-~~~--~,，，,,，G） №.….….….. .
º.º.º.）ſaeį，， ， ， p




A BIOLOGICAL ANALYSIS OF THE YELLOW PERCH POPULATION
IN THE LES CHENEAUX ISLANDS, LAKE HURON
by
David O. Lucches i
A thesis submitted in partial fulfillment
of the requirements for the degree of
Master of Science
School of Natural Resources
The University of Michigan
1988
Committee members
Associate Professor James S. Diana, Chairman
Adjunct Professor W. Carl Latta
Ex-officio Examiner James C. Schneider
ACKNOWLEDGEMENTS
This study was funded by the Michigan Department of
Natural Resources, Division of Land Resource Programs,
Coastal Management Program Contract LRP-8C-7. First, I
would like to thank the fisheries personnel from District 4
of the Michigan Department of Natural Resources for their
help in the field. John Schrouder, Bill Gruhn, and Chuck
Payment spent weeks tagging perch and provided much
appreciated guidance with this study. Working with them was
a real pleasure. In addition, I am indebted to Cheryl
Holbrook for entry of tagging and creel census data into the
computer.
The people of Cedarville-Hessel We re also very
cooperative and helpful with this study. Thanks go out to
all resort owners, and especially to Bob's Bait and other
bait stores for recording thousands of tag returns. Special
thanks to Mary Torsky, Bob and Helen Marks, and the
VanSloten's for adopting me during my stay in the Les
Cheneaux Islands. Also, I appreciate the thorough creel
census work and friendship of clerks Betty Sheffield and
Jerry Broda.
All personnel of the Institute for Fisheries Research
of the Michigan Department of Natural Resources deserve
thanks for their many contributions to every aspect of this
study. Without their help and guidance, completion of this
i i
project would have been extremely difficult. Moreover, they
made working on this thesis fun, and I enjoyed their company
immensely. A special thanks to Jim Schneider who put an
extraordinary amount of time into this study and provided
needed guidance, especially during the early stages, and to
Drs. Jim Diana and Carl Latta. They provided financial
assistance and encouragement throughout my work on this
project. Jim also spent hours helping me with large data
sets and critiquing the many drafts of this thesis. It was
a pleasure to work with him.
Finally, I thank my parents, Oreste and Shirley, wh9
were always fully supportive of my desire to work in the
field Of fisheries. Without their understanding and
support, I know this would never have been possible.
ii i
TABLE OF CONTENTS
ACKNOWLEDGEMENTS O O O O Q O Q O O O O Q O © © O O C i i
L I ST OF TABLES O © Q O O Q © © Q O O C , C. O © O © Q O v
LIST OF FIGURES O Q O © O © Q O Q O O O O O O O O Q vi i
ABSTRACT O O O O O O Ö O © O Q C O O O Q © O O Q O O i X
INTRODUCT I ON © O © O O C Q O O Q © O Q © O. O © Q O © l
METHODS O O O © O G O C O O O O O Q Q © C O C © © Q 4.
Cree l Cens llS © O O O O O O O © Q O C O © O O O 6
Yellow Perch Tagging . . . . . . . . . . . . . 7
Population Estimates . . . . . . . . . . . . . 9
Growth and Age Structure . . . . . . . . . . . l3
Mortality and Exploitation . . . . . . . . . . 17
Hooking Mortality . . . . . . . . . . . . . . . l 8
Fishery Simulation . . . . . . . . . . . . . . 19
RESULTS O O O © © © O © O O Q O O O © g O O O Q © O © 2 3
Creel Census . . . . . . . . . . . . . . . . . 23
Distribution O © © O © © O O © O Q O O O © O O 30
Population Estimates . . . . . . . . . . . . . 34
Growth and Age Structure . . . . . . . . . . . 38
Mortality and Exploitation . . . . . . . . . . 47
Hooking Mortality . . . . . . . . . . . . . . . 49
Fishery Simulation . . . . . . . . . . . . . . 51
D I SCUSS I ON © O O O Q O Q O © © O © Q © O © O Q Q O O 5 9
Management Recommendations . . . . . . . . . . . 78
L I TERATURE C I TED Q O O Ç © O O © © O O C Q O O Q Ö O 8 l
iv.
Table
LIST OF TABLES
Percent composition of species in the catch
for anglers in the Les Cheneaux Islands from
catch estimates for the l886 creel census. .
Yellow perch catch, catch per angler hour
(CPH), and total angler hours and trips from
creel census estimates from 1979-81 and l986
in the Les Cheneaux Islands. (Two standard
errors in parentheses.) . . . . . . . . . . .
Deviation in growth (mm) from the 1969–84
average for years (A) and year classes (B) of
yellow perch collected during fall assessment
in the Les Cheneaux Islands. . . . . . . . .
Number of yellow perch caught per l, 000 feet
experimental mesh gill net (CPUE) by age for
l969-86 fall assessment nettings in the Les
Cheneaux Islands. . . . . . . . . . . . . . .
Percent of total catch by age for yellow
perch collected during l 969–86 fall
assessments in the Les Cheneaux Islands.
Ages deviating by more than one standard
deviation above and below the l869-86 average
percent of total catch for each age are
indicated by (*) and ( - ), respectively. . .
Hooking mortality of yellow perch in the Les
Cheneaux Islands with percent mortality of
perch at the time anglers reached the boat
launch (immediate) and after 24 hours in a
live cage (cumulative) . . . . . . . . . . .
Predicted sport fishery catch in number (per
l,000 yearling recruits) under 0-, 7-, and 8-
inch minimum size limits (MSL) for l year and
6 years (equilibrium conditions) after
implementation of a MSL. Percent change in
catch under 7- and 8-inch MSL's relative to
l986 catch was calculated for all ages and
for age-5+ yellow perch. . . . . . . . . . .
Page
25
26
43
45
46
50
52
10.
ll.
12.
l3.
lá .
Predicted sport fishery yield in kilograms
(per l,000 yearling recruits) for yellow
perch under 0-, 7-, and 8-inch minimum size
limits (MSL) for l year and 6 years
(equilibrium conditions) after implementation
of a MSL. Percent change in yield under
7- and 8-inch MSL's relative to 1986 yield
was calculated for the perch fishery. . . . .
Predicted population size (number per l,000
yearling recruits) for yellow perch under 0-,
7-, and 8-inch minimum size limits (MSL) for
l year and 6 years (equilibrium conditions)
after implementation of a MSL. Percent
change in population size under 7- and 8-inch
MSL's relative to l886 population size was
calculated for age-3+ and age-5+ perch. . . .
Predicted catch in number of yellow perch
(per l,000 yearling recruits) for sport and
commercial fisheries in the Les Cheneaux
Islands for l986 and for 7- and 8-inch
minimum size limits (equilibrium conditions).
Predicted yield in kilograms of yellow perch
(per l, 000 yearling recruits) for sport and
commercial fisheries in the Les Cheneaux
Islands for l886 and for 7- and 8-inch
minimum size limits (equilibrium conditions).
Total lengths (mm) at the end of each year of
life for yellow perch in different locations
throughout the Great Lakes . . . . . . .
Total annual mortality rates for yellow perch
in different locations throughout the Great
Lakes and other large inland lakes. . . .
Predicted percent change in catch in numbers
and yield of yellow perch under 7- and 8-inch
MSL's (equilibrium conditions) relative to no
MSL for the present study and Schneider's
study. . . . . . . . . . . . . . . . . . o
vi
53
55
56
58
63
66
74
l0.
LIST OF FIGURES
The Les Cheneaux Islands area of northern
Lake Huron O O O © O © O O © O O C O O © O & ©
Best estimates for monthly catch of yellow
perch from January 1986 through September
l386 for anglers in the Les Cheneaux Islands.
Residence of anglers interviewed by the creel
census clerk for the l886 winter and summer
fisheries in the Les Cheneaux Islands. . . .
Response of Les Cheneaux perch anglers
interviewed during the l886 creel census to a
question concerning catch preference for
Yellow perch . . . . . . . . . . . . . . . . .
Response of Les Cheneaux perch anglers
interviewed during the l886 creel census to a
question concerning their desire to support
fishing regulations on the yellow perch
fishery. . . . . . . . . . . . . . . . . . .
. Distribution of tag returns, April 1985
through April 1986, of yellow perch tagged in
Mackinac Bay, April 1985. . . . . . . . . . .
Distribution of tag returns, April 1985
through April 1986, of yellow perch tagged in
Flower Bay, April 1985. . . . . . . . . . . .
Number of tags returned by anglers each month
April 1985 through October 1986, for yellow
perch tagged in 1985 (orange tags) in the Les
Cheneaux Islands. . . . . . . . . . . . . . .
Mean length (with 95% confidence intervals)
for ages 2- to 5-male yellow perch collected
during l969-86 fall assessments in the Les
Cheneaux Islands. Length-at-age derived from
samples containing only one fish is indicated
with an asterisk (*). . . . . . . . . . . . . .
vi i
Page
24
29
31
31
32
33
36
39
ll.
l2.
l3.
Mean length (with 95% confidence intervals)
for ages 2– to 5-female yellow perch
collected during l969-86 fall assessments in
the Les Cheneaux Islands. Length-at-age
derived from samples containing only one fish
is indicated with an asterisk (*). . . . . .
Mean growth (with 95% confidence intervals)
for male and female yellow perch collected
during 1969-86 fall assessments in the Les
Cheneaux Islands . . . . . . . . . . . . . . .
Percent of yellow perch tagged in 1985
(upper) or 1986 (lower) in the total
estimated catch (mi/ni) for 1986 in the Les
40
42
Cheneaux Islands. . . . . . . . . . . . . . .
vi i i
70
ABSTRACT
The yellow perch (Perca flavescens) fishery in the Les
Cheneaux Islands, which supports the tourism-based economy
of the area, was reputed by local residents to have
declined. Complaints about the fishery prompted this study.
The purpose was to estimate catch, mortality, exploitation,
and growth statistics from a creel census and tagging study.
These estimates were compared to estimates from previous
years and were applied in a fishery simulation model to
predict changes in the fishery produced by 7- to 8-inch
minimum size limits (MSL). From the creel census, the best
estimate for yellow perch catch in l986 was approximately
438, 000 fish. For an estimated la 2,000 trips, anglers
fished slightly over 400,000 total hours. The summer
fishery accounted for 89% of the total perch catch. Counts
of boats made from an airplane were on average 2.5 times
greater than counts made from the ground, and were used to
construct best estimates of catch and effort. Estimates of
growth, age structure, and mortality rate for perch were
constant over time, indicating that the population has
remained stable since sampling began in 1969. In addition,
Creel census estimates provided little indication of trends
in perch catch. Simulation of the perch fishery predicted
that yields (in weight) would only increase slightly (2-3%)
under 7- and 8-inch MSL's relative to the l886 fishery.
ix
Number of older perch (age 5+) was predicted to increase
substantially (7-54%). Commercial fishing in 1986 accounted
for only a small proportion of the predicted total catch in
numbers (4%) and yield (6%). Relative importance of the
commercial fishery was predicted to increase with increasing
MSL, but catch in numbers and yield would still comprise
less than l O% of the total.
INTRODUCTION
Yellow perch, Perca flavescens, is an important species
to both the commercial and sport fisheries of Lake Huron.
In 1986 yellow perch ranked first in numbers caught, with
sportfishermen taking an estimating catch of over 3 million
(Rakoczy and Rogers lº&7). Many small coastal towns along
Lake Huron rely on tourism generated by this popular sport
fishery for yellow perch.
Overfishing, which results in a decrease in numbers and
size of perch caught by anglers, can severely affect the
local economy by discouraging tourism. Unfortunately,
localized populations of yellow perch in the Great Lakes are
readily susceptible to overfishing by commercial fishermen
(Hile and Jobes l84l; El-Zarka l859; Eshenroder lº 77; Wells
l977; Wells and Jorgenson l983). Surprisingly, overfishing
of Great Lakes perch has never been attributed to
sportfishing, though annual yields for the sport fishery may
be many times greater than for the commercial fishery (Weber
l986; Diana et al. 1987; Keller et al. l.987).
Public perception Of overfishing also can be
detrimental to tourism. Public perception that a fish
population is being overfished is in itself sufficient to
discourage anglers, though the quality of a fishery may have
changed little. Under these circumstances, regulatory
agencies are often called upon to improve the quality of the
fishery with hopes of improving tourism.
Although problems such as overfishing of Great Lakes
yellow perch are relatively common, regulatory agencies
often lack adequate information for making sound management
decisions. Seldom have recent studies on Great Lakes yellow
perch completely analyzed both the status of the population
and the impact of the fishery upon it (Weber 1985, 1986).
Furthermore, long-term data necessary for assessing trends
in the perch population and catch are often scant or
nonexistent.
In recent years, residents of the Les Cheneaux Islands
have consistently complained that the yellow perch
population has declined and that tourism is suffering as a
result of this decline. Anglers attribute the apparent
decline in their catch to commercial fishing by Native
Americans, to increased predation by introduced species of
fish such as brown trout (Salmo trutta) and chinook salmon
(Oncorhynchus tshawytscha), and to heavy angling pressure
during the spawning period. Despite angler complaints,
evaluation of the rather uneven data available suggested
that the perch population and catch have remained relatively
stable since the first assessment in 1969 (Michigan
Department of Natural Resources interoffice communication
from J. R. Ryckman and J. C. Schneider to W. C. Latta,
December 1, 1983). However, more complete data on the perch
population and on exploitation of that population were
necessary to test results from previous analyses.
Therefore, a cooperative study between the Michigan
Department of Natural Resources (MDNR) and The University of
Michigan was organized to gather information on the perch
fishery and to determine its importance to the local economy
of the Les Cheneaux Island area.
Specifically, this thesis covers the evaluation of the
yellow perch fishery in the Les Cheneaux Islands. A creel
census and tagging study were conducted to estimate catch,
mortality, exploitation, and growth statistics. These
estimates were:
l. Compared tC) estimates obtained from previous
studies to determine whether declining trends in
the population and catch had actually occurred;
2. Utilized in a fishery simulation model to predict
changes in the fishery under different regulations;
and
3. Used to assess the impact of the Native American
fishery on the perch population and sport catch.
The findings from this study were then used to
determine an optimal strategy to manage this yellow perch
fishery.
METHODS
The Les Cheneaux Islands are located along the northern
shoreline of Lake Huron approximately 23 km northeast of the
Straits of Mackinac (Figure l). The bays and channels of
the Les Cheneaux Islands cover approximately 5,000 acres of
water and contain a variety of aquatic habitats, including
shallow (0-2 m) bays densely populated with submergent and
emergent aquatic macrophytes, deeper (2-8 m) channels and
open bays, and deep (8-20 m) outer bays and rocky shoals.
Over 70% of the original shoreline has been disrupted by
development in the resort communities of Cedarville and
Hessel (Jaworski and Raphael l978). Still, most of the
aquatic habitat continues to retain its oligotrophic
character.
Little information on the Les Cheneaux yellow perch
fishery prior to 1979 is available. Spot censuses were
conducted prior to sampling in 1969, but no estimates of
annual catch are present in the literature. However, resort
owners stated that yellow perch have attracted vacationing
anglers to the area since the l820s. Since then the perch
fishery has supported as many as 50 resorts. In 1986 the
number of resorts in the area was about one-half that
figure. Still, tourism generated primarily by the yellow
perch fishery was estimated to increase local income by
approximately 2–4 million dollars (Diana et al. 1987).
º u O. In H Đx{ef I u u Đu! qu OUI JO eÐ.I e Spuſe I SI X neau euſ) S3T 3UĻI,• I3. In 6! 3
Spue|S|
XneºuÐųO SÐT
ue6ļųoIW
SJeļeuJO||×
}）;
uuolun!！!
!G”
exſen
GD = -
��
«…» «， ，
∞∞∞
źģ
?%
∞
Ķ%ğ3！¿?
�ğſae
Šğ
°′ģ
ſae
ſae
�%ſae
ģģ
?ffffffffffffffffff
$:$2&§::：:：:：:Œ
*
&
«»
*…
。ſae
∞
�
∞
¿%žģ
žģğģ¿№，
ģģķi…
$%%%%};ĶĶ
ģ
§3
*
§§§
$ķ
∞●$$∞ ×
∞∞∞، ،ſae» ſ.:
ae3.%ſae
sae
º
ķ
%
ſae∞∞∞
！ſae«
&&&&&
Ķ%
¿?∞∞∞%ź
ğ∞∞∞ģ
∞
sº
º
º
º
cº
&
∞�_ae
?ķ
e
*
º
º
SS
§
8.
3&
（&3&&3&&3
žģĶķ
&&&&&&&&
ģ
∞∞∞§©®°¶
ſ.<ſſººſ：
§§Ķ
$%$ſae
%%
%$%%
ſae
∞
$$$$$$$$
%ğ¿?
ķŽ2
§©®°¶√≠√∞∞∞$%%
%
ź%
}%%;$.¡№ººººººģ。
ģs_（?žģ
%%%%№žģ。%%
ae
&
$$$$$$$$$
**º
33:、、、。
%
ğ
$§$%%%%%ģğ«»ğĶķ¿Œ
„F§§§§§§§©®$（№№:3，。::::::：::::::::::::::%
Dģžģ§ ğ&gģğģ» №
};。§§ģ§ſ.ğ、žģ。žģģğģžź
ź%$%$%$§$%;§%%%%%%%%%%%%%ģź$。
}ģţģğğğ






















































Creel Census
A creel census was conducted in the Les Cheneaux
Islands from ll January through 25 October 1986 (for
detailed methods, see Ryckman and Lockwood l 985). The
census clerk interviewed anglers 5 days per week on a random
schedule with 8-hour shifts. The weekly schedule always
included weekends and holidays. Census shifts were arranged
so that on any given day, either morning or evening, anglers
were randomly sampled. Anglers were interviewed at the end
of a fishing trip to determine catch (species, number, and
size), hours fished, species sought, gear, and residence.
Anglers were also asked if they would prefer to catch seven
7.5-inch perch or five 9-inch perch, and if they would
support regulations to produce larger, but fewer, perch in
the catch.
Periodic counts of anglers were made throughout the
census period to estimate angler effort in the Les Cheneaux
Islands. From ll January through la April the census clerk
made counts of shanties and open-ice anglers from various
access points. Biweekly counts of shanty use by anglers
were made to determine percent of shanties occupied, which
were stratified by month and location. From 23 April
through 25 October 1986, the census clerk made similar
periodic counts of fishing boats, pier anglers, and shore
anglers. Counts were randomly scheduled and made twice
daily. From 19 May through 7 November 1986, counts of
fishing boats were also made from an airplane five times a
week, weather permitting.
Total effort and total catch were estimated by strata
(weekday and weekend, and month) for each category of
angling (shanty, open ice, boat, shore, and pier). Angler
hours, mean catch per hour (CPH), and their variances were
calculated as outlined by Ryckman (1981). CPH was
multiplied by total angler hours for each stratum to
calculate total catch by Stratum. Total catch and total
angler hours were summed across strata to give an annual
estimate for total catch and total effort for each category
of angling. These estimates were summed for all categories
of angling to give overall estimates of total catch and
total effort.
While interviewing fishermen, the clerk also measured
total length for a representative sample of five yellow
perch from the angler's creel. From these samples, length
frequency of yellow perch in the catch was estimated.
Yellow Perch Tagging
Data used in estimating exploitation rates,
distribution, and population size were collected from a
tagging experiment conducted by fisheries personnel from
District 4 of the Michigan Department of Natural Resources.
During the last 2 weeks of April, trap nets were set in
Mackinac, Sheppard, and Flower bays (Figure l) to collect
spawning yellow perch. Yellow perch 7.0 inches and longer
8
were measured, sexed, and then tagged using serially
numbered Floy FD-68B anchor tags. The T-bar anchor end of
the tag was inserted between interneural bones of the dorsal
fin. Stebo (1972) showed that mortality of yellow perch
tagged using this method was not significantly greater than
for untagged fish in the Ottawa River, Ontario. Tag colors
were orange in l985 and yellow in 1986, however, tag numbers
overlapped. Scale samples were collected from five fish per
inch group per sex to determine growth rate.
In 1985 and l386, 4,969 and 6,680 yellow perch were
tagged, respectively. Tag returns were collected and
recorded at local bait stores and resorts. Several measures
were taken to encourage returns. Flyers describing the
tagging study, and providing information regarding return of
tags, were posted at resorts and local bait and tackle
stores. Local newspapers published accounts of the project.'
Chippewa tribal fishermen were also asked for their
cooperation in returning tags from yellow perch taken in
their assessment nets. Also, in 1985, a $l. 00 reward was
offered for each tag returned, and a lottery for $500.00 in
prizes was held following the summer fishery. In 1986 the
$l.00 reward was dropped and a lottery for ten $100.00
prizes was held. Census clerks also recorded tag returns
while interviewing anglers.
All tagged fish and tag returns were recorded on
computer file by District 4 employees, and all entries were
rechecked against the original tag-return sheets. Sex,
9
length, date tagged and recaptured, and tagging and
recapture location were entered for each tagged perch.
To analyze the distribution of tag returns, return
entries were sorted by tagging and return site. Numbers of
returns by site for each tagging location were then plotted
on a map and distributions of tag returns were assessed
visually.
Population Estimates
Population estimates were made using both single and
multiple census techniques. Single C e IſlSUIS techniques
included two applications of the Petersen estimate (Ricker
1975) :
N = (M* n) /m
where N = estimated size of population at time of marking
M = number of fish marked
n = catch or sample taken for census
m = number of recaptured marked fish in a sample.
variance for the simple Petersen estimate was calculated as:
V (N) = [M** n ° (n-m) 1/m . .
The first application of the Petersen estimate compared the
April 1986 trap net catch of untagged yellow perch to perch
tagged in 1985. The second application compared number of
tagged to untagged yellow perch observed by the census clerk
in the June 1986 sport catch. Individual estimates were
10
calculated under assumptions of either negligible or l-inch
annual growth. Stebo (1972) found that Floy FD-67 anchor
tags on yellow perch were retained well in his study of less
than a year. Therefore, negligible tag loss was also
assumed when estimating population size.
Multiple CeIlSUIS techniques used for population
estimation included mean of Petersen estimates (Ricker lº 75)
and regression estimation (Paloheimo l963). The l886
recaptures of yellow perch tagged in l985 and l986, and 1986
creel census catch estimates were used in both of these
population estimators.
Mean of Petersen estimates took the average of simple
Petersen estimates (Ni = Mi ni/mi) over a total of i periods
(s):
N
=
(Ni-N) */(s-l).
Variance for N was calculated as:
v(N) = Y(Ni-N) */[ (s-1) - (s-2) I or v(N) = v (Ni)/(s-l)
where v(Ni) is variance for the simple Petersen estimate for
the ith period. Similar values for variance estimates
calculated from these tWO equations indicated that
assumptions underlying this estimator had been met.
Advantages of using mean of Petersen estimates stem from the
fact that individual estimates of population, made for each
period, permit calculation of error based on observed
ll
variability, and can reveal trends indicating departure from
assumptions underlying this estimator (Ricker 1975).
Using the regression estimator (Paloheimo 1963),
population size was estimated as:
N = M”) ni/) mi.
variance for 6 (), mi / ni) was calculated as:
V (3) = 0
x:y:/. ni, where
0x-y a - [Y nº tº- (: mi) 1/[.. ni ' (s-l) 1.
From V (3), confidence intervals were calculated as:
(Mº i ni)/ſi mi i t .2|{v(9): (1 ni)' ſ ].
Advantages of using the regression estimator are that it
weighs each estimate of 6; according to how it varies with
6, and it is a robust estimator with respect to underlying
assumptions (Paloheimo 1963).
Several modifications were made to determine ni from
catch estimates. First, since only perch 27 inches were
tagged, catch estimates had to be modified from catch for
all sizes to catch for perch 27 inches. Therefore,
percentage of perch 27 inches was calculated from sport
catch length frequency, and total catch was then multiplied
by this percentage.
Further alteration of catch estimates was necessary to
account for seasonal growth. Only perch in the catch which
l2
were 27 inches at time of tagging were to be included in
estimates of ni •
Although no data were available to
approximate seasonal patterns in growth, Jobes' (1952)
observations on seasonal growth of Lake Erie yellow perch
appeared reasonable for application to perch in the Les
Cheneaux Islands. Thus, annual growth, which totaled about
l inch, was added by monthly increment (using Jobes' data)
to the initial minimum size of 7 inches, and only the
percentage of perch which exceeded this minimum size were
included in the estimate of ni for each month.
Absence of catch estimates for lS85 required that
multiple census population estimates, based upon l985 tagged
perch, be constructed using lº&6 returns. Therefore, M and
n; had to be adjusted in order to estimate abundance from
l985 tagged fish captured in 1986. To estimate M as of
January l986, number of perch tagged was multiplied by
survival rate (s) where: s = m + n – m ºn. Exploitation (m)
for April–December 1985 was estimated from percent of tags
returned for that period. Conditional natural mortality (n)
was also estimated for that period under the assumption that
expectation of death from natural causes remained constant
throughout the year. In addition, minimum size for perch
included in monthly ni estimates was set at 8 inches and
seasonal growth during summer was calculated as before.
13
Growth and Age Structure
From 1969 to the present, scale samples were collected
by MDNR (Alpena Great Lakes Fisheries Station and
District 4) during their fall assessment of the fish
population in the Les Cheneaux Islands. Yellow perch and
other species of fish were netted using graded-mesh gill
netS. Aging of many of the scales was checked to verify
that aging done by technicians in Alpena and Newberry had
been consistent. All scales were impressed on cellulose
acetate slides and examined under a microfiche projector at
44 times magnification.
To analyze for annual variations in growth of Les
Cheneaux yellow perch, mean length-at-age for l969–86
collections was calculated for each age and vear. Then, a
95% confidence interval for each mean length was constructed
using the pooled estimate for variance of the population
(s”). This pooled variance was derived from a two-way
analysis of variance (ANOVA) which tested the effect of year
and age on length. The mean square error (MSE) from the
ANOVA is the best estimate for s” (Neter et al. 1985).
Variance of the mean (s; *) was then calculated for each mean
length by dividing s” by sample size for each age and year.
values for standard error of the mean (s;) were multiplied
by *0.05, n-l (Remington and Schork lº&5), where Il Was
sample size at each age, to obtain 95% confidence intervals
for each mean length. All statistical analyses were
l4
performed on the Michigan Terminal System using MIDAS (Fox
and Guire lº'76).
Length-age data were also analyzed to determine whether
year class strength had a noticeable effect on growth rate.
Annual growth was calculated for ages 2–7 by subtracting the
mean length at age t-l from that at age t for the following
year. For example, mean length of age-2 perch in 1971 was
subtracted from mean length of age-3 perch in 1972 to
calculate growth for 3 year olds in 1972. Average growth
for l869-86 was calculated by age, and then summed for ages
2-5 and ages 2-6. Growth was also summed for ages 2-5 and
ages 2-6 for each year and year class of perch. All of
these values were subtracted from 1969–86 average growth to
calculate deviation from the mean. Deviations for years and
year classes were not normally distributed. Therefore, a
non-parametric procedure, the Wilcoxon sum rank test
(Remington and Schork lS85), was applied to test the
hypothesis that average deviation of growth between year
classes and years was not significantly different.
Catch per unit effort (CPUE) for assessment nettings
was analyzed to search for trends in CPUE at each age and
for the presence of exceptionally strong year classes.
Annual CPUE was calculated by age. Values of CPUE were not
normally distributed for all ages, so comparisons of mean
CPUE for different periods were made using the Wilcoxon sign
rank test.
15
To test for the presence of exceptionally strong year
classes, percentage of total catch at each age was
calculated for each assessment from 1969 to 1986.
Percentages for total catch at each age were then averaged
over all assessments, and standard deviations Were
calculated by age. Deviation from the mean was calculated
by age and year. Ages having a deviation greater than one
standard deviation Were considered exceptional. The
repeated appearance of an exceptional year class over
consecutive years was considered to indicate a strong year
class of yellow perch. -
Finally, mean lengths of perch netted in the fall were
compared with back-calculated mean length-at-ages to
determine whether fall length-age data could be used in
constructing a von Bertalanffy growth model. Assuming
negligible growth occurred between time of capture in the
fall and time of annulus formation, then fall lengths for
each age and back-calculated length-at-age should not
produce significantly different von Bertalanffy growth
parameters. To test this hypothesis, two von Bertalanffy
growth coefficients (k and Lee) calculated from fall length-
age data and back-calculated length-at-age, were compared. -
Scale samples from 1971, l080, and l386 were used in
back calculations. For each year, five scale samples per
age were taken at random using the ages originally
determined by the fisheries technicians. For ages 6 and
older, all available scale samples were used. The scale
l6
samples were aged and distance from focus to each annulus
and to scale margin was measured. Perch length was
regressed upon length of the scale radius using MIDAS. The
intercept of this regression approximates length of yellow
perch at time of scale formation (Carlander l'977), and
equals the intercept (a) in the Fraser-Lee modification of
the direct proportion back-calculation procedure:
Li = a + I (Le G- a) 'Sil/se
where Le = length of fish at capture; Se = length of scale
at capture; Si = length of scale at age i ; and Li = length
of fish at age i (Bagenal 1978). A FORTRAN program,
utilizing the Fraser-Lee modification, back-calculated
length-at-age for each age and then used a weighted average
to compute mean length-at-age for each year's sample of
perch. To compare mean lengths of fall fish with back-
calculated mean length-at-ages, Ford-Walford plots were
constructed using the Least Squares Method (Walford l 946).
Slope of the Ford-Walford plot is related to Ford's growth
coefficient k [slope = exp (-K) l, and intercept equals
Le’ (l-k), where Le is asymptotic length of the fish. For
Cº)
l971, 1980, and 1986, differences in these growth parameters
for Ford-Walford plots, generated from fall length-age data
and back-calculated length-at-age, were tested for using an
analysis of covariance based on a 5% level of significance
(Neter et al. 1985).
17
Mortality and Exploitation
Total instantaneous mortality rate (Z) was estimated
using the Least Squares Method to fit a straight line to the
descending limb of a catch curve (Ricker lS75; Everhart and
Youngs 1981). Catch curves were derived from spring tagging
and fall assessment catch data. To determine total catch by
age, all yellow perch taken during fall assessment gill
netting were scale sampled and aged. For the spring tagging
study, subsamples of five perch per sex, per inch group,
were collected for scale analysis.
From the aged scales, percentages of perch at each age
for each inch group were determined. These percentages were
then multiplied by total catch for each inch group to give
total catch by age within an inch group. Total catch by age
per inch group was then summed across inch groups to give
total catch at each age. Perch older than the first age
class which had declined to lo fish were not included in the
catch curve (Everhart and Youngs 1981). A one-way analysis
of covariance was used to test for significant differences
between Z values calculated from spring and fall catch data.
Similar values for Z would permit pooling of fall and Spring
catch data to estimate Z.
Instantaneous rate of fishing mortality (F) WaS
estimated using the relationship:
18
Z/A = F/u
where (A) is actual total mortality rate and (u) is rate of
exploitation (Ricker lST5). This relationship holds for
Ricker (1975) Type-II fisheries where fishing and natural
mortality were assumed to operate concurrently. Actual
mortality rate is (l − eTºº), and rate of exploitation was
estimated from the percentage of tagged yellow perch
returned within a year after tagging. The estimate for
instantaneous rate of fishing mortality was subtracted from
instantaneous total mortality rate to Calculate
instantaneous rate of natural mortality (M) : M = Z – F.
Hooking Mortality
Les Cheneaux yellow perch anglers were asked to
participate in a study to estimate hooking mortality for
yellow perch. Anglers volunteered to utilize various
angling techniques (bobber fishing, tight-line fishing, and
vertical jigging) and baits (minnovs, earthworms, and
leeches) to catch yellow perch. All sizes of perch caught
were kept and were placed in wire live baskets, which were
submerged alongside their boats. Transport of perch from
fishing site to the boat launch was in 20-liter plastic
buckets and length of time for transport was less than 20
minutes. Perch were rapidly measured, their overall
condition noted, and were placed into l,000-liter live
Cages. Dead yellow perch were culled and measured last.
Live cages were totally submerged in a well-oxygenated,
19
partially shaded location, and the water temperature was
measured daily. After 24 hours, condition of the yellow
perch was checked again, and they were returned to the
participating angler or released.
Fishery Simulation
The dynamic pool model is one of the most commonly used
estimation procedures for stock assessment (Tyler and
Gallucci l980). Unlike the surplus production model, the
dynamic pool model explicitly incorporates growth and
mortality rates into its calculation for yield. Though its
consideration of biological parameters makes the dynamic
pool approach intuitively more appealing to biologists,
parameters utilized in this model are sometimes difficult to
estimate.
The Beverton and Holt (1957) approach to the dynamic
pool model is often used for stock assessment. In their
approach to yield modeling, Beverton and Holt made several
simplifying assumptions. These assumptions included using
constant rates of natural and fishing mortality which were
immediately applied to cohorts of fish at their respective
ages (knife-edge recruitment), using the von Bertalanffy
growth curve to represent growth in length, and taking the
simple cube of the von Bertalanffy parameters to convert
them from length to weight. A modified version of the
Beverton and Holt approach, the multiple-cohort population
simulator (Clark and Huang 1985), was used to assess yellow
20
perch stock in the Les Cheneaux Islands. Modifications over
Beverton and Holt's model included:
l. Using instantaneous rates of natural (M) and
fishing (F) mortality calculated by age;
2. Using separate F values for different fishing
gears;
3. Using gradual recruitment of a cohort into fishing
gears;
4. Using length-weight coefficients to convert von
Bertalanffy parameters from growth in length to
growth in weight; and
5. Using the Baranov catch function to compute catch
by age.
These modifications for this fishery yield model were
suggested by Tyler and Gallucci (1980), and were described
in greater detail by Clark and Huang (lºg 5).
The multiple cohort population simulator was used to
simulate catch, yield, and population size for the perch
fishery under 0-, 7–, and 8-inch minimum size limits (MSL).
The l986 fishery (0-inch MSL) provided much of the baseline
data that the model is based on. In order to simulate the
fishery under different MSL's, instantaneous rates of
mortality for sportfishing had to be estimated by age.
F values estimated from angler tag returns were assigned to
ages greater than 4, which were fully exploited. TO
estimate F values for ages l-3, which were not yet fully
2l
exploited by the sport fishery, F values were adjusted by
trial and error until age frequency of predicted catch was
similar to that of observed sport catch (as calculated from
length-frequency data).
F values were also adjusted for ages of perch protected
under 7-inch and 8-inch MSL's. Percentages of age-3 perch
greater than 7 inches in total length were estimated from
age frequency for the l886 sport catch. This percentage was
then multiplied by the F value for that age. The same
procedure was performed for age-4 perch greater than 8
inches in total length.
The effect of hooking mortality on protected ages of
perch was also included in the simulation. To represent the
effect that hooking of sublegal fish had on yield under any
MSL, hooking mortality rate was multiplied by F values
calculated under no MSL, and these values were added to
natural mortality rates for sublegal ages.
Simulations were also run to show the effect of
commercial fishing on sportfishing. F values for the
commercial fishery were calculated by age. To calculate F
values by age, l086 fall assessment data were used to
estimate age frequency for commercial catch from length
frequency for 1986 (yellow) tags returned by Native American
netters. Since a majority of commercial netting was done in
May, length-at-age t for fall fish was assigned to age tºl
for netted fish. This correction was necessary to adjust
for the fact that commercially netted fish were assumed to
22
be one year older with no appreciable growth. Age frequency
for perch tagged in 1986 was also estimated. At each age,
percentage of tagged fish returned by commercial netters was
calculated to give an estimation of exploitation by age. F
values were then calculated using these exploitation rates.
For all simulations, recruitment was held constant at
l, 000 perch entering the fishery annually. Growth, natural
mortality, and fishing intensity were assumed to remain
constant in spite of regulations or changes in density of
perch. Calculations were made on an annual basis, beginning
in October and extending through September.
23
RESULTS
Creel Census
In l986 yellow perch comprised 85% of the total
estimated sport catch in the Les Cheneaux Islands (Table l).
Rock bass (Ambloplites rupestris) and cisco (Coregonus
artedi i) were second and third in total numbers caught. The
best estimate (made using airplane counts) for total catch
of yellow perch was about 438,000 fish (Table 2). Perch
catch peaked in June (Figure 2). Total angler hours were
estimated at slightly over 400,000 (Table 2). Using ground
counts, total yellow perch catch and total angler hours were
estimated at 213,000 and l84,000, respectively. The
estimates for total catch in l980 and l386 were similar,
however, the estimate for angler effort was significantly
lower in 1980 (Table 2). Therefore, catch per angler hour
of yellow perch was significantly higher in 1980 than in
l 986.
Yellow perch catch during winter (49, 283) represented
only ll # of the total for lS86 (Figure 2). Yellow perch
comprised 95% of total catch in winter, with remainder of
the catch comprised mostly of cisco and northern pike (Table
l). Total effort for the winter fishery (33,499 hours)
represented 8% of the yearly total. Effort for shanty
anglers comprised the greatest proportion (63%) of this
estimate. Effort for the winter fishery was similar in
24
150-r-

1
O
0-
5
O
sm
º
º
º
4.
%
M
A
Month
Figure 2. Best estimates for monthly catch of yellow perch
from January 1986 through September 1986 for
anglers in the Les Cheneaux Islands. *
25
Table l. Percent composition of species in the catch for anglers in the
Les Cheneaux Islands from catch estimates for the l986 creel
CellSUlS •
Percent of total catch
Winter Summer
Species fishery fishery Combined
Yellow perch 95 85 85
Cisco 3 3 3
Northern pike l 2 2
Menominee º-º tº ſº sº l l
Rock bass - tº- º 4 4
Pumpkinseed - --> a º º 2 2
Smallmouth bass - * - sº º l l
Chinook salmon - - - --> l l
Other l l * l
26
Table 2. Yellow perch catch, catch per angler hour (CPH), and total
angler hours and trips from creel census estimates from 1979–
8l and 1986 in the Les Cheneaux Islands. (Two standard errors
in parentheses . )
Season Catch Total angler
and per
year Catch hour Hours Trips
Winter
19801 lC)9, 145 3.44 3l, 752 7,687
(24,826) (l.0l) (5, 857) (l, 497)
l98l. 50, 718 l. 86 27, 281 6, 22.l
(6,776) (0.25) (386) (452)
1986 49, 283 l. 47 33,499 8,646
(6, 196) (0.24) (3,607) (889)
Summer
l.979 83, 818 0.96 87, 526 22, 137
(35, 822) (0.44) (l4, 464) (3,657)
l980 91, 929 2 - 27 40, 506 l2, 302
(ll, 270) (0.33) (3,282) (l,090)
19862 l64, 047 l.02 lSO, 287 54, 195
(28, 147) (0.22) (l7, 102) (5,659)
l.986 3 389, 444 l.04 372, 781 l33, 871
(68, 40l) (0.22) (46, l28) (l5, 466)
Combined
l980 201,074 2.78 72,285 l9,989
(27, 264) (0.46) (6, 714) (l, 852)
l9862 213, 330 l. 16 l83, 786 62, 84l
(28,821) (0. l.9) (l7, 478) (5,728)
1986 3 438, 727 l.08 406,280 l42, 517
(68, 402) (0.21) (46,261) (lS, 491)
* Estimates for l979-81 taken from Ryckman and Lockwood (1985).
* Estimates based on counts of boats from shore.
* Estimates based on counts of boats from airplane.
27
1980, 1981, and 1986 (Table 2). Perch catches during the
winter of 1981 and l886 were also similar. However, yellow
perch catch during winter l980 was substantially better than
in 1981 and 1986. Therefore, catch per angler hour was also
significantly greater in 1980 (Table 2).
During the open-water fishery, airplane counts of
fishing boats were on average 2.52 (+0.225) times greater
than counts made from different vantage points on the
ground. This surprisingly large difference between air and
ground counts was attributed to the large percentage of
boats hidden in the myriad of islands and bays, or located
too far offshore to be seen by the creel clerk counting from
vantage points along shore. Consequently, estimates of
total catch and total angler hours were based on both ground
and air counts. Estimates calculated using air count data
were believed to be more accurate, while estimates utilizing
ground counts were valuable for comparison with estimates
from previous surveys, which used only ground counts. The
pilot was unable to accurately count pier and shore anglers,
therefore, catch-and-effort estimates made for pier and
shore anglers were based on ground Counts.
The majority (89%) of yellow perch catch was made
during summer (Figure 2). The best estimate for total
summer catch was about 390,000 yellow perch. Species of
lower frequency during the summer fishery included rock
bass, cisco, northern pike (Esox lucius), menominee
(Prosopium cylindraceum), and pumpkinseed (Lepomis gibbosus)
28
(Table l). Angler hours for summer totaled 373,000, of
which 97.8% was by boat anglers, l. 6% by shore anglers, and
0.6% by pier anglers. The estimate of yellow perch catch
based on ground counts was about 164,000 fish; this was
significantly greater than comparable estimates for summer
1979 and l380 (Table 2). In fact, the l986 estimate was 39%
higher than the 1979-80 average. Similarly, the estimate
for angler effort during summer lº&6 was also significantly
greater than estimates for l979 and l380 (Table 2). Total
catch and total angler hours were significantly greater for
the l886 summer fishery, however, resultant catch per angler
hour was similar to that for summer lº 79, and both these
rates were significantly below catch per angler hour for
l980 (Table 2).
An analysis of angler residence from 2,530 interviews
showed major differences in demography of anglers
participating in winter and summer fisheries (Figure 3). A
majority (77%) of winter fishing effort was by local anglers
(Mackinac and Chippewa counties), and combined efforts for
residents of the Lower Peninsula and out-of-state anglers
constituted a substantially smaller proportion of the total
effort (Figure 3). However, for the summer fishery, anglers
residing in the Lower Peninsula contributed the largest
proportion to total angling effort, out-of-state anglers
were second, and Upper Peninsula residents contributed the
smallest portion to total effort. Overall, local anglers
constituted 21% of the angling effort, out-of-state anglers
'ſ， N?





30
contributed 22%, and anglers residing in the Lower Peninsula
contributed the majority of effort (51%).
- Most anglers (68%) fishing in the Les Cheneaux Islands
stated that they were fishing for yellow perch. The
percentage of anglers targeting yellow perch varied over the
fishing season with 95% of winter anglers and 61% of summer
anglers fishing for perch. During summer, anglers also
fished for pike (15%), cisco (12%), and salmonids (4%).
A majority of anglers stated that they would prefer to
catch five 9-inch perch over seven 7.5-inch perch (Figure
4). Anglers during winter, who were predominately local
residents, often preferred the many small-perch option,
while summer anglers, predominately tourists, were more
inclined to select the few large-perch option. Likewise, a
majority of anglers interviewed supported restrictive
regulations on the fishery, which would produce larger, but
fewer perch (Figure 5). Once again, response varied with
fishing season.
Distribution
Perch tended to remain close to where they were tagged
and formed fairly discrete stocks. A majority of perch
tagged in Mackinac Bay were recaptured by anglers in Hessel
Bay and adjoining bays on the west side of the islands
(Figure 6). Similarly, perch tagged in Flower Bay tended to
remain on the east side of the islands (Figure 7). Returns
for perch tagged in Sheppard Bay indicated that these fish
ſā
3.
82
±
©
Q.
9.
�^
3-4
(0.(1)
„C:
O 4–2 UN
* 4）) %） • r-4
O \!-!
Un Q,
25 Qų „C;
Ú). Un 2) U
%）, C. Un 3-4
Q!) (1)(1)
r-1 U) O Q,
On4-3
C -4>
•(Ó Q) (1) O
Q) 3-4 r-4
„C 3-4 • ++ r-4
U) U UN Q)
3-4Q) >，
QU \O TO
Q, COCl)
m.-OY 3–4 „C.
× r-$ • •-， 4-)
:-)Q!)
^{C} (1) „C. №
NQU „G 4-2 O
£ 4-3
Q!)Un Un
„C; OD G C
U) C • r + O
••-， OE • ++
Úſ 3-4 3-4 4-)
Q) 2，3 G) (0
, ! TO U r-+
£
\!-\! 70 O
O Q!) O
>
Q!) Q) C
Un • r + O
p : ?> ••+
O $-）�



32
º G86 I Ț ț u dv ' Keg Deuț （țoew u ! p3.66e4
ųou 3d Mo II ĐÃ go º 986 I I ţ adv. q6nou qą G86 I T ſ udv ' suunqa u 6.e4 go uoſ qnq ! a q sſ q º 9 e-Infiţă
S40ļ0UUO]])|
© --
!G”
3：3，
§©®
aerºsº.
*ķ。
ĶĶ *¿
©
?ź
ſae
×
�■■<！«■
§§
«»
！， º
ae
§§2:33%%%
ğğź3&&3&&
%%%%%%%%%%ğ%22%;
ģ№
№ ſae
?
%%%%%ģ
、、。
ſº：
ſae
№
-->$ſaeź
ź
ģžģ
�！
$$$$$$$$$$
§§§§§§3%
。$ſ;
ģğ
¿?ŠĶ:::::::::::::
ſae?ź
ź•
ae
●ſaeſae
�&&3&&3&&ķ
šķ£
šķĶĶ
ģķ
ſº：��
$3&&3&&
§
№
�
。«…»
%%
*_*）
%$§$%
*
��
、&&
§§§§§
¿?
∞TºTºT;， &]*）>（.*
«…» «， ！» º
«»， «» ， «» ，
?
ģ
|Ķ%%%%%%%%;%%
šķčģ
��33ģ
ſe，%$
�ae&×�●
§§
ğ
&$¿?%
ģ。
*&ķģ×
£:
$3&&
ğ
§
¿№
。
ſae�
*ĶſaeĶ
�、。、、。*）:¿?
ģģ
§§ºš¿
。%%%%%%
źģ}}
&&&&%$§$%%%%%%
&šķ。&%$3
¿3%sae
«^
<&，
ſae
%&&
Š%
Žģſº
！
。
??
$
ğ?
?ķ�ģż}}
§§ș
ž3$%%
32:2§@₪
∞∞∞∞∞∞
§&&&&ģsae.
saešķ};§§§�%%�� ž%
%ſae
∞∞∞
&ģ
3:33:
Ōſ©
$$$$ğź}
$%$&#%;ģģğ
§$%$§$%%%%%
。
Zģģ%ģģ£82 @
&&
ģ<aeº，
&&&&&&&&?%∞ķÁeg&}
�&&&&%%źģ
�
?
Ģ$%$§$%&
ſae≡
¿ķ$3ķ}}§§©®
ĶNºgº
$ğ。
ſº«T，，，，
■■■■
ae«»），ºsº，
。
Ķ%
§©®°¶
©®
ģ
ģ
∞∞∞
ķ%�、3%eae
$%$%%%%%ģķ(4&ºeupſoB&&
∞№Ō[…]�&º！
?ĢģĞğģģW №
šķ%%&&ķ}Ķ::
$$$$$$$$$%
$$$$
§:
º sº
º
sº
Sº
ķĶ
§§§。�■■$�
Dae��ģ§§§©®
ķģ！。ķ
§ğğğššğ
№ae
«»
-->źſae
ſaeDE!§§¿?ģèg
$ĶķOCģ、、。！Ō%
Ķ;%%&&DUſºĶ:º：ķ。ģģ
&Œ №©C§§¿?Ķ3$2$33%$%
№
。∞∞∞
%%ſae§§
Už$žģĶ$ģ§©
¿
■3×3×3№aeae
$$$$$$$$$$$$$$$$$§$%ź%


























































































33
* G96Ț Ț ț udv ' Keg 13 AOI） uț pe 66.e3
ųou ad ao II ĐÃ go º 996 I T Į a dy q5nou q \ G 86 I T ſ a dw ' suuną 3 a 6e3 Jo uoſ 3 nqſ u 3 s ſ CI º L3.un 6ță
SIÐļºuJOIX
ºm
O --
įG”
ķ%
ķ
%%�
ſae?§）,
%%%%%%;&%Ž%
。
?
%%
¿№
ſººſ¿№
|-%:2%&
ae%%%
ģ
！»
ºººººººººººººººººººººº
。
、。
ź。źģ
。
ğžģ
¿
|-
…….…
|-
ğ
%
º.
ż
ſaeaeaeſae
§§$$$$%ģ
ģ•
∞∞∞Ç
∞∞∞
،%žģ
-º-2–
Œſae
∞
º，ģ
ģģ
ğŒŒ3%
}}
%}
ģ
ķģ
%%%%
（±√∞
*：ſae
ģĶ
。
º.ķ%%
Dr.
ģ
&&*№.ſaeŻĘ
·
&#3$%
∞∞∞
、、3。、、。
%%%%
3:33%
Š&∞
∞ſae
S
）
•∞
• T •∞ →→•¿•
••%32）;
%
§№ae，ķ%ģğ
∞∞∞šķ%
%%Ç3333333.23%
№ºººº·žģ&（，，，，-
ķaºĶŠĶķ•
&&&&&&%%&
ºººſae!！！
saeº
∞º）
•
… --
?
，，ſae···
ğD، ſae
∞%%%%%%%%%
ģ
$%№.$:$$§$%
}}};};};&
Ķºº：：、،}%%
、§$%&!。șğ
Fø.Ķ%ź$$$$•I，ķ?،•Dae ：Ľ£&
}}%;&ķ$%šķ&99Ķģ\ （&&•Dw№•ģ}
∞■■■■∞∞∞∞∞∞∞∞∞∞∞∞∞∞， -∞ √∞∞∞∞•ººººººººººº¿?ae
{%%%%%%%%%%%žğèMÈ�، ، ،（* * * *，ģ•）%
§§ģU3\,ź?89∞ģ€.ģ
$$$ģ%83%}º.¿ĶĶ%%！ģ�·¿?
ğģ&%ķ$→ģ! Aeg º Xè€.ź
。waeD3ſºſ$§« » «»），$%ģĶ©3%
ķſaeĶ
&’« »ſº:º
。%ģĢeupțOeyN \，ź
ģģ∞ģººººººººººººº%
?sgfă*ğ
ģğ].*ģ
$$$$$$$$$$$$$ģ
\，
----
II
∞
،
¿??¿
ķģ
-Yº
ģ}$$$$$$$$$$$§§§§§●
$ķ§$%$§$%
ģ、
§：N Þžģķ
ſae �&
��
Ľ！，
●
;2∞∞∞
¿№º：∞º
ķģ-
§2№aeaeĶ
%º：ķ、?źž%%%%%Ķ·%ğ¿?§§ſaeſº?*：•;ģ
%Ķ&$%$%ſaeº，%%%%ſºnº：DºmaeĶ%3%žģ3}}&&ž%
№§§}ķtae3;};&&ķ$ģ?>，4%ż?●}
%%%%%%%%%%;%%$$$$$$ſaeĶĶ%ğ&&%;
Ź，3%::
ººººº-º-º-º-ºĻº）
%%%%！ºnº
�§§∞ •
Ľ∞∞∞、、。
。
ķ$ģ3&2&&
ķ}$%$§$%&&
(）、čĶ
Ķ%;&ģ
■ģ§:ģsº
§§
ſºs
�■■
w：∞∞∞ſae∞
Ķ%;№ºPaeº，�.-.■3×3×3;&&&&&&&…lae
*№.Ľsº¿©%;&3%。、。}，
§;}}%%■■■■ ■ ■ ■،Caeſae■•§3ģ•¿ģ、、、。%%&&žģ%%%、、。
►aeae¿№D…！，�•$№:%%%%%%%%%&3&2&×3，2×。、、
ĶSsaeaeJº：ſaeŽŇ،%&！！?！！?!！?！！?！！?！ae、、、、、、、、」，
ķ¿;ķģžģģ
ķſººſ！ 1vae，ĶĶ%$§$%
* „º.ººº..ººº,，,*，*,*\ſ*：）*）*）*）*）*）*）*)*）*）*ķķ%;ķčķ

















































































34
ranged over a wider area than perch tagged in Mackinac or
Flower bays (Figure 8). Even among fish tagged in Sheppard
Bay, a substantially greater percentage of these perch were
located in adjacent Muscallonge Bay and Middle Entrance than
other areas. Returns for 1986 tagged perch demonstrate
similar distributions.
A large percentage (96%) of the 253 perch tagged in
1985 and recaptured in 1986 tagging study trap nets were
caught in the same bay where the fish had been tagged. This
suggests that perch may home to spawning bays. Homing of
yellow perch to spawning bays further supports the idea that
discrete stocks may exist within the Les Cheneaux Island
alſ e a .
Population Estimates
Angler returns for orange tags from April 1985 to April
1986 totaled l, 40l. A large percentage of recaptures were
made during the summer (Figure 9). Returns for yellow tags
from April 1986 to April 1987 totaled l, 126, and the
seasonal distribution of returns was quite similar to 1985.
Thus, annual return rates for l985 and 1986 were 28% and
l7%, respectively.
The two applications of the Petersen estimate made
using tag return results gave very different estimates for
population size of yellow perch. The population of perch 27
inches was estimated at approximately 80,000 fish using
ratio of perch tagged in 1985 to untagged perch captured in
35
q2.1 ed AOI I ĐÃ go º 986 I
Suºļºu/O||×
O --
!G”
%%
ğ
I Į udvſ
ģ
（?
�
ģ%§©®
�
�
§
ºz
ģ}O
、。
׿№
�$ſ;&ģ
* G86 I I
q6nou u q G86 I T ſ udvſ
Į udv ' Keq pueddau S u ț pe 66e q.
* Su. In q3 u
6.e4 go uO Į qnq ! a q S Į GI
！ºſº
%žģ
Ķğ
ğ
�， � �ae_&
№ºººººººººººººººº
?
∞∞∞
ſºs
。2。
**，№rae
%%&&
Ď*、≡ae
}}Ķ
źŽž;
&%Ź！
«»
•∞*
ſaeſae！&&
%%%%%%%%§%3%
%%;$%$§$%$§$%�№º&&
■■■■∞ºŌ
ģ
、。
∞; ∞$$$
∞∞∞ →∞ √
¿№●*
ſºº. №
。
%%
3:&&
%&%$&#%
ģ
。
¿
ģ
::::::::::::::&&&&&&&&&&&&
žğ
·ſae
&ſae
ſaeſae
·23%
�！%
ģ
ן∞∞∞
3，333$
Äğķ
žģğģ%$
%%%%%%&&%$$ğ%3，3%
ķģģź}
∞∞∞Ķ
*šķžģ
。
*
！
ſaeae
*�
¿
ſae
∞
≡
%&！！&&
§§
ģ
ģ
%&&*
*§§©®
。
§%Ķ2%
}
ģ
、
§§ae
№ğ
%
ź
©�ķ
≤≤
�
%%%%%
%%%%%%
ģ
$%&
£§§§§
aeº-aeae
º_º_ſeģ¿¿.
ğ
�ģ
%
38
ģ
、
ģ！
√≠√∞， √æ√≠√∞
ſae
ģķ
«»
�
∞∞∞
∞ż%ģğ«;ºs
、&&&&&&&&&&&&::3
！
¿?
žģ?
?ğ
№ №.ģ
��ğķžģ。
ģ
?
ğ
ae
Ķ23%
%3&3&&3&&
Ķ
%
¿C）
∞∞∞
ķğ
��$%&&&&&&&&&&
Ķ、。
ģ。
«…»ź
¿º
«¿TºT« （Te）， «（.*?）（ *) ≤
ſae
∞∞∞
%%&&
§§§），
ķ
�
§§-
Ķ2：5
%$§$%ķ«（…»（&
§§©®
£ģ
%
¿№º:
、、、。、3。
3×3×3;&&&&
ĶĶ$§），
¿?
žģ
。
�
* ſº •
saeſºxae:
·º·：·º·：·º·：·º
¿
��¿?
、、。
::::::：:::::::::::::::::：::::::::::3
«» ，ź。
Źź�ģ
�ž%
ģ
ģ
2×
ğ
≡
■
č¿
。
8 3_In6! 3
•
ſae
ģ
。
∞∞∞∞∞∞
ķ
•
ſae：
ķ
aeae
¿•
∞ √Æ√≠√∞£;
$&Ķ
§§§§¿
。§§
ģ
3;
¿?±
∞ź





















































































36
400-

me
J50-
300-
2
5
O
º
º
;
2
O
O
1
5
O
ºl
- e.
100-
50-4-
.
.
A
A.
S
T
.
J
F
M
Á
M
j
j
Å
S
Figure 9. Number of tags returned by anglers each month
April 1985 through October 1986, for yellow perch
tagged in 1985 (orange tags) in the Les Cheneaux
Islands.
37
the 1986 trap nets. This estimate appears extremely low,
considering that catch of perch 27 inches was estimated at
approximately 360,000 fish in 1986.
The Petersen estimate, which compared tagged to
untagged perch actually observed in the June sport catch,
gave a more realistic estimate of population size. Assuming
negligible growth in May, the population of perch 27 inches
was estimated at about 524,000. This population estimate
was considered the most accurate of the estimates because it
was relatively precise (coefficient of variation (CV) =
l9%), and assumptions underlying it did not appear to be
violated. If perch were assumed to grow 0.2 inch in May,
this estimate was reduced to 35l,000 fish.
The four estimates of population size derived from mean
of Petersen estimates or regression estimators were not.
significantly different (p > 0.05). Using mean of the
Petersen estimates, population size (perch 27 inches),
estimated from 1986 returns of l985 tagged perch, was about
900, 000 fish. This estimate increased to l, 600,000 fish
when 1986 returns of lº& 6 tagged fish Were used.
coefficient Of variation for both estimates WaS
approximately 25%. Regression estimates of population size,
derived from 1986 returns of 1985 and l386 tagged perch,
were 1,000,000 and l, 800,000 perch 27 inches, respectively.
These population estimates were similar to mean of the
Petersen estimates, however, they were less precise (CV =
26% and 42%).
38
Growth and Age Structure
Analysis of scale samples collected from l969 to l886
revealed large discrepancies in ages previously assigned to
yellow perch by fisheries technicians. Aging of l985 and
1986 samples differed substantially from aging done for
1969–84 samples. Scales for lSS6 perch were re-aged, while
scales from 1985 and l381-82 (which were not available) were
omitted from the growth analysis.
Overall, growth of yellow perch in the early 1970s was
better than the l869-86 average (Figures lo and ll). Age-4
and age-5 perch were significantly larger in l969–74 than in
1975–86 (p <0.05). Age-3 perch also showed better than
average size in the early lS70s, but they were about average
in 1969–70. Although average size was smaller in 1975–86,
perch growth for individual years during this period was
equal to or better than in 1969–74. Unlike l'969-74, perch
did not show better than average growth for more than two
consecutive years during this period.
Growth of age-3 and older yellow perch declined sharply
by 1975 (Figures lo and ll). Size began to decline for
younger perch prior to l975. Size of age-3 perch started to
decline in 1973 and continued to decline through l875. Size
of age-2 perch appeared to decline slightly from 1972
through 1975, however, mean lengths for age-2 perch during
this decline were not significantly different from years
prior to it (p <0.05).
39
350 ºr º, D
Age 2 Age 3
300 +


250 +
200 + f E f E sºo f
*H*H.H { { | º Ef
E 100+++++++++++++++++++ 4––––––––––––––––––.
5.
+ 350 r


#
Age 4 * Age 5
300 + tº º
250 + | |#
l º £If |
### ## ++ £ f |fft|{# FI
+
200 +
150 +
100 !------------------, ------------------
70 75 80 85 70 75 80 85
Year
Figure lo. Mean length (with 95% confidence intervals) for
ages 2- to 5-male yellow perch collected during
l969-86 fall assessments in the Les Cheneaux
Islands. Length-at-age derived from samples
containing only one fish is indicated with an
asterisk (*).
40
350+ ºn 9
- Age 2 - - Age 3
300 + -
250 !
- f
200 + | #: "g | ſ f
150 ! T # H - l *H
jº H-H+++++++++++-H +++++++++++++++++-H
#350.
5 Age 4 Age 5
300 +




200 +
150 +
100 !------------------ +-H–H–H–H–H–H–H–H–H
70
Figure ll.
75 80 85 70 75 80 85
Year
Mean length (with 95% confidence intervals) for
ages 2- to 5-female yellow perch collected
during 1969–86 fall assessments in the Les
Cheneaux Islands. Length-at-age derived from
samples containing only one fish is indicated
with an asterisk (*).
4l
From 1975 to 1986, average size of perch showed no
trend. However, sizes of age-3 and age-4 perch were highly
variable, and on occasion, IIle alſ) length-at-age was
significantly different for consecutive years (Figures 10
and ll). Sizes of age-5 perch in 1975–86 also varied, but
variation among mean lengths was not as great as for age-3
and age-4 perch. Size of age-2 perch varied least. NO
significant differences were present between mean lengths
for age-2 perch (p > 0.05).
Female perch grew significantly faster than males, (p
<0.05) for ages 2-7 (Figure lz). Sexual differences in mean
size increased with increasing age, but were insignificant
at age 8+, due to high variability and small sample sizes.
Also, size among individual female perch, ages 2-7, was
significantly more variable than for males (F-test, p
<0.001).
Overall, growth Of yellow perch showed greater
deviation among years than among year classes (Table 3).
Deviation of yearly growth from the 1969–86 average was
significantly greater than deviation of year classes from
that average for both age-2 to age-5 females (p = 0.0l ) and
age-2 to age-6 males (p = 0.03). The difference between
deviation for growth for years and year classes was
marginally significant for age-2 to age-5 males (p = 0.06)
and insignificant for age-2 to age-6 female perch (p =
0.13). Average deviation for growth between year classes
was not significantly different between sexes (p 20.85).
42
350-
300-
250-
200-
:
100 | |
Age
Figure l 2. Mean growth (with 95% confidence intervals) for
male and female yellow perch collected during
l969-86 fall assessments in the Les Cheneaux
Islands.

43
Table 3. Deviation in growth (mm) from the 1969-84 average for years
(A) and year classes (B) of yellow perch collected during fall
assessment in the Les Cheneaux Islands.
(A) Deviation from average yearly growth.
Males Females Males Females
Year ages 2-5 ages 2-5 age 2-6 age 2-6
1970 +24 +l6 +9 +20
197l +34 +5 +6O +l3
1972 +ll +l4 +28 +l4
1973 -10 +13 -30 O
1974 +5 -9 -15 tº-º º sº tº
1975 -68 -l.00 -39 -140
1976 +17 º +29 +44 +35
1977 +18 +22 +12 +4l
1978 -52 -24 -63 –57
1979 O -36 –8 –36
l980 +36 +47 +36 +47
l981 tº-º º º GEs tº tº tº-, -ºº º tº gº tº
l982 - ºs tº E_2 -ºº º sº - - º tº-º-º: sº-
1983 --> º º tº ºr -º cº- ſº-ſº sº-3 -> tºp º ºs
1984 -34 –26 —5.3 +26
Average
deviation* 34 39 40 57
(B) Deviation from average growth for a year class.
Males Females Males Females
Year age 2-5 age 2-5 age 2-6 age 2-6
1968 + l2 + 16 -8 º +3
1969 O +4 -20 tº º cº-> -º
1970 +10 +ll +39 * --> ----> -->
1971 –23 -lá +4 -8
1972 -16 -l -22 +18
1973 +l -21 -l.0 -54
1974 + 17 -ll +9 -8
1975 –3 –3 -3 +30
1976 -l2 -l2 tº- tº-º-e ‘E-º tº º ºſmºs
Average
deviation* l4 l3 26 29
* Calculated from the absolute value of deviations for years and year
classes.
44
Likewise, average deviation for growth between years was
also not different for males and females (p > 0. 84).
CPUE for any age from 2-8 (Table 4) was not
significantly different between 1969–74 and 1975–86 (p
>0.07). CPUE in 1985-86 was extremely small for all ages
and, therefore, comparisons were also made between 1969–74
and 1975–84. CPUE for age-5 and age-6 perch was
significantly greater in 1975–84 (p <0.02). Differences
between CPUE for the two periods were insignificant for all
other ages.
With the exception of the l87l year class, year classes
failed to demonstrate greater than average representation
within the catch for consecutive years, which indicated a
lack of exceptionally strong year Classes for this
population. For the strong lS7l year class, CPUEs for age-
2, -3, and -4 yellow perch were greater than one . standard
deviation above the mean (Table 5). However, most of the 38
ages which deviated by more than one standard deviation from
the average were randomly distributed among different year
classes (Table 5). Thus, a large proportion of the
differences in relative strength for ages from year to year
appeared to be a result of random variation in catch. The
insignificant level of variability in year class strength
supported the idea of pooling catch data to obtain better
estimates for total mortality and age composition.
Homogeneous catch data also imply constant recruitment of
45
Table 4. Number of yellow perch caught per l,000 feet experimental mesh
gill net (CPUE) by age for l969-86 fall assessment nettings in
the Les Cheneaux Islands.
Age
Year l 2 3 4 5 6 7 8 9 lO Total
1969 --- 3.6 5.3 8.6 5.0 3.6 0.7 0.3 --- --- 27. l.
1970 O. 4 5.3 40.9 4.9 3.7 l. 7 2.4 l.2 0.8 --- 6l. 3
1971 O. l. 31.4 22.2 l'7.3 2.6 l. l 0.9 0.2 --- --- 75.8
lg72 O. l 8.8 6. 8 6.6 3.6 0.7 l. l 0.3 --- ---- 28.8
1973 O. 3 24. 4 26.4 lº. 6 10. l 2.5 --- 0. 4 --- --- 77.7
lg74 --- 9.8 65. O la .9 4.0 l.2 0.4 --- --- --- 95.3
lo'75 0.2 7.5 l2.2 23.5 ll. 8 l.5 0.5 0.7 --- --- 57.9
lo'76 --- la .5 ll. 2 lo. 8 8.0 l. 5 0.5 --- --- --- 45.5
lo'77 O.2 2.9 Ll. 9 ll. 4 5.6 2.4 0.8 --- --- --- 35.2
lo'78 ––– 4.6 33.7 l.2.4 8.5 5.6 3.8 l. l 0.3 --- 70.0
lg79 --- le. O le. 3 21.3 9.6 3.2 0.6 --- 0.3 --- 67.3
1980 1.6 16.2 6.8 ll.0 9.2 5.4 0.8 0.3 --- --- 51.3
1981 --- --- 43.7 l9.6 26.2 23.8 9.4 2.7 l. 6 --- 127.0
1982 0.9 LS.2 4.7 ll.l 12.5 4.l. l. 3 0.5 --- 0.3 50.6
1983 --- 0.3 l9.0 lo. 8 9.8 6.3 l. 3 l-3 0.3 --- 49.1
1984 ––– 17.9 7.6 21.3 13.0 3.7 l. 4 --- --- --- 64.9
1985 ––– 1.8 4.5 4.2 3.3 l. l 2.1 --- --- ~~~ 17.0
1986 --- 3.1 5.1 9.3 2.6 1.6 0.1 0.3 --- --- 26.5
46
Table
5. Percent of total catch by age for yellow perch collected
during l969-86 fall assessments in the Les Cheneaux Islands.
Ages deviating by more than one standard deviation above and
below the l969-86 average percent of total catch for each age
are indicated by (*) and ( - ), respectively.
Age
Year l 2 3 4. 5 6 7 8 9 lC)
1969 --- 13. l l 9.7 31.7 l8.6 lº . 1 + 2.7 l. l --- ---
1970 O. 7 8. 7 66.7° 18. O - 6 s O - 2 .. 6 4. O 2. O l. 3 * ---
lg7l 0.2 41.3 + 29.4 22.9 3. 4- 1. 4- 1.1 0.3 --- ---
1972 0.5 31.3° 24.2 23.6 l2.9 2.5 4. O L - O --- ---
1973 O. 4 31.5* 34.0 l7.4 l2.9 3.2 --- O - 6 --- ---
lo'ſ 4 ––– lo. 3 68.2° 15.6- 4.2- l. 3.- 0.4 --- --- ---
l975 0.3 lix.0 21.0 40.6* 20.5 2.5 O. 9 l. 2 --- ---
lo'76 --- 29.7 24.5 23.8 l7.6 3.3 l. 1 ——— --- ---
1977 O. 7 8.3 33.7 32.3 lS. 9 6.9 2.2 --- --- ---
1978 --- 6.5 48.2 * l'7.7 l2. l 8. O 5 : 5 L. 6 O. 4 ---
1979 --- 23.9 24.3 31.4 lá .3 4.8 O. 8 --- 0.5 ---
1980 3. lt 31.5* l3.3 - 21.5 l7.9 lo .. 5 l. 5 O. 7 --- ---
1981 --- ––– 34.4 LS.5- 20.6 l8.7° 7.4 ° 2. lt l. 3" ---
l982 0.6 30.4+ 9.5 - 22.3 25. lt 8. l. 2.5 L - O --- 0. 5*
1983 --- 0.5- 39.0 22. l 20.0 l2.7* 2.6 2.6* 0.5 ---
l984 --- ––– 27.6 ll. 7 - 32.9° 20.0° 5.7 2. lt --- ~~~
1985 --- 10.6 26.5 24.7 19.4 6.5 l2. 3 + --- --- ---
1986 --- l3.7 22.5 40.7° ll. 6: 7.2 2. 7 O. 6 1. O ---
47
perch into the fishery, which was an underlying assumption
in fishery simulation.
No significant differences were found between von
Bertalanffy coefficients (k and Lee) generated from fall
length-age data and back-calculated length-at-age (p >0. 54).
The von Bertalanffy parameters calculated for l971, 1980,
and l386 from fall length-age data were more consistent when
compared to those calculated from back-calculated length-at-
age. However, R* values indicated that back-calculated mean
length-at-age fit the von Bertalanffy model slightly better
than mean lengths for fall fish. Pooled length-age data for,
l969–86 fit the von Bertalanffy growth model well (R- =
0.99), and growth parameters generated were reasonable for
yellow perch. Therefore, pooled length-age data were used
to estimate von Bertalanffy growth parameters used in
fishery simulation.
Mortality and Exploitation
Total instantaneous mortality rates and their 95%
confidence intervals, calculated from 1986 spring and fall
catch data, We re 0.97 (+0. 75) and 0.86 (+0.32),
respectively. Values for Z were not significantly different
(F-test, p = 0. 73). Catch-curve regression for pooled
spring and fall catch data resulted in a Z value of 0.97
(+0.66). The Z value calculated from the catch-curve
regression of average CPUEs for 1969-85 was 0.80 (+0.17),
which was not significantly different from Z calculated from
48
1986 catch data (F-test, p = 0.39). Therefore, all
available catch data from lº S9 through l986 were then pooled
to give an overall estimate for Z of 0.80 (+0.17). Actual
total mortality calculated from this overall estimate for z
was 55% (+l 6%) per year.
Rate of exploitation (u) was estimated using lo&5 tag
return S. Tag returns by anglers for l986 appeared
unreasonably low. Comparison of l985 and lº&6 yellow perch
sport catch estimates for May–June showed a twofold increase
in catch in 1986, yet return rate in 1986 of yellow tags was
only 56% of return rate in 1985 of orange tags for that same
period. Relative to catch, return rates by sport anglers
were over 3.5 times greater in 1985. Decrease in rate of
return was probably a result of the altered reward system
and decreased interest in the tagging study after the first
year.
The annual return rate of 28% for orange tags was
considered the minimum sport angling exploitation rate.
Although little published information is available on tag
reporting rates, a tagging study on walleyes in the
Minnesota waters of Lake of the Woods did show nonreporting
- by anglers to be slightly less than one-half the number of
tags reported (Payer et al. 1987). Under this assumption,
nonreporting in this tagging study was estimated at about
14%, which inflates annual exploitation rate by sport
anglers to 42%.
49
Indian assessment netters returned only 23 orange tags
in 1985, which represented a 0.5% exploitation rate for
commercial fishing. Returns increased to l8l yellow tags in
l986, representing a 2.3% exploitation rate. Overall, rate
of exploitation (u) upon the perch population from both
sport and commercial fishing was estimated at 43% for 1985.
From this estimate of exploitation rate, instantaneous rates
of fishing mortality and natural mortality were estimated at
0.63 and 0.17, respectively.
Hooking Mortality
For the hooking mortality study, anglers caught lb 3
yellow perch varying from 5.0 to 9.5 inches in total length.
Mortality ranged from about 20-30% for perch of all sizes
(Table 6). Hooking mortality during winter should be
considerably less. With lower water temperatures, fish are
less susceptible to disease or stress induced mortality from
handling or injuries sustained from hooking. Also, the
majority of Les Cheneaux winter anglers fish with ice jigs
and spoons. Perch taken using this method of fishing were
usually hooked in the lip rather than in the gills or
esophagus. Although hooking mortality rate was probably
lower during winter, the winter fishery takes a relatively
small percentage of the total perch catch (Figure 2).
Therefore, annual hooking mortality was assumed to remain at
about 20% to 30%. Hooking mortality for yellow perch
.<7 inches was not significantly different from hooking
50
Table 6. Hooking mortality of yellow perch in the Les Cheneaux Islands
With percent mortality of perch at the time anglers reached
the boat launch (immediate) and after 24 hours in a live cage
(cumulative).
Immediate Cumulative
Species Number Number
and Number • Of Percent of Percent
Size caught deaths mortality deaths mortality
Yellow perch 153 3l 20.3 45 29.4
(all sizes)
Yellow perch 34 6 l'7. l 8 23.5
(<7.0 inches)
51
mortality for all sizes of perch (Table 6). Thus, a hooking
mortality rate of 25% was used in the fishery simulation.
Fishery Simulation
Simulation of the perch fishery under a 7-inch MSL
predicted that, at equilibrium, there would be little change
in the anglers' catch in numbers, yield, or older (age 5+)
yellow perch compared to the existing fishery under no MSL
(Tables 7 and 8). On the other hand, simulation of the
perch fishery under the 8-inch MSL predicted a significant
decline in number caught, a small increase in yield, and a
substantial increase in number of older perch in the catch
compared to l886 catch.
The first year after new regulations were imposed,
predicted catch in numbers would decline l.2% under the 7-
inch MSL and 49% under the 8-inch MSL. As the fishery
approached equilibrium, catch in numbers under the 7-inch
MSL would remain slightly below lS86 catch, whereas, the
severe decrease in catch under the 8-inch MSL would be cut
almost in half (Table 7).
Predicted yields in kilograms for perch under both
7- and 8-inch MSL's (equilibrium conditions) demonstrated
only al slight increase over yield for l886. Upon
implementation of a 7-inch MSL, yield would decline slightly
(6%). However, with time, yield would slowly improve until
it was slightly better than yield for 1986 (Table 8). Yield
under the 8-inch MSL initially showed a substantial decline
52
Table 7. Predicted sport fishery catch in number (per l,000 yearling
recruits) under 0-, 7-, and 8-inch minimum size limits (MSL)
for l year and 6 years (equilibrium conditions) after
implementation of a MSL. Percent change in catch under 7- and
8-inch MSL's relative to l986 catch was calculated for all
ages and for age-5+ yellow perch.
Predicted catch
0-inch MSL 7-inch MSL 8-inch MSL
Age 1986 l year 6 years l year 6 years
l O O O O O
2 l2 O O O O
3 l38 91 92 O O
4 - l92 l92 207 90 l.08
5 90 90 98 90 l40
6 42 42 45 42 65
7 20 20 2l 20 32
8 9 9 lO 9 l4
9 4 4 4 7
10+ 4 4 4 4 4
Total 5ll 452 48l 259 370
Percent change
(all ages) in º º tº -12 -6 - 49 –27
Percent change -
(age 5+) gº tº Ee O +7 O +54
53
Table 8. Predicted sport fishery yield in kilograms (per l,000 yearling
recruits) for yellow perch under 0-, 7-, and 8-inch minimum
size limits (MSL) for l year and 6 years (equilibrium
conditions) after implementation of a MSL. Percent change in
yield under 7- and 8-inch MSL's relative to l986 yield was
calculated for the perch fishery.
Predicted yield (kg)
0-inch MSL 7-inch MSL 8-inch MSL
Age l986 l year 6 years l year 6 years
l O O O O O
2 l O O O O
3 l5 9 10 O O
4 33 33 36 l5 19
5 24 24 25 24 36
6 l6 l6 l8 15 24
7 l0 lC) lC) lC) l4
8 6 6 6 6 9
9 3 3 3 3 5
10+ 2 2 4 2 6
Total ll.0 lC)3 ll2 75 ll3
Percent change º- ºr º Eº -6 +2 -32 +3
54
(33%), but increased rapidly to 3% above 1986 yield (Table
8).
Greatest increase under MSL regulations, relative to
the 1986 fishery, was the number of older perch (age 5+) in
the catch (Table 7). As for overall catch and yield,
increase in number of age-5+ perch under the 7-inch MSL was
relatively small (Table 7). This increase was substantially
greater under the 8-inch MSL, with number of age-5+ perch in
the catch approximately 54% greater than in l986.
Predicted population size under the 7-inch MSL also
demonstrated a relatively small increase compared to l886,
while increase under the 8-inch MSL was considerably
greater. Age-3 and older perch in the population would
increase relatively little under both MSL regulations (Table
9). Number of older perch increased significantly under
- higher MSL's, with number of age-5+ perch increasing 8% with
a 7-inch MSL and 55% with an 8-inch MSL at equilibrium.
Increases for perch over 7 inches and under 7- and 8-inch
MSL's were predicted to be 7% and 29s, while those for perch
over 8 inches were predicted at 8% and 42%, respectively.
Addition of the commercial fishery resulted in only
small declines for sport catch in numbers of perch. Catch
in numbers for commercial netters constituted only a small
proportion (4-8%) of total catch under different MSL’s
(Table lo). As MSL increased, proportion of total catch
harvested by the commercial fishery did increase slightly
(Table lo). Participation by the commercial fishery in 1986
55
Table 9. Predicted population size (number per l,000 yearling recruit)
for yellow perch under 0-, 7-, and 8-inch minimum size limits
(MSL) for l year and 6 years (equilibrium conditions) after
implementation of a MSL. Percent change in population size
under 7- and 8-inch MSL's relative to l986 population size was
calculated for age-3+ and age-5+ perch.
Population size
0-inch MSL 7-inch MSL 8-inch MSL
Age 1986 l year 6 years l year 6 years
l l, 000 l, 000 l,000 l, 000 l,000
2 844 844 84.4 843 843
3 7Ol 709 710 709 709
4 466 496 501 555 562
5 217 217 235 280 338
6 lC2 102 ll.0 lC2 l58
7 48 48 5l. 48 74
8 22 22 24 22 35
9 lO lC) 10 lC) l6
10+ 9 9 10 9 l3
Total
(age 3+) l, 575 l, 6l3 l, 65l. 1,735 l, 905
Percent change
(age 3+) - º -º +2 +5 +10 +21.
Total
(age 5+) 408 408 440 47 l. 634
Percent change
(age 5+) - sº -º O +8 +15 +55
56
Table lC. Predicted catch in number of yellow perch (per l, 000 yearling
recruits) for sport and commercial fisheries in the Les
Cheneaux Islands for l986 and for 7- and 8-inch minimum size
limits (equilibrium conditions).
Predicted catch
0-inch MSL 7-inch MSL - 8-inch MSL
Sport Commercial Sport Commercial Sport Commercial
Age fishery fishery fishery fishery fishery fishery
l O O O O O O
2 l2 O O O O O
3 iss O 91 - O O O
4 192 O 2O7 0 108 O
5 87 - 9 94 lC) 136 15
6 38 5 42 5 59 7
7 17 3 l8 3 26 4
8 7 2 7 l lC) 2
9 3 O 4 l 5 l
10+ 2 O 2 l 3 l
Total 496 l9 465 2l 347 30
Percent
of annual
total catch 96 4 96 4 92 - 8
57
reduced predicted catch for anglers by only about 3%. This
reduction in predicted catch would remain constant under the
7-inch MSL, but would increase under the 8-inch MSL to about
6%.
Fishing at a rate similar to that for lº&6, the
commercial fishery would account for 6% to 10% of predicted
total yield (Table ll). Competition from assessment netters
reduced predicted sportfishing yield for lº& 6 (no MSL) by
approximately 7%. Yields under 7- and 8-inch MSL's would be
reduced by about 7% and l C#, respectively. For this
simulation, Indian assessment nets would effectively negate
Small gains in predicted yield achieved through
implementation of 7- and 8-inch MSL's.
For the competing fisheries, effect of the commercial
fishery upon catch of age-5+ perch was similar to its effect
upon yield. Commercial fishermen would harvest .
approximately ll?, of the total catch in numbers of age-5+
perch under all three MSL's. Therefore, reduction in angler
harvest of age-5+ perch as a result of Chippewa assessment
netting was predicted to be about 9%.
58
Table ll. Predicted yield in kilograms of yellow perch (per l, 000
yearling recruits) for sport and commercial fisheries in the
Les Cheneaux Islands for lo&6 and for 7- and 8-inch minimum
size limits (equilibrium conditions).
Predicted yield (kg)
O-inch MSL 7-inch MSL 8-inch MSL
Sport Commercial Sport Commercial Sport Commercial
Age fishery fishery fishery fishery fishery fishery
l O O o O O O
2 l O O O O O
3 - is O lC) O O O
4 34 O 36 O l.9 O
5 23 3 25 3 36 4
6 lá 2 l5 2 22 3
7 8 l - 9 2 l2 2
8 4 l 5 l 7 l
9 2 O 2 O 4 l
10+ l O 2 O 3 O
Total l02 7 104 8 103 ll
Percent
of annual
total yield 94 6 93 7 90 10
59
DISCUSSION
Analysis of growth, age structure, and mortality
suggests that the yellow perch population has remained
relatively stable. Growth of yellow perch was significantly
greater in 1969–74. However, analysis of length-age data
for this period suggests that growth was at a peak then.
Growth in 1975-86 was variable, but stable. Age structure
for fall assessment catch varied little from l 969-86.
Mortality for yellow perch demonstrated no significant
trend. The combination of constant growth, age structure,
and mortality strongly indicates that the perch population
has remained relatively stable since sampling began in l969.
Length-age and CPUE suggest that growth in early
samples was exceptional, and may not represent average.
growth prior to sampling in l969. First, although size of
age-4 and older perch was greater in 1969-74, size of age-2
and age-3 perch was about average throughout this time
(Figures l () and ll). Also, yearly growth summed for perch
Was not greater during this period than in 1975–86 (Table
3). However, yearly growth was above average for several
consecutive years, which resulted in significantly greater
mean sizes for perch in 1969–74. Unlike 1969-74, years of
good growth during lS 75-86 were interspersed with years of
poor growth. Thus, overall size for 1973-76 year classes Of
yellow perch was about average to slightly below average.
60
Finally, high densities of age-3 and older perch, along
with poor growing conditions, probably resulted in the sharp
decline in size for 1975. Fall CPUE data suggested that a
strong year class of perch was produced in 1971 (Table 4).
Year classes produced in 1969–70 also were slightly larger
than average. As a result, greatest CPUEs for age-3 and
age-4 perch for l969-86 fall assessments were recorded in
l974 and l375, respectively. Large declines in growth for
perch age-3 and older coincided with these extraordinary
CPUEs, which implies that the decline in growth through 1975
was density related. -
Analysis of length-age and CPUE data made it apparent
that both density dependent and density independent factors
are important in determining growth rate for yellow perch in
the Les Cheneaux Islands. An inverse relationship between
density and growth in yellow perch has been demonstrated in
several studies (Beckman l950; Bardach l85l; Buck and Tho its
lS70; Schneider lS72, l 973, 1979, l983; Kempinger et
al. l.982). Over the past 60 years, density dependent growth
has been quite evident for perch in Saginaw Bay. El-Zarka
(1959) noted that growth for Saginaw Bay perch in 1949-53
had declined substantially below that reported by Hile and
Jobes (lº 4l) for lS29–30. He attributed this decline to a
sevenfold increase in density of perch over that period. In
l966, growth for Saginaw Bay perch once again improved in
response to a reduction in stock size from increased
commercial fishing (Eshenroder lº 77). Presently, excellent
6l
recruitment and an 8.5-inch MSL on perch for the commercial
fishery have once again increased stock size, and perch
growth has declined (Weber 1985).
Although density dependent factors appear important in
controlling growth, further analysis of fall length-age data
indicated that density independent factors may play a
greater role in determining growth rate of perch in the Les
Cheneaux Islands. Comparisons between yearly growth and
growth of year classes showed significantly greater
fluctuations for yearly growth (Table 3). These large
fluctuations in yearly growth were especially evident for
1975–86, while fluctuations in growth of year classes were
substantially less during that period. This would imply
that environmental conditions during the growing season had
a greater effect on growth than did size of a year class.
In addition, analysis of fall assessment CPUE demonstrated
insignificant variation between size of most year classes.
One explanation for significant variation in annual
growth for Les Cheneaux yellow perch concerns large
differences in water temperature during the growing season.
Water temperature in the Les Cheneaux Islands is strongly
dependent upon wind direction and water temperature in the
open lake. Strong offshore winds will push warm surface
water offshore. In addition, slow-moving currents flowing
through the Les Cheneaux Islands continually replenish bays
and channels with colder water from the open lake.
Consistent offshore winds and cold temperatures in the open
62
lake can reduce temperatures in the bays and channels
substantially below the temperature for optimal perch growth
of 24.7 °C (Hokanson 1977). Coble (1966) found that growth
in adult female yellow perch in Lake Huron was correlated to
water temperature at a depth of 6 meters. Le Cren (1958)
found that two-thirds of the variation in annual growth for
European perch, Perca fluviatilis, in Lake Windemere was
accounted for by surface water temperature during the
growing season. He believed that the primary effect of
temperature on growth was its direct effect on metabolism,
rather than its effect on the food SOUllr Ce ,
Growth rate for Les Cheneaux yellow perch is not
exceptional among Great Lakes perch populations, but similar
to rateS for slower-growing populations (Table l?).
Historical growth rates of Lake Erie perch (Jobes l952) and
of Saginaw Bay perch (Hile and Jobes lº ql) prior to l840
were significantly greater than growth rates for LeS
Cheneaux perch and other Great Lakes populations. Growth
rate of perch currently is declining in Lake Erie and
Saginaw Bay (Weber 1985; Hayward and Margraf l987). Growth
rates of yellow perch in Green Bay and northern Lake
Michigan (Hile and Jobes lºq2), areas of a latitude similar
to the Les Cheneaux Islands, were not substantially
different from growth rate for perch in the Les Cheneaux
Islands. Recent studies analyzing growth of yellow perch
around Ludington (Brazo et al. 1975) and in the southern
63
Table l2. Total lengths (mm) at the end of each year of life for yellow
perch in different locations throughout the Great Lakes.
Year (S) Age
Of Sample
Location Study Sex Size l 2 3 4 5 6 7 8
Les Cheneaux
Island, l969-86 M l, 559 --- l23 la 9 l70 lo3 227 249 272
Lake Huron* F l, 187 --- lá5 lS4 i88 221 254 290 312
Lake Erie.” l.927-37 M --- 91 l68 213 239 257 --- --- ---
F --- 94 l'70 218 249 2.72 --— — — — — — —
Saginaw Bay, l.929–30 M 203 76 lºS 196 236 269 292 325 —--
Lake Huron * F 600 76 lºš 206 2.44 272 292 307 ---
Saginaw Bay, l949-55 M 2,096 66 lo'7 l42 l70 los 216 236 ---
Lake Huron 4 F 1, 305 69 log lSO 191 224 256 282 –––
Saginaw Bay, 1983 M 671 76 lºS lo O l?5 l88 208 231 ---
Lake Huron * F 724 76 l47 l73 193 216 24 l 272 ---
Green Bay, 1932-38 M. l65 73 ll:7 lS4 l86 2.17 246 264 ---
Lake Michigan * F 587 72 ll 7 162 203 231 266 292 ---
Northern Lake
Michigan, 7 1937 M+F 509 72 ll 4 lS2 l8O 214 247 --- ---
(Beaver Island)
Saugatuck, l97l-79 M l, 604 85 lb 2 207 235 255 267 279 ---
Lake Michigan? F 2,084 85 lb 7 227 26.l. 288 310 326 ---
* Lengths of yellow perch at time of capture
Growth between time of
annulus formation was assumed negligible; and therefore,
for age t was assigned to age t-l for comparison with back-calculated
mean lengths from other studies.
length-at-age.
* Hile and Jobes (l'952).
* Hile and Jobes (lo 4l).
* El-Zarka (l.959).
* Weber (l'985).
were used to estimate mean
capture in fall and time of
mean length
• Hile and Jobes (1942); sexes combined for Northern Lake Michigan age
analysis.
7 Wells and Jorgenson (l.983).
64
basin of Lake Michigan (Wells and Jorgenson l983) show
substantially greater growth for perch in those areas.
Age structure of yellow perch in the Les Cheneaux
Islands has also undergone little change from 1969 to l986
(Table 4). Comparison of CPUE for all ages between 1969–74
and 1975-86, periods of substantially different growth,
showed no significant differences in CPUE. Furthermore,
annual total mortality has remained relatively constant. A
significant trend in mortality should produce an inverse
trend in growth in density affected populations. Trends in
growth would produce changes in age structure. Absence of
trends for CPUE, along with relatively constant growth and
mortality, strongly support that the perch population has
remained relatively stable. -
Year class strength of yellow perch has been relatively
constant when compared with other Great Lakes perch
populations. Variability in recruitment and consecutive
years of poor reproductive success have been associated with
perch populations in Saginaw Bay (Eshenroder lS77) and Lake
Erie (Hartman l973; Leach and Nepszy l876; Nepszy lS77; and
Spangler et al. 1977) which were subject to severe
biological stresses such as overexploitation or cultural
eutrophication. Low variability in recruitment and
continued production of successful year classes in the Les
Cheneaux Islands imply that high exploitation rate is not
severely impeding reproductive success.
65
Several problems inherent in the fall assessment data
created difficulties for growth and CPUE analyses. First,
yearly growth at each age was estimated from mean lengths,
which were calculated from very uneven sample sizes. Uneven
sample sizes, in addition to unequal variances, resulted in
estimates of pooled variance that were relatively large.
Another problem with the length-age data was absence of
young-of-the-year and yearling perch in fall assessment
catch. . Only nine yearling perch were caught from 1969–86.
All other perch netted during fall assessments had three
seasons growth and were partially recruited into the sport
fishery. Samples of younger perch would have provided
useful information for assessing recruitment, growth, and
impact of density on growth for unexploited ages.
CPUE from fall assessments was not an extremely
reliable estimator of yellow perch density. Assessments
were conducted over only a 2-day period, and net hours for
l,000 feet of experimental mesh gill net often totaled less
than a l ()0. Since nets were in the water for such a short
period of time, catch of perch was inconsistent.
Mortality rates for yellow perch in the Les Cheneaux
islands were similar to mortality rates estimated for perch
elsewhere in the Great Lakes and other large inland waters
(Table ls). In several cases, resultant total mortality
rate from the intensive sport fishery was similar to total
mortality rates for commercially fished populations. From
assessment gill net catch in the Les Cheneaux Islands, total
66
Table lº. Total annual mortality rates for yellow perch in different
locations throughout the Great Lakes and other large inland
lakes.
Year (s)
of Total annual
Location study mortality rate
Les Cheneaux Islands, l969-86 O. 55
Lake Huron
Saginaw Bay, 1960-74 0.61
Lake Huron*
Saginaw Bay, 1983-84 0.64
Lake Huron?
Saugatuck, - lg78 0. 56
Lake Michigan *
Benton Harbor, 1979 . 0.60
Lake Michigan?
Red Lakes, l94l-56 O. 57
Minnesota.” :
1 Estimated from catch data for Eshenroder (l'977).
*Keller et al. (l.987).
*Rybicki (1985).
“Heyerdahl and Smith (197l).
67
annual mortality of yellow perch was estimated at 55%.
Fishing mortality was directly estimated from tag returns at
42% and natural mortality was estimated at about lé+
annually. Estimates of total annual mortality for perch in
Saginaw Bay, made from commercial trap-net data for 1960–74
(Eshenroder lº 77) and l383–84 (Keller et al. 1987) were
similar to the estimate for Les Cheneaux perch. Rybicki
(1985) obtained similar estimates of total annual mortality
for Lake Michigan perch populations off Benton Harbor and
Saugatuck, Michigan. He found a significantly greater
mortality for females than males in both populations. Only
Lake Erie yellow perch had a total annual mortality rate
substantially greater than in the Les Cheneaux Islands
(Hartman et al. 1977). -
Estimates of mortality rates for yellow perch in the
commercially fished Red Lakes in north-central Minnesota
showed greatest similarity to those for perch in the Les
Cheneaux Islands (Table l3). For Red Lakes perch, Heyerdahl
and Smith (1971) estimated total mortality (55%) from
commercial gill-net data collected from 1943–55. Annual
natural mortality for ages not exploited by the commercial
fishery (22%) was directly estimated from experimental gill-
net data. Fishing mortality, estimated at 45% annually, was
only slightly greater than in the Les Cheneaux Islands, and
could almost entirely be attributed to commercial fishing.
Prior to tagging in 1985, exploitation of Les Cheneaux
yellow perch was believed to be relatively low (MDNR
68
interoffice communication from J. R. Ryckman and J. C.
Schneider to W. C. Latta, December 1, 1983). Returns from
the 1985–86 study indicated that exploitation was
surprisingly high, especially for a yellow perch sport
fishery on the Great Lakes. In fact, tag returns for l985
(28%) and l886 (17%) were exceptional, when compared with
returns for other perch tagging studies (Mraz l'95l; Clady
l977; Weber and Les l982; Rybicki l985).
Although some yellow perch traveled a relatively long
way from the tagging site (up to l8 km), the majority of
perch remained within a limited distance from bays where
they spawned. Most tagged perch returned by anglers were
taken close enough to their spawning bays that distributions
of perch tagged in Mackinac, Sheppard, and Flower bays
(Figures 6-8) appeared discrete. Similar results were
obtained for other tagging studies on the Great Lakes (Mraz
1951; Rybicki lº&5) and on Oneida Lake, New York (Clady
1977).
A large percentage (96%) of the 253 recaptures of lS85
tagged perch in l986 tagging study trap nets occurred in the
same bay where the fish had been tagged, which suggests
that perch may home to spawning bays. Results from other
tagging studies conducted on large inland lakes also support
the idea of yellow perch homing to spawning sites (Clady
1977); Weber and Les l982). Honing implies that separate
spawning stocks of yellow perch may exist within the Les
Cheneaux Islands.
69
The best estimate of population size for perch
>7 inches was between 400,000 and 900,000 fish. Despite
exceptional tag return rates, coefficient of variation for
population estimates ranged from slightly less than 8% to
over 42%. Estimates from mean of Petersen estimates were
relatively more precise than regression estimates. Still,
difficulties in meeting underlying assumptions resulted in
imprecise population estimates for this estimator, also.
Several problems with using mean of Petersen estimates
and regression estimators to determine population size for
this tagging study were evident. Although mean of Petersen
estimates were most precise, this estimator relies on the
assumption of random mixing. Values for variance estimated
from observed variance for each replicate of Ni and from a
variance estimator dependent upon assumptions underlying the
model were substantially different. Furthermore, observed
distribution for tag returns implied that the assumption of
random mixing was violated. Likewise, trends for the value
of mi /n; (Figure 13) suggested that relative catchability of
perch changed over time. Large variance for mi/ni
significantly increased size of confidence limits for
estimates made from both mean of Petersen estimates and
regression estimators. A decreasing trend for values of
mi/n} can indicate immigration into a population, which will
result in serious overestimation of population size
(Paloheimo l863).
70
0.030+

0.020+
0.
O
1
O
0.
O
O
O
1985 togged perch
º
%
%
%
// 7.
44 ºn 2
1986 togged perch
0.020+
%
0.01 O-F Z %
-Ho %, %
% Z Z %
0.000 —ſº 4. -4–144–21–K+
Q J F M A M J J A S
Month
Figure lº. Percent of yellow perch tagged in 1985 (upper)
or l886 (lower) in the total estimated catch
(mi/ni) for 1986 in the Les Cheneaux Islands.
71
For this tagging study, fluctuations in tag return rate
with respect to estimated catch (mi/ni) were probably a
result of factors other than varying catchability. A large
proportion of variability for mi /ni appeared attributable to
incomplete tag returns. Analysis of mi /n; provided evidence
that non-reporting of tags was a significant problem.
First, (mi/ni) for 1985 tagged perch in May–June 1985 was
about 3.5 times greater than for 1986 tags during the same
period in 1986. Decreased return rate in 1986 probably was
in part due to alteration of the reward system in 1986.
Further evidence of incomplete returns was the continual
decline in mi /n; for April–August 1986 (Figure l3).
Fishermen appeared to lose interest in the tagging study as
SUlſtliſle rº progressed. Finally, mi /n; for both 1985 and 1986
tags was disproportionately high in September 1986, which
was probably due to the prize lottery held near the end of
the month. Extremely high return rates in September suggest
that tags accumulated over summer were returned just prior
to the lottery. Since catch data were not available for
most of 1985, mi /n; could not be calculated for 1985.
Initial interest in the new study and a $1.00 reward for
each tag returned should have resulted in greater and more
consistent values for mi/ni in 1985.
Equilibrium yield for perch was predicted to only
increase slightly under a 7- to 8-inch MSL (Table 8). Under
these MSL's, fishermen would experience a moderate decline
in numbers of perch Creeled. However, numbers of larger
72
perch (age 5+) and overall stock density would increase
substantially (Table 9).
Lowest predicted yield under no MSL was only 3% less
than maximum predicted yield under the 8-inch MSL for
equilibrium conditions. Minimal variation among yields for
different MSL's has been found for many other fisheries
(Ricker 1975; Hartman et al. 1980). Ricker (1975) found
this condition for bluegills in Muskellunge Lake. Hartman
et al. (1980) predicted only minor changes for yield as
minimum age of exploitation increased for yellow perch in
the western basin of Lake Erie. Ricker (1975) concluded
that stability for yields under different MSL's has a number
of implications which can apply to the perch fishery in the
Les Cheneaux Islands. First, considerable leeway is allowed
for errors in data used to compute MSL's for maximum yield.
Second, it is not important to determine the exact optimum
MSL for maximum yield. Third, if a particular MSL was best
so as to obtain optimal recruitment, then present MSL may be
adjusted considerably to meet this requirement without
sacrificing yield. Fourth, if either individual size of
fish caught or CPUE are important considerations for the
fishery, either of these goals can be favored by a MSL
without sacrificing a considerable loss in yield.
Schneider (MDNR º Fisheries Division Interoffice
Communication to John schrouder, November 15, 1985) also
simulated the yellow perch fishery under 0-, 7-, and 8-inch
MSL's. In his simulation of the perch fishery, Schneider
73
used catch data from 1969–82, length frequency data from the
1979-81 creel census, and an exploitation estimate from
preliminary tag returns for the l985 tagging study.
Overall, changes in catch, yield, and population size under
the 7- and 8-inch MSL's were greater for his simulation of
the fishery (Table l4). Schneider's predicted decline in
equilibrium catch in numbers, and increase in yield and
catch of older perch for the 7-inch MSL were over double
that for the present study. For the two simulations,
difference between predicted catch and yield under the 8–
inch MSL was not as great, however, Schneider's fishery
simulation still predicted a significantly greater increase
for overall yield and catch of age-5+ perch.
For 7- and 8-inch MSL's, increased gains in yield and
catch of older perch for Schneider's fishery simulation over
the present simulation were largely attributable to
different length frequencies used in estimating F values for
ages protected by the MSL. Schneider used length
frequencies from the l879–81 sport catch, from which it was
estimated that over 45% and 80% of the catch were less than
7 and 8 inches, respectively. For length frequencies from
1986 sport catch, used in the present simulation, estimated
percentages of perch catch less than 7 inches (17%) and
8 inches (65%) were substantially smaller. Thus,
implementation of a 7- or 8-inch MSL in Schneider's fishery
simulation resulted in the protection of a much larger
proportion of the perch catch.
74
Table l4. Predicted percent change in catch in numbers and yield of
yellow perch under 7- and 8-inch MSL's (equilibrium
conditions) relative to no MSL for the present study and
Schneider's study.
Predicted percent change
7-inch MSL - 8-inch MSL
This . This -
Harvest study Schneider study Schneider
Number -6 -20 - –27 –30
Weight +l +7 +3 +ll
Number of age 5+
+7 +36 & +54 +93
75
Reanalysis of Schneider's simulation demonstrated that
the greater rate of instantaneous fishing mortality assigned
to fully recruited ages of perch in his study did not result
in increased yield and catch of older perch for MSL's. For
fully recruited ages of perch, F values used in Schneider's
fishery simulation and in the simulation for this study were
0.70 and 0.59, respectively. If the fishery is simulated
using these F values, while all other variables are held
constant, then net gain in yield under the 8-inch MSL
becomes slightly greater (3%) for the simulation in this
study. Actually, as fishing mortality increased for ages
not protected under MSL's, yield decreased slightly.
Accuracy of predictions for the fishery depends on
assumptions that growth, natural mortality, recruitment, and
Overall fishing intensity do not change significantly with
implementation of a MSL. Analyses of available data suggest
that these parameters, except possibly growth, should remain
relatively constant. Analysis of fall CPUEs indicated
relatively COnStant recruitment and total mortality.
Constant total mortality, which is predominantly a function
of fishing mortality, would imply that fishing intensity has
been relatively constant, despite annual variation in the
quality of the fishery.
Growth must also remain constant after addition of a
MSL for predictions in catch and yield to be accurate.
Since implementation of a MSL often results in a substantial
increase in stock density, this assumption implies that
76
growth of perch is density independent, or that population
size is substantially below carrying capacity. Although
growth of perch in the Les Cheneaux Islands appears to be
strongly influenced by density independent factors, effects
of density upon growth were apparent in the data and deserve
consideration. Increasing numbers of age-3 and age-4 perch
through addition of al MSL could further increase
fluctuations in growth from year to year. For example, the
collapse of growth in l375 implied that a combination of
exceptionally strong year Classes and poor growing
conditions can drastically reduce, annual growth. However,
strong year classes of perch did grow well other years.
Therefore, increasing stock density of perch through MSL's
may further slow growth auring years with poor growing
conditions, which would result in decreased overall growth
for perch.
Native American commercial fishermen presently have a
relatively small impact on the Les Cheneaux yellow perch
fishery. Commercial catch in 1985–86 represented only a
small fraction of total catch. Catch in numbers and yield
for the commercial fishery in 1986 comprised less than 5%
and l (); of respective totals for the fishery (Tables lo and
ll). Importance of commercial fishing with respect to the
sport fishery in the Les Cheneaux Islands is small, and is
similar to the situation presently found for the perch
fishery in Saginaw Bay (Keller et al. 1987). Commercial
fishing pressure in the Les Cheneaux Islands and Saginaw Bay
77
is only a fraction of that imposed by commercial perch
fishermen elsewhere in the Great Lakes (Hartman et al. 1980;
Keller et al. 1987; Rakoczy and Rogers l987).
Relative importance of the Native American commercial
fishery was predicted to increase with increasing MSL
(Tables lo and ll). As perch catch for sport fishermen
declined with increasing MSL, catch for commercial fishermen
increased, though commercial fishing effort remained
COnStant. The commercial fishery, which seldom exploited
yellow perch less than 8.0 inches total length, benefited
from restrictions on sport catch of smaller perch. Still,
commercial catch in numbers and yield under a 8-inch MSL
remained at less than 10% of the respective totals.
Native American commercial fishermen, fishing at their
l986 rate, would effectively nullify small gains in yield
realized under 7- and 8-inch MSL's. The presence of this
commercial fishery also reduces the predicted increase in
catch of larger perch (age 5+) by almost half. Predicted
declines in catch and yield were estimated under the
assumption that mortality of yellow perch smaller than
8 inches is negligible for 2-3/4-inch mesh . gill net.
Substantial mortality for perch less than 8 inches would
further reduce gains in catch and yield for the fishery
under MSL's. Simulation for the competing fisheries
demonstrates that, even at a relatively low intensity of
fishing, the commercial fishery must be given consideration
78
when determining a management strategy for perch in the Les
Cheneaux Islands.
Management Recommendations
A 7-inch MSL on perch was implemented in the Les
Cheneaux Islands as of l October 1986. Catch data collected
during the l887 creel census have not yet been analyzed, so
predictions for catch in numbers and yield under the 7-inch
MSL cannot be verified. However, the 7-inch MSL appears to
be fully accepted by local residents. Of 50 local resort
owners, businessmen, and anglers attending a presentation on
the l986 yellow perch study in March 1987, 49 voted in favor
of retaining the 7-inch MSL. Response to increasing the MSL
to 8 inches was not nearly as favorable. Therefore, the
proposed l October l987 implementation of an 8-inch MSL was
cancelled. The 7-inch MSL will continue until its impact on
the fishery can be evaluated.
Implementation of further regulations on the perch
fishery will depend on success of the 7-inch MSL, attitudes
of local businessmen and anglers, continued good growth of
perch, and the future of commercial fishing operations.
Results from questions on catch preference posed in both
creel and economic censuses clearly established that anglers
consider size of perch in their catch most important (Diana
et al. 1987). On the other hand, local businessmen
vehemently oppose the 8-inch MSL, insisting that its
implementation would substantially reduce business.
79
However, anglers have indicated in the l886 economic survey
that their participation in the fishery would not decrease
regardless of regulations (Diana et al. 1987).
Growth must continue to be monitored under the 7-inch
MSL. With the predicted increase in stock density for perch
under the 7-inch MSL, good growth must continue for this
regulation to produce significant increases in catch of
larger perch. Consistently poorer growth than that found
prior to l886 would imply a density dependent response to
increased stock size. Significantly slower growth would
more than likely nullify any increases achieved under this
MSL.
If gains under different regulations are to be more
accurately assessed, length frequency of the sport catch
should also be monitored. Predicted increases under
regulations rely heavily upon proportion of catch estimated
to be protected under that regulation. For MSL's this is
determined from length-frequency data. Analyses of length
frequency from l S79-81 and l 986 creel censuses suggest that
size of perch in the catch varies from year to year,
possibly depending on size of the year class being recruited
into the fishery. Therefore, average trends for length
frequency must be assessed if increases in yield under MSL's
are to be accurately predicted.
Finally, catch for the Native American commercial
fishery for perch must be monitored. Simulations
demonstrated that presence of a commercial fishery, even at
80
relatively low rates of exploitation, can substantially
reduce gains achieved under regulations. If rates of
fishing for the commercial fishery should increase, catch in
numbers and yield under a regulation should be reevaluated.
81
LITERATURE CITED
Bagenal, T. B., editor. 1978. Methods for assessment of
fish production in fresh waters. Blackwell
Scientific Publications. Oxford, England.
Bardach, J. E. 1951. Changes in the yellow perch
population of Lake Mendota, Wisconsin, between 1916
and 1948. Ecology 32: 719-728.
Beckman, W. C. l.950. Changes in growth rates of fishes
following reduction in population density by
winterkill. Transactions of the American Fisheries
Society 78: 82–90.
Beverton, R. J. H., and S. J. Holt. l.957. On the dynamics
Of exploited fish populations. Fishery
Investigations Series 2, Marine Fisheries, Great
Britain, Volume l8. Ministry Of Agriculture,
Fisheries, and Food. London, England.
Brazo, D. C., P. I. Tack, and C. R. Liston. l.975. Age,
growth and fecundity of yellow perch, Per Ca
flavescens, in Lake Michigan near Ludington,
Michigan. Transactions of the American Fisheries
Society l()4: 726–730.
Buck, J. D., and C. F. Tho its III. 1970. Dynamics of one-
species populations of fishes in ponds subjected to
cropping and additional stocking. Illinois Natural
History Survey Bullet in 30: 69-165.
Carlander, K. D. 1977. Handbook of freshwater fishery
biology. Volume 2. Iowa State University Press,
AmeS, Iowa.
Clady, M. D. l.977. Distribution and relative exploitation
of yellow perch tagged on spawning grounds in Oneida
Lake. New York Fish and Game Journal 24: l 68-l'77.
Clark, R. D., Jr., and B. Huang. 1985. Conflict between
sportfishing, commercial fishing, and rehabilitation
of lake trout in Lake Michigan. North American
Journal of Fisheries Management 5:26l-276.
Coble, D. W. l966. Dependence of total annual growth in
yellow perch on temperature. Journal Of the
Fisheries Research Board of Canada 23: 15-20.
82
Diana, J. S., C. A. Jones, D. O. Lucchesi, and J. C.
Schneider. 1987. Evaluation of the yellow perch
fishery and its importance to the local economy of
the Les Cheneaux Islands area. Final Report Grant
LRP-8C-7, Coastal Zone Management Program. Michigan
Department of Natural Resources, March 1987, Ann
Arbor. -
El-Zarka, S. E. —D. 1959. Fluctuations in the population
of yellow perch, Perca flavescens (Mitchill), in
Saginaw Bay, Lake Huron. U. S. Fish and Wildlife
Service, Fishery Bullet in 59: 365-415.
Eshenroder, R. L. l.977. Effects of intensified fishing,
species changes, and spring water temperatures on
yellow perch, Perca flavescens in Saginaw Bay.
Journal of the Fisheries Research Board of Canada
34 : 1830–1838 .
Everhart, W. H., and W. D. Youngs. 1981 . Principles of
fishery science. Cornell University Press, Ithaca,
New York.
Fox, D. J., and K. E. Guire. 1976. Documentation for
MIDAS, 3rd edition. Statistical Research
Laboratory, Ann Arbor, Michigan.
Hartman, W. L. 1973. Effects of exploitation,
environmental changes, and new species on the fish
habitats and resources of Lake Erie. Great Lakes
Fishery Commission, Technical. Report No. 22, Ann
Arbor, Michigan.
Hartman, W. L., S. J. Nepszy, and R. L. Scholl. l 980.
Minimum size limits for yellow perch (Perca
flavescens) in western Lake Erie. Great Lakes
Fishery Commission, Technical Report No. 39, Ann
Arbor, Michigan.
Hayward, R. S., and F. J. Margraf. 1987. Eutrophication
effects on prey size and food available to yellow
perch in Lake Erie. Transactions of the American
Fisheries Society llé: 210-223.
Heyerdahl, E. G., and L. L. Smith, Jr. l.97 l. Annual catch
of yellow perch from Red Lakes, Minnesota, in
relation to growth rate and fishing effort.
University of Minnesota Agricultural Experiment
Station, Technical Bullet in 285.
Hile, R., and F. W. Jobes. l94l. Age, growth, , and
reproduction of the yellow perch, Perca flavescens
(Mitchill), of Saginaw Bay. Transactions of the
American Fisheries Society 70:l 02-122.
83
Hile, R., and F. W. Jobes. 1942. Age and growth of the
yellow perch, Perca flavescens (Mitchill), in the
Wisconsin waters of Green Bay and northern Lake
Michigan. Papers of the Michigan Academy of
Science, Arts, and Letters 27:24l-266.
Hokanson, K. E. F. l S77. Temperature requirements of some
percids and adaptations to the seasonal temperature
cycle. Journal of the Fisheries Research Board of
Canada 34: 1524-1550.
Jaworski, E., and C. N. Raphael. 1978. Fish, wildlife, and
recreational values of Michigan's coastal wetlands.
Great Lakes Shorelands Section, Division of Land
Resources Programs, Michigan Department of Natural
ReSOurces.
Jobes, F. W. 1952. Age, growth, and production of yellow
perch in Lake Erie. U. S. Fish and Wildlife
Service, Fishery Bullet in 52: 205–266.
Keller, M., J. C. Schneider, L. E. Mrozinski, R. C. Haas,
and J. R. Weber. l 987. History, status, and
management of fishes in Saginaw Bay, Lake Huron,
1891 – 1986. Michigan Department Of Natural
Resources, Fisheries Technical Report No. 87–2, Ann
Arbor.
Kempinger, J. J., A. M. Forbes, S. L. Serns, and H. E. Snow.
l 982. Population development of re-introduced
Smallmouth bass and yellow perch in Nebish Lake.
Wiscons in Department of Natural Resources, Research
Report ll.9, Madison.
Leach, J. H., M. G. Johnson, J. R. M. Kelso, J. Hartman,
W. Numann, and B. Entz. 1977. Responses of percid
fishes and their habitats to eutrophication.
Journal of the Fisheries Research Board of Canada
34 : 1964–1971 .
Le Cren, E. D. 1958. Observations on the growth of perch
(Perca fluviatilis L.) over twenty-two years with
special reference to the effects of temperature and
changes in population density. Journal of Animal
Ecology 27: 287-334. ‘.
Mraz, D. 1952. Movements of yellow perch marked in
southern Green Bay, Lake Michigan, in l950.
Transactions of the American Fisheries Society
81: 150-l 6l.
Nepszy, S. J. 1977. Changes in percid populations and
species interactions in Lake Erie. Journal of the
Fisheries Research Board of Canada 34: l861–1868.
84
Neter, J. , W. Wasserman, and M. A. Kutner. l985. Applied
linear statistical models. Richard D. Irwin Inc.,
Homewood, Illinois.
Paloheimo, J. E. 1963. Estimation of catchabilities and
population sizes of lobsters. Journal of the
Fisheries Research Board of Canada 20: 59-88.
Payer, R. D., D. L. Pereira, M. L. Larson, J. A. Younk,
R. V. Frie, D. H. Schupp, and T. C. Osborn. 1987.
Status and simulation model of Lake of the Woods,
Minnesota, walleye fishery. Minnesota Department of
Natural Resources, Division of Fish and Wildlife,
Investigational Report No. 389.
Remington, R. D., and M. A. Schork. l.285. Statistics with
applications to the biological and health sciences.
2nd edition, Prentice-Hall Incorporated, Englewood
Cliffs, New Jersey.
Rakoczy, G. P., and R. D. Rogers. 1987. Sportfishing catch
and effort from the Michigan waters of lakes
Michigan, Huron, and Erie, and their important
tributary streams, April 1, 1986-March 31, 1987.
Michigan Department of Natural Resources, Fisheries
Technical Report No. 87-6a, Ann Arbor.
Ricker, W. E. 1975. Computation and interpretation of
biological statistics Of fish populations.
Fisheries Research Board of Canada Bullet in l91,
Ottawa, Canada. ſº
Rybicki, R. W. 1985. Feasibility of trap nets for
harvesting and observations on distribution, growth,
and survival of yellow perch in southern Lake
Michigan. Michigan Department of Natural Resources,
Fisheries Research Report No. 1930, Ann Arbor.
Ryckman, J. R. l.98l. Creel census methods in general.
Appendix VI-A-9 in Manual of Fisheries Survey
Methods, J. W. Merna et al. Michigan Department of
Natural Resources, Fisheries Management Report
No. 9, Ann Arbor.
Ryckman, J. R., and R. N. Lockwood. 1985. On-site creel
surveys in Michigan, l 975–1982. Michigan Department
of Natural Resources, Fisheries Research Report
No. 1922, Ann Arbor. -
Schneider, J. C. l.972. Dynamics of yellow perch in single-
species lakes. Michigan Department of Natural
Resources, Fisheries Research Report No. 1791, Ann
Arbor.
85
, Schneider, J. C. 1973. Density dependent growth and
mortality of yellow perch in ponds. Michigan
Department of Natural Resources, Fisheries Research
Report No. 1795, Ann Arbor.
Schneider, J. C. l979. Survival, growth, and vulnerability
to angling of walleyes stocked as fingerlings in a
small lake with yellow perch and minnows. Michigan
Department of Natural Resources, Fisheries Research
Report No. 1875, Ann Arbor.
Schneider, J. C. 1983. Experimental walleye-perch
management in a small lake. Michigan Department of
Natural Resources, Fisheries Research Report
No. 1905, Ann Arbor.
Spangler, G. R. , N. R. Payne, J., E. Thorpe, J. M. Byrne,
H. A. Regier, and W. J. Christie. l.977. Responses
of percids to exploitation. Journal of the
Fisheries Research Board of Canada 34: l983-1988.
Stebo, W. T. l.972. The effects of dart tags on yellow
perch. Transactions of the American Fisheries
Society lol: 365-366.
Tyler, A. V., and V. F. Gallucci. 1980. Dynamics of fished
stocks. Chapter 5. Pages lll-l 48 in R. T. Lackey
and L. A. Nielson, editors. Fisheries Management,
Blackwell Scient if ic Publications, Boston,
Massachusetts.
Walford, L. A. 1946. A new graphic method of describing
growth of animals. Biological Bullet in 90:14l-l47.
Weber, J. J., and B. L. Les. 1982. Spawning and early life
history of yellow perch in the Lake Winnebago
system. Wiscons in Department of Natural Resources
Technical Bullet in No. l 30, Madison. *
Weber, J. R. l.985. Calculation of growth and mortality of
yellow perch, l075 to l884, in Saginaw Bay. Pages
172-195 in Michigan Dingell–Johnson Annual Report
l985, Study 434, Ann Arbor.
Weber, J. R. 1986. Comparison of sport and commercial
harvest of yellow perch in Saginaw Bay. Pages l85–
205 in Michigan Dingell–Johnson Annual Report l'986,
Study 434, Ann Arbor.
Wells, L. 1977. Changes in yellow perch Perca flavescens
populations of Lake Michigan, l954-75. Journal of
the Fisheries Research Board of Canada 34: l82l-l329.
86
wells, L., and S. C. Jorgenson. 1983. Population biology
of yellow perch in southern Lake Michigan. U. S.
Fish and Wildlife Service Technical Paper 109.
*T*----~-...--__ --~~
UNIVERSITY OF M
||||III.
3 9015 01363"536
III.
<■
º
№
|×
ºſºg
ſå
ſº º
ºſº



x, , ， º º
* :: º ;
-§.
¿??¿
№-*g-
· · · · * * · · · · · · · ·%¿
g gaeae， ſººvae→---
wae-*** þð
ſ. » （， ，,s，{-∞*、、、）, **
· · · · · · ·，·，≤）
Rusº • • • • • •*... ».
* （ ）-ºrº， º， º * * • • •
！ 3***， ** *****« », ºs x， ~~~……… ）
**********æ， , ，'$3&-&-ºpºs. , ！
• • • • • • • • • .
* * * * * *， º*， e-, ſae
！=）
• ！ ！ -， ºrº º g. .
; * *****， *） &&t，
… *
* *= * * · * * · · · ·