gear, small sample sizes, or a seasonal segregation
by age classes.

Healey (1980) noted a complete removal of the older
ages of lake whitefish and subsequent increases in the
younger ages associated with increased exploitation. In
this study, the North Shore and Beaver Island stocks are
characteristic of heavily exploited stocks while the Leland

stock is characteristic of a lightly exploited stock.

Table 8.

Mean age of lake whitefish from the sampled catch

in the North Shore, Leland, Beaver Island, and Grand Traverse
Bay areas of Lake Michigan.

 

 

Sample Age
Location Grids Date Size Mean S.Dev.
North Shore 115,116 10/29/80 513 -3. 24 0.56
117 6/29/81 107 3. 90 0.48
218 8/24/81 300 3. 91 0.41
8/25/81
116,117 10/17/81 331 3.99 0.39
10/24/81
216,218 5/17/82 261 4.85 0.44
Leland 714,814 10/30/80 114 3.81 1.46
812,814,912 6/15/81 81 5.53 2.52
812,814,912 8/27/81 94 4.82 1.90
812 10/21/81 244 4.60 1.65
10/22/81
812,814,912 5/20/82 232 5.83 1.97
5/24/82
Beaver Island 316 6/16/81 219 4.45 0.96
317,418 5/18/82 169 4.76 0.67
5/19/82
Grand Traverse 615 6/14/81 140 5.63 1.44

Bay

 

Population and Biomass Estimates

Population estimates using Petersen mark-recapture
techniques indicate that the North Shore stock is
considerably larger than the Leland stock. There were an
estimated 1.7 million legal sized fish in the North Shore
stock in November 1980 (Appendix 7). The stock boundaries
have been delineated to include statistical grids 115-118
and 213-220 (Figure 3). The biomass estimate for this stock
was 1.4 million kilograms. Ninety-five percent confidence
intervals using a poisson approximation (Ricker 1975) were
1.5-1.8 million individuals and 1.3-1.6 million kilograms.
In Leland there were an estimated 264 thousand whitefish
weighing 329 thousand kilograms in November 1980.
Ninety-five percent confidence intervals are 206-337
thousand individuals and 257-420 thousand kilograms. Leland
stock boundaries include grids 615, 714, 812-814, and
911-912.

The population estimates were based soley upon the
reported catch in pounds, converted to kilograms, and the
number of recaptures from the fishermen known to be
cooperating. These fishermen are the ones who reported all
tags that were recaptured, allowed their catch to be
sampled, and/or donated their time and equipment during the
tagging operations. The catch was adjusted for the
recruitment of those fish that were sublegal in November
1980, yet grew into the catchable portion of the population

during the 1981 fishing season (Appendix 8).

“7

48

In the Leland area thirty-nine tagged fish were
captured during the five weeks immediately following the
closed fishing season (November 1-30). These fish were
returned from the same location where they were tagged. It
is believed that these fish did not get a chance to
thoroughly mix with the remainder of the Leland stock.
These recaptures provided an estimate of 7.6 thousand fish
on the spawning grounds. The estimate of the spawning
population represents 2.9 percent of the total estimated
numerical size of the Leland stock. This suggests that this
particular spawning population was a minor portion of the
entire Leland area spawning stock. These fish were not
included in the total population estimate, since it was
believed that they were captured before they could disperse
and mix with the entire untagged portion of the population.

Population numbers and biomass were dominated by the
1977 year class in all areas (Table 9). The percent
representation of each age class in the North Shore area
differs little between the numbers and biomass columns,
where as in the Leland area, the older fish make up a larger
percentage of the total biomass than the total numbers. For
example, fish aged V and older represent 30.5 percent of the
total harvestable biomass, yet only 14.5 percent of the
total numbers. The closing of the Leland area to commercial
fishing from 1970-1976 has allowed the survival of a larger
percentage of older fish in this population, compared to the

North Shore area. Historically, the North Shore area has

“9

Table 9. Population numbers and biomass (kg) of lake
whitefish by age class for November 1980 in the North Shore,
Leland, and Leland spawning areas. All figures were derived
from the October 1980 percent age compositions in the
respective areas.

 

North Shore Leland Leland Spawners
Age Number Biomass Number Biomass Number Biomass

 

3 1,272,535 1,058,329 148,945 149,558 4,295 4,313

4 365,906 338,895 76,849 79,391 2,216 2,290
5 16,632 22,976 19,278 37,544 556 1,083
6 0 0 0 0 0 0
7 3,326 4,308 . 7,130 18,448 206 532
8 0 0 4,754 16,471 137 475
9 0 0 4,754 18,118 137 523
10 3,326 11,488 0 0 0 0
11 0 0 0 0 0 0
12 2,377 9,883 69 285

 

Total 1,663,207 1,435,996 264,087 329,422 7616 9501

50
supported a larger number of fishermen and has produced
considerably larger yields than the Leland area (Figure 1).
The larger size of the North Shore population is the reason
that this is so. Why this area is more productive requires
detailed examination of the spawning habitats, food
availability, and fecundity characteristics of the whitefish

in these areas.

Mortality Estimates

The estimated annual exploitation rate (u) of the North
Shore tagged fish (49.6%) was more than double that of the
Leland tagged fish (24.2%). The rates of exploitation were
estimated from those tags returned by the cooperating
fishermen in each area. These rates were expanded by the
ratio of the total catch to that of the cooperating
fishermen in each area to obtain values of the total
exploitation rates for each stock (Appendix 9).

An estimate was not possible for the Grand Traverse Bay
tagged fish. Too few tags were returned to adequately
define the stock boundaries. In addition, since the fish
tagged in this region were all shorter than the minimum size
limit (19 inches), several of the recaptures were returned
to the lake.

The annual survival rate (S), estimated from tag
returns, was 29.3 percent for the North Shore stock and 41.4
percent for the Leland stock. The corresponding
instantaneous total mortality rates (2) were 1.229 and
0.881, respectively (Table 10).

The instantaneous total mortality rate (1.229)
estimated for the North Shore stock was approximately that
estimated for ages III and IV (1.240) by Patriarche (1974)
for whitefish in northern Lake Michigan (Table 11). Healey
(1975) reported a value of 1.022 as the mean of 14 exploited
populations of whitefish and a value of 0.673 as the mean of

13 unexploited whitefish populations. The North Shore

51

52

 

 

Ammm.nmpv Aammfi mcflpmmv
mmH.H 500.0 Awumv o>pso copmo ham om9o>whe ccmso
Ao:m.umpv Ammcfihmmv
mno.fi mmw.o Amudv m>hso nopmo GCMHmH hm>wmm
Awmm.nmnv Amaammv
:wm. 5mm.o Awuév o>hso nopwo
Anam.u My Amwcflpmmv
mnm.m :No.o Awlmv m>pso SOPMO
mam.o mom.o Ham.o omm.o mchspmh was acmamq
Aooo.fiumhv AmHH65v
do5.m omm.o Aoudv o>hso sopmo
A555.umpv Ammcflpamv
cM5.H :Nm.o Awudv o>pso :OPMo
mwm.o How.o mmm.H mo5.o mauspop may mnonm cpmoz
coapmeflpmm
E m N < mmw< Mo CoflmeoA
66:96:

 

.cmmflcoas mxmg campmMmsppoc ma nmflmopfics oxma mo mmpmu hpwddpuoe AEV Hahspm: msomsmp
Icmpwcfi ccm .Amv mcflnmfl% msoocmPCMPmcH .ANV Hapop msomsmpcmpmcfi .A<v Hmscc¢ .49“ wands

53

 

 

Table 11. Annual (A), instantaneous total (Z), instantaneous fishing (F), and instanta-
neous natural (M) mortality rates for lake whitefish from Lake Michigan and
other lakes.

Location
(Method of Estimation) Ages A 2 F M Reference
Northshore
tag returns 0.708 1.229 0.861 0.368 This study
catch curve 4-6 0.824 1.736 This study
(springs)
catch curve 4-6 0.939 2.794 This study
(falls)
catch curve 3-“ 0.711 1.2“0 0.820 0.420 Patriarche, 1974
Leland
tag returns 0.586 0.881 0.363 0.518 This study
catch curve 5-8 0.624 0.979 This study
(springs)
catch curve 4-8 0.587 0.884 This study
(falls)
Beaver Island
catch curve 4-9 0.602 0.920 This study
(spring 1981)
catch curve 5-6 0.935 2.736 This study
(spring 1982)
catch curve 4-9 0.658 1.073 This study
(springs)
Grand Traverse Bay
catch curve 5-8 0.697 1.195 This study
(spring 1981)
catch curve 5-13 0.340 0.416 0.416 Rybicki and Keller,
1977
Big Bay de Noc
catch curve 4-8 0.772 1.478 1.064 0.414 Ebner, 1980
(springs)
catch curve 3-4 0.936 2.789 Roelofs, 1958
North-Moonlight Bays
catch curve 4-8 0.610 0.941 0.464 0.477 aner, 1980
(springs)
catch curve 4-10 0.668 1.102 0.634 0.468 Humphreys, 1978

Lesser Slave Lake 0.529 0.753 0.124 0.629 Bell et a1., 1977

Canada (age composi-

tion)

Average of 14

exploited populations 0.640 1.022 Healey, 1975

Average of 13 unex-

ploited populations 0.490 0.673 0 0.673 Healey, 1975

 

54
estimate is larger than both of these estimates. The Leland
estimate (0.881) is intermediate between the two. Both the
Leland and North Shore estimates are considerably less than
the value of 2.813 calculated by Roelofs (1958) for the fish
in Big Bay de Noc. A large proportion of his total was

attributed to sea lamprey, Petromyzon marinus Linnaeus,

 

predation.

The instantaneous rates of total mortality for the
Beaver Islands and Grand Traverse Bay stocks, estimated from
age compositions, were 1.073 and 1.195 respectively. These
estimates were derived from catch curves (Robson and Chapman
1961), as are most reported values of Z in the literature.
A catch curve is a plot of the natural log of frequency
against age. Total instantaneous mortality is estimated
from a linear regression of the points in the descending
limb of this plot. The slope of this regression is the
total instantaneous mortality rate (2). An estimate of this
type assumes that the rate of recruitment is constant for
all ages present in the fishery. This assumption is
obviously not valid when strong year classes, such as the
1977 year class, are present in the catch. Robson and
Chapman (1961) suggest averaging the percentages of the fish
in each age class over several years to offset the effects
of unequal recruitment. The relative frequencies of ages
were averaged over the two years of the study, when data was
available (Table 11, Figure 7). The range of ages used for

each estimate is indicated by the horizontal range of the

55

Figure 7. Catch curves for stocks of lake whitefish
from the North Shore, Leland, Beaver Island, and Grand
Traverse Bay areas of Lake Michigan, in spring and fall.

- North Shore, springs of 1981 and 1982 combined
- North Shore, falls of 1980 and 1981 combined

- Leland, springs of 1981 and 1982 combined

- Leland, falls of 1980 and 1981 combined
Beaver Island, spring 1981

- Beaver Island, spring 1982

- Beaver Island, springs of 1981 and 1982
combined

- Grand Traverse Bay, spring 1981

3‘ unwound!»
I

DOG

Figure 7.

56

 

l-1.73B

O\
O

l=2.704

O

9

 

.‘q

I I I I I

l-0.079

O
O
O
O 000

'(D

T r

#1

z-o.ssa d

 

.
.
q
.
4

Z=0.920

o
o
0 o o

I I

Z ' 2.736

I r I

 

 

o
0 o o
o

 

 

T
2

I I

.-
O

1
8 10 12

AGE

‘1
.q

and
.s
O
.6
fl

 

57
regression line in each plot. Estimates were calculated for
the Beaver Island samples independently, and combined.
Separate estimates were determined for spring and fall
samples.

Although the relative age frequencies were averaged
over the two years of the study, the effects of the 1977
year class were still prominent. Therefore, the estimates
for the North Shore area and for the spring 1982 Beaver
Island sample are considered to be overestimates. The
estimates for the Leland area are quite similar to the tag
return estimates for that area. This indicates that the
catch curve estimates are reasonably accurate when a broad
age composition is present in the catch.

Examination of the descending limb of the fall catch
curve for the Leland stock reveals an obvious concavity.
This area was closed to commercial fishing in 1970 and was
reopened in 1977 (Figure 2). Consequently, the older fish
in the population had higher survival rates during their
early life and were still present in the catch. The younger
ages have lower survival rates, as a result of fishing
mortality in addition to natural mortality. This is
indicated by the steeper slope in that portiOn of the curve
representing these ages. In order to obtain the best
estimate of the mortality rate of the Leland fish from catch
curves, ages IX and older were omitted from regressions.

The first age class considered to be fully vulnerable

to the gear was that age class which was the most abundant

58
in the catch. Considering the results of the ‘trapnet
selectivity study conducted on whitefish in Hammond Bay,
Lake Huron (Eschenroder et al 1980), there are probably no
age classes which are fully vulnerable to the gear in the
North Shore and the Spring 1982 Beaver Island samples.

Catch curve estimates are generally not suitable for
comparison between studies, due to the somewhat arbitrary
manner by which the first age class that is fully vulnerable
to the gear is selected. Commonly the most abundant age
class is considered to be the youngest age used in the
construction of catch curves, however these fish may not be
captured with 100 percent efficiency by the gear. Cucin and
Regier (1965) found that the narrow range of ages present in
single mesh gill net catch overestimates whitefish total and
natural mortality rates. This is due to the sharp decrease
in efficiency of gill nets beyond a certain size of maximum
efficiency of capture.

The instantaneous rate of fishing mortality (F) was
higher for the North Shore stock (0.861) that for the Leland
stock (0.364) (Table 10). The instantaneous fishing
mortality represents 70.1 percent of the total instantaneous
mortality in the North Shore area and 41.3 percent in the
Leland area.

Ebner (1980) reported F values of 1.064 for the
whitefish from Big Bay de Noc and 0.464 for whitefish from
North-Moonlight Bays (Table 11). These values represented

72 percent and 49 percent of the total instantaneous

59

mortality for each area, respectively. The 2 values for
this study were estimated from catch curves that exhibited
concavity of the lower portion of the descending limb, as
was illustrated in the Leland sample. The concavity of the
descending limb of a catch curve tends to underestimate the
total instantaneous mortality rates, thus overestimating the
contribution of fishing mortality as a relative frequency of
z. This in turn causes the instantaneous natural mortality
rate (M), which is the difference Z-F, to be underestimated.

The instantaneous rates of natural mortality were
estimated to be 0.368 in the North Shore area and 0.518 in
the Leland area. These represent annual natural mortality
rates of 29.9 percent in the North Shore area and 58.7
percent in the Leland area. The lower rate of natural
mortality in the North Shore area may be a result of the
larger removal of individuals by the fishery. This may act
to reduce the mortality from non-fishing sources. Healey
(1975) reported an instantaneous natural mortality rate of
0.629 as an average for 13 unexploited whitefish
populations. Rybicki and Keller (1977) reported a value of
0.416 for M in an unexploited portion of Grand Traverse Bay.
Additional values of 2, F, and M from the literature are

presented for comparison in Table 11.

Length and Weight

The mean length and weight of the average fish in the
sampled catch was larger in the Leland stock than the North
Shore and Beaver Island stocks at the time of each sample
(Table 12). This is a consequence of the older age
structure of the Leland catch. The increase in the mean
size of the average fish in the Leland and North shore areas
during the sampling period reflects the dominance of the
1977 year class in the catch. The 1977 year class comprised
73-86 percent of the North Shore catch. This year class
also dominated the Beaver Island and Leland samples,
representing 63-64 percent and 32-56 percent of the catch in
these areas respectively. The presence of this strong year
class overshadows the effect of the recruitment of sublegal
fish, thus the mean size increased from sample to sample.

The deviation of the June 1981 North Shore sample, from
this trend of increasing mean size, is probably due to the
nature of this particular sample. The June 1981 sample is
not truly representative of the dockside commercial catch.
It was obtained on a day when the Michigan Department of
Natural Resources fisheries biologists were obtaining a
sample of lake run catch, i.e. sublegals were included, for
sex and maturity estimates. Consequently, fish that were
smaller than those that would ordinarily be kept by the
commercial fishery, were present in disproportionate

numbers.

60

Table 12.

61

Mean length (mm) and weight (g) of lake whitefish

from the sampled catch in the North Shore, Leland, Beaver

Island, and Grand Traverse Bay areas of Lake Michigan.

 

 

Sampling Length Weight
Grids Date N Mean S.D. Mean S.D.
North Shore
115,116 10/29/80 513 459.9 25.9 848.3 202.5
117 6/29/81 107 441.4 35.3 855.8 243.2
218 8/24/81 300 460.5 21.6 934.4 167.7
8/25/81
116,117 10/17/81 331 473.3 23.6 950.8 184.0
10/24/81
216,218 5/17/81 263 473.6 24.4 969.7 179.6
5/18/81
Leland
714,814 10/30/80 114 486.0 58.9 1231.3 728.2
812,814,912 6/15/81 81 501.4 68.9 1457.8 817.4
812,814,912 8/27/81 94 501.3 63.8 1549.0 899.2
812 10/21/81 244 511.8 62.9 1526.2 838.4
10/22/81
812,814,912 5/20/82 232 525.3 68.5 1695.9 899.7
5/24/82
Beaver Island
316 6/16/81 219 479.0 35.6 1099.6 275.7
317,418 5/18/82 169 475.5 31.1 1049.1 331.5
5/19/82
Grand Traverse Bay
615 6/14/81 140 499.4 42.8 1329.2 494.9

 

62

Gear selectivity may have had some effect on the June
1981 sample means. If the faster growing members of the
1977 year class were vulnerable to the fishing gear as age
III fish in the fall of 1980, then the four year olds in
June 1981 would represent the slower growing members of this
year class. This could also explain the decrease in the
mean size between these sampling dates. Since no decrease
in the mean size was noted between fall 1981 and spring 1982
samples, the decrease in mean size is believed to be a
result of the nature of the spring 1981 sample.

The presence of proportionately more older fish in the
spring 1981 sample compared to the spring 1982 Beaver Island
sample is reflected in the higher mean size of the whitefish
in 1981. Either two distinct stocks exist in the region
southeast of Beaver Island, or the samples represent
segregation by age class.

The spring 1981 catch sampled from the Beaver Island
area resembled that from the Leland area. These samples
were characterized by the presence of numerous fish age VI
and older in the catch. The older fish, which are also the
larger fish, result in higher mean length and weight values
in Leland than those for the North Shore sample. In the
spring of 1982, the sample from the Beaver Island area
contained few old fish. The mean length and weight of this
sample resembled those from North Shore in May 1982.

The length composition of the sampled catch from the

Leland area was significantly different (p<.001) than that

63
from the North Shore and Beaver Island areas (Table 13).
The larger representation of old fish in the Leland catch
resulted from more fish being in the larger length
categories (Table 14).

Less than one percent of the fish sampled in the North
Shore area were longer than 540 millimeters in total length.
Thirteen to twenty precent of the Leland catch was‘ longer
than 540 millimeters. Five and a half percent of the Beaver
Island spring 1981 sample and 1.2 percent of the 1982 sample
was longer than 540 millimeters. The closure of the Leland
fishery between 1970-1976, and the lighter exploitation rate
in this area, account for the presence of numerous large
fish.

The North Shore and Beaver Island (grid 316) spring
1981 length compositions were significantly different
(p<.001), yet the spring 1982 contrasts were not (p>.10).
Despite the lack of statistical differences, these samples
are believed to have originated from separate genetic stocks
based on the distribution of tag returns.

Plots of the spring Beaver Island samples, with the
North Shore samples included as a reference, illustrate the
differences between the stock structure of the catch sampled
from grid 316 in 1981 and that sampled from grids 317 and
418 in 1982 (Figure 8). The 1982 samples are quite similar,
but the 1981 samples are not. The presence of older fish in
the 1981 Beaver Island sample distinguishes it from the

North Shore sample and the 1982 Beaver Island samples.

6“

Table 13.

Length composition comparisons among sampling
locations using a Chi square goodness of fit test.

 

Sampled Compared

Length Range

 

(grids) (20mm intervals) X2 a
Fall 1980
North Shore (115,116) vs. <440 - >559 72.15 <.001
Leland (714,814)
Spring 1981
North Shore (117) vs. <440 - >559 66.87 <.001
Leland (812,814,912)
North Shore (117) vs. <440 - >539 81.23 <.001
Beaver Island (316) “
Leland (812,814,912) vs. <440 - >559 23.82 <.001
Beaver Island (316)
Summer 1981
North Shore (218) vs. <440 - >559 95.19 <.001
Leland (812,814,912)
Fall 1981
North Shore (116,117) vs. <440 - >559 141.77 <.001
Leland (812)
Spring 1982
North Shore (216,218) vs. <440 - >559 121.84 <.001
Leland (812,814,912)
North Shore (216,218) vs. <440 - >539 3.63 >.250
Beaver Island (317,418)
Leland (812,814,912) VS. <44O - >559 85.95 <.001

Beaver Island (317,418)

 

65

 

 

 

 

 

 

 

ov5 5. v. N.H «.H ov5
95.5 a. w. 85
oo5 m. H.H oo5
one m.a «.H ~.H a.v m.H cam
cow 5. m. m.n o.~ ~.m ~.~ o.~ cue
ave v.~ m.v o.H ~.~ ~.H 0.5 ovo
cma 5. m. m. H.~ omo
coo m.m a. Nu a. com
com 5. o.H v.m 5.5 m. can
own H.~ n.~ v. w. ~.m a.v H.~ N.w m.u com
com o.~ e. v.~ v. o. 5. o. v.m w.m n.m ~.w w.~ cvm
o~m n.m H.v m.m 5.5 o.n m.a o.H m.m n.~H m.o «.0 a. o~m
com o.o~ v.m~ w.va ~.md c.aa n.m a. m.v n.m~ «.5H m.m o.na o.H com
omv o.mm m.- c.o~ m.m~ v.~n 5.mH H.ma m.o~ m.mH H.o~ H.¢H o.m m.ma cav
8v v.3 o.m~ 1mm m.on o.~n mdn 6.: n.5N 9nd v.m mdn mdN 1mm 03.
O¢v m.v n.v~ ~.v~ o.m~ o.o~ m.nm ~.o~ A.mm m.o v.5 w.oa m.~a v.m~ ovv
o~v A; o.v o.v m6 n.” odm m.5H @.N m.v ~.m «.w m.n omv
ocv m.a m.~ m. H.ma n.~ v. o.~ «.H m.~ m. ocv
omm 5. m.~ v.m can
cc... m . v. own
cvm cvn
omn a. own

AEEV AEEV
b.8030 33 «mad Sad mama dwma Sad Hum." 83 mama Sad god Sad coma b.8950
59.3 958m 3.06m 5:08 69.88 Sum 055m 69.0% :65 6:08 Sam 855m 550% Zn... 583

8m .a. .o 9:33 6933 985. 5.82 953

.NmmfiuomoH CH cmwflcowz mxwa cuopmmozuso: song noumo ooaasmm ocp mo mCOHvaoasoo camcoq .3” wanna

66

E 321112.117. "
,,. _

16-

10-

,-

° r} 1 r1 (ELI

320 340 360 380 400 420 440 480 480 500 520 >540

so-

25- 1

zo- ‘

15- {

1o-

5 - .

° 1 r’ Trl . j
440 430

320 340 300 380 400 420 480 500 520 >540

LENGTH (mm)

 

1981

 

 

 

 

,
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I.

.\\\f\\\\\\\\\\\\\\\\

 

 

 

 

 

 

 

 

 

 

 

1982

PERCENT

      
 

A\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

 

 

 

 

 

 

 

 

Figure 8. Length compositions of the spring 1981 and spring
1982 sampled catch of lake whitefish from the North Shore and
Beaver Island areas of Lake Michigan.

67

The mean length of the whitefish comprising the 1977
year class from the Leland samples (Figure 9) were larger at
each age than those from corresponding samples for the North
Shore area (p<.05)(Tab1e 15) (Appendix 10). The June 1981
whitefish in the North Shore area from the 1977 year class
were shorter than the corresponding fish in the Beaver
Island area (p<.05). The May 1982 comparison of the mean
length of the fish in this year class, between these two
areas, was not significant (p>.10). The opposite was found
for the comparisons of the mean length of the fish from the
1977 year class between the Leland and Beaver Island spring
samples. The 1981 contrast was not significant (p>.10). yet
the 1982 contrast was (p<.05). This further exemplifies the
idea that the Beaver Island samples were obtained from
different stocks, or subpopulations of the same stock, in
subsequent years.

The plots of the mean length at age for the
representatives of the 1976 year class exhibit trends
similar to those observed for the 1977 year class (Figure
9), especially for those fish sampled in August 1981,
October 1981, and May 1982. The November 1980 means were
not significantly different (p>.10) between the North Shore
and Leland stocks, however the August 1981 samples exhibited
differences in mean length (P<.01), with the Leland fish
being larger. These are the only samples where the 1976
year class was present in sufficient numbers to justify

statistical comparison (n>19).

68

 

“s ‘ coco-0.----.” ...... .

.. G—ONorth Shore If
no s---aLeland ..’

t-«sBeaver lsland ,0

405-

“II-

 

 

405‘

400-

455 cl

450-

 

«s- 1977 year class

 

440

 

 

2..
0.4
g—
a.
3..
)5;
:5.
L-
L...
>9
m...
0.4
:—
o~
c...
1".-
:—
,J
5:-
L

 

550
O
545 - ”a

0 " '0'
5‘ s—o North Shore
1135 - ou-s Leland [-

«k-«tBeaver Island .o'
530 - '_,,..c'

LENGTH (mm)

520 -
515 .
s10 -
sos -
soo -
495.
490 -
485 .

 

«0-
415. 1976 year class

 

 

470

 

 

Figure 9- Mean length of the 1977 and 1976 year classes in
the sampled catch from northeastern Lake Michigan in
October 1980 through May 1982.

69

Table 15. Results of comparisons of mean length (mm) at age for lake whitefish from north-
eastern Lake Michigan using Sgheffé tests. The hypothesis of equality of means
was rejected if the interval gk 1 95% M50 did not include zero.

 

 

 

Year Mean Scheffé Test
Class Length Contrast 95% 90%
Samples Compared (Age) (mm) qk MSD MSD Decision
November 1980
North Shore vs. 1977 453.4 10.35 4.97 reject
Leland (III) 463.7
North Shore vs. 1976 478.4 6.05 12.31 10.82 accept
Leland (IV) 472.3
June 1981
North Shore vs. 1977 444.8 21.32 9.31 reject
Leland (IV) 466.1
North Shore vs. 1977 444.8 23.26 6.36 reject
Beaver Island (IV) 468.1
Leland vs. 1977 466.1 1.93 8.69 7.61 accept
Beaver Island (IV) 468.1
August 1981
North Shore vs. 1977 461.3 12.05 5.45 reject
Leland (IV) 473.2
October 1981
North Shore vs. 1977 474.1 19.76 6.02 reject
Leland (IV) 493.8
North Shore vs. 1976 508.8 21.62 16.49 reject
Leland (V) 530.4
May 1982
North Shore vs. 1978 445.2 17.11 10.20 reject
Leland (IV) 462.3
North Shore vs. 1977 478.3 17.47 4.41 reject
Leland (V) 595.8
North Shore vs. 1978 445.2 11.29 8.77 reject
Beaver Island (IV) 456.5
North Shore vs. 1977 478.3 1.33 4.59 3.98 accept
Beaver Island (V) 479.6
Leland vs. 1978 462.3 5.82 9.89 7.03 accept
Beaver Island (IV) 456.5
Leland vs. 1977 595.8 16.14 5.12 reject

Beaver Island (V) 479.6

 

70

The mean length of those fish in the October 1980 and
the June 1981 samples which comprise the 1976 year class,
deviated from the pattern observed for the 1977 year class
in the Leland area (Table 16). The Leland fish in these
samples were shorter than those fish from the other two
areas. The Leland fish were much larger in the later
samples. The first two samples from Leland may not have
been truly representative, due to their small total sample
sizes (n=33, n=18).

The assumption of normality, which is essential for
valid probability statements concerning means using an
analysis of variance, was tested and the hypothesis accepted
for' 17 of the 18 distributions (p<.05) (Appendix 11). A
five percent probability of type I error implies that the
hypothesis of normality will be rejected one time out of
twenty, when the hypothesis is indeed correct. Hence, all
contrasts were tested as though the hypothesis was valid.
Adjustments of the degrees of freedom of the test statistics
for the Scheffe interval were made, prior to testing
differences between means, when the assumption of
homogeneous variance was rejected (Box 1954). Results of
the tests for homogeneous variance are in Appendix 11.

The mean length of the fish in the study area from the
1978 year class was compared between the May 1982 samples.
This is the first sample where this age class was present in
the catch in any appreciable numbers. The mean length of

the fish in the 1978 year class from the Leland area were

71

 

 

 

 

 

 

 

 

H 0m:: mH5 OH
H 055H 005 5
m N55H N05 0
m momH :mm 5
5 005H Hmm NH 5:5H NH5 0
00H 0:0H 00: H: 0:NH mom 5
mm 055 55: 5:H mHoH 00: :
0 05m 50: m
ccmHmH Ho>mmm
H omnn 550 0H
H 00:H 0H5 5
0 momH 5H5 H 0mmH 0mm H ommH 0:5 0
505 055 05: om mmmH 505 5 :5mH 505 0 m5HH 05: m 05mH 5:5 5
m: N50 5:: :0m 055 :5: H55 055 H0: mm 050 m:: 50H 555 05: :
H 050 5H: 0m 0H5 05: 55 5H0 ::: 5H :00 HH: 05m :00 mm: m
H 5H5 mam H 000 0H: m
06030 30603
3 3 H z 3 H z 3 H z 3 H z 3 H 000
~05H 30: Hm5H Hmnopoo H05H Hmsms< H05H menu 005H 6000300 003<

 

.mhmpmeHHHHe 0H 0H Haw00H

.mesnw CH ma pnmwmz

.smwanows oxmq cumpmanpuoc :5 6906 wcwagsmm 3086 vs moms
msoflpm> pm :mHMmPHSB oxma mo sz nonssc Ucm ABV Pcwwms came .AQV spwcoa cams .49H wands

72

 

 

 

 

 

 

 

 

 

H 0550 000 HH
0 00H0 0:0 0H
0 000H 000 0
0 050H 000 5
H: 000H 0H0 0
00 05HH 00: 0
0 000H 00: :
Ham mmh m>dh .H. “0th U
0 0000 :05 0H
: 050: 050 H 0H00 0:5 0 0000 000 : 0:50 000 H 00H: 500 0H
0 0000 :50 : 0:0: 050 0 0000 000 HH
5 5500 000 0 0:0: :00 : 0050 000 H 00:0 050 0H
5 0000 000 0 0:H0 0:0 H 000: 050 0 0500 000 5
H 0000 0H0 0 0000 050 H 00:0 000 0 0000 000 0 0000 550 0
50 HH00 050 0 0000 000 0 0000 0H0 0 0:0H 0:0 0 HH00 H00 5
0H 505H 0:0 0H 5000 H50 5 000H 5:0 0H 000H 500 0
00H :H0H 05: 50 050H 000 0 005H 000 0H 00HH H5: 0 555H 500 0
00 050H 00: :0H H50H :5: 00 H5HH 05: 00 000H 00: 00 000H 05: :
: 000 00: 0H 500 H0: 0 0H0H 50: 0 005 0:: 00 H05 :0: 0
H 00: 000 0
0:0H0H
z 3 H z 3 H z 3 H z 3 H z 3 H 000
005H 303 H05H 0000000 H05H 000000 H05H 0000 005H 0000000 0000

 

50000H00000 .0H 0Hp03

73
larger than those from the North Shore area (p<.05). The
Beaver Island fish from the 1978 year class were larger than
the North Shore fish (p<.05), but were not significantly
different (p>.10) from the Leland fish.

The available sex data indicates that the mean length
at age for females is larger than for males in the Leland
and North Shore areas (Table 17). The females are also
heavier than males at any given age. This would be expected

since many were in their prereproductive condition.

74

Table 17. Mean length at age in millimeters (L) and numbers
(N) of male and female lake whitefish from the Leland and

North Shore areas of Lake Michigan.

 

 

 

 

 

 

 

 

Leland1 North Shore2
Age Males Females Males Females
L N L N L N L N
2 365.0 1 322.0 1
3 425.0 7 431.0 5 402.3 4 413.6 11
4 494.3 40 498.5 24 436.9 33 450.2 50
5 525.0 1 462.0 2 503.5 4
6 558.0 5 596.3 4
9 630.0 1
11 630.0 1
13 720 0 1
Total 54 36 39 66

 

1 sampled October 1981

2 sampled June 1981

Weight-Length Relationships

The transformed least squares predictive regressions of
weight on length varied considerably between areas and among
sampling dates (Table 18). Ricker (1973) suggests that a
functional regression be fit to weight-length data because
the dependent variable, length, is not fixed. A functional
regression fits a line by minimizing the squared distances
of the points from the regression line with respect to both
variables simultaneously. A better predictive relationship
would result from using a predictive least squares
regression where the distances of the points from the
regression line are minimized with respect to the dependent
variable. If the joint distribution of the dependent and
the independent variables is bivariate normal, least squares
estimators are valid for testing hypotheses (Brownlee 1960).

A quadrant test (Cramer 1946) was implemented to test
the joint distributions of weight and length for bivariate
normality (Appendix 12). Dr. John Gill (personal
communication) indicated that tests of the hypothesis of
equality of regression slopes, using analyses of covariance
techniques (Snedecor and Cochran 1967), result in valid
conclusions when the probability of type I error is very
small, even if the hypothesis of bivariate normality is
rejected.

Since all contrasts of slopes were either highly
significant (p<.001) or the variables were jointly

distributed as bivariate normal, least squares regressions

75

76

Table 18. Predictive weight-length relationships for lake
-whitefish from the North Shore, Leland, Beaver Island, and
Grand Traverse Bay areas of Lake Michigan. Total length was
measured in millimeters and weight was measured in grams.

 

Sampling
Location Date Predictive Regression r

 

North Shore

115,116 10/29/80 Log W = '13.15 + 3.24(Log L) .80
117 6/29/81 Log W = -13.49 + 3.32(Log L) .90
117(>400mm) 6/29/81 Log W = -12.83 + 3.20(Log L) .76
ll7(males) 6/29/81 Log W = -12.12 + 3.09(Log L) .84
ll7(females) 6/29/81 Log W = -l3.89 + 3.39(Log L) .92
218 8/24/81 Log W = -13.11 + 3.25(Log L) .78
8/25/81
116,117 lO/l7/81 Log W = -13.35 + 3.28(Log L) .78
10/24/81
216,218 5/17/82 Log w = —11.72 + 3.02(Log L) .76
5/18/82
all samples Log W = -12.37 + 3.12(Log L) .78
Leland
714,814 10/30/80 Log W = -14.79 + 3.53(Log L) .96
812,814,912 6/15/81 Log W a -12.72 + 3.21(Log L) .96
812,814,912 8/27/81 Log W = -13.89 + 3.40(Log L) .94
812 10/21/81 Log W = -14.03 + 3.41(Log L) .94
812(males) 10/21/81 Log W = ~12.53 + 3.17(Log L) .95
812(females) 10/21/81 Log W = -13.78 + 3.38(Log L) .95
10/22/81
812,814,912 5/20/82 Log w = -13.61 + 3.35(Log L) .96
all samples Log W = -13.93 + 3.40(Log L) .95

Beaver Island

316 6/16/81 Log w = -12.60 + 3.17(Log L) .88
317,418 5/18/82 Log W = -13.08 + 3.25(Log L) .85
all samples Log W = -12.82 + 3.21(Log L) .87

Grand Traverse Bay

615 6/14/81 Log W -12.16 + 3.11(Log L) .88

 

77
were chosen over geometric mean regressions due to their
better predictive abilities. Examination of residual plots
indicated the validity of the assumption of homogeneous
variance of the predicted weights around the regression
line, over the range of observed lengths for each area
(Anscombe and Tukey 1963, Behnken and Draper 1972).

The increase in weight (W) of lake whitefish can be
described as a function of length (L) to the power b from
the relationship, W=a1b, where a is a constant. A natural
log transformation of this equation results in a linear
relationship with the slope b. Comparison of slope values
of the transformed relationships is one manner of
determining differences between fish stocks. A higher slope
value for one population compared to another indicates that
that the fish from the former population are heavier at a
given length than the fish from the poplulation with the
lower slope value. The same is true for comparisons of the
same stock at different times of the year.

The slopes of the weight-length regressions were
significantly different (p<.001) between the North Shore and
Leland areas at all sampling dates (Table 19). The same was
true for all contrasts between the Leland and Beaver Island
areas. The fish from the Leland stock are heavier at a
given length than the fish from the Beaver Island and North
Shore area, with one exception (Table 20). The June 1981
sample from the North Shore had a higher slope value (3.32)

for the weight-length regression than the Leland sample

78

Table 19. Tests of equality of slopes of the weight-length regressions between sampling
areas using an analysis of covariance (Snedecor and Cochran, 1967).

 

 

 

Sample Residuals F
Location Size Slope df MS ratio d
October 1980
North Shore 513 3.24 511 3.877 0.0076
Leland 114 3.53 112 0.787 0.0070
Sum 623 4.664 0.0075
Pooled 627 3.57 624 6.042
Difference 1 1.378 1.3776 183.91 < .001
June 1981
North Shore 107 3.32 105 0.818 0.0078
Leland 81 3.21 79 0.592 0.0075
Sum 184 1.410 0.0077
Pooled 188 3.40 185 1.603
Difference 1 0.193 0.1927 25.16 < .001
North Shore 107 3.32 105 0.818 0.0078
Beaver Island 219 3.17 217 1.587 0.0073
Sum 322 2.405 0.0075
Pooled 326 3.21 323 2.417
Difference 1 0.011 0.0118 1.58 > .10
Beaver Island 219 3.17 217 1.587 0.0073
Leland 81 3.21 79 0.592 0.0075
Sum 296 2.179 0.0074
Pooled 300 3.27 297 2.584
Difference 1 0.406 0.4056 $5.10 < .001
August 1981
North Shore 300 3.25 298 1.874 0.0063
Leland 94 3.40 92 0.909 0.0099
Sum 390 2.784 0.0071
Pooled 394 3.71 391 4.139
Difference 1 1.355 1.3550 189.77 < .001
October 1981
North Shore 331 3.28 329 2.484 0.0076
Leland 244 3.41 242 2.639 0.0109
Sum 571 5.123 0.0090
Pooled 575 3.69 572 7.717
Difference 1 2.594 2.5943 289.22 < .001
May 1982
Beaver Island 169 3.25 167 1.198 0.0072
North Shore 263 3.02 261 1.963 0.0075
Sum 428 3.161 0.0074
Pooled 432 3.14 429 3.505 0.0082
Difference 1 0.344 0.3437 46.51 < .001
Leland 251 3.35 249 2.384 0.0096
Beaver Island 269 3.25 167 1.198 0.0072
Sum 416 3.582 0.0086
Pooled 420 3.51 417 4.385 0.0105
Difference 1 0.803 0.8029 93.25 < .001
North Shore 263 3.02 261 1.963 0.0075
Leland 251 3.35 249 2.384 0.0096
Sum 510 4.347 0.0085
Pooled 514 3.64 511 6.848 0.0134
Difference 1 2.501 2.5010 293.55 < 001

 

79

Table 20. Slopes (b) of the predictive weight-length regression for lake whitefish
from northern Lake Michigan.

 

 

Month and Year Sample
Location of sample Size Reference
Leland May-June, 1981-1982 313 3.27 This study
Leland August 1981 94 3.40 This study
Leland October, 1981-1982 358 3.48 This study
North Shore May-June, 1981-1982 370 2.92 This study
North Shore August, 1981 300 3.25 This study
North Shore October, 1981-1982 844 3.31 This study
Beaver Island May-June, 1981-1982 388 3.21 This study
Grand Traverse Bay May-June, 1981 140 3.11 This study
North and Moonlight Bays June-July, 1977-1979 1,832 3.36 Ebner, 1980
North and Moonlight Bays Sept.-Oct., 1975-1979 1,623 3.51 Ebner, 1980
Big Bay de Noc May, 1977-1979 1,462 2.95 Ebner, 1980
Big Bay de Noc Sept.-Oct., 1976-1979 1,668 3.23 Ebner, 1980
Grand Traverse Bay (outer) June, 1978-1979 97 3.03 Rybicki, 1980
Ile aux Galets June, 1978 125 3.21 Rybicki, 1980
Northern Green Bay May, 1979 63 2.95 Rybicki, 1980
Peshtigo Reef June, 1977 269 2.96 Gunderson, 1978
Chambers Island June, 1977 298 3.14 Gunderson, 1978
Grand Traverse Bay (lower) unknown, 1971-1973 486 3.46 Patriarche, 1977
North Shore unknown, 1971-1973 unknown 3.28 Patriarche, 1977
North Shore May, June, Oct., 1966 683 2.91 Piehler, 1967
East of Seul Choix October, 1966 328 3.12 Piehler, 1967
Gull Island Aug.-Sept., 1950 254 2.99 Caraway, 1951
High Island Aug.-Sept., 1950 174 '2.82 Caraway, 1951.

 

80

(3.21). This sample from the North Shore area contained a
disproportionate number of sublegal fish, as mentioned
previously. These smaller fish influenced the slope of the
regression line considerably. When the June 1981 sample
(b=3.32) was combined with the May 1982 sample (b=3.02) from
the North Shore area, the resultant regression slope (2.92)
was lower than either of the separate regression slopes.
Also, if only those fish larger than 400 millimeters are
used in the regression, the slope was 3.20. The influence
of the presence of the sublegal fish on the slope value for
the June 1981 North Shore sample is almost as large as the
differences in slope being tested. Therefore, tests
involving comparisons of slopes from this sampling date
should be interpreted with caution.

The North Shore and Beaver Island comparisons of slope
were not consistent with previous results involving these
samples. The June 1981 slopes were not significantly
different (p>.10). The May 1982 slope from the Beaver
Island sample was larger than that for the North Shore
sample (p<.001). Previous comparisons of vital statistic
parameters between these areas showed similarities for the
May 1982 samples and differences for the June 1981 samples.
The nonrepresentative nature of the spring 1981 North Shore
sample most likely biased this sample, as noted previously.
Significant differences between May 1982 samples are not
surprising since the stocks in these areas are believed to

be distinct.

81

Slope values can be used as an index of the relative
condition of a poplulation of fish. They may be influenced
by factors such as the relative abundance of food, genetic
characteristics of the stocks, optimality of local
temperature regimes, and reproductive conditions of the
fish.

Seasonal variability of the values of the slopes from
the weight-length regressions was obvious within each
sampling area (Table 20). The higher slope valueS' for the
fall samples suggest the presence of whitefish, which are
heavier at a given length, onto the spawning grounds.
Presumably this heaviness is a result of the developement of
reproductive products. It appears from the available sex
data that the females are largely responsible for the higher
fall slope values. The slope value for the weight-length
relationship of the females in the North Shore June 1981
sample was 3.39 compared to the slope value of 3.09 for the
males in that sample (Table 18). The corresponding slope
values for the Leland October 1981 sample were 3.38 for the
females and 3.17 for the males.

Since the market preference for whitefish is in the
round (not eviscerated). only two samples containing sex and
maturity information were available. Because samples
earlier in the year were not available, it was not possible
to discern the true magnitude of the differences between
male and female weight-length regressions, that is solely a

result of ovarian egg developement.

82

Values of the slope of the predictive weight-length
relationships from other investigations of lake whitefish in
northern Lake Michigan are provided for comparison in Table
20. The wide range of reported values further illustrates
the variability of this parameter both seasonally and
between locations. The spring 1981 Grand Traverse Bay
sample has a small slope (3.11), similar to that reported
for the outer portion of the Bay (3.03) (Rybicki 1980), yet
is quite different from the slope (3.46) reported for the
lower Bay (Patriarche 1977). The North Shore slope values
closely resemble those reported from this area by Piehler
(1967) and Patriarche (1977) and are similar to those values
reported for the heavily exploited Big Bay de Noc stock
(Ebner 1980).

Mean Back-Calculated Growth

The legal-sized lake whitefish from the Leland stock
tend to be larger at the time of annulus formation than
those from the North Shore, Beaver Island, and Grand
Traverse Bay areas (Figure 10, Table 21). It appears that
the fish from the North Shore and Beaver Island stocks are
larger at the prerecruited ages than those fish from Leland.
The mean back-calculated lengths at the earlier ages
represent an extrapolation beyond the values of the
dependent variable (total scale radius) used in the
regression, and should be regarded with caution.

The growth curves are nearly identical for the North
Shore, Beaver Island, and Grand Traverse Bay areas for ages
I-VI. Only two fish, out of the 1514 sampled in the North
Shore area, exceeded six years of age, thus no comparisons
were made beyond this age. The Beaver Island and Grand
Traverse Bay samples exhibit similar mean back-calculated
lengths for ages VI-VIII. Insufficient sample sizes (n<5)
for fish aged Ix or older do not allow for meaningful
comparisons in these areas.

The scale radius-body length equations calculated from
the combined samples for each area are given in Table 22.
Regressions were computed for the Beaver Island samples
separately to determine if obvious differences in
back-calculated lengths resulted. The strong similarity
between the separate regressions and associated mean

back-calculated lengths for the Beaver Island samples

83

84

mac

—
_
—
-
—
-
—

_

 

 

tum um¢m>c~= 025.5 II...
ozcgmu ¢m>¢mm l..l
ochuq 2.....-

mmozm 1:32 I

.
I “
“ .“
“‘-“ ‘

0.---I8I"“888

 

 

 

cow

omN

can

own

cow

one

can

own

com

cum

och

.cmm0:002 wxmq sumpmmmnppo:
:0 @mHQEMm zmfiwmpficz 6x00 00 0mm 90 AEEV :pwsma vaMHSOHMOnxomn cams

(wufl H19N31 031V1n31V0-XOV8 NVSW

.0H 0L00H0

85

mmohsommm 0003002 00

pcmsph0mma c0m0:002 .0006 umnmflansmcs

Anomfiv hmanmwm

:

Ammmav czopmm

 

 

 

0

00000 0H00H
mmothwmmm
05: :0: 0Hm mmfi :NH 0050 dwppmmsonm
05: :0: 000 00H 000 005H :0030H000z

MPh0>mhm

H00 000 :0: 0H0 500 05: 505H 0:0 003000003
HH: mom 00H mmH mmumomH mppmmsomm

N 50m
000 5:0 500 H50 000 0H0 00: 00: 00: 000 :00 05H 00-005H 00000500 00000
0H5 0:0 000 0:0 000 00: 50: 05: 005 505 50: 00-005H H000HwH 00>000
:00 500 000 0:0 H00 550 500 0H0 05: H0: 0:5 0:0 H05 00-005H H000H0H
000 000 0:0 050 000 00: 00: 05: 055 :55 :H0H 00-005H H00000 00003

0H HH 0H 5 0 5 0 0 : 0 0 H
2 00009 :00P0ooq
mw<

 

pom £00908000

.C0wwcoas 0x09 00000005000: 8000 300000023 0x00
0303::0 00 AEEV msmemH cmp0asoa0ouxo0n 0008 Ho mComH00mEoo .Hm 00909

86

.00000 000000000 000 000 0o 000000 000 00 0 H

 

 

050. 0 5H.0 + 00.500 00 50H 00H: .5H5V
000. 0 00.0 + 00.000 00 5H0 AOH5V
00H: .5H5 .0H50

000. 0 50.0 + H0.000 00 500 000000 00>000
o A0H0 .0H0 .5HH .0HH .0HH0

050. 0 05.H + 00.000 0 550H 000:0 00002
0 AOH00

0H0. 0 50.0 + 50.500 0 00H 000 0000>000 00000
0005 .0H0 .0H0 .0H5v

005. 0 05.5 + H5.H5H 00 005 000000

N0 0000000m0m 0>00o0000m z 00000ov 00000000

 

00>00m 0000 080000 000

0x00 00 0000800 Nmm0 u
0:0 80 Ao0v 00000800008

.0000000800 000 0000000000 000000000 000 0009800 080000

H05H 000000 000

.80000002 0x00 800000030000 8000 300000003

0000 00009800 000 Amv 000000800008 :0 050000 00000 00000000
00 000800 5000 00000 00 00000000m00 0>00o0000m .NN 0000B

87
governed the choice of using the combined spring data for
all subsequent discussion regarding mean back-calculated
growth.

Mean back-calculated growth may be influenced by
differential fishing mortalities among locations,
availability of food, differential competitive interactions,
‘or genetic differences between stocks. Healey (1980)
investigated the effects of various levels of experimental
exploitation on previously unexploited stocks and concluded
that growth rates of whitefish increase with increased
exploitation.

The higher back-calculated growth rates in the Leland
stock, compared to the more heavily exploited North Shore
stock, indicates that some factor, other than exploitation
rate, was responsible for the differernce observed. The
abundant 1977 year class, which was more prevalent in the
catch from the North Shore and Beaver Island areas, may
exhibit slower growth as a result of more prevalent
intraspecific density-dependent factors, such as
intraspecific competition for food and space. Also, the
higher fishing pressure in the North Shore and Beaver Island
areas would tend to remove a significant portion of the
faster growing members of a cohort as age III and age Iv
fish, thus leaving the slower growing fish to be used for
the back-calculation of lengths at the later annuli.

Lee's phenomenon, a commonly observed systematic error

associated with back-calculation techniques, was not present

88
in my calculations (Appendix 13). This effect, which can be
attributed to the selection of the faster growing fish at
earlier ages or improper aging, results in calculated
lengths that are higher at the early annuli for younger fish
than for older fish.

Conversely, a reversed Lee's phenomenon appeared in
this study. The same trend was present in the calculations
for whitefish in northwestern Lake Michigan (Humphreys 1978,
Gunderson 1978). The higher mean back-calculated lengths at
the early annuli, when computed from the older fish, appears
to represent a loss of predicting power of the scale-length
regressions at the upper range of the values of scale radii.
This in turn results in higher correction factors, which act
to increase all back-calculated lengths that require

adjustment.

Instantaneous Growth Rates

The instantaneous true growth rates (G) were larger for
ages II-VII in the Leland samples than in the North Shore
and Beaver Island samples (Table 23, Figure 11).
Instantaneous growth rates, for the combined spring samples,
ranged from 0.523 to 0.043 for ages III-XII from the Leland
area and 0.435 to 0.085 for ages II-VI from the North Shore
area. The instantaneous growth rates for the combined
spring samples from the Beaver Island area ranged from 0.315
to 0.022 for ages III-XI. The values of G from Grand
Traverse Bay are higher than the values from the North Shore
and Beaver Island stocks. The instantaneous growth rates
for fish from Grand Traverse Bay are less than those from
the Leland stock.

An occasional fish older than age XII was encountered
in the samples, yet difficulty in discerning the annuli
while aging and their relative scarcity did not warrant the
calculation of growth rates.

The instantaneous growth rates increased from spring to
fall (Table 23). The change is primarily due to the higher
slope values (b) of the predictive weight-length regressions
for the fall samples (Table 18), a result of the development
of reproductive products.

The growth rates computed in this study were within the
range of those reported from northwestern Lake Michigan
whitefish (Table 23). The more lightly exploited

North-Moonlight Bays stock (u=30%) showed higher growth

89

9C)

 

 

 

 

.pwm. npaA
anaA .comumoeso «so. mac. ~AA. -A. maA. mAn. mAm. can. bop.A mA.n maAuam
panama
muonsunu a
mama omwucmmm
.NmmA. mhmA-maaA
ommA .uocnm Ase. ewe. BAA. ooA. mam. VAm. New. mma. who.A mm.n mocAuam
mamA-mhmA
.nmmA. mAAum
ommA .uwcnm who. sac. ~AA. hmA. «AN. own. mmm. onA.A ~A~.~ Am.n axon unmAA
ucooz a cube:
.NovA. mhnhhmA
ommA .umenm «so. moA. NAA. va. oA~. mhw. Ann. mmm. Noo.A mm.~ mucAuam
annonmA
.mooA. uAAmm
002 up
ommA .umcnm moo. ewe. mac. mmA. omA. mmN. Aha. mAm. me~.A n~.m sum oAm
AmaA
.OVA. chuam
scuba mAce ovo. mmo. Ame. ANA. mwA. emA. mmN. ow.m >un umum>mue
0mu0>mu9 CGMHU
Na a AomA
.mmm. Lo mchuam
scuba mAne «No. ova. pmo. awe. ooA. «NA. m-. mAm. A~.m ocuAmA um>mmm
.nmn. Na a AaaA
scuba mAse mvo. Nqo. one. amo. ”mo. mAA. AmA. cam. mam. aha. e~m.A mq.n mAAom
.va . AmmA
scuba mAze omo. “ma. Aoo. GMA. aoA. MQA. Aom. «mm. mam. oq.n nosesm
.cAn. «a a AmaA
wagon mAne mqo. nvo. mmo. mmo. who. AAA. AvA. onA. nun. nmm. -.m we mchuam
oeoAmA
.vem. ~m a AmaA
away» mAna moo. ooA. one. GMA. mmA. «Am. nAm. An.n mo nAAum
.oon. AmaA
xenon mAce ~mo. «AA. NoA. «mm. m~.n “mesam
~o . AmmA
.ohm. Lo mchuam
xenon «Ace mac. NoA. mhA. Aon. mmq. ~a.~ chasm :uuoz
mocmumumz nAumA ~A-AA AA-oA oA-m mum m-h pus e-m m-v «In m-~ ~-A .2. Amouma.
odoAm :oAuMUOA
.AmhmA .umonm.

AAA mm3 - NA omoA. n u o

unassu0u on» :w com: mum: msumcoa voumasvouuxomn can ucocoaxm unowmzuzuocma
m>Aquumum one .caoAnoA: mxaq cuwnuuo: eouu :mAuouAns mxmA How .0. mouuu nuzoum wauu mzoocaucmumeA .m~ mAnce

91

 

 

 

 

 

 

 

 

 

 

 

 

 

05- a
North Shore
0.4- -------- Leland
.. ..... Beaver Island
-—---Grand Traverse Bay
03%
ul 01-
.-
<
c: it.“
0.1- ‘
1: "xx-'-
; "
2 oo \‘
, I l I I I
o ii4yéiss1ouiz
U)
:3 L4 b
o ‘x
m 0
:2 ll- \
g "a North Shore
‘ “m-..
i 1.0~ k Leland
.— ‘.
m i.
z: 03% '
_ ‘r
‘.
I I 7* T T I 1
53455189101112

Figure 11.

AGE (years)

Instantaneous growth rates from spring (a) and

fall (b) samples of lake whitefish at various ages from
northeastern Lake Michigan.

92
rates than the more heavily exploited Big Bay de Noc stock
(u=56%)(Ebner 1980). This is consistent with the findings

from the North Shore and Leland stocks.

Age at Recruitment

The average lake whitefish from the North Shore and
Beaver Island stocks was recruited into the trap net fishery
at an earlier age than the average fish from the Leland
stock (Appendix 14). The estimated age at recruitement for
the North Shore and Beaver Island stocks was the same (3.1
years). Fish from the Leland stock were recruited at 3.2
years of age.

Ebner (1980) found whitefish from the more heavily
exploited Big Bay de Noc stock to be recruited at an earlier
age (3.25 years) than fish from the more lightly exploited
North-Moonlight Bays stock (3.40 years). My results are
consistent with the findings from this study.

The back-calculated lengths at the prerecruited ages
are larger in the North Shore area than in the Leland area.
At approximately 3.25 years of age, the Leland fish are
longer than the North Shore fish and retain this advantage
throughout their lives (Appendix 14). The fish from the
Beaver Island area show nearly identical back-calculated
growth as the North Shore stock, thus accounting for the
similar ages at recruitment.

The fish from the Grand Traverse Bay Sample are not
directly comparable, due to the different minimum size limit
and selectivity characteristics of the purse seine. These
fish are recruited into the purse seine fishery at 5.1 years

of age.

93

Sex and Maturity

The data pertaining to sex and maturity are grather
limited. This information was collected only when the
whitefish were not marketed in the round (not eviscerated).
The two available samples were not obtained at the same time
of the year, thus not lending themselves readily to
comparison.

The male-female ratio of the fish in the North Shore
sample was 1:1.7. This sample was collected in August 1981.
The Leland sample, which was obtained in October 1981,
exhibited a male-female ratio of 1.5:1 (Table 24). The
observed differences in sex ratio are probably a result of
the different sampling dates. Hoagman (1973) found that
whitefish segregate by sex and arrive at the spawning
grounds at different times. This may explain the
differences in sex ratio observed between the areas.

Ninety-one percent of the North Shore sample was
mature. All whitefish longer than 410 millimeters were
mature. Essentially the same percentage of males (90%) and
females (91%) were mature in the North Shore sample (Table
24).

Eighty-two percent of all fish in the Leland sample
were mature. The Leland sample consisted of larger fish
than the North Shore sample, yet only 66.7 percent of the
females and 92.6 percent of the males were mature. All
males longer than 460 millimeters were mature and 50 percent

of the males 420-439 millimeters were mature. All females

94

95

Table 24. Percent maturity of male and female whitefish in 10 millimeter length
categories from the North Shore and Leland areas.

 

 

 

 

Females Males

Length (mm) Immature Mature 3 Mature Immature Mature % Mature
North Shore

(June 1981)
320 1 0 0.0
360 1 0 0.0
380 4 0 0.0 l 0 0.0
390 0 2 100.0 1 1 50.0
400 l 3 75.0 1 3 75.0
410 0 3 100.0 0 2 100.0
420 0 6 100.0 0 4 100.0
3430 0 46 100.0 0 25 100.0
Total 6 60 90.9 4 35 89.7
Leland

(October 1981)
360 l 0 0.0
400 l 0 0.0 l 0 0.0
410 l 0 0.0 0 1 100.0
420 l 1 50.0
430 l 1 50.0
440 2 0 0.0 0 1 100.0
450 3 0 0.0 1 4 80.0
460 2 0 0.0 0 2 100.0
470 1 0 0.0 0 3 100.0
480 0 2 100.0 0 6 100.0
490 1 4 80.0 0 4 100.0
3500 0 18 100.0 0 27 100.0
Total 12 24 66.7 4 50 92.6

 

96
shorter than 480 millimeters were immature.

The fish from the North Shore sample matured at an
earlier age than those from the Leland sample (Table 25).
Eighty percent of the age III fish and 93 percent of the age
Iv fish in the North Shore sample were mature, compared to
33 percent and 89 percent of the age III and age IV fish,
respectively, in the Leland sample. All fish age V and
older were mature in both areas.

Early maturation is one mechanism by which the
whitefish from the North Shore stock may compensate for the
removal of the older spawning fish by the fishery. This
allows the stock to reproduce in large numbers prior to
being fully vulnerable to the fishing gear. More data
concerning length at maturity are necessary to determine the
proportion of each stock of interest that reproduces prior
to reaching legal size. The data suggest that the
difference between the length at recruitment and the length
at maturity may be quite small. Since the trap nets used
for whitefish only exhibit 31 percent efficiency of capture
for fish 432 millimeters in total length and exhibit a peak
for fish 489 millimeters in total length (Eschenroder et al
1980), the margin of safety for reproduction is broadened.
It appears from the consistently high yields of whitefish
from the North Shore area that reproduction is not a factor
limiting whitefish abundance in that area.

Piehler (1967) found 41 percent of the females and 77

percent of the males in northern Lake Michigan to be mature.

97

AmmA oCSh cmAQEMm N
AmmA Amnopoo emAasmm A

 

 

om mm oA mm as @A Ampoe
ooA A MA
ooA A AA
ooA‘ A m
ooA A a
03 a 0
00A 0 00A A o m
3 t. me am .3 N. A
om NA m mm m a m
o A o A m
chaps: A“ chaps: % mASAMEEH % cusps: & chaps: % oASPmEEH %

 

 

whoam apnoz

(D
(:0
<1:

AecmAmA

 

.cmwAQOAz oxmq mo macaw panama can

mpocm zppoz may CA own some Pm chapme 0A0; wasp cmAAmPAca mxma Mo depCmonmmn .mm manna

98

His samples contained primarily age II fish. 0f the age III
fish in his samples, all of the females (n=42) and 92
percent of the males (n=22) were mature. Healey (1975)
noted an inverse relationship between age at maturity and
the level of exploitation for whitefish in several northern
lakes. The results of the present study are consistent with
this trend.

Bell et al (1977) found mature whitefish of a given age
to be considerably larger than the immature fish of the same
age. They suggest that maturity is reached at a given
length, irrespective of age. Spangler (1970) found male
whitefish in Lake Huron to mature for the first time at age
III and females at age IV. Growth rates for female versus
male whitefish at the prerecruited ages are necessary to
discern whether maturity is truly reached at a given length,
or at a given age.

Dryer (1963) found that Lake Superior whitefish which
were shorter than 368 millimeters were immature, and those
which were larger than 442 millimeters were mature. These
studies indicate that exploited populations of whitefish
tend to mature when they reach a total length of 400-450
millimeters in total length, corresponding to three or four
years of age. Also, males tend to mature at an earlier age

and smaller size than females.

SUMMARY

1. The results of this investigation suggest the
existence of three discrete stocks of whitefish within the
study area. These include one North Shore stock, one Leland
stock, and at least one Beaver Island stock.

2. The samples from the Leland, Grand Traverse Bay,
and Beaver Island (grid 316) areas contained broader age
compositions than those from the North Shore and Beaver
Island (grids 317 and 418) areas. The differences between
the age compositions for the 1981 Beaver Island sample (grid
316) and the 1982 Beaver Island sample (grids 317 and 418)
suggest that separate subpopulations, or distinct stocks,
were sampled at these times.

3. The 1977 year class was extremely abundant and
dominated the catch in all areas on all sampling dates.
This year class ranged from 32-56 percent of the Leland
catch, 63-64 percent of the Beaver Island catch, and 74-86
percent of the North Shore catch (Table 6). The mean age
and the mean size of the sampled whitefish were strongly
influenced by the presence of the 1977 year class.

4. The North Shore stock is considerably larger than
the Leland stock. Population numbers were estimated at 1.7
million legal sized whitefish in the North Shore area and
260 thousand in the Leland area. No population estimates
were possible for the Beaver Island and Grand Traverse Bay

areas.

99

100

5. Annual exploitation rates were estimated at 49.6
percent in the North Shore stock and 24.1 percent in the
Leland stock. The annual mortality rates were estimated at
70.7 percent and 58.6 percent in these areas, respectively.

6. The instantaneous total mortality rate (2)
estimated from tag returns, was larger in the North Shore
stock (1.229) than in the Leland stock (0.881). Estimates
of the total instantaneous mortality in the Grand Traverse
Bay and Beaver Island areas, computed from catch curves,
were 1.195 and 1.073, respectively.

7. The instantaneous fishing mortality (F) represented
a larger percentage (70.1%) of the total instantaneous
mortality in the North Shore area than in the Leland area
(41.3%). Instantaneous natural mortality rates (M) were
estimated to be 0.368 in the North Shore area and 0.517 in
the Leland area. These mortality figures are well within
the range of the values reported for other stocks of
whitefish in the Great Lakes and Canada.

8. The slopes of the natural log transformed
weight-length relationships were higher in the Leland area
than in the North Shore and Beaver Island areas. The slope
values were higher in the fall samples than in the spring
samples.

9. The length compositions were broader, and the mean
lengths were higher in the Leland, Grand Traverse Bay, and
Beaver Island (grid 316) samples than in the North Shore and

Beaver Island (grids 317 and 418) samples. The 1977 year

101
class stongly influenced the mean lengths and the‘ length
compositions of the sampled catch in all areas, throughout
the study. I

10. The mean length at age of the members of the 1977
year class in the Leland area were significantly larger, at
all sampling dates, than the corresponding mean lengths in
the North Shore area.

ll. Comparisons of length compositions between the
North Shore and Leland samples all showed significant
differences (p<.001). The Beaver Island (316) length
composition was not different from the Leland sample in June
1981 (p>.10), but was different from the North Shore sample
(p<.001). The May 1982 Beaver Island sample (grids 317 and
418) was similar to the North Shore sample (p>.25) and
significantly different from the Leland sample (p<.001).

12. Back-calculated growth and the instantaneous
growth rates indicate that the Leland stock is faster
growing than the other stocks for all of the recruited ages.
The mean back-calculated lengths at age are nearly identical
for the North Shore, Grand Traverse Bay, and Beaver Island
areas.

13. The North Shore and Beaver Island fish are
recruited into the trap net fishery at earlier ages than are
the fish from the Leland area. The Leland fish are not only
recruited at an older age but also mature at a larger size

than the North Shore fish.

APPENDICES

 

 

 

A. omo.NNN an NAN.mev aN on.mNN voN.NN o NoN.NA mam.mA Aauop

n A N A cAuo egocxca

e. on NN m Aches cam

men

Non

NoN.m AAoN

A ANN.A awn woo

moo.n 66A moN.N «co

N Nme .mA.vA oao.n Ame meo.N we.

ooN can meA eNn

on NNA.A Ama.A mNo.¢ Nmm.. oAn

ems va.a NNo.v an aAm

ooA.N N mmo.A NAM

nm~.Av 98.3 A v8.3. 3n

oAa.m A on.mA veN.¢ mAn

oqe.N co. mea.v NNA.A .NN.A .AN

mu Nam Amo.A AAA

41 ANA an N oN oNo.n eNN

N moo.nA v mNN.mv oN aAN

oNo.on aoN.oN Nsm.m NAN

vA NmN.mn N man.qN o eac.NA NAN

AN nvN.on NA Nem.Ac .A ova.ma m N oAN

ooh.» oAN.vA . nmN.oA mso.v mAN

moo.o o.N.AA vAN

omc.nn A Aoc.veA NAo.NN an. owe NAN

Amo.N omm.q onN NAA

aoo.A Aoo.N NAA

N NNN.NA N ow awn von.AA va.N 6AA

NNo.n mv¢.A mAA

a u m u m o o z u m 0 cans
mean an: .IIIAAmmmIIIIII span; .Ilnnmmmmwmmnu. .nnuxmmmmmmlll

 

.muonfisc :A man mouauamomu can moaned :A aw guano
uOu AmmA chusp vAum AmoAuuAumum an zmAuquzz och we .0. :Uuao AnAuuwfieoo can Am. ocusuou on» hAcucoz

.uuuu ouosm :uuoz 0:»

.A xAccmaa<

103

.AAAAA eons: AAAA mum AAA AAA .NAA .AAA

 

 

 

 

 

A
AAA. AAA. .NAAJ «A «AA .AA A AAA.AN SA www.mwv vN ANNJAA AN Siomm AA. ANAJAA A38.
NA AA AA AA AA Aqua egoexca
AAN AA A AA AA A AN AS;
A A AAA
N A A NAA
A AAA.AA AN AAA.A AAA.A ANA.A AAAA
A AAA.AA AAN.AA AAA.A AA AAA.A AAA
A AAA.AA AAA.A AAA AAA
N AAA.AA ANN.AA ANA NAA.A AAN.A ANA.N AAA
A ANA.AN AAA.AN AAA.A AAA AAA ANA
A AAA.AN AAA.A AAA «AA AA AAA
A AAA.AN AAA.A AAA.A AAA.A AAA.A AAA
N AAA.AA AAA.AA AAN.A AAA AAA AAA
A AAA.NAA AAA.A AAN AAN.AA AAA.AA ANA.AA ANA.AA AAA
A AAA.AA AAA AAA.AA AAA.AA AAA.N AAN.A AAA
A AAA.AA NAA.AA AAA.A AAA.AN AAA
A AAA.A AA ANA.A AAA
A NAA.A ANA.A AA NNA AN ANN
AA NAA.ANA NAA AAA.A AAN.AA A AAA.AA NA AAA.AA AAN
A ANA6AN AAA.N 25.2 3A.: www.mm AAN
AN AAA.AAA A AAAK 813 AAA. 3 AAN
AAA AAA.AAA A AAA.AN A A A AAA.AA AA NAN.A A ANA.AA AA NAA.NA AAN
AA NAA.AA A NAA.A N ANA.AN ANA.A AAA.NA ANA.A AAN
A AAA.AN AAN.A AAN
A AAA.AAN AAA.A A AAA.AA AAA.AN AAA AAA.A AAN
A AAA.AA AAA.A ANA AAA
AA AAA.AAN AAA.NA NAA.AA AA ANA.AAA AAA.AA AA AA A AAA
AA AAA.AAA A AAA AAA.A AA NAA.AN ANA.AA AAA.AA NNA.AA AAA
AA AAA.AA AAA.A AA AAA.AA AAA AAA
A o A o A o A o A o A o A o AAAu
AAAuos exocxcs .awuawumu Amwaazga. umnouoo “maemumum .Iulmmmmmmnlnu NAAA
.AmaeAuaoo. .A oncmaac

101+

 

 

 

 

 

AAAoA wom.~AA va.mAA mm Amo.mm A ono.mA o Ano.h A och.oA Adan?
cAv.OA mmm.w~ ANA.MA A mvn.0A ~¢m.o omm.m .umsz web
A vcm.hn NNm.Am aoa.m~ oAmA
v com.OA omo.A OAm.AA moo NAm
mmm.m moo.~ AAm
A AAA.~ AAA.v AN vme.w «Am
hwn wNm.h mhm nAm
A A o mmn.m mNA.~ «Am
mnn.~A hmv.vA mvm.~ NAm mmw.A mmp.~ AAA
A Ann va.~ oo¢.o A cow AAA
va.A AAA
mvm ms mA NAA
mm¢.mA Awh.o Amw.OA omm.A hm~.A er
AAvn vnoavw th.~ vac mAm
mom «Aw
m U U m U m U m U m U “HAG
95A. IIAWAAMIII no.3. Ila llama)
.mmmmnucoumm CA mum unusumu on» awn omum>mua
cacao .uumnadc :A one mousuamoou can mvcaom :A mA coumo .ummum awn mmum>uua venue can ccmAwq any

uou AAAA AAAAAA AAAA AquuAAuAuA an AAAAmuAss AAAA Ac .0. guano AaAoumeeoo one AA. Aausuwu AA» AAzueo:

.m xAvcmmmd

105

 

 

 

 

 

AAAAA AAA.AAA A A AAA.AA A AAA AAAA AAA.AAN A AAA.AA .N.A AAA.AA AAAAA AAN.AAA Auupa

A AAA.AAA AAA.AA NAA.AAA AAA.AA AAA.A AAA.A .oAAz. AAA

A AAA.AAN AAA.AA AAA.AN ANA.AA A AAN.AA AAAA

A NNA.AA N AAA.A N AAA.A A AAA.A AAA.AN NAA

A ANA.A AAA

AN AAA.AA A AAA AA AAA

A AAA.A AAA NAA AAA

N AAA.AA AAA.A AAA.A AAA.AN A N NAA

.N.A AAN.AAA AAA.A aA AAA.A AAA.AA .A.A AAA.AA A AAA.AN AAA

A AAA.AA A AAA AAA

A NAA.A AAA

A AAA AAA NAA

.A.A AAN.AA A AAA.A A AAA.A A AAA.A A NAA.A AAAN AAA.AN AAA

ANAA AAA.AA A N .A. A A AAA.AA AAA

A AAN AAA

A o A A o A o A o A o A o A o AAAu
AAAAAA AAAAAAA .AmmeMAA canam>oz umnouuo umasmumum IIIAAAAAAIIII .IINAmmIIIIII

.AmaeAucoo. .N xAueoAAc

106

Appendix 3. Estimation of the instantaneous rate of tag
loss (L) for the 1980 Floy dart tags.

 

Relationship: (1 -a')

 

where

Information required

Estimate of L:

 

e-Lt

n = percentage of double-tagged fish
which are returned with only 1 tag

9 = base of the natural logarithnl
L = instantaneous rate of tag loss
t = time in years

to estimate L:

r = 1 /15 = .0667

Date of return of the incomplete
double tag: 6/21/82

Date of double tagging: 11/3/81

Number of days between 11/3/81 and
6/21/82: 231 days

Midpoint of interval between 11/3/81
and 6/21/82: 115.5 days
115.5 days/365 days = 0.316 years

(1 - .0667) = e'L('316)
.9333 = e'L('316)
.0690 = L(.316)

L ‘ .2184

107

Appendix 1+. Comparison of ages assigned by the two principal
scale readers, including calculation of the index of average
percent error.

 

Age assigned Agreement by reader #2

 

 

 

 

by reader #1 Total +/- 1 year

N Number Percent Number Percent

3 74 71 95.9 74 100.0

a 63 51 81.0 63 100.0

5 33 23 69.7 33 100.0

6 13 8 61.5 13 100.0

7 6 4 66.7 6 100.0

9 1 0 0.0 1 100.0

11 1 O 0.0 1 100.0
14 1 1 100.0 1 100.0
7-14 9 5 55.6 9 100.0
Total 192 158 82.3 192 100.0

 

Index of average % error (I): 2.35 %

 

R
I=1/N£ 1/R Z (Xij'le x100
i=1 i=1 x.
J
I = (4.51/192) x 100 = 2.35 %

108

 

A AAA u AAA x AAQAANAA ... A
AA AuA AuA

AAA x AA .. AAA; N A\A IN z\A .... A
A z

A mm.o .AAV AoAAm A mwmnm>m mo xmch

 

 

 

 

 

o.ooA mm o.ooA mm mm :
0.03 A 3AA A A A
Hemopmm Amnssz pamopmm pwasz
Amok 6:0 :\+ Anyoe z mw<
mAAom
PCAEmmpw< Adm CAm

 

.Aoupm Acmonmm
mwmpm>m 9o xmpcA 02A mo CoApmAsoAmo chpsAocA .cmwAnOAS AAAA mo «ohm onocm nvuoz 029
CA :mAAmpAnz mxmA Sony mzmu CAM paw mmAmom hp pmcmAmmm mmmm Mo ComAAMQEoU A“ xAcCAQm<

109

 

ANAA AAAAA

AAAAA AAAA AN NAA. A AAA.A + A.AA- u A AAAAAAAA AAAA .AAAAAAIAAAAAA
AAAN .AAN AAAAAA

AAAAA AAAA AA AAA. A AAA.A + AA.NA- u A AAAAAAAS AAAA .Anocm Apnoz

AAAA .AA AA AAAAAAAAAAA NAAN AAA. A AAA.A + A.AN- u A coAAA AAAA .AAA AcoaeA:
monopomom .z A :oApmsqm m>APoAcoAm noAPwooA

 

.AAAAAaAAAAs AA AAA AAA prcAA AAA AAA AAAAA

.Copsm mAAA .hmm unoeemm cam cmwAnoAE

mAmA :Aopmmocpuoc Cmmspmp mmAcmcoAponA :chAA I npnAw Mo ComAthEou .mA AAUCAQQ<

110

Apfiendix '7. Population estimates of lake whitefish

(2 30 mm) in Lake Michigan for the North Shore, Leland and
Leland Spawning grounds during November 1980. The North
Shore estimate includes rids 115-119, 213-220. The Leland
estimate includes grids 15, 714, 812-814 and 911-912.
Estimates include only catch and recaptures from cooperating
fishermen.

 

 

North Shore Leland Leland
area area spawning
grounds
M 1603 376 415
C 797.130 126,872 2039
Ca 717,338 121,840 -
R 313 57 39
Ra 363.22 63.08 40.40
Mean weight (kg) 0.8634 1.2474 1.2474
Biomass (kg) 1,435,996 329,422 9501

95% Confidence (1,295,697- (256,884- (6986-
Intervall 1,591,501) 420,294) 13272)
Numbers 1,663,207 264,087 7616
95% Confidence (1,500,708— (205.907- (5601-
Interval 1,843,316) 336,942) 10640)

 

M = number of fish tagged in November 1980 that were
larger than 430 mm.

C - total catch of cooperating fishermen in pounds
during 1981

Ca = C adjusted for recruitment of fish that were less
than 430 mm in November 1980

R = total number of recaptures of M marked fish
Ra = R adjusted for tag loss

 

1 Confidence intervals are Poisson approximations (Ricker 1975)

111

Appendix 8. Calculations of the adjustment of the 1981

commercial harvest by cooperating fishermen, of lake white-
fish in the North Shore and Leland areas for the recruitment
of those fish that were sublegal ( 430 mm) in November 1980,

yet grew into the harvestable portion of the population
during 1981.

 

Mean Length of legal portion of the 1977 year class
in sampled catch. Standard deviation is in parentheses.

Sampling dates

 

 

Area
11/4/80 6/29/81 8/24/81 10/17.24/81
North Shore 456.2 458.9 461.9 474.0
(17-3) (20-7) (19-0) (19-0)
Leland 465.9 469.0 474.0 494.3

(15.5) (23.7) (19.4) (31.0)

Estimated mean length of fish that were 430 mm in November,
1980, at subsequent sampling dates.

Sampling dates

 

 

Area
11/4/80 6/29/81 8/24/81 10/17.24/81
North Shore 430.0 432.7 435.7 447.8
Leland 430.0 433.1 438.1 458.1

 

Recruitment estimate: Proportion of the catch at each

sampling date that was shorter than the estimated mean length
of those fish that were 430 mm in November 1980.

 

Sampling dates
6/29/81 8/24/81 10/17,24/8I

North Shore .31711 .1015 .0981
Leland .0321 .0249 .0913

 

112

Appendix 8. (continued)

Catch adjustment: The total catch of cooperating fishermen
was adjusted to estimate only those fish that were in the
legal portion of the November 1980 population. This was done
by multiplying the catch during each interval between samp-
ling dates, times the proportion of fish larger than the
estimated mean length of a November 1980 sublegal fish, at
the time of the later sample.

 

Area Date Proportion Catch 2 Adj. Catch

North Shore 8/24/81 .8985 468,730.4 421,154.3
10/20/81 .9019 328.399.6 296,183.6
TOTAL 717,337.9

 

 

Leland 6/29/81 .9679 37,230.5 36,035.4
8/24/81 .9751 65,478.8 63,848.1
10/20/81 .9087 24,163.0 21,956.9
TOTAL 121,840.4
1 The June 1981 sample was omitted from adjustment
calculations due to the atypical disproportionate
numbers of sublegal fish in the catch.
2

Total harvest in pounds for the cooperating fish—
ermen during the interval between sampling dates.
The catch for partial months was estimated by the
fraction of the month times the monthly catch.

 

113

 

25> u 2mm em

Amw>a w 85 a 3.1,

AA\ AVA:\ A .numv u s

z\om .nom u :

:oEuonmwm Ham mom £0980 Havop n ”J

Cosmonmwm mcflpmumaooo mo mccson ca 20900 Havop u o

mmoH wmp pow umpmswcm om u om .wom

Aowwmp :0:; as on: cusp pomuma who; wasp :thoanm wcwpmnonooo an mouspnwoon u om
.58 on: camp Amwuma sway comma» Mo 9098:: u 2

 

Asmmo.v ofidm. ANHNO.V mmwa. :NN.NmH th.wmfi mo.mm mm mum Noam UcmHmA
«ohm

AAAAA.V mmAA. AAAAA.A AANN. mmo.::m.A AAA.AAA mm.mom mAm noAA muosm spuoz

 

>mA as >mA o: AA A om AA 2 coApmooA
am am .now

 

.mmpm comm Cw copwo Hmpop map op nopmo uwocp mo owvmn 029 an em

nvcmaxo :0:» .cAEnmzmHm wcflpmhmaooo czocx Eon“ mausgop paw :opdo one song covmasoamo who;
mosam> .smwwnofiz mqu ho mmopm 090cm canoz cam unmamq 029 CH common Ass omzAv_:mwmmpw:3
oxma pom .mmoa may now nofipoohnoo pmpmm .mopmp Cowpmpwoanxo HM:£C¢. .m xHACAQQA

114

 

 

 

oov.h- one Amuoe
hv.mom nw.m>v ooa panama uo>oom
.Hoo. v a. How moo.uc~ hnv :«zuwz An.nmn on.mhv mow .>. ouozm :uuoz
~mm.n~ omn.~a Hah.mN N coo3u0n mm.vmm bh.mmm mma puma panama
ham.hv ANA Amuoa
om.mn~ Hm.omv mm panama uw>mom
.Hoo. v a. can vah.av mNH casuaz mn.mn~ ~m.mvv me .>H. o.uocm :uuoz
mam.o ~vm.~ nmo.m ~ coozuon mm.a~m mm.~mv an chad psoAmA
a. ~A¢A an:
AA A: A: Angoe
.mo. v a. nNm ovm.~v mv :«zudz mm.mva~ hm.omm n~ .>. panama
mam.m ohn.m onm.m A coozuoa ma.-o mh.oom ow mhwa muonm :uuoz
AAA.AAA “A. Amuoa
.Aoo. v a. can no~.mo~ ovv :Azuqs ao.hmm Hm.nmv eoa A>H. unnamA
mh.mo oom.ov coo.ov H :mw3umn m~.Aom mo.vhv «mm puma ouosm :uuoz
Hume AQQOuoo
m~a.o~a mam Huuoe
Aaoo. v a. non amm.o~a mam Casuaz wu.v~v wa.mnv no .>H. panama
ww.o~ vom.n ton.h A cowsuon mv.omm ma.~mv Hmm FBAH ouozm nuuoz
AmmA umsmsa
mvm.aa~ mom Huuoa
mm.wmm mo.mwv Ava panama um>bwm
.aoo.v m. m.nno anm.oma new :Anuwz no.mno AH.omv mm .>H. panama
mm.- mom.m~ mao.om ~ cmmzuon we.m~m mh.vvv mm puma muonm :uuoz
Anon 0:56
~n~.mm and anuoa
.oa. A 3. «mm HA~.hm AMA cwnufiz mm.oom om.~hv mm .>H. OcquA
hmv.a o~m omm a cowzuwn wo.mvm mm.mhv baa whoa ouozm :uuoz
-h.cna wwv Houoa
.Aoo. v a. hmm mmh.vo~ Hoe :Azuaz ha.o~m ~n.mov mo .HHH. panama
on.wa wmm.m mmm.m a cowzuwn mm.~em nm.mmv can hnma ouozm :uuoz
cmma quOuoo
oAumu m: mm up cofiu0wum> mocmwum> AEE. numcmq 2 .00m. acofiuoooq
m uo wousom cam: mmmHU uno» can mama
.nmAw cm M 2 mums: madmawm Ham :0 omuospcoo cums
momwacz< .mmmauomma mcwusc cmmwnowz oxmA :uouwconuuo: :A nmwumuflzz oxma mo mommmHo umm> mnaa can

.nhma .enma on» uo mo>fiumuc0moumou u0u mom um .55. camcoa some so moccauo> mo uwm>aocm so no magnumo .oa xwpcomaé

115

now. can. AN A>V GCMHAA

 

 

 

pomwmu AAAAA.~ AAA. AAA. AA AAAA muonm npuoz

AAA. AAA. AAA A>HV AcmamA

pomwmu AAAAA.AA AAA. AAA.A AAA AAAA muonm :puoz
AAAA nmnopoo

AAA. AAA. AA A>AA AcAAmA

Agmoom AAAAA.A AAA. AAA.A AAA AAAA macaw :ppoz
AAAA pmsAsA

NAA. AMA.A AAA osmHmH nm>mmm

AAA. AAA. AA A>HA AcmAmA

pomnmp AAAAA.A AAA. AAA. AA AAAA macaw Apnea
AAAA AAAA

AAA. AAA.A AA A>AA AAAAAA

pomflmp AAAA.A AAA. AAA. AAA AAAA muonm Appoz

AAA. AAA. AA AAAHA AcmAmA

AAAAAA AAAAA.A AAA. AAA. AAA AAAA AAAAA spuoz
AAAA umnopoo

AA."
swflmwowm a v0 N
z

A xv mocmwum> msoACAMoSom pmwa hpwamspoz Amway msofipmooq

N pow pmme pwmapumm Mkocuwsmn>ohoEoHox mmmao ham» opmn

 

.mwm pm :pMCma cmms so Aocmwum> Mo mmmhamcm map you mamscwmmn
Ampsnmupmfic adamshoc Asa mocmwpw> msomcmwoso: Ho mcoflpQEAAAA Asp mo mamms .AAH xwocmnm<

116

 

 

 

 

AAN. AAA. AsmamH Am>mmm

AAA. AAA.A A>A AAAAA spuoz

pommmp NAAA.A AAA. AAA.A AAAA AAAAAA

AAA. AAA. A . AAAAAA Am>mmm

AAA. AAA. A>AA macaw Appoz

poommu AAA.A AAA. AAA. AAAA AAAAAA
NAAA an:

AWN. "iv G v0 N

coflmfioma Amw<v msoflpmooq

A Nxv mocmAAA> AsoACAonoz pmme zpwamshoz mmAHo AAA» mama

AAA AAAA AAAAAAAA

AmoCAHEmI>voEoHox

 

AcmscflpCoov

.AA AAAAAAAA.

Appendix 12.

117

Results of the quadrant test of the joint

distribution of length and weight for bivariate normality

(Cramer 1946).

The hypothesis was rejected when the

calculated Chi square value exceeded the tabular value at
the 0.05 level of Significance.

 

 

Sample
Sampling Date Grids Size r X2 Decision
October 1980 115,116 513 .886 11.91 reject
714,814 114 .967 53.89 reject
June 1981 117 197 .931 2.41 accept
812,814,912 81 .970 9.70 reject
316 219 .930 2.09 accept
August 1981 218 300 .885 17.90 reject
812,814,912 94 .943 26.87 reject
October 1981 116,117 331 .872 17.29 reject
812 244 .933 38.92 reject
May 1982 216,218 263 .862 3.54 accept
812,814,912 232 .949 45.08 reject
317,418 169 .904 7.59 accept

 

118

 

 

 

. u : .A.AA. .m.Am. .A.A~. .A.A~. .A.A~. .A.A~. .~.AA. coAuuA>oA
Auuceoem
A.AAA «.AAA A.AAA A.AAA A.A~A A.~Av A.AAA A.A~A A.AAA ~.vmn AAmuoe
counvoz
A.AAA ~.nAA A.AAA A.AAA A.A~A n.AAA «.AAA A.AAA A.AAA A.A~e A AA
~.vAA A.AAA A.AAA A.A~A A.AAA A.~An A.A~m A A
.A.A~. .A.A~. .A.v~. .A.A~. .A.AA. .A.A~.
A.AAA ~.AAA ~.AAA A.AAA A.AAA A.Avm AA A
.A.A~. .A.-. .A.A~. .m.AA. .A.AA.
A.AAv A.AAA A.Amv A.AAA A.Avm AAN A
.A.A~. .A.-. .A.A~. .A.AA.
A.~Av A.A~A ~.AAA A.~Am AAA A
.~.AA. .A.AA. .A.AA.
A.A~A A.AAA A.~nm AAA A
.A.Av. .~.Aq.
A.AAA A.AA~ A A
AA AA AA A A A A A v A A A 2 AAA
mamaszd

 

.mwmmzucouaa :« who n:0wuna>op pumpcmum
.cumAnow: oxmq no noun ouonm suuoz on» Eouu :mwuouas3 oxmA new anumcmu poumasoHnonxoon new:

.mma xwpcoamc

1.159

 

 

 

.A.AA. .A.AA. .A.AA. .A.AA. .A.AA. .A.AA. .A.AA. .A.AA. .A.AA. AA.AA. .A.AA. .A.AA. :oAqu>mA
Guava—mum
A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA AAAAoA
AAAAAAoz
.A.AA. .A.AA. .A.AA. .A.AA. .A.AA. AA.AA. .A.AA. .A.AA. .A.AA. .A.AA. .A.AA. .A.AA.
A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA AA AA
.A.AA. .A.AA. .A.AA. .A.AA. .A.AA. .A.AA. .A.AA. .A.AA. .A.AA. .A.AA. .A.AA.
A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA AA AA
.A.AA. .A.AA. .A.AA. .A.AA. .A.AA. AA.AA. .A.AA. .A.AA. .A.AA. .A.AA.
A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA AA AA
.A.AA. .A.AA. .A.AA. .A.AA. .A.AA. .A.AA. .A.AA. .A.AA. .A.AA.
A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA AA A
.A.AA. .A.AA. .A.AA. AA.AA. .A.AA. .A.AA. .A.AA. .A.AA.
A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A A
AA.AA. .A.AA. AA.AA. AA.AA. .A.AA. .A.AA. .A.AA.
A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA AA A
.A.AA. .A.AA. .A.AA. .A.AA. .A.AA. AA.AA.
A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA AA A
AA.AA. AA.AA. .A.AA. .A.AA. .A.AA.
A.AAA A.AAA A.AAA A.AAA A.AAA AAA A
AA.AA. .A.AA. .A.AA. AA.AA.
A.AAA A.AAA A.AAA A.AAA AAA A
.A.AA. .A.AA. .A.AA.
A.AAA A.AAA A.AAA AA A
A.AAA A.AAA A A
AA AA AA A A A A A A A A A 2 AAA
AsAs::<

 

.mmmmsucoumm :A mun unawuuw>ou vuuvcmum
.cmoAnoA: oxmq wo mono ccmAmq may Eouu :mAumuAnz meA new mnumcmA vmunAsoAMOnxomn cum:

.nAA xAvcoAAA

120

 

 

 

AA.AA. AA.AA. .A.AA. .A.AA. .A.AA. .A.AA. AA.AA. .A.AA. :oAAoA>AA
Aunccuum
A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA AAAAoa
AAAAAAA:
A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A A
AA.AA. AA.AA. .A.AA. .A.AA. .A.AA. .A.AA. .A.AA. .A.AA.
A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A A
.A.AA. .A.AA. .A.AA. AA.AA. AA.AA. AA.AA. .A.AA.
A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A A
.A.AA. .A.AA. AA.AA. .A.AA. AA.AA. .A.AA.
A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA AA A
AA.AA. .A.AA. .A.AA. .A.AA. .A.AA.
A.AAA A.AAA A.AAA A.AAA A.AAA AA A
.A.AA. .A.AA. .A.AA. .A.AA.
A.AAA A.AAA A.AAA A.AAA AAA A
.A.AA. AA.AA. .A.AA.
A.AAA A.AAA A.AAA A A
AA AA AA A A A A A A A A 2 AAA
msA:::<

 

.mmmonucouua :A can acoAuMA>mc cumonmum

.cmmwsoAz meA no mono ocmamm u0>mmm mnu Eouw :mAwmquz mme AOu usumcmA vouuuaoAAOIxoun cums .onA xwvcommd

 

121

 

 

AA.A A AA.A A .A.AAA .A.AAA .A.AAA .A.AAA .A.AAA .A.AAA AA.AAA .A.AAA coAAAA>AA
AAAAAAAA
A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA AAAAAA
AAAAAAos
A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA AA
.A.A A .A.A A .A.A A .A.AAA .A.AAA .A.AAA .A.A A .A.AAA .A.AAA .A.AAA
A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA AA
.A.AAA .A.AAA AA.AAA AA.AAA .A.AAA AA.AAA .A.AAA .A.AAA
A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A
.A.AAA AA.AAA AA.AAA AA.AAA AA.AAA AA.AAA AA.AAA
A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A
.A.AAA AA.AAA .A.AAA .A.AAA .A.AAA AA.AAA
A.AAA A.AAA A.AAA A.AAA A.AAA A.AAA A
.A.AAA .A.AAA .A.AAA .A.AAA .A.AAA
A.AAA A.AAA A.AAA A.AAA A.AAA A
.A.AAA .A.AAA AA.AAA .A.AAA
A.AAA A.AAA A.AAA A.AAA A
AA AA AA A A A A A A A A A AAA
msAscc<

 

.mmmmnucouma :A one acoAqu>0© vuuvcmum
.comAcoA: 0x04 uo noun awn omuo>mua ccmuo 0:» Eouu :mAumuwsz mxmA ken mauocmu cmunAsvoouxomn cum: .CAA xAccwmm<

122

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

voo-
eso-
soo-
550-:
sou-1 ,v" 4mm sgelimit
450- 432 mm sizellmit
400-
E 3504 Leland
E «mu-Grand Traverse Bay
soo-
::
8 25° L’S.‘ years
3 = 3.2 years
J 200 I I i ] r I I I
a o I 5 3 4 s T s 4.5 1o 11 12
700- '
a
.A
5 550- /x
8 son. {/1
g 550- "K
i ’/
g 500-
450- " 432 mm size limit
400-
350 4 North Shore
------- Beaver Island
300-
mm-
--3.1 years
200 I . gr I l I l l l I l T7
o 1 2 4 5 6 7 8 9 In 11 12

Appendix 14.

AGE Iyearsl

Graphical estimation of age at recruitment from

the mean back-calculated length (mm) at age of lake
whitefish from northeastern Lake Michigan.

 

LI ST OF REFERENCES

LIST OF REFERENCES

Anscombe, F. J. and J. W. Tukey. 1963. The examination
and analysis of residuals. Technometrics 5:141-160.

Baldwin, N. S., R. W. Saafield, M. A. Ross, and H. J.
Buettner. 1979. Commercial fish production in the
Great Lakes 1867-1973. G. Lakes Fish. Comm. Tech.
Report Number 3. 187p.

Beamish, R. J., and D. A. Fournier. 1981. A method for
comparing the precision of a set of age _
determinations. Can. J. Fish. Aquat. Sci. 38:
982-983.

Beamish, R. J., K. Mills and D. Barnes. 1976. Evaluation
of the scale method of age determination for routine
ageing of lake whitefish. Pages 89-94 in R. J.

Beamish, L. M. Blouw, and G. A. McFarlane. A fish and
chemical study of 109 lakes in the experimental lakes

area (ELA), northwestern Ontario with appended reports

on lake whitefish ageing errors and the northwestern
Ontario baitfish industry. Fish. Mar. Serv. Tech. Rep.
607. 116p.

Behnken, D. W., and N. R. Draper. 1972. Residuals and their
variance patterns. Technometrics 14:101-112.

Bell, 6., P. Handford, and C. Dietz. 1977. Dynamics of an
explioted population of lake whitefish (Coregonus
clupeaformis). J. Fish. Res. Board Can. 34:.
942-953.

 

Box, G. E. P. 1954. Some theorems on quadratic forms
applied in the study of analysis of variance problems.
I. Effect on inequality of variance in the one way
classification. Ann. Math. Stat. 25:290-403.

Brown, R. J. 1968. Population structure and growth char-
acteristics of the whitefish in northern Lake Michigan
1929-1967. M. S. Thesis. University of Michigan,
Ann Arbor. 61p.

Brownlee, K. A. 1965. Statistical Theory and Methodology in

Science and Engineering. 2nd ed. New York: Wiley.
590p.

123

124

Caraway, P. A. .1951. The whitefish, Coregonus clupeaformis
(Mitchell), of northern Lake Michigan, with special
reference to age, growth and certain morphometric
characters. Ph. D. Thesis. Michigan State University,
East Lansing. 142p.

 

Carlander, K. D. 1981. Caution on the use of the regression
method of back calculating length from scale measure-
ments. Bull. Am. Fish. Soc. 6:2-4.

Chapman, D. G. 1951. Inverse, multiple, and sequential
sample censuses. Biometrics 8:286-306.

Christie, W. J. 1963. Effects of artificial propagation and
the weather on recruitment in the Lake Ontario white-
fish fishery. J. Fish. Res. Board Can. 20:597-646.

Cramer, H. 1946. Mathematical Models of Statistics. Princeton
University Press.

Cucin, D. and H. A. Regier. 1966. Dynamics and exploitation
of lake whitefish in southern Georgian Bay. J. Fish.
Res. Board Can. 23:221-274.

Dryer, W. R. 1963. Age and growth of whitefish in Lake
Superior. U. 5. Fish. Wildl. Serv. Fish. Bull. 63:
77-95.

Ebner, M. P. 1980. Population dynamics of lake whitefish,
Coregonus clupeaformis, in Green Bay and Lake
Michigan east of Door County. M. S. Thesis. University
of Wisconsin, Stevens Point. 106p.

Ebner, M. P., and F. A. Copes. 1982. Loss of Floy anchor
tags from lake whitefish. North Am. J. Fish. Mgmt. 2:
90-93.

Eschenroder, R., D. B. Jester, Jr., A. J. Nuhfer, and A. T.
Wright. 1980. Selectivity of deep trap nets for lake
whitefish. Unpub. manuscript, Michigan Dept. Nat. Res,
Fisheries Division. 16p.

Gunderson, J. L. 1978. Vital statistics of the lake white-
fish in three areas of Green Bay, Lake Michigan with
comparison to Lake Michigan east of Door County,
Wisconsin. M. S. Thesis. University of Wisconsin,
Stevens Point. 95p.

Healey, M. C. 1975. Dynamics of exploited whitefish pop-
ulations and their management with special reference
to the Northwest Territories. J. Fish. Res. Board Can.
32:427-448.

125

Healey, M. C. 1980. Growth and recruitment in experiment-
ally exploited lake whitefish (Coregonus
clupeaformis) populations. Can. J. Fish. Aquat.
5C1. 37:255-267.

 

Hoagman, W. J. 1968. Annulus formation on scales of four
species of coregonids reared under artificial
conditions. J. Fish. Res. Board Can. 25:2111-2112.

Humphreys, J. D. 1978. Population dynamics of lake white-
fish, Corggonus clupeaformis, in Lake Michigan
east of Door County, Wisconsin. M. S. Thesis. Univ-
ersity of Wisconsin, Stevens Point. 69p.

Lawler, G. H. 1965. Fluctuations in the success of year
classes of whitefish populations with special refer-
ence to Lake Erie. J. Fish. Res. Board Can. 22:1197-
1227.

McCombie, A. M. and F. E. J. Fry. 1960. Selectivity of
gill nets for lake whitefish, Coregonus clupeaformis.
Trans. Am. Fish. Soc. 89:176-184.

 

Mills, K. H. and R. J. Beamish. 1980. Comparisons of fin
ray and scale age determinations for lake whitefish
(Coregonus clupeaformis) and their implications for
estimates of growth and annual survival. Can. J. Fish.
Aquat. Sci. 37:534-544.

Ovchynnyk, M. M. 1962. The use of scales and bones for age
determination of the Great Lake whitefish, Coregonus
clupeaformig (Mitchell). Zoologischer Anzeiger
169:198-217.

 

Patriarche, M. H. 1974. Whitefish in Lake Michigan. pages
15-20 13 Michigan Dept. Nat. Res. Fish. Div.,
Estimates of biomass of principal species in the Great
Lakes. Fish. Res. Reprint 1813, 38p.

Patriarche, M. H. 1977. Biological basis for management of
lake whitefish in the Michigan waters of northern Lake
Michigan. Trans. Am. Fish. Soc. 106:295-308.

Piehler, G. R. 1967. Age and growth of the common white-
fish, Coregonus clupeaformis (Mitchell), of
northern Lake Michigan and Bay de Noc areas. M. S.
Thesis. Michigan State University, East Lansing. 94p.

Ricker, W. E. 1973. Linear regressions in fishery research.
J. Fish. Res. Board Can. 30:409-434.

Ricker, W. E. 1975. Computation and interpretation of bio-
logical statistics of fish populations. Bull. Fish.
Res. Board Can. 191:382p.

126

Robson, D. S., and D. G. Chapman. 1961. Catch curves and
mortality rates. Trans. Am. Fish. Soc. 90:181-190.

Robson, D. S., and H. A. Regier. 1964. Sample size in
Petersen mark-recapture experiments. Trans. Am.
Fish. Soc. 93:215-226.

Roelofs, E. W. 1958. Age and growth of whitefish,
Coregonus clupeaformis (Mitchell), in Big Bay
de Noc and northern Lake Michigan. Trans. Am. Fish.
Soc. 87:190-199.

Rybicki, R. W. 1980. Assessment of lake whitefish
populations in northern Lake Michigan. Mich. Dept.
Nat. Res. Fish. Div. unpub. report (PL88-309). 27p.

Rybicki, R. W., and M. Keller. 1977. Progress report on
whitefish in Lake Michigan in 1976. Presented at Great
Lakes Fishery Commision, Lake Michigan Committee
Meeting, Milwaukee, Wisconsin, February 22-24, 1977.

Seber, G. A. F. 1973. The Estimation of Animal Abundance
and Related Parameters. Chas. Griffin and Co. Ltd.,
London.

Smith, S. H. 1968. Species succession and fishery exploit-
ation in the Great Lakes. J. Fish. Res. Board Can.
25:667-691.

Snedecor, G. W. and W. G. Cochran. 1967. Statistical
Methods, 6th Edition, Iowa State University Press,
Ames.

Spangler, G. R. 1970. Factors of mortality in an exploited
population of whitefish, Coregonus clupeaformis,
in northern Lake Huron. Pages 515-529 12 C. C.
Lindsey and C. S. Wood, eds. Biology of Coregonid
Fishes. University. of Manitoba Press, Winnepeg, Man.

van Oosten, J. 1923. A study of the scales of whitefish of
known ages. Zoologica 2:380-412.

Wells, L. and A. L. McLain. 1973. Lake Michigan: Man's
effects on native fish stocks and other biota. Great
Lakes Fish. Comm. Tech. Report 20. 55p.

"11111111111?