'é’HESJs

r 3
‘ _\

!
a-«A’Lr aka/iizy 1
Xfiémdcuca 5:3 I:

..= .l ‘: “1:3: ‘ WNW

’ I.
, 217.: . ' ',:~*.;av.ma-.W W

  

This is to certify that the
dissertation entitled
Saginaw Bay Suckers: Their Dynamics and
Potential for Increased Utilization

presented by

Douglas Wayne Kononen

has been accepted towards fulfillment
of the requirements for

Ph. D. degree in Department Of

Fisheries and Wildlife

Major professor

Date December, 1981

MS U is an Affirmative Action/Eq ual Opportunity Institution 0-12771

 

 

 

MSU
LIBRARIES
-_

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

 

SAGINAW BAY SUCKERS: THEIR DYNAMICS AND
POTENTIAL FOR INCREASED UTILIZATION

By

Douglas Wayne Kononen

A DISSERTATION

Submitted to

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

DOCTOR OF PHILOSOPHY

Department of Fisheries and Wildlife

1981

 

 

 

 

ABSTRACT
SACINAw BAY SUCKERS: THEIR DYNAMICS AND
POTENTIAL FOR INCREASED UTILIZATION
By

Douglas Wayne Kononen

Yield potential estimates from underutilized Great
Lakes fish stocks are lacking for most species. This
investigation was undertaken to determine potential limits

on the commercial exploitation of white suckers (Catostomus

 

commersoni) in Saginaw Bay, Lake Huron. Principal data

 

sources included historical fishery catch and effort data
along with biological data on the size, sex and age
composition of the commercial catch. Fish of different
sizes and sexes were sampled for analyses of PCB's and DDT.
Commercial fishery cost and revenue data were incorporated
into a logistic surplus production model function and used
in a rudimentary bioeconomic analysis of the fishery.
Approximately 1,000,000 pounds of white suckers were
harvested annually from Saginaw Bay during the first half of
the 20th century. Current harvest levels range from 100,000
to 150,000 pounds per year. Catch-effort analyses indicate
a general pattern of decreased exploitation and increased
abundance of Saginaw Bay suckers since the early 1950's.

Cautious interpretation of catch-effort data indicates that

 

 

Douglas Wayne Kononen

landings could be increased by as much as 500% without
danger of stock overexploitation.

Surplus production model estimates Of mean exploitable
population biomass approximate 4,000,000 pounds. Bio-
economic analysis indicates that an ex-vessel price increase
of $0.05/1b would generate a short term economic rent of
about $11,000 to the commercial fishery. Contaminant
analySes indicate that the mean levels of PCB's (0.05 pmn)
and EDDT (0.015 ppm) in Saginaw Bay white sucker fillets are
well below the FDA'S 5 ppm tolerance limit for these
compounds. Adult white suckers experience low annual growth
rates (from 1 to 2 cm in total length), high annual survival
rates, £3 (from 0.59 to 0.72) and long life spans (up to 19+
years old). Females grow faster and larger and, on the
average, live longer than males.

The Beverton-Holt dynamic pool model, incorporating a
linear age-weight relationship, was used to estimate the
equilibrium yield per recruit at different rates of fishing.
Optimal utilization of this fishery, in terms of maximizing
yield per recruit, can only occur with a six to ten-fold
increase in the rate of fishing. The current absence of
profitable markets for suckers render immediate increases in

the rate of exploitation unlikely.

 

ACKNOWLEDGMENTS

This endeavor was made possible through the efforts of
several individuals and organizations. I wish to express
special thanks to my wife, Laureen, for her understanding
advice, moral support and perserverance throughout the
tenure of this project. Dr. Niles R. Kevern suggested this
problem and provided direction and logistical support. The
BayPort Fish Company, BayPort, Michigan, furnished
invaluable commercial fishery data and commercial catch
samples. The Michigan Department of Natural Resources
supplied historical fishery data and a much needed microtome
for fish aging studies. Cynthia Merrill, from Michigan
State University's Department of Food Science, conducted the
pesticide analyses. Additional thanks go to the members of
my dissertation committee: Drs. Darrell L. King; Milton H.

Steinmueller; Daniel R. Talhelm and William W. Taylor.

ii

 

TABLE OF CONTENTS

Page
List of Tables. . . . . . . . . . . . . . . . . . . v
List of Figures . . . . . . . . . . . . . . . . . . vii
I. Introduction
A. Background. . . . . . . . . . 1
B. Purpose of this InveStigation . . . . . . 8
C. Methods . . . . . . . . . . . . . . . . . 10
II. Saginaw Bay
A. The Physical Setting. . . . .. . . . . . 13
B. Limnology . . . . . . . . . . . . . . . 15
C. The Fishery . . . . . . . . . . . . . . . 17
III. Catch and Effort Data Analysis
A. Introduction. . . . . 21
B. The Basic Theory of Catch and Effort
Data Analyses . . . . . . . . . . . . . . 22
C. The Leslie Model. . . . . . 27
D. Sources of Commercial Fishery Catch
and Effort Data . . . . . . . . . 29
E. The Saginaw Bay Sucker Fishery. . . . . . 32
1. Background . . . 32
2. The Sucker Fishery, I929 to 1956. . . 35
3. The Sucker Fishery, 1960 to 1979.. . 51
IV. Fishery Bioeconomics
A Introduction. . . . . 67
B. The Logistic Surplus Production Model . . 69
C Application of the Logistic Surplus
Production Model. . . . . 75
D. Saginaw Bay Sucker Fishery Economics. . . 82
E Discussion. . . . . . . . . . . . . . . . 91
V. PCB's and DDT in Saginaw Bay Suckers
A. Introduction. . . . . . . . . . . . . . 98
B. Sampling Procedures . . . . . . . . . . . 100

iii

C. Analytical Procedures. . . . . . . . . . . 100
D. Results . . . . . . . . . . . . . . . . . 102

VI. Saginaw Bay Sucker Population Dynamics

A. Introduction. . . . 105
B. A Brief Review of White Sucker Biology. . 106
C. Saginaw Bay White Sucker Characteristics. 108

1. Fall 1979 . . . . . . . . . . . . . . 108

2. Spring 1980 . . . . . . . . . . . . . 115
3. Summer 1980 . . . . . . . . . . . . . 123
A. Fall 1980 . . . . . . . . . . . . . . 126
5. Spring 1981 . . . . . 133
6. Composite Length- Frequency Distributions 139
7. White Sucker Fecundity. . . . . . . . 139
8. Miscellaneous Observations. . . . . . 142
9. Adult White Sucker Survival . . . . . 1A3
D. Estimation Of Yield From a Given
Recruitment . . . . . . . . 1A8

E. Estimating Mortality Rates of Pre-recruits 155
VII. Summary and Recommendations

A. Recapitulation. . . . . . . . . . . . . . . 16A

B. Recommendations . . . . . . . . . . . . . . 168

C. Critique. . . . . . . . . . . . . . . . . . 170

Appendix. . . . . . . . . . . . . . . . . . . . . . . ix

References . . . . . . . . . . . . . . . . . . . . . xxiv

iv

Table
11-1
III-1

III-2

III-3

III-fl

III-5

III-6

III-7

III-8

III-9

IV-1
IV-2
IV-3
V-1
VI-l
VI—2

LIST OF TABLES

Saginaw Bay Physicochemical Data.

Saginaw Bay Commercial Sucker
Production (1929-1956).

Saginaw Bay Sucker Fishery Summary
Statistics (1929-1956).

Catchability and Mean Biomass Estimates
(1929-1945).

Catchability and Mean Biomass Estimates
(19u6-1956)0 o o o o o o o o

Saginaw Bay Commercial Sucker Production
(1960-1979).

Saginaw Bay Sucker Fishery Summary
Statistics (1960-1979)

Saginaw Bay Monthly Catch and Effort
Data (1960-1967). . . . . . .

Catchability and Mean Biomass Estimates
(1960-1967).

Saginaw Bay Monthly Catch and Effort
Data (1968-1979) . . . . . .

Surplus Production Model Results

Cost and Revenue Data.

Average and Marginal Cost Data

PCB's and ZDDT in Saginaw Bay Suckers.
Fall 1979 Summary Statistics

Spring 1980 Summary Statistics

Page
18

39

HO

47

49

52

53

56

58

60
78
85
87
101
109

117

 

VI-3
VI-4
VI-S

VI-6

Summer 1980 Summary Statistics.
Fall 1980 Summary Statistics.
Spring 1981 Summary Statistics.

Female Age-Specific Fecundity and
Survival Rates.

124
128
134

159

Figure
II-l
III-1

III-2

III-3

III-4

III-5

III-6

IV-4

VI-l

VI-2
VI-3
VI-4

LIST OF FIGURES

Saginaw Bay.

Lake Huron Statistical Catch
Districts. . . . . . . . . .

Participation in the Lake Huron
Commercial Fishery (U.S. Waters,

Saginaw Bay Commercial Sucker
Production (1929-1956).

Saginaw Bay Sucker Abundance
(1929-1956).

Saginaw Bay Commercial Sucker
Production (1960-1979). . .

Saginaw Bay Sucker Abundance
(1960-1979). 0 o

Logistic Surplus Production
Model (1930-1955). . . .

Logistic Surplus Production
Model (1968-1978). . . .

Average and Marginal Cost of the
Saginaw Bay Commercial Sucker
Fishery, 1979. . . . . . .

Cost and Revenue in Terms of Effort.

Fall 1979 Length- Frequency
Distributions. . . . .

Fall 1979 Relative Condition Factors
Fall 1979 Length-Age Relationships

Spring 1980 Length- Frequency
Distributions. .

vii

 

Page
14

33

37

41

42

54

55

81

83

88

92

110
113

116

119

 

VI-5 Spring 1980 Relative Condition

Factors. . . . . . . . . . . . . . . . . . . 121
VI-6 Spring 1980 Length-Age Relationships. . . . 122
VI-7 Summer 1980 Length- Frequency

Distributions. . . . . . . . . . . . . . . 125
VI-8 Summer 1980 Relative Condition Factors . . . 127
VI-9 Fall 1980 Length-Frequency Distributions . . 129
VI-lO Fall 1980 Relative Condition Factors . . . . 131

VI-ll Fall 1980 Length-Age Relationships. . . . . . 132

VI-12 Spring 1981 Length-Frequency
Distributions . . . . . . . . . . . . 136

VI-l3 Spring 1981 Relative Condition Factors. . . . 137

VI-14 Spring 1981 Female Length-Age
Relationship. . . . . . . . . . . . . . . 138

VI-15 Composite Length-Frequency
Distributions. . . . . . . . . . . . . . . . . 140

VI-16 Combined Fall 1979 and Fall 1980
Catch Curve. . . . . . . . . . . . . . . . 146

VI-l7 Combined Spring 1980 and Spring
1981 Catch Curve . . . . . . . . . . . . . 147

VI-l8 Equilibrium Yield Per Recruit at
Different Rates of Fishing . . . . . . . . . . 154

viii

CHAPTER I

Introduction

A. Background

 

The Laurentian Great Lakes contain abundant populations
of such commercially underutilized fish as carp (Cyprinus

carpio), suckers (Catostomus and Moxostoma spp.), alewives

 

(Alosa pseudoharengus) and rainbow smelt (Osmerus mordax).

 

 

The principal reason that these and other so-called rough
fish species are underutilized is the absence of dependable
and profitable markets for rough fish. Changes in the
species composition of the Great Lakes fisheries have had a
dramatic negative impact on the size and scope Of commercial
fishing Operations. The survival of the remaining
commercial fishery hinges, in part, on the development of
abundant stocks of heretofore underutilized fish. Estimates
of the yield potential from latent or underutilized Great
Lake fish stocks are lacking for most species. This
investigation was undertaken to determine potential limits
on the commercial exploitation of suckers in Saginaw Bay,
Lake Huron.

The species composition of the Great Lakes fisheries
has changed markedly over the past forty to fifty years.

Formerly plentiful stocks of lake trout (Salvelinus

 

namaycush), lake whitefish (Coregonus clupeaformis), lake

 

herring (Coregonus artedii), walleye (Stizostedion vitreum),

 

 

lake sturgeon (Acipenser fulvescens) and burbot (Lota lota)

 

 

have been supplanted by abundant stocks of alewives, rainbow
smelt, carp and other so-called rough fish species.
Commercial overfishing” the introduction and proliferation

of such exotic species as the sea lamprey (Petromyzon

 

marinus) and alewife, environmental enrichment and pollution
have combined to help bring about this transformation in
species composition. Documentation and interpretation Of
the changes which have occurred in the Great Lakes fisheries
during the last four tolfive decades are provided by: Hile
and Buettner (1959); Buettner (1968); Smith (1968); Berst
and Spangler (1973); Smith (1973); Regier (1973) and
Christie (1974).

The virtual elimination of commercially exploitable
stocks of lake trout, lake whitefish and lake herring in
many areas of the Great lakes has contributed to a
substantial reduction in the size and scope of Michigan's
Great Lakes commercial fishery over the past twenty to
thirty years. Scott (1974) reports a decrease in the number
of Michigan commercial fishing licensees from approximately
900 in 1963 to about 660 in 1967. After 1970, according to
Borgeson (1972), state-imposed restrictions on gear and
entry reduced the total number of Michigan Great Lakes
commercial fishermen by an additional 50%. At present,

Michigan licenses from 100 to 150 commercial fishermen,

exclusive of non-licensed Native American fishermen (Asa
Wright, personal communication).

The Michigan Department of Natural Resource's
management objectives for Michigan's Great Lakes commercial.
fishery include restricting new entries and encouraging the
exit of marginal fishing operations. The efficient and
equitable allocation of Great Lakes fishery resources
between sport and commercial fishermen is a complex social,
political, economic and biological problem with no immediate
satisfactory solution. The principal thrust Of Great Lakes
fisheries management is currently oriented toward a large,
fairly recently established sport fishery for introduced
salmonids. Suspicions, real or otherwise, that the
commercial fishery poses a threat to the survival Of highly
prized artificially maintained sport fish stocks have
resulted in widespread misunderstandings and feelings of
animosity between sport and commercial fishermen. State
licensed commercial fishermen are prohibited from using gill
nets for capturing fish and must return all designated sport
fish (e.g. lake trout, Pacific salmon, walleye pike) and
'undersized commercial species (e.g. lake whitefish, yellow

perch, Perca flavescens) to the water in an unharmed

 

condition. The entrapment gear used by non-Indian
commercial operators allows for selective exploitation1of
Great Lakes fishery resources without causing significant
damage to the sport fishery. At present, regulations

prohibiting the use of gill nets do not extend to non-

A
licensed Native American commercial fishermen Operating in
Michigan's Great Lakes waters. The unregulated use of gill
nets poses a potentially severe threat to the sport fishery,
exacerbating the problems of efficient and equitable
resource management.

Annual fish production within the Great Lakes
represents a large potential source of high quality protein
for consumers in the United States and Canada. Only a small
proportion of this production finds its way into the
marketplace due to the reduced size of current commercial
fishery operations and the relative paucity of profitable
markets for underutilized rough fish species. The existence
of abundant potentially exploitable populations of
underutilized fish (e.g. carp, alewives, suckers and smelt)
in Michigan's Great Lakes waters is documented by Hile and
Buettner (1959), Borgeson (1972), Galloway and Kevern (1976)
and Rybicki (1979). With the exception of such local
enterprises as the lake whitefish fishery of Northern Lake
Michigan, the traditional Great Lakes commercial fishery,
which had been based largely upon lake trout, lake whitefish
and lake herring, has all but disappeared. The survival of
the remaining commercial fishery hinges upon the
reestablishment of a viable coregonid fishery and the
enhanced development of fisheries for underutilized species.

According to Frick (1965), if the Great Lakes
<zommercial landings are to be maintained in volume and

value, then product development, quality improvement and

market promotion are necessary to increase the demand for
those species available in abundance. Great Lakes
commercial fishermen have been reluctant to change the focus
of their operations to underutilized, lesser valued species
due to the absence of remunerative markets for these fish.
The lack of profitable markets for underutilized fish may,
in part, be the fault of the commercial fishing industry
itself, since, as Anderson (1973) has pointed out, fishing
industries generally lag behind competitors in other food
processing industries in developing markets for their
products. In recent years it has become evident that
assistance from outside the Great Lakes commercial fishing
industry is necessary to provide both the incentive and the
technological support for increased development of
underutilized fishery resources. In response to
developmental assistance requests from the commercial
fishing industry, a Sea Grant-supported program conducted by
Michigan State University's Departments of Fisheries and
Wildlife and Food Science has been designed to stimulate the
development of underutilized Great Lakes fish resources,
primarily via production of a frozen minced fish product
(Michigan State University, December 1975 and March 1978).
The development of viable markets for underutilized
fish and fish products is, perhaps, the single biggest
obstacle confronting potential developers of underutilized
Great Lakes fish stocks. According to Anderson (1973),

before consumers become adventurous and try unfamiliar

 

6

species and products, they must be taught to accept fish as
a prime food. In 1978, according to the U.S. Department of
Agriculture's Economic, Statistical and Cooperative Service,
Americans consumed an average of 149.7 pounds of red meat,
57.1 pounds of chicken and turkey and 13.4 pounds of fish
per person. This low per capita consumption of fish
relative to other meats indicates that Americans, as a
whole, may not regard fish as a primary dietary item. If
this observation holds, then the initial introduction of
underutilized species into the marketplace may be met with a
less than enthusiastic response.

Estimates of the net all or none values (social
surplus) of Great Lakes sport and commercial fisheries have
been made by Talhelm, et a1 (1979). These estimates
indicate that the sport fishery's net all or none value
(i.e. approximately $525 million) greatly exceeds the
correspondent value of the commercial fishery (i.e.
approximately $12 million). These estimates are in terms of
constant 1979 dollars. Establishment of new, profitable
markets for underutilized fish and fish products could
noticeably improve the economic importance of the Great
Lakes commercial fishery relative to the Great Lakes sport
fishery. Increased exploitation of underutilized Great
Lakes fish should not interfere with overall fishery
management goals by the State of Michigan and other
governmental agencies. It is possible that the controlled

harvest and concomitant reduction of large populations of

7

rough fish may favor or enhance the growth and development
of some highly valued sport fish stocks.

Large numbers of suckers, predominately white suckers,
(Catostomus commersoni) are captured annually by Saginaw Bay
commercial fishermen in entrapment gear set for more

commercially valuable species (e.g. channel catfish,

 

Ictalurus punctatus, yellow perch, and lake whitefish).
Historical catch data summarized by Hile and Buettner (1959)
show annual commercial landings Of Saginaw Bay suckers
averaged approximately 1,000,000 pounds during the late
1800's and up through the 1930's. During this period,
fishermen received up to $0.30 per pound for suckers at
dockside. This relatively high ex-vessel price for suckers
was attributed to the existence of a large, stable Eastern
United States market for these fish. After the 1930's, this
market disappeared, resulting in markedly decreased per
pound revenues to the individual fisherman (i.e. from $0.30
to $0.05 per pound). A large proportion of the suckers sent
to the East Coast had been processed into gefilte fish.
Changes in traditional ethnic lifestyles or increased
availability Of acceptable substitutes helped shift the
demand for large quantities of Saginaw Bay suckers. From
1968 to 1979, annual landings of Saginaw Bay suckers
averaged approximately 120,000 pounds. Conversations with
Saginaw Bay fishermen reveal that most (i.e. from 50% to
75%) of the current sucker catch is returned to the water

due to the absence of a profitable sucker market. This

information, combined with historical sucker fishery catch
statistics, reveals a great potential for increased

utilization of Saginaw Bay suckers.

B. Purpose of this Investigation

 

This paper presents results of a two year investigation
designed to evaluate the population dynamics of Saginaw Bay
white sucker stocks. Samples of the commercial catch
revealed that white suckers Comprise more than 90% of the
‘botal sucker catch in terms of numbers and weight from the
waters of eastern Saginaw Bay. Other sucker species present

in Saginaw Bay are the longnose sucker (Catostomus

 

catostomus) and the redhorse sucker (Moxostoma spp.).

 

 

Information concerning sucker growth and mortality is needed
to help estimate the potential for increased utilization of
these fish. This study was conducted in conjunction with
lWichigan State University‘s Sea Grant-sponsored efforts to
stimulate the development of underutilized Great Lakes fish
via production of a frozen, mechanically debOned minced fish
product.

Aside from commercial fishery catch and effort data,
little information is available concerning the population
dynamics and abundance of Saginaw Bay suckers. In order to
evaluate the potential impacts of increased harvest levels
on population stability, baseline data is needed regarding:
(1) the size, age and sex composition; (2) age and size
specific growth and mortality rates and (3) the relative

magnitude of existing stocks. Prior to fishery development,

9

levels of environmental contaminants in underutilized fish
should be determined to insure compliance with applicable
governmental health regulations.

The principal question addressed by this investigation
is, what are the potential limits on Saginaw Bay sucker
exploitation? Crutchfield (1973) cites an urgent need for
reasonably accurate yield estimates from latent or
underutilized fisheries prior to instigation of multiple
objective management policies. The delineation of a
reliable production function, relating measurable inputs to
corresponding outputs, is one of the most important aspects
of any resource development evaluation. Without an accurate
prOductiAHT function, application of biological and economic
management techniques is much more difficult. Much of the
present study involves attempts to relate fishing effort to
catch. Additional questions addressed by this study are:
(1) what are the magnitudes of the likely benefits to the
commercial fishery of increased sucker utilization and (2)
what are the levels of the common environmental
contaminants, DDT and PCB'S, in the usable flesh of Saginaw
Bay'suckers?

In addition to possible biological constraints, other
impediments to the enhanced development of the Saginaw Bay
sucker fishery are: (1) the short (less than two day) shelf
life of whole iced ffixfli; (2) the large number of small y-
bones found in the edible flesh (fillets) of suckers and (3)

the generally poor reputation enjoyed by suckers and other

10
rough fish species among the consuming public.
Establishment of a dependable and profitable market for
suckers would increase the annual landings of these fish and
help strengthen the local commercial fishery. A greatly
expanded and profitable market for suckers cannot be
established unless the existing impediments to sucker
fishery development are removed. The'short shelf life of
iced whole suckers precludes their transport to distant
markets. The geographic scope of potential markets could be
broadened by adoption of suitable processing techniques
(e.g. smoking or mincing). Remediation of the y-bone
nuisance could be achieved by a mincing process (Michigan
State University, December 1975 and March 1978). A minced
product could be more readily packaged, stored, transported
and marketed than whole iced or frozen fish. With proper
marketing methods, negative consumer attitudes regarding

rough fish may be altered.

C. Methods

An ideal fish population dynamics investigation shOuld
include a comprehensive mark and recapture study to
determine fish movements, distribution and estimates of
populatdxni density. In addition, timely and accurate catch
and effort data for each commercially harvested species must
be available. Funding constraints fem this project ruled
out the possibility of any meaningful mark and recapture
program. Fortunately, a long term commercial fishery data

base is available for each principal species landed by

11

Michigan commercial fishermen (Hile, 1962). These catch and
effort data, if interpreted carefully, allow an investigator
to calculate estimates of relative population abundance.
The analysis and interpretation of Saginaw Bay sucker
fishery catch and effort data is detailed in Chapter III.

Catch and effort data, along with basic cost and
revenue information furnished by individual operators, were
used to develop a rudimentary bioeconomic model of the
Saginaw Bay commercial sucker fishery. The logistic surplus
production model was utilized to estimate a production
function for this fishery. Some of the problems associated
with the bioeconomic analysis of multi-species fisheries,
particularly with regard to optimization of single species
harvest levels, are discussed in Chapter IV.

Previous work on samples of the commercial catch from
Lake Huron and Saginaw Bay (Zabik, et.al, 1978) indicates
that the levels of the environmental contaminants, PCB's and
DDT,in suckers are within the 5 ppm allowable tolerance
limits established by the U. S. Food and Drug Administration
for these compounds. Continued monitoring of environmental
contaminant levels to insure compliance with applicable
health regulations is a necessary component of any fisheries
development effort. Samples of the commercial sucker
landings, stratified by sex and size, were obtained for
analysis of DDT and PCB's. Methods and results of these

analyses are presented in Chapter V.

-12

Representative samples of commercially caught suckers
ownn the east central portion of Saginaw Bay were Obtained
from September 1979 through April 1981. Data from these
samples were used to determine the size, age and sex
composition and age-specific growth and mortality rates of
Saginaw Bay white suckers. Age-specific survival and
fecundity data were incorporated into 3 Leslie Matrix model
(Vaughan and Saila, 1976) in an effort to obtain estimates
<Df age 0 survival rates. ' Estimates Of relative yield per
recruit were calculated for different, hypothetical rates of
fishing via a modification of the dynamic pool model
described by Beverton and Holt (1957). This model,
incorporating a linear age-weight function, was used to
estimate the relative susceptibility of Saginaw Bay sucker
stocks to different rates of exploitation. These efforts
are described in Chapter VI.

Chapter VII contains a summary of the biological and
economic analyses contained in Chapters III through VI. In
this section, recommendations for future management of
Saginaw Bay sucker stocks are discussed. Problems and
deficiencies in the analysis of sucker population dynamics
encountered in the preceding chapters are discussed along

with recommendations for future work.

CHAPTER II

Saginaw Bay

A. The Physical Setting

The followingidescription of the physical
characteristics of Saginaw Bay is taken from the monograph
by Beeton, et al, (1967). Saginaw Bay (Figure II-l) is a
large indentation on the western shore of Lake Huron (42 km
wide at the mouth between Point Aux Barques and Au Sable
Point and 82 km long from the midpoint of a line between
these points to the mouth of the Saginaw River). The bay
narrows to a width of 21 km between Point Lookout on the
west shore and Sand Point on the east shore. A broad shoal
at this constriction (between Charity Island and Sand Point)
effectively divides the bay into outer and inner zones. The
total area of 2960 km2 is divided about equally between the
two zones, but the outer zone is considerably deeper (mean
depth, 15.6 m, maximum depth, 40.5 n0 than the inner zone
(mean depth, 4.6 m, maximum depth, 14.0 m) and contains
about 70% of the total volume of the bay. The volume of
water above the 5.5 m contour is only 32% for the outer bay,
but 80% for the inner zone. The east shore Of the outer bay
is rocky; the west shore has extensive sandy areas, although

some rock and clay occur near Point Lookout. The shoal

13

14

 

 

if 41'. ..

-' - 7mm" fl ' 0 N
”L

, 0m in!
wmmnsou-
wmwuuasM-azm
~Wm

 

 

 

 

 

 

 

 

 

 

Figure II-l. Saginaw Bay (from Beeton, et a1,

1967).

15

around Charity Island and most of the points hitme outer
bay are rocky. Tawas Point and Sand Point are sand spits.
The inner bay has extensive shallows. A broad sandy flat
extends southward from Wildfowl Bay. Another irregular
sandy flat extends from the Saginaw River along the west
shore to Point Au Gres. Several shallow spits off the
mouths of small rivers extend over this flat perpendicular
to the shore. The two large flats have extensive marshes
near shore. Coryeon Reef is a sand and gravel bar between
the shore southeast of the Saginaw River and the Charity
Islands. Its ridge is about 1.8 m below the surface and it
is separated over most of its length from the sandy flat
near shore by water deeper than 3.7 m. The bay has several
islands, the most prominent of which is Charity Island. A
group of marshy low-lying islands (North, Stony and
Katechay) lies southwest of Sand Point. These islands are
surrounded by marshy shallows from which there is no clear

line of demarcation.

B. Limnology

 

The most recent comprehensive limnological survey of
Saginaw Bay was a cooperative investigation by the Michigan
Department of Conservation and the U.S. Bureau of Commercial
Fisheries in 1956 reported upon by Beeton, et a1, (1967).
Federal and State of Michigan monitoring of Saginaw Bay
tributary water quality has been sporadic over the past
twenty years and is summarized in the U.S. Environmental

Protection Agency's STORET data system (Freedman, 1974).

16

Biological investigations of Saginaw Bay have focused on
fish (Carr, 1962, and Eshenroder, 1977) and benthic
macroinvertebrates (Schuytema and Powers, 1966, and
Schneider, et al, 1969). A summary of'existing physical,
chemical and biological information on Saginaw Bay is
provided by Freedman (1974).

Beeton, et a1 (1967) report that water circulation
within the bay is generally counterclockwise, however,
surface circulation patterns are highly variable, depending
primarily'cni the direction of the prevailing winds. Winds
ffiwnn the northeast, east and southeast produce a clockwise
icirculation whereas winds from the northwest and west
generate a counterclockwise surface circulation. Little is
known concerning the direction of subsurface circulation
patterns within Saginaw Bay. The inner bay (mean depth, 4.6
m) becomes homothermous during late July, according to
Beeton, et a1, (1967), whereas the outer bay (mean depth,
15.6 m) exhibits a more complex summer thermal
stratification, influenced by intrusions of Lake Huron
water. Dissolved oxygen concentrations vary according to
water temperature and wind velocity. Beeton, et a1, (1967)
found dissolved oxygen concentration at 9 m in the inner bay
at a low value of about 70% of saturation after a period of
calm during mid-June. The pH of Saginaw Bay waters
increases from approximately 7.8 - 8.1 in June to about 8.4
in July, after which it gradually decreases to lower values.

Alkalinity varies from 110 - 125 ppm CaCO3 in the inner bay

17

and from 90 - 110 ppm CaCO3 in the outer bay. A brief
summary of some major physicochemical parameters for Saginaw
Bay is listed in Table II-l.

According to Freedman (1974), large summertime blooms
of blue green algae are a common occurrence in Saginaw Bay.
These algal blooms have been attributed to increased
concentrations of phosphorus resulting from agricultural
activity in the Saginaw Bay watershed. Schneider, et a1,
(1969) investigated the distributiOn and abundance of
benthic fauna in Saginaw Bay. These workers found that
oligochaetes and chironomids dominate the benthic fauna,
comprising approximately two thirds of the average standing
crop of 4.43 g/m2. The deeper portions of the inner and
outer bays and Wildfowl Bay were found to be the most
productive areas in terms of benthic fauna whereas the
Shallow, sandy Coreyon Reef was the least productive zone.
.Oligochaetes comprised 50% of the standing crop in the zone
closest.tx> the mouth of the Saginaw River while chironomids
and amphipods dominated the benthos in the middle and outer
portions of the bay, respectively. Pronounced decreases in

Observed numbers of burrowing mayflies (Hexagenia spp.)

 

occurred after the mid—1950's.

C. The Fishery

 

According to Beeton, et a1, (1967), Saginaw Bay
supported a prosperous recreational fishery during the first
half of the 20th century and also produced about 40% of the

U.S. commercial catch from Lake Huron although it

'18
Table II-L

Saginaw Bay Physicochemical Data

Parameter Value Source
Climate
type maritime Eshenroder (1977)
latitude (°N) 43° 50'
avg. max. temp. (0C)
July 28.6o
January 0.1O
annual precipitation 76.6 cm
growing season 218 days
Drainage basin 2
area 21000 km Freedman (1974)
human population
1940 0.6 million
1970 1.2 million
Effluent streams
mean discharge 3
8 largest streams 115 m3/s Freedman (1974)
Saginaw River 96 m /s
Morphometry
length 83 km Beeton, et a1,
width at mouth 42 km2 (1967)
total area 2960 km
mean depth
inner bay 4.6 m
outer bay 15.6 m
maximum depth
inner bay 14.0 m
outer bay 40.5 m 3
‘ volume 27 km
flushing rate 186 days
Chemistry
specific conductance
inner bay 469 umhos Beeton, et a1,
outer bay 253 umhos (1967)
Saginaw River mouth 800 pmhos
Lake Huron proper 174 umhos
alkalinity
inner bay 110-125 ppm CaCO3
outer bay 90-110 ppm CaCO
pH 3
June 7.8-8.1
July 8.4
phosphorus (total)
inner bay 0.041 ppm

outer bay 0.018 ppm

19

constituted only 5% of the total area of Lake Huron's U.S.
waters. During this period the commercial fishery depended
upon abundant stocks of lake herring, walleye pike, yellow
perch and lake whitefish. Of these species, only yellow
perch are currently taken in appreciable quantities by
commercial fishermen. The most widely distributed species
of fish within Saginaw Bay, according to Carr (1962), are
alewives, rainbow smelt, yellow perch and white suckers.
Approximately 74 species of fish are known to occur in the
bay. At present, Saginaw Bay commercial fishermen rely upon
harvests of channel catfish for the major portion of their
incomes. Carp, yellow perch, sheepshead (Aplodinotus

gigggiegg), quillback, (Carpiodes cyprinus) and lake

 

whitefish are the other principal commercially harvested
species.

Saginaw Bay is a productive, enriched body of water
which supports a thriving recreational fishery and a small,
declining commercial fishery. The vast potential of this
body of water for fish production was heavily exploited
during the latter portion of the 19th century and up through
the first half of the 20th century. Changes in the species
composition of the fishery have markedly diminished the
commercial utilization of the bay's fish producing
capability. The future of the commercial fishery hinges
upon carefully regulated efforts to capitalize upon the
bay's productivity through increased exploitation of

underutilized species. Management goals for Saginaw Bay's

20

latent fisheries should consist of a mixture of biological,
economic and social objectives designed to enhance the
productive capacity of existing stocks and the region's
economic and social well-being. Prior to formulation of an
underutilized fishery management policy, we need to know the
fishery yield potential. Chapter III is concerned with this

problem.

CHAPTER III

Catch and Effort Data Analysis

A. Introduction

 

The distribution and abundance of commercially
exploited fish populations depend upon a set of n
genetically regulated physiological and behavioral rate
variables, rk(t), k = 1,...,n, which determine a
population's ability to respond to changes in environment
(biotdx: and abiotic) and level and pattern of exploitation.
Some examples of these rate variables, rk(t), are: age,
size and sex specific growth and mortality; age and size
specific fecundity; recruitment; immigration and emigration.
The study Of fish population dynamics includes Obtaining
reasonably accurate measurements of these variables and
relating this information to past, present and projected
future levels of population abundance. Much of the data
used to determine the dynamics of commercially exploited
fish populations is acquired from: mark and recapture
studies; samples obtained from research vessel surveys and
commercial landings; and commercial fishery catch and effort
records.

The acquisition of biological data constitutes the

most costly portion of any fish population dynamics

21

22

investigation. Most studies of the dynamics of commercially
exploited fish populations depend, therefore, to a large
extent, upon the information contained in the records of

total landed catch, C and nominal effort, f provided at

t’ t’
little cost by the commercial fishery. With reliable catch
and effort statistics, estimates of relative population
abundance can be inferred from the ratio of total catch, Ct’
to total effort, ft' The accuracy of population abundance
estimates calculated from catch per unit effort analyses
depends, among other things, upon the degree to which

records of: (1) total landed catch, C reflect actual

t,

catch and (2) nominal effort, f reflect actual fishing-

t 7
induced mortality.

B. The Basic Theory of Catch and Effort Data Analyses

 

The change in stock biomass due to mortality during
the time interval, At, is given by,

III-l dB(t)/dt

-[M(t) + F(t)] B(t),

where: B(t) = stock biomass at time t
M(t) = instantaneous rate of natural mortality
F(t) = instantaneous rate of fishing mortality.

In practice, M(t), the instantaneous rate of mortality
attributable to all factors other than fishing (e.g.
predation, disease, parasitism, senescence, etc.) and F(t),
the instantaneous rate of mortality attributable to the
effects of fishing, are regarded as constants during the
time interval, At. This is a lumped parameter model which

assumes that each unit of biomass within the exploitable

23
portion of a given population is subject to the same rates
<3f mortality, M(t) and Fit), during a given time interval,A
t.
The solution to equation III-l when M(t) and F(t) are
constant during time, at, is given by,
e-(M + F)At,

III-2 Bt = BO

where: B - stock biomass at tO

ED
cf
l I

- stock biomass at t.

The exponential expressiOn in equation III-2 is the

probability (S) that a unit of biomass in the portion of the

population which is vulnerable to exploitation will survive

to the end of the time interva1,A t, or

III—3 S = e-(M + F)At.

Conversely, the probability (A) that a unitcfi‘biomass hi

the exploitable portion of the population will succumb to

the forces of mortality during time, At, is

III-4 A = 1 - S = 1 - e‘(M * F)At.
The proportion of total removals due to fishing during

time, At, is given by, -

III-5 u = FA/(M + F),

where u is referred to as the rate of exploitation.

Theoretically, total removals due to fishing (i.e. total

<3atch Ct) during time, At” equal the rate of exploitation,

u, multiplied by the size Of the stock present at the

beginning of the time interval or,

III-6 C : UB

t = FABO/(M + F) : F(l - S)BO/(M + F)

-(M + F)At

O
FEl - e

lBo/(M + F).

24

Now, in accordance with equation III-2, we know that stock
biomass decreases exponentially during time, At. By
integrating equation III-2 with respect to time we obtain an

estimate of mean stock biomass during time,At,

1- _ -(M + F)At
III-7 Bt - B011 - e 1

Substituting equation III-7 into equation III-6 leads to the

/(M + F).

following expression for total catch, Ct’ during time, At,

III-8 Ct = uBO = EEt.

F, the instantaneous rate of fishing mortality, is usually
assumed to be directly proportional to the amount of nominal
fishing effort, ft’ applied to the fishery during time, At,
III-9 F : qft'

The proportionality constant, q, in equation III-9 is
referred to as the catchability constant or the probability
that a given unit of'fish biomass within the exploited
portion of a population will be captured by a unit of
nominal fishing effort, f, during time, At. The units of q
are in terms of f'1.

A production function is a mapping which relates a set
of identifiable, measurable inputs (e.g. land, labor,
capital, management) to a corresponding set of identifiable,
measurable outputs (i.e. goods or services). Perhaps the
simplest model of a fishery production function is given by
the following equation,

III- 10 Ct : Q(ft,‘3't),
In this equation, total catch, Ct’ during the time interval,A
t, is hypothesized to be a functhmucfi‘the total amount of

25

runninal effort exerted by the fishery, ft’ and the average
size of the exploitable portion of the population, 5,. In
economic analyses of commercially exploited fisheries, the
function, Q(ft:§t), is sometimes represented in the
following form (Clark, 1976).
III-ll Ct = q(rt)a(E,)b,
where q is the catchability constant and a and b are
positive constants.

The well-known assumption that the ratio of catch per
unit effort, Ct/ft’ is directly proportional to the mean
size of the vulnerable portion of the population present

during the period when fishing takes place, B is described

t’
by Beverton and Holt (1957). This assumption leads to the
following formulation of equation III-ll,
III-12 Ct = q(ft)(E¥),
where the constants a and b in equation III-11 are both set
equal to one. Equation III-l2 is also derived by
substituting equation III-9 into equation III-8.

A more general form Of‘a fishery production function,
Q(ft{§£), is given by,
III-l3 Ct = Q(ft’§t) = th(ft)a(Ft)b,
where catchability, q, is a time varying function and x, a
and b are positive constants. The analytical difficulties
involved in estimating the parameters of equation III-13 are
formidable, so most catch and effort data analyses use

equation III-12 to describe the short term production

function of a commercially exploited fishery. Equation III-

26

12 represents the catch, Ct’ obtained during time, At, by
applying ft units of effort to the fishery. Long term
fishery production functions incorporate specific
assumptions about stock biomass growth (e.g. logistic growth
models) to help describe the assumed long term equilibrium

relationships between and among catch, C effort, f and

t’ t’

stock biomass, B An example and discussion of a long term

to
fishery production function is contained in Chapter IV.

Nominal effort, f measured in terms of standardized

1;!
gear lifts or sets, is not a unidimensional measure of

effective fishing mortality, F, as may be erroneously

inferred from equation III-9. Nominal effort, ft’ is a

function of several variables, each of which plays an
important role in determining the effective rate of fishing
mortality exerted upon an exploited population during time,

t. For example, ft can be represented by,

III-l4 ft = f(g1,...,gm).

Some of these variables, g i = 1,...,m, are: size,

i!
capacity and power of the fishing vessels; knowledge and
experience of the crews; selectivity of the fishing gear;
and per unit harvest costs and prices. The enumeration and

quantification of these effort variables, g along with the

i!
determination of their exact relationship to effective
fishing effort, F, would involve an exceedingly complex form
of economic input analysis, the scope of which is probably

not justified by the value of most fisheries. Equation III-

14 indicates that the assumption of a constant effective

27

fishing mortality rate, F, may be unrealistic for long time
periods. The potential difficulties and costs associated
with a multi-dimensional analysis of fishing effort lead
most fisheries investigators to conditionally accept the

assumption that nominal effort, f is a reliable measure Of

t’
effective fishing mortality, F(t). This assumption will be
adopted for the purpose of the present analysis.

In addition to the assumption of constant catchability,
q, some additional assumptions upon which equation III-l2 is
based are: (l) instantaneous fishing and natural
mortalities occur uniformly throughout the year and from
year to year; (2) fish in the exploitable portion of the
population are uniformly distributed throughout the area of
exploitation and (3) individual fishing vessels and units of
fishing gear are distributed so they are not in direct
physical competition with one another.

Mean exploitable population size, Bt’

the time interval covered by available catch, C

present during

t’ and

nominal effort, f data can be calculated from equation

t!
III-l2 provided a reasonably reliable estimate of

catchability, q, can be obtained.

C. The Leslie Model

 

If a population is fished until enough fish are removed
to significantly reduce the catch per unit effort ratio,
Ct/ft’ then estimates of catchability, q, and initial
population size, 80’ can be Obtained from the following

relationship between catch per unit effbrt and cumulative

28

catch, K (of. Seber, 1973, and Ricker, 1975).

t
III-15 Ct/ft = th.

The population, B present during the period covered by

t!
available catch and effort data can be equated to the
initial population, 30’ minus the cumulative catch, Kt’ or
III-l6 Bt 2 BO - Kt'

By substituting equation III-16 into equation III-15 we get,
III-17 Ct/ft : q(BO - Kt) : qBO - th.

Equation III-17 indicates a negative linear relationship
between catch per unit effort, Ct/ft’ and cumulative catch,

I( ‘with slope, -q, and y-intercept, qBO. This model was

t’
first introduced by Leslie and Davis (1939) and subsequently
modified by Braaten (1969). The reliability of this method
of determining catchability, q, and initial population
abundance, B0 depends upon the following assumptions: (1)
the population is closed with respect to recruitment,
natural mortality and migrations, or equivalently, these
processes are in equilibrium; (2) catchability, q, is
constant for all sizes of fish appearing in the catch
throughout the time period covered by available catch and
effort data; (3) catch and effort records are completely
reliable; (4) fishing effort is randomly distributed
throughout the range of the population; (5) individual units
of fishing effort are independent; (6) the entire catch or
most of the catch is captured by the same type of fishing

gear and (7) there is no variation in catching efficiency

between and among the various units of fishing gear employed

29

by the fishery. The range and severity of the above
restrictions greatly reduce the applicability of the Leslie
method to commercial fishery population dynamics studies.
Despite these severe limitations, however, the Leslie and
other similar methods of population enumeration can be used
to obtain rough estimates of catchability and mean
population size for those fisheries in which: (1)
catchability, q, is reasonably constant; (2) catch and
effort records are fairly reliable and (3) the fishery is~

dominated by a single, uniform type of fishing gear.

D. Sources of Commercial Fishery Catch and Effort Data

 

Since 1929, Michigan's Great Lakes commercial fishermen
have been required to submit detailed monthly reports to the
Michigan Department of Natural Resources describing, on a
daily basis, the total landed weight of each species
harvested along with the total number of units of effort
employed by each type of fishing gear. A comprehensive
description of the current Great Lakes commercial fishery
catch and effort data reporting system is provided by Hile
(1962).

According to Hile and Buettner (1959), about 70% of the
total Saginaw Bay sucker landings from 1929 to 1956 were
Obtained with shallow trap nets (i.e. trap nets fishing in.
water less than 20 feet in depth). During the period from
1960 to 1979, approximately 90% of all commercially landed
suckers from Saginaw Bay were captured by shallow trap nets.

After the commercial use of gill nets was banned in the

30

early 1970's, the Saginaw Bay commercial fishery for all
species has been exploited almost exclusively by means of
shallow trap nets. One unit of nominal effort for this gear
corresponds to one net lift.

A uniform analysis of catch and effort data requires
that all nominal effort from all types of gear (if more than
one type of gear is employed) be standardized against a
specific gear (usually the gear which consistently captures
more fish than any other type of gear). Prior to analysis
of Saginaw Bay sucker fishery catch and effort data, all
reported effort was standardized against the shallow trap
net as follows,

III-18 f = (Ct)(fS)/(CS),

where: f total (standardized) effort

t
Ct = total catch from all types of gear
f3 = total number of units of standard effort
(i.e. total number of shallow trap net
lifts)
CS = total catch obtained with the standard gear.

No single gear utilized by any given fishery is likely
to be entirely non-selective with respect to the size, sex
or age composition of a particular exploited population.
Therefore, when obtaining fish samples from the commercial
catch for purposes of size, sex and age composition
analyses, sampling effort should be stratified, if posSible,
according to the various types of gear used to capture the

species under investigation. Calculations of catch indices,

31

population estimates, growth rates and mortality rates can
be biased if all fish sampling is restricted to a single
size selective gear.

According to Patriarche (1968), stationary trapping
devices are usually selective with respect to the species
and sizes of fish they capture and retain. Latta (1959)
found the vulnerability of white suckers to trap net capture
increased with increasing fish size, however, Laarman and
Ryckman (1980) reported that trap net size selectivity was
not evident for this same species. Patriarche (1968),
observing that good estimates of population dynamics
parameters can be systematically biased if certain species
exhibit strong tendencies to escape confinement, found that
28% of the total number of white suckers initially captured
had escaped from experimental trap nets within 48 hours.
The tendency of suckers to escape from trap net confinement
is corroborated by Saginaw Bay commercial fishermen, who
observe that the absence of suckers in a given trap net's
catch often indicates the presence of a hole in the net.
Captured suckers work so persistently in exploring possible
openings in the trap nets' pot end (i.e. confinement area)
mesh that they often abrade away the papillose flesh of
their snouts and lips. This phenomenon of snout abrasion is
particularly evident in suckers obtained from trap nets
which have been set for prolonged periods (e.g. more than 5
' days) between lifts.

Since the Saginaw Bay sucker fishery is prosecuted

 

32

almost exclusively by means of shallow trap nets, any size
selective bias for suckers inherent to this gear plus any
marked tendency for these fish to escape confinement could
significantly bias estimates of population parameters
calculated from trap net samples. Determination of possible
size selectivity or escape tendencies could be assessed via
limited mark and recapture studies. In the absence of such
information, the assumption must be made that estimates of
growth and mortality rates obtained from samples of the trap
net catch are representative only of that portion of the
stock which is most vulnerable to trap net capture and
confinement.

Saginaw Bay commercial sucker fishery yearly catch and
effort data for the period from 1929 to 1956 were obtained
from the monograph by Hile and Buettner (1959). Monthly
catch and effort report summaries for the period from 1960
to 1979 were obtained from the Michigan Department of

Natural Resources' Fisheries Division, Lansing, Michigan.

E. The Saginaw Bay Sucker Fishery

 

1. Background.

 

For purposes of commercial fishery catch and effort
data reporting, the area designated as Saginaw Bay, Lake
Huron Statistical District MH-4 (Figure III-1), extends into
a wedge-shaped region of Lake Huron proper (Smith, et a1,
1961). The adjacent open-lake waters of Lake Huron were
included in the Saginaw Bay district, according to Hile

(1962), to separate the whitefish grounds off the mouth of

33

 

 

   

 

 

 

l.
Bodevcr
I. x; I
. ' pg; -.
CHEW I 03-1.. ‘
l \. .. ,-"Lonol t.‘~
,. \\W.' " Fitz-«Noint‘a 63-3
- .—" ~ Covet.
Mom?! \'\ 9' 33.2 x‘ W
tau-2 . 001-: o, . ~WL WM
‘1 ‘ti 3:. ”.3515"
s *‘P'
............. .\\ - 00-4
‘Afl
1 E
tau-3 \ 0'"! t
. '5
\ 7
Anson. ................. l
urn-4 ‘.
. o’ l: -------- new:
m '-'°" um \ LAKE HURON
IA
" ‘ OH ----------- DISTRICT aouuoam
5" C RlcmouoanLE ------- , -""’ — "‘” summon» aouuouu
moo!
I! ........ Gamma
m I no so co [0
Statute Mun ‘
m" mm

Figure III-l.

 

 

Lake Huron Statistical Catch
Districts (from Smith, et a1, 1961)

34

the bay ownn the more northerly grounds of the Oscoda area
and the southerly grounds of the Harbor Beach region.
Almost no suckers are taken from the deep waters of this
wedge-shaped region, therefore, reported catch and effort

2 area of

statistics correspond almost exactly to the 2960 km
Saginaw Bay described by Beeton, et a1, (1967).
Multi-species fisheries present a perplexing set of
problems to any investigator studying the population
dynamics of a particular species' population within such
fisheries. The Saginaw Bay trap net fishery is a relatively
non-selective harvester of an assemblage of different

species including: channel catfish; yellow perch; carp;

suckers; lake whitefish; sheepshead (Aplodinotusggrunniens);

 

quillback (Carpiodes cyprinus); black crappies (Pomoxis

 

nigromaculatus); white bass (Morone chrysops) and others.

 

 

The fishery has some ability to target on desired species
(e.g. yellow perch and lake whitefish) during specific times
Of the year. At present suckers represent an incidental
catch and are not specifically sought by commercial
fishermen. The species composition of the commercial catch
varies according to the seasons. For example, lake
whitefish, yellow perch and suckers are most available Us
the fishery in the early spring and late fall whereas carp,
sheepshead and quillback are most available during the warm
summer months. Availability, as used here, is a term which"
refers to that portion of a given stock which is actually

vulnerable to a fishery during a particular time. If a

35

species' vulnerability to the fishery varies during
different times of the year (e.g. due to temperature-induced
behavioral movements), then changes in relative abundance
indices (e.g. Ct/ft) from season to season within a given
year are probably unreliable indicators of changing
population abundance. In such instances only yearly
relative abundance indices are likely to be of any value in
assessing population abundance. A particularly vexing
problem associated with the analysis of Saginaw Bay sucker
fishery catch and effort data is that, due to the poor
market for these fish, reported landings are usually a
subset of the actual catch. The reliability of catch per
unit effort indices calculated from such data is
questionable unless it can be assumed that a reasonably
constant proportion of the actual catch is landed throughout
the fishing season from year to year. With such problems in
mind, we will cautiously proceed with an analysis Of Saginaw

Bay commercial sucker fishery catch and effort data.

2. The Sucker Fishery, 1929 to 1956

 

An excellent description of the Saginaw Bay commercial
fishery for all major species harvested during the period
from 1929 to 1956 is provided by Hile and Buettner (1959).
Up until the mid-1940's, the principal species harvested by
the fishery were lake herring, walleye, suckers, carp and
yellow perch. Changes in the species composition of the
Saginaw Bay fishery paralleled similar changes which

occurred throughout the Great Lakes since the early 1940's.

36

Declining abundances of highly valued species (e.g. lake
herring and walleye) contributed to corresponding reductions
in the size of the commercial fishery'(cfu Figure III-2).
From the group of highly prized species harvested in
abundant quantities during the early part of the 20th
century, only yellow perch were caught in commercially
significant amounts by the mid-1950's. Increased abundances
of carp, channel catfish and suckers were observed after the
mid-1940's, however, these species remained underutilized
due to the absence of sufficiently profitable markets.

For the Great Lakes fisheries, according to Hile
(1962), the period from 1929 to 1943 is regarded as the so-
called base or reference period of abundance for all
commercially harvested species. During this time, the
abundances Of the principal commercially harvested species
populations were considered relatively stable. An index of
relative abundance for a given species can be calculated by
dividing each year's catch per unit effort ratio, Ct/ft,'by
the base period's mean catch per unit effort for that
species in a particular catch district. The resulting
abundance index, when multiplied by the amount of nominal
effort, ft’ expended during a given year, provides a measure
of the expected catch for that year (i.e. the catch that

would result if the abundance of the current year's stock

37

 

 

 

 

 

 

IOO 1- ' 1000
801 < 800
60 - 1 600
Total Total
Vessels Fishermen
(—) (--)
40+ ~ 400
20- . 200
C 1 1 L l l O
60 64 68 72 76
Year

Figure III-2. Participation in the Lake Huron
Commercial Fishery (U.S. Waters, 1960-1976).

38

were equal to the abundance of the stock during the base
period). The assumption of constant catchability, q, is
implicit in this particular measure of relative abundance.
The years from 1929 to 1956 can be divided into two
separate periods, one of lower abundance, from 1929 to 1945,
and one of higher abundance, from 1946 to 1956. Table III-
1, Table 111-2, and Figure III-3 contain summaries of catch,
Ct’ effort, ft’
for these years. The mean catch per unit effort from 1929

and catch per unit effort, Ct/ft, statistics

to 1943 (i.e. the base period of abundance) was
approximately 27.7 pounds of suckers per shallow trap net
lift. Figure III-4 depicts the abundance of Saginaw Bay
suckers as a percentage of mean 1929 to 1943 catch per unit
effort. A period of slightly decreasing relative abundance
from 1929 to 1945 is followed by a period of steadily
increasing relative abundance from 1946 to 1956.
Equation III-12 can be rewritten as,
III-19 Ct = dfitrt = b(ft),

which suggests that a simple linear regression of total

catch, C versus nominal effort, f forced through the

t’ t’
origin will yield an estimate (i.e. the slope parameter, b)

of mean biomass, E multiplied by catchability, q. The

t,
'regression Of'Ct on ft for the 1929 to 1945 period resulted
in the following,
- - .2-
III-20 Ct . b(ft) - 26.9(ft), r - 0.97.

If an independent estimate of catchability, q, could be

<ibtained, we could estimate‘B.t simply by dividing the slope

Year

1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
19u2
1943
19uu
1945
1946
1947
'1948
1949
1950
1951
1952
1953
195a
1955
1956

* Source:

39

Table III-l
Saginaw Bay Commercial Sucker Production

(1929-1956)*

Catch (lbs) Effort Catch Per Effort
Ct ft** Ct/ft
1193000 36000 33.1
1687000 53000 31.8
1183000 47000 25.2
1292000 36000 35.9
1069000 32000 33.4
1136000 33000 34.4
803000 40000 20.1
936000 39000 24.0
922000 38000 24.3
780000 29000 26.9‘
748000 30000 24.9
880000 31000 28.4
820000 29000 28.3
683000 31000 22.0
892000 36000 24.8
740000 37000 20.0
813000 39000 20.8
1214000 34000 35.7
932000 26000 35.8
821000 24000 34.2
630000 19000 33.2
642000 18000 35.7
904000 19000 47.6
941000 21000, 44.8
971000 23000 42.2
1060000 19000 55.8
889000 17000 52.3
536000 14000 38.3
Hile and Buettner (1959)

** Standardized Effort, f = (Ct)(fS)/(CS)

Note:

t
where: C

total catch from all types of

t gear

f = total effort with standardized
3 gear

C = total catch with standard gear

8

One unit of standardized effort equals one shallow
trap net lift.

40

Table III-2
Saginaw Bay Sucker Fishery

Summary Statistics, 1929-1956*

 

 

 

 

 

Statistics 1929-1945 1946-1956 1929-1956
Ct range (lbs) 683000- 536000- 536000-
1687000 1214000 1687000

Ct 975000 867000 933000
SCt 258000 200000 239000
ft range (shallow 29000- 14000- 14000-
trap net l1fts) 53000 34000 53000

Ft 36200 21300 30400
5ft 6400 5400 9500
Ct/ft range (lbs per 20.0- 33.2- 20.0-
trap net 11ft) 35.9 55.8 55.8

Ct/ft 26.9 41.5 32.6
SCt/ft ' 5.1 7.7 9.5

*Data source: Hile and Buettner (1959)

 

 

41

 

 

 

20- 400
l64 -80
12« -60
Catch Effort
(105 lbs) ,'\ (103 an.)
(_) l ‘1 (---)
I \‘ 40
8~ .
1' ,"1 , \
\\’/ \ I \
1”\/
/\
, 4 . \I \ »2O
\.
O . . . . L t . O
30 35 4O 45 50 55
Year

Figure III—3. Saginaw Bay Commercial Sucker
Production (1929-1956).

42

200‘

|60+

1201

 

tAUuukflmn*

80-

40-

 

 

it 35 4O 45 50 55 60
Year

um». «pressed as a mucootm 1929-1943 c,/f,

Figure III-A. Saginaw Bay Sucker Abundance (1929-1956).

43

parameter, b, calculated from equation III-19 by q.
One of the first steps undertaken when examining a time
series of commercial fishery catch.and effort data is the

calculation of the following simple linear regression,

III-21 Ct = a + b(ft).
A significant positive correlation between catch, Ct’ and
effort, ft’ indicates that the stock under examination has

not been‘overexploited since, according to Ricker (1975),
catch can only increase with an increasing effort if there
are reserves Of stock to draw upon. Equation III-21,
applied to the 1929 and 1945 catch and effort data series
resulted in the following,

2

III-22 C 1896.69 + 26.86(ft); r = .45; r : .67*

t
(p : .01).

According to Ricker (1975), a large significant

correlation between catch, C and effort, ft, indicates

t’
that the rate of exploitation, u, has been less than 75%
over most or all Of the range of efforts represented by the
data. Recall equation III-5,
III-5 u = FA/(F + M) : F(1 - S)/Z : F(1 - S)/(1n S).

u represents the fraction Of the total mortality rate,
A, attributable to fishing. The upper bound for u, then, is
the value of A. By definition, total instantaneous
mortality rate, Z, equals F + M, and survival, S, equals
'EZAt, during time, At. For the sake of convenience, assume

that the duration of the time period, At, is equal to one.

Previous investigations of the dynamics of white sucker

 

44

populations (e.g. Geen, et a1, 1966 and Coble, 1967) plus
the present investigation (Chapter VI) indicate that annual
white sucker survival rate (S) ranges from 0.50 to 0.75.
Taking natural logarithms, we find that Z ranges from
approximately .228 to .693. We are now in a position to
calculate rough point estimates of instantaneous fishing

mortality rate, F.

Case (1)

Assume annual survival rate, S, equals .50. Then, A

l - S = .50. The upper bound for rate of exploitation, u

.50. Rearranging equation III-5 we solve for F as follows,
III-23a F = u(F + M)/A = u(lnS)/(l - S)
: .50(.693)/(l - .50)
= .693.
Now, since for S : .50, lnS = Z = F + M = .693, equation
III-23a implies that all mortality is attributable to
fishing.
If u = .35, we obtain,
III-23b F : u(F + M)/A = u(lnS)/(1 - S)
= .35(.693)/(1 - .50)
= .485.
If u = .20, we obtain,
III-23c F = u(F + M)/A = u(lnS)/(l - S)
= .20(.693)/(1 - .50)
= .277.

 

45

Case (2)
If annual survival rate, S, equals .625, then A = l -
= .375. The upper bound for rate Of exploitation, u : .375.

Solving equation III-5 for F we obtain,

III-24a F : u(F +M)/A u(lnS)/(l - S)

.375(.470)/(1 - .625)
= .470.

If u = .250, we Obtain,

III-24b F : u(F + M)/A u(lnS)/(1 - S)

.250(.470)/(1 - .625)
.313.

If u = .125, we Obtain,

III-24c F : u(F + M)/A

u(lnS)/(l - S)

.l25(.470)/(1 - .625)

.157.

Case (3)

Assume annual survival rate, S, equals .75. Then, A

1 - S = .25. The upper bound for rate of exploitation, u

.25. Solving Equation III-5 for F we Obtain,

III-25a F = u(F + M)/A (lnS)/(1 - S)

.25(.288)/(1 - .75)
= .288.

If u = .175, we obtain,

III-25b F : u(F + M)/A u(lnS)/(l - S)

.175(.288)/(l - .75)

= .202.

46

If u : .100, we obtain,

III-25c F = u(F + M)/A u(lnS)/(l - S)

.lOO(.288)/(1 - .75)
= .115.

Equation III-23a represents our worst assumptions about
survival and exploitation rate whileequation III-250
represents our best assumptions about these parameters.

By rearranging equation III-9 we can solve for the
catchability constant, q, as follows,

III-26 q = F/ft.

If we substitute mean effort, f; = 36,200 trap net lifts,
during the 1929 to 1945 period into equation III-26 along
with each of the values for F obtained from equations III-23
through III-25 we obtain a range of possible estimates for
catchability, q. Table III-3 contains a summary of these
catchability estimates which range from q : 3.18 x 10.6 to q
= 1.91 x 10'5. These estimates for q, cOupled with the
result of equation III-20, enable calculation of rough point

estimates of mean biomass, Bt’ ranging from Bt

: 8,459,000 pounds (Table III-3).

= 1,408,000

pounds to 8;

Application of Leslie's model (equation III-17) to the
1929 to 1945 catch and effort data resulted in the
following,

qBO — qKt = 33.5044-.000000715(Kt); r2 = .42

III-27 Ct/ft

where: q

7.15 x 10-7 lift-1

BO : qBO/q : 46,845,000 pounds.

Recall that the prerequisite for application of the Leslie

Equation

III-23a
III-23b
III-23c
III-24a
III-24b
III-24c
III-25a
III-25b
III-25c

cf
I

**b

47

Table III-3

(1929-1945)

F Estimate q
.693 l
.485 l
.277 7
.470 1
~313 8
.157 4
.288 7
.202 5
.115 3

- 36,200 trap net lifts

26.9 (see equation III-20)

.91
.34
.65
.30
.65
.34
-96
.58
.18

Catchability and Mean Biomass Estimates

writ,

X

X

X

10'5
10'5
10‘6

10"5

E

t
1408000

2007000
3516000
2069000
3110000
6198000
3379000
4821000
8459000

: b/q**

48

model is that enough fish must be removed from the
population to significantly reduce the catch per unit effort
ratio, Ct/ft’ Examination of the catch per unit effort
column in Table III-l reveals no Obvious, consistent decline
in Ct/ft‘ This indicates that the fundamental prerequisite
necessary for application Of the Leslie model has not been
met. The estimate of catchability, q, obtained from
equation III-27 appears to be an underestimate (compared to
q estimates listed in Table III-3), the effect of which
results in an overestimate of mean stock biomass, 8;.

The regression of Ct versus ft’ forced through the
origin, for the 1946 to 1956 period resulted in the
following,

III-28 Ct = 40.0(ft); r2 = 0.97.
Equation III-21 applied to the 1946 to 1956 catch and

effort data resulted in,

III-29 Ct = 297694.46 + 26.78(ft); r2 = .53; r = .73*
(p = .05).
The significant correlation between Ct and ft for the 1946

to 1956 period indicates the sucker stocks were probably not
overexploited.

If we substitute mean annual effort, f;
net lifts, during 1946 to 1956 into equation III-26 along

: 21,300 trap

with each of the values for F obtained from equations III-23
through III-25 we Obtain a range of possible catchability
estimates. Table III-4 contains a summary of these

estimates which indicates catchability ranges from 0 = 5.40

49
Table III-4
Catchability and Mean Biomass Estimates

(1946-1956)

Equation F Estimate q = F/E¥* §£ = b/q**
III-23a .693 3.25 x 10"5 1231000
III-23b .485 2.28 x 10‘5 1754000
III-23c .277 1.30 x 10'5 3077000
III-24a .470 2.21 x 10'5 1810000
III-24b .313 1.47 x 10‘5 2721000
III-24c .157 7.37 x 10‘6 5427000
III—25a .288 1.35 x 10'5 2963000
III-25b .202 9.48 x 10"6 4219000
III-25c .115 5.40 x 10‘6 7407000
* E; = 21,300 trap net lifts

** b = 40.0 (see equation III-28)

5O

6

x 10' to q = 3.25 x 10'5.

These estimates for q, coupled
1aith the result of equation III-28, enable calculation of

rough point estimates of mean stock biomass, Bt’ ranging
from 1,231,000 pounds to 7,401,000 pounds (Table III-4).

Application of the Leslie model (equation III-l7) to
the 1946 to 1956 catch and effort data failed to result in a
model of the proper form (i.e. with a negative slope
parameter,-q).

In the above analyses of catch and effort data for the
periods, 1929 to 1945 and 1946 to 1956, we have seen how the
assumption of a constant instantaneous total mortality rate,
Z, necessary for calculating point estimates of F via
equations III-23 through III-25, resulted in higher
estimates of catchability, q, during the latter period than
for the earlier period. The apparent increase in relative
abundance during 1946 to 1956 (as measured by the increase
in mean catch per unit effort over the previous period)
could be a result of increased trap net catchability, which,
in turn, results in lower estimates of mean population

biomass, Bt’ during 1946 to 1956 than during 1929 to 1945.
Whether or not such an increase in trap net catchability
actually occurred from one period to the next is a matter
for conjecture, however, such an idea may be given some
credence by the fact that more nets with nylon mesh were
used during 1946 to 1956 than during 1929 to 1945, when the

mesh of most nets was made from cotton or linen fibers.

Different types of fibers exhibit different catchabilities.

51

Since nylon is stronger and more durable than natural
fibers, nylon nets may possess greater catching potential

than natural fiber nets.

3. The Sucker Fishery, 1960 to 1979

 

_ The Saginaw Bay sucker fishery during the years fmmn
1960 to 1979 can be divided into two periods, one of higher
relative abundance, 1960 to 1967, and one of lower relative
abundance, 1968 to 1979. Table III-5 and Table III-6
contain summaries of catch, C

effort, f and catch per

t’ t’
unit effort, Ct/ft’ statistics for these years. Figure III-
5 depicts the declining sucker catch from 1960 to 1979.
Figure III-6 shows the relative abundance of Saginaw Bay
suckers during 1960 to 1979 expressed as a percentage of
mean 1929 to 1943 catch per unit effort. A period of
steadily decreasing relative abundance from 1960 to 1967 is
followed by a more or less stable period of relative
abundance from 1968 to 1979.

During 1960 to 1967, 84.5% of the total sucker catch
*was taken with shallow trap nets, 7.3% was harvested with
haul seines, 3.6% was taken with fyke or hoop nets and 4.6%
‘was harvested by a mixture of other, miscellaneous gears.
Table III-7 contains a monthly summary of total reported

catch, Ct’ and nominal effort, f during 1960 to 1967. In

t!
terms of total landed catch, the fishing year for this
period can be divided into the following quarterly intervals
(the percentage Of the total catch harvested during each

quarter is contained in parentheses); December - February

52

Table III-5
Saginaw Bay Commercial Sucker Production

(1960-1979)*

Year Catch (lbs) Effort Catch Per Effort

ct ft Ct/ft
1960 378700 9350 40.5
1961 517300 10540 49.1
1962 662800 9250 71.7
1963 472100 6730 70.1
1964 392800 6770 58.0
1965 353400 6330 55.8
1966 299300 6040 49.6
1967 219500 5000 43.9
1968 148900 4310 34.5
1969 123200 3740 32.9
1970 137700 3670 37.5
1971 132400 3620 36.6
1972 90600 3640 24.9
1973 144700 4540 31.9
1974 110600 4210 26.3
1975 108600 4360 24.9
1976 124800 3750 33.3
1977 98400 2950 33.4
1978 132000 3480 37.9
1979 107600 2590 41.5
*Source: Mich. Dept. Nat. Res., Fish. Div.

53

Table III-6
Saginaw Bay Sucker Fishery

Summary Statistics, 1960-1979*

Statistics 1960-1967 1968-1979 1960—1979
ct range (lbs) 219000— 90600- 90600-
662800 1u8900 662800
at u12000 121600 237800
Sct 137500 18500 168700
f range (shallow 5000- 2590- 2590-
trap net lifts) 10540 H540 105u0
Ft 7500 3740 5240
Srt 1950 570 2270
CL/f‘t range (lbs 40.5— 24.9- 24.9-
” per trap net 71.7 ”1.5 71.7
lift)
Ct/ft 54.8 33.0 41.7
S
Ct/ft 11.u 5.3 13.6

* Data source: Mich. Dept. Nat. Res., Fish. Div.

Sh

 

 

 

 

 

 

IO .20
8 .l6
6‘ ' -I2
Cgfch Effort
“0 “’5" (103 lifts)
(--) (~—-)
4. . a
2. 14
o 1 . . . o
60 64 68 72 76 80

Year

Figure III—5. Saginaw Bay Commercial Sucker
Production (1960-1979).

55

250-

200 -

l50 -

Abundance"

100-

50-

 

Year
*Abundanco expand a a Mo. of mom 1929-B43 Ct/f,

Figure III—6. Saginaw Bay Sucker Abundance (1960—l979)

Month

January
February
March
April
May

June
July
August
September
October
November
December

2
Ct

Month

January
February
March
April
May

June
July
August
September
October
November
December

2
ft

* Source:

56

Table III-7

Saginaw Bay Monthly Catch and Effort Data

(1960-1967)

Catch (lbs) Catch Range Ct/ZCt
Ct (lbs)

77248 100- 16605 2.
115375 5635- 29801 3.
402814 20684- 94933 12.
515445 18068-177428 15.
522142 21887- 12552 15.
211581 5636- 47457 6.

64292 3218- 14874 2.

52613 912- 20301 1.
483029 21610- 79989 14.
449336 29296- 82651 13.
338267 13905- 77130 10.

63615 38- 18362 1.

3295757
Standardized Effort Range ft/zft
Effort (trap net lifts)
f
t
1458 74- 336 2.
1481 136- 309 2.
3728 191- 725 5.
7366 381-1400 11.
7772 619-1259 12.
4414 107- 794 7.
2030 173- 430 3.
1290 50- 377 2.
8146 412-1335 13.
13089 933-2615 20.
10296 203-3141 16.
1817 213- 655 2.

62887

Mich. Dept. Nat. Res., Fish. Div.

X

towoxsloxozoooxmmw

\D-BmOHNO-zNO-ZL»

100

100

 

57

(7.7%); March - May (43.6%); June - August (10.0%) and
September - November (38.6%). Similarly, in terms of the
amount of‘runminal effort expended, the fishing year can be
divided into the same quarterly intervals (the percentage of
total effort expended is contained in parentheses):
December - February (7.6%); March — May (30.0%); June -
August (12.3%) and September - November (50.2%).

The regression of Ct versus ft’ forced through the
origin, for the 1960 to 1967 data, is,

III-3O ct = 54.8(ft); r2 = 0.96.
Regression of catch on effort for the 1960 to 1967 catch and
effort data results in,
III-31 ct = 23008.09 + 51.86(ft); r2 = .54; r = .74*
(p = .05).

If we substitute mean annual effort, f; = 7,500 trap
net lifts, during 1960 to 1967 into equation III-26 along
with each of the values for F obtained from equations III-23
through III-25 we obtain a range of possible catchability

estimates. Table III-8 contains a summary of these

estimates which indicate catchability ranges from q = 1.53 x
10"‘5 to q = 9.24 x 1075. These catchability estimates,
coupled with the result of equation III-30 enabled
calculation of crude point estimates of mean stock biomass,
'8t, ranging from 593,000 pounds to 3,582,000 pounds (Table
III-8).

Application of the Leslie model (equation III-17) to

the 1960 to 1967 catch and effort data failed to result in a

Equatio
III-23a
III-23b
111-230
111-243
III-24b
1114246
III-25a
III-25b
III-250
* ft
** b

58

Table III-

Catchability and Mean Biomass Estimates

8

(1960-1967)

n F Estimate q :
.693 9.24
.485 6.47
.277 3.69
.470 6.27
-313 4.17
.157 2.09
.288 3.84
.202 2.69
.115 1.53

7,500 trap net lifts

54.8 (see equation III-30)

F/f

X

4
t

10‘5

10"5

10‘5

10‘5

10

10'5

10’5

10‘5

10'5

ml

= b/q**
593000
847000
1485000
874000
1314000
2622000
1427000
2037000
3582000

 

59

model of the proper form (i.e. with a negative slope
parameter,—q). A glance at the catch per unit effort column
in Table III-5 reveals that, for the period from 1962 to
1969, the Saginaw Bay sucker fishery experienced a steady
decline in catch per unit effort. Recall that a steady
decline in catch per unit effort is a necessary prerequisite
for the application of the Leslie model. It would seem
instructive, therefore, to apply the Leslie model to the
1962 to 1969 catch and effort data in order to obtain an

additional estimate of catchability, q.

. 2 -
111-32 Ct/ft - qBO - th - 82.2482 — .OOOOl77(Kt), r - .94
where: q = 1.77 x 10'5 lift’l
BO : qBO/q = 4,647,000 pounds.

These estimates of catchability, q, and biomass, B0,
are close to those obtained from calculations summarized in
Table 111-8.

During 1968 to 1979, 92.1% of the total sucker catch
was taken with shallow trap nets, 4.0% was harvested with
haul seines, 1.9% was taken with fyke or hoop nets and 2.0%
was harvested by a mixture of other, miscellaneous gear.
Table 111-9 contains a monthly summary of total reported

ciatch, C and nominal effort, f during 1968 to 1979. In

t’ t’
terms of productivity, the fishing year during this perhmi
can be divided into the following quarterly intervals (the
Percentage of the total catch harvested during each quarter

is contained in parentheses): December - February (11.4%);

March - May (56.1%); June - August (9.7%) and September -

 

Saginaw Bay Monthly Catch and Effort Data
(1968-1979)*

Month

January
February
March
April
May

June
July
August
September
October
November
December

2
Ct

Month

January
February
March
April
May

June
July
August
September
October
November
December

2ft

* Source:

Catch (lbs)

73348
69533
317826
337652
163011
68998
39799
34208
113449
149260
69693
22699

1459476

Standardized

Effort
ft
1235
1556
3487
5661
3709
3247
3720
4212
6921
6286
3433

749

44216

Mich. Dept.

60

Table III-9

Catch Range

1150-10166
1167-17687
5120-50985
9474-55466
4558-35305
1851-11820
1003- 6373

614- 5284
5216-18199
6589-18843
2329-10179

15- 5715

Effort Range
(trap net lifts)

20-199
33-270
100-616
141-732
150-487
117-414
163-562
49-534
312-941
343-689
168-476
1-110

Res., Fish. Div.

ft/Zft

H

FJH
H-qsuanx-anv-aunv
O C O O O O O C 0 O O .

\zoomslmzouzwxomoo

O‘CDNCDUUNNNI—‘QDCDO

x 100

x 100

61

November (22.8%). In terms of the amount of effort expended
during these years, the fishing year can be divided into the
same quarterly intervals (the percentage of total effort
expended is contained in parentheses): December - February
(11.4%); March-May (56.1%); June - August (25.2%) and
September - November (37.7%).

The regression of Ct versus ft’

origin, for the 1968 to 1979 catch and effort data is,

.2-
III-33 ct = 32.2(ft), r - 0.98.

Equation III-21 applied to these data resulted in,

111-34 ct : 66545.16 + 14.73<rt); r2 : .21; r = .46

forced through the

(N.S., p. = .05).
Previous calculations of catchability and mean stock biomass
(of. Tables III-3, III-4 and 111-8) depended upon the
assumption that the rate of exploitation, u, was less than
75% (refer to the discussion preceding and accompanying
equations III-23 through III-25). This assumption, in turn,
depended upon a significant positive correlation between
catch, Ct’ and nominal effort, ft (Ricker, 1975). The fact
that the catch-effort regression described by equation III-
34 is not significant (p = .05) could indicate that the rate
of exploitation, u, during 1968 to 1979 was greater than
that assumed for other periods (14L.1929 to 1945, 1946 to
1956 and 1960 to 1967, respectively). This is most unlikely
in view of the relatively stable picture of abundance during

1968 to 1979 shown in Figure III-6. To maintain

consistency, the method used to calculate point estimates of

62

catchability and mean stock biomass for the 1929 to 1945,
IU946 to 1956 and 1960 to 1967 periods will not be used for
‘bhe 1968 to 1979 period. It seems reasonable, however, to
assume that catchability, q, during 1968 to 1979 was not too
different from catchability during 1962 to 1969 (as
determined by equation III-32). If so, then an estimate of

{mean stock biomass, B during 1968 to 1979 can be obtained

t’
by dividing the slope coefficient, b, from equation IITQ33
by the 1962 to 1969 estimated catchability or,
111-35 8,: = b/q = (32.2)/(1.77 x 10'5) = 1,819,000 lbs.
Application of Leslie's model (equation III-17) to the
1968 to 1979 catch and effort data failed to result in a
model of the proper form (i.e. with a negative slope
parameter,-q). It is interesting to note that Leslie's

model, applied to catch and effort data from 1960 to 1979,

resulted in the following estimates,

- I — 0 2 -
III-36 Ct/ft _ qBO - qht _ 63.0615-.00000673(Kt), r - .43,
with: q = 6.73 x 10'6 lift"1
80 : qBO/q : 9,370,000 pounds,

The following approximate 95% asymptotic confidence
limits for E0 (biomass present at the start of 1960) were
calculated after the method described by Ricker (1975),

Lower limit 6,310,000 pounds

Upper limit 66,043,000 pounds.
Tending toward the conservative, suppose that the lower
limit of approximately 6,310,000 pounds is a reasonable

estimate of stock biomass available at the beginning of

63

1960.

The Saginaw Bay trap net fishery captures a number of
species simultaneously. It is reasonable to assume that the
relative catching efficiency or catchability, q, of trap
nets f2”: two species present in abundance in the commercial
«catch should be roughly similar. To test this hypothesis,
catch and effort data from 1968 to 1975 for yellow perch and
suckers were analyzed by means of equation III-17. The
following result was obtained for yellow perch,

2

III-37 Ct/ft : qB - qK = 79.8146 - .0000128(Kt); r :.83.

O t
The following result was obtained for suckers,

. 2
III-38 Ct/ft = qBO - th : 37.5793 - '0000123(Kt)’ r

:.51.
Comparison of equations III-37 and III-38 reveals that
estimated yellow perch catchability (q_= 1.28 x 10-5) and
estimated sucker catchability (q = 1.23 x 10-5) are very
close for the 1968 to 1975 time period.

A reasonable estimate of survival rate for suckers in
the vulnerable portion of the population is S = .60 (of.
Chapter VI). This estimate is very close to the survival
rate S = .625, assumed by equations III-24. With a constant
survival probability, S = .625, and hypothetical rate of
exploitation, u = .25, estimates of mean stock biomass, 8;,
for the periods from 1929 to 1945 and 1946 to 1956 are
3,110,000 pounds and 2,721,000 pounds, respectively (of.
Tables III-3 and III-4). The same survival rate and

exploitation rate result in a mean biomass estimate of

1,314,000 pounds for the 1960 to 1967 period. An estimate of

64

mean biomass for 1968 to 1979 is provided by equation III-
35 and is equal to approximately 1,819,000 pounds. A
conservative estimate of mean stock biomass during the 1929
to 1956 period is about 3,000,000 pounds. Similarly, a
conservative estimate of mean stock biomass during the 1960
to 1979 period is approximately 2,000,000 pounds.
Rearranging equation III-8 we get,

III-39 F = ct/fit.

During 1929 to 1956, mean annual catch, 'C't, equalled about
933,000 pounds. Dividing this by our above estimate of mean
biomass (i.e. 3,000,000 pounds) for the same period we
arrive at an estimate of effective fishing mortality rate, F
or,

III-40 F = ct/E£ = (933000)/(3000000) = .311.

This compares favorably with our estimates of F obtained
from equations III-23 through III-25. During 1960 to 1979,

mean annual catch, Ct’ equalled approximately 238,000
pounds. Dividing this by our above estimate of mean biomass
(i.e. 2,000,000 pounds) for the same period we arrive at an
estimate of effective fishing mortality rate, F or
III-41 F: ct/E; = (238000)/(2000000) = .119.
‘This is close to the estimate of F obtained from equation
III-25c.

The above estimates of mean annual sucker biomass of
3,000,000 pounds for the 1929 to 1956 period and 2,000,000
pounds for the 1960 to 1979 period are probably overly

conservative. In his study of white suckers in South Bay,

 

65

Lake Huron, Cable (1967) estimated a standing stock of
approximately 17.5 kg/hectare for regions less than 36 m
deep. If the standing stock of white suckers per hectare in
Saginaw Bay is comparable to that of South Bay, then an
estimate of standing stock is 17.5 kg/hectare times 2.96 x

5

10 hectares (the area of Saginaw Bay) or approximately 5.18

x 106 kg (11,396,000 pounds). Suppose that one third (i.e.
3,799,000 pounds) of this estimated standing stock is
vulnerable to the commercial fishery. As previously
indicated, a plausible estimate of survival rate, S = .60
for the vulnerable portion of Saginaw Bay sucker
populations. From equation III-5 we calculate a mathmI
exploitation rate, u = .40,which obtains when all mortality
is attributable to fishing. If we assume a more realistic
exploitation rate, u : .20, then a point estimate of
effective fishing mortality rate, F = .255 which, when
multiplied by 3,799,000 pounds, yields an estimate of
potential annual catch (i.e. 969,000 pounds). This estimate
is very close to the 1929 to 1956 mean annual harvest of
933,000 pounds.

Mean 1968 to 1979 annual reported sucker landings were
approximately 122,000 pounds. Discussions with commercial
fishermen reveal that up to 75% of the actual sucker catch
is regularly returned to the water due to the absence 0f
profitable markets far suckers. If this is so, then annual
landings could increase by a factor of some 400% without an

apparent increase in nominal fishing effort.

 

66

The results of the present analysis of catch and effort
data indicate that the Saginaw Bay commercial fishery could,
with careful monitoring and regulation, harvest
approximately 500,000 pounds of sUckers per year. Harvest
levels beyond 500,000 pounds per year could probably be
obtained only with an expansion of fishing effort. Initial
"fishing up" efforts for suckers could effectively reduce
the usefulness of the catch per unit effort ratio, Ct/ft’ as
an indicator of relative stock abundance. The catch per
Inuit effort ratio is a useful indicator of relative stock
abundance when the fishery is not subject to marked
fluctuations of its characteristic rate variables, rk(t)
(e.g. exploitation and recruitment). Monitoring of stock
responses to changes in exploitation levels during the
fishing up stage would probably be best accomplished by
observing changes in the size, age and sex composition and

growth and mortality rates of the exploited populations.

 

CHAPTER IV

Fishery Bioeconomics

A. Introduction

 

The concept of a fishery production function was
introduced in Chapter 111 where equation III-12 was employed
to represent the relationship between total catch, Ct’ and

nominal effort, f assuming constant levels of

t 7
catchability, q, and mean eXploitable population size, 8t.

Recall equation III-12,

III-12 Ct : q(ft)(Bt) = q(Bt)(ft)’

which implies that for each level of mean exploitable

population biomass, B total catch, C is determined

t’ t’
directly by the amount of effort, ft, applied by the
commercial fishery. Equation III-12 is a short term, non-
equilibrium fishery production function used to describe the
effects of changes in effort on total catch during a period
of time when mean catchable stock biomass is assumed
constant.

Population size, Bt’ is a dynamic entity responding to
cxnrtinual changes in its determining rate variables, ri(k),
i = i, ..... ,n. For an exploited fish stock, the most

important of these rate variables, ri(k), are: growth rate;

Inatural mortality rate; fishing mortality rate; recruitment

'67

68

rate and net migration rate. The rate of change in the size
of an exploited fish population can be represented by,
IV-1 dBt/dt = G + R - M - F:tI,
where: G = rate of stock growth

R = rate of recruitment

M = rate of natural mortality

F = rate of fishing mortality

I : rate of migration (immigration-emigration).
The long term persistence of an exploited fish population
depends upon the establishment of some sort of equilibrium
relationship between the rates of biomass addition and
subtraction. In practice, unless there is evidence to the
contrary, the net migration rate, I, can be assumed equal to
zero for most freshwater fish populations. Suppose that the
rate of population change, dBt/dt, is equal to zero (i.e.
the population is in an equilibrium or steady state). From
equation IV-1 we have,
IV-2 dBt/dt : G + R - M - F = 0,
which, when rearranged gives,
IV-3 F = G + R - M.
The right side of equation IV-3 indicates that, in
equilibrium, the rate of population decrease due to fishing
is exactly offset by the aggregate rates of growth,
recruitment and natural mortality. In equilibrium,
mortality rates are balanced by compensatory changes in the
rates of growth and recruitment. The determination of an

accurate and realistic functional form for equation IV-3

 

69

constitutes one of the most difficult problems confronting
an investigator studying the dynamics of exploited fish

populations.

B. The Logistic Surplus Production Model

 

It is usually the case, in studies of the dynamics of
exploited fisheries, that precise information concerning
rates of growth, natural mortality and recruitment is
unavailable. In such instances, investigators are often
obliged to use some form of lumped parameter model to
describe the rate of change in fish population biomass,
dBt/dt. For example, the so-called logistic growth model is
given by the following,

IV-4 dBt/dt : r(Bt)Bt,

where r(Bt) is a decreasing function of population size, Bt'
In accordance with conventional assumptions regarding
density dependent population growth, r(Bt) is given by,

IV-5 r(Bt)

where: r the intrinsic population growth rate

K = maximum population size.
Equation IV-5 substituted into equation IV-4 results in,
IV-6 dBt/dt : rBt(1 - Bt/K),

with solution given by

_ rt
IV-7 Bt - BO Ke .

 

rt
K - 80(1 - e )
Now, the rate of change in catch, Ct’ during time, A t,
can be described by,

70

IV-8 dCt/dt qf B

t t’
where: q = catchability
ft = effort during time, At.

The integrated form of equation IV-8 is the short term
production function represented by equation III-12.

By subtracting equation IV-8 from equation IV-6 we get
an expression for rate of change in stock biomass as a
function of growth rateg r(Bt)Bt, and harvest rate, qftBt’
or

IV-9 dBt/dt = rBt(1 - Bt/K) - qftBt’

At equilibrium, this equals zero, implying that,

IV-1O qftBt = rBt(1 - Bt/K),

which, upon solving for B results in the following nonzero

t,
equilibrium,

IV-11 Bt = (K/r)(r - qft).

If we substitute equation IV-11 into equation III-12 we get,

IV-12 Ct : qftBt

qft(K/r)(r - qft)

qftK - (K/r)(qft)2
qK(ft) - q2(K/r)(ft2),

‘which describes the long term equilibrium relationship
between catch, Ct’ and effort, ft' If we differentiate
equation IV-12 with respect to effort” ft’ set the result

equal to zero and solve for f we obtain,

t,

which is the amount of fishing effort necessary to produce

the maximum sustainable physical yield.from the exploited

stock. ILf we substitute equation IV-13 into equation IV-12

71

we obtain,

IV-14 Ct = qu/2q — q2(K/r)(r2/4q2)
: Kr/2 - Kr/4
= rK/4,

which is the maximum physical yield or MSY (maximum
sustainable yield).

The following linearization procedure, adopted from
Jensen (1976), was used to estimate the parameters of
equation IV-9. From equation III-12, mean exploitable stock

biomass during year t is given by,

IV-15 Bt : Ct/(qft)
= (1/q)(Ut).
where, Ut = Ct/ft'

A rough approximation of the derivative in equation IV-9 is,

IV-16 dBt/dt = (Bt+1 - Bt_1)/2
= (1/q)(Ut+1 - Ut_l)/2
: (1/(1) AUt’
A - _
where, Ut - (Ut+1 Ut-1)/2'

By substituting equations IV-15 and IV-16 into equation IV-9

we get,
_ 4 A' - - -
IV 17 (./q) Jt - (r/q)(Ut) [1 (Ut)/(qK)] Ct‘
By multiplying through by q we obtain,
- A - _ -
IV 18 Ut - (rUt) [1 (Ut)/(qK)] th,

which can be written as the following linear equation,

VI-19 Y = a
t 1X1 + a2X2 + a3X3,
2
. - A . _ e _ . _ . - o -
where. Yt - Ut’ X1 _ Ut’ X2 - (Ut) , X3 - Ct’ a1 - r, a2 -

-r/(qK) and a3 : -q. The constants a1, a2 and a3 can be

72

estimated by least squares multiple regression analysis.

The following examples from the fisheries literature
are indicative of the widespread use and application of
surplus production models to commercially exploited
fisheries: Schaefer (1968); Fox (1971); Walter and Hogman
(1971); Walter (1973); Fox (1975); Walter (1975); Jensen
(1976); Hilborn (1979); McGaw (1980); Deriso (1980); Gatto
and Rinaldi (1980); Mohn (1980) and Uhler (1980). Surplus
production models, particularly the logistic surplus
production model, have been utilized to describe and manage
virtually every significant commercially exploited fishery
in the world. The attractiveness of surplus production
‘models is attributable to their simplicity and minimal data
requirements (i.e. yearly total catch, Ct’ and nominal
1effort, ft data). The simplicity of these models, however,
is overshadowed by the restrictive assumptions necessary for
their successful application. Some of the critical
assumptions cfi'logistic surplus production models are that:
(1) the time lag between spawning and subsequent recruitment
has a negligible effect on population growth: (2) each unit
<3f biomass in the exploited stock is identical with respect
to growth and mortality factors (i.e. the age structure of
the stock is unimportant); (3) the model is applied to a
<3Losed homogeneous stock; (4) the exploited stock is in a
state of equilibrium with respect to the various factors

which affect stock biomass change and (5) the catch per unit

73

effort ratio, Ct/ft’

population biomass, Bt'

is an accurate and reliable proxy for

Surplus production model users have, in recent years,
identified the need to integrate biological data on the
processes of growth, mortality and recruitment with
traditional catch and effort statistics. The combination of
biological data with catch and effort data is necessary,
according to Jensen (1976) and Hilborn (1979), to confimn
and expand upon results obtained from surplus production
models. Dependence upon the catch per unit effort ratio, in
one form or another, as a proxy for biological variables
results in models whose independent variables lack
sufficient contrasts. The usefulness of multiple regression
estimation systems may be severely compromised by the
presence of independent variable multicolinearity. Surplus
production models estimated by linear or nonlinear
techniques continue to enjoy widespread use despite such
potential statistical intractabilities.

Deriso (1980) describes a difference equation stock
production model which incorporates age-structure,
recruitment, survival and growth in an attempt to estimate
stock dynamics parameters and optimal harvesting strategies.
A similar approach is employed by Gatto and Rinaldi (1980)
in their effort to develop a difference equation stock
production model for the Gulf of Venezuela commercial

fishery. These models are, perhaps, more realistic than

 

74

other surplus production estimation systems in that they
recognize the necessity for including biological information
directly into the estimation algorithm. The shortcomings of
these efforts stem from the fact that they rely too heavily
on the catch per unit effort ratio, Ck/fi, to approximate
the biological processes of growth, survival and
recruitment.

The effects of random variability introduced into catch
and effort time series data on surplus production model
parameter estimates are explored by Fox (1971), Uhler (1980)
and Mohn: (1980). These studies emphasize the difficulty of
obtaining useful results from surplus production models when
the data are subject to varying degrees of instability.
Stock production model estimates of optimal effort and
maximum sustainable yield (MSY) are statistical point
estimates and, as such, may not be significantly different
from zero. Therefore, when using any of the stock
production model variants, according to McGaw (1980), care
should be exercised in the interpretation of the results for
purposes of fisheries management.

The proper perspective of surplus production model
analysis, according to Fox (1975), is that it is little more
than a regression model, yet often very useful for making
"first estimate" projections of the relationships between
the level of exploitation and expected equilibrium yield.
The assumption of equilibrium conditions precludes

application of surplus production models to either the

 

75

ascending or descending stages of a fishery. Hilborn (1979)
observes that fisheries control systems that utilize catch
and effort data frequently fail to provide reasonable
estimates of logistic growth model parameters and will
produce poor catches, especially when managing long-lived
low productivity species. Christie (1974) asserts that
catch and effort approaches to population estimation appear
to be too slow and retrospective for dynamic systems like
those of the Great Lakes. It would seem, then, that the
applicability of stock production models to real world
dynamic fishery systems is extremely limited. However, due
to the all too prevalent lack of comprehensive biological
data, these models are often the only available means of
obtaining information about exploited stock dynamics. With
this in mind, we will proceed to apply the logistic surplus
production model to Saginaw Bay commercial sucker fishery

catch and effort data.

C. Application of the Logistic Surplus Production Model
Recall equation lV-12,

- _ 2 2
1v-12 ct - qK(ft) - q (K/r) (ft ),

which suggests that, under equilibrium conditions, a

parabolic relationship exists between equilibrium yield, Ct’
’ 2

and nominal effort, ft' If we let a = qK and b = q K/r,

then equation IV-12 can be written,

_ 2
IV-20 Ct - a(ft) - b(ft ).

If we divide through by ft we obtain,

IV-21 Ct/ft : a - b(ft)’

 

76

which indicates another means of surplus production model
estimation (under assumed equilibrium conditions).
Differentiating equation IV-20 with respect to nominal
effort, ft’ setting the result equal to zero and solving for
ft yields,

IV-22 fMSY = a/(2b),

which is the optimal level of effort (i.e. the level of
nominal effort which results in MSY). By substituting
equation IV-22 into equation IV-20 we obtain an expression
for maximum sustainable yield (MSY).

IV-23 CMSY = a2/(4b).

Recall the linearization procedure (equations IV-15
through IV-19) used to estimate the logistic surplus
production model (equation IV-9). In order to estimate
population growth parameters from time, t to t+n, catch and
effort data is needed from time, t-1 to t+n+1. For example,
catch and effort data from 1929 to 1956 is required to
obtain growth parameter estimates for the period from 1930
to 1955. In effect, the linearization procedure "swallows"
the first and last years' data in order to derive population
growth parameter estimates for the intervening years. The
equilibrium-centered model represented by equations IV-2O
through IV-23 uses all of the available data. For
example, catch and effort data from 1929 to 1956 is used to

obtain estimates of optimal effort and MSY for the same time

period.

 

77

Equations IV-19 and IV-21 were applied to catch and
effort data from the following time periods: 1930 - 1945;
1946 - 1955; 1930 - 1955; 1961 - 1967; 1968 - 1978 and 1961
- 1978. Population dynamic parameter estimates resulting

from these applications are listed in Table IV-1.

1. 1930 to 1945

Although both the multiple regression and simple linear
regression representations of the logistic surplus
production model yielded parameter values with the expected
signs, the estimates of fMSY and MSY seem too large when
compared to observed mean catch, Ct, and mean effort, fk,
during this period. In addition, the estimate of
catchability obtained from the linearization method, q = 2.3
x 1077, appears to be too small when compared to independent
estimates of catchability for this period (Table III-3).
The surplus production model fails to adequately describe

observed catch and effort data from 1930 to 1945.

2. 1946 to 1955
The linearization procedure (equation IV-19) failed to
generate parameter estimates with the expected signs.
Estimates of fMSY and MSY obtained from equation IV-21 were
reasonably close to observed mean catch, Et’ and mean effort

ft, during this period.

3. 1930 to 1955
Both estimation procedures yielded parameter estimates
with the expected signs. The estimates of fMSY and MSY

obtained from both methods are in relatively close agreement

78

Table IV-l
Surplus Production Model Results
1930-1945
t r(Ut) - r/(qK)(Ut2) - q(ct>
.101186; r/(qK) = .003739; q = 2.3 x 10‘7

 

AU

'1
II

 

 

K = 117,662,000 pounds; fMSY = r/(2q) = 220,000 lifts
MSY = rK/4 = 2,976,000 pounds
2 2 _ 2
Ce - qK(ft) - q K/r(ft ) - 27°0623(ft) - .00006151(ft )
Ct/ft : a - b(ft) : 28.2137 - .00004521(ft)
f = a/(2b) : 312,000 lifts; MSY : a2/(4b) : 4,402,000
MSY
pounds
E, = 962,000 pounds; 7; = 36,300 lifts
1946-1955
Ct/ft : a - b(ft) : 58,3428 - .000075513(ft)
fMSY : a/(2b) = 38,600 lifts; MSY = a2/(4b) = 1,127,000
pounds
E, = 900,000 pounds; f, = 22,000 lifts
1930-1955
_ 2
AUt - r(Ut) - r/(qK)(Ut ) - q(Ct)
r = .092137; r/(qK) = .000961; q : 1.5 x 10‘6
K : 63,917,000 pounds; fMSY = r/(2q) : 30,700 lifts
MSY : rK/4 : 1,472,000 pounds
_ 2 2 _ 2
Ce - qK(ft) - q K/r(ft ) - 95.8762(ft) - .00156087(ft )
Ct/ft : a - b(ft) : 54.4734 - .00071726 (ft)
fMSY = a/(2b) = 38,000 lifts; MSY = a2/(4b) = 1,034,000
‘ pounds
C, : 938,000 pounds; ft : 31,000 lifts

79

Table IV-l (cont.)

 

1968-1978
2
AUt _ r(Ut) - r/(qK)(Ut ) - q(Ct)
r = .313292; r/(qK) = .001112; q = 7.2 x 10‘5
K = 3,913,000 pounds; f = r/(2q) = 2,200 lifts

MSY
MSY = rK/4 = 306,000 pounds

c = qK(ft) — q2K/r(ft2) = 281.7374(ft) - .06474821(ft2)

ct/ft = a — b(ft) = 47.7471 - .00404820(ft)
fMSY = a/(2b) = 5,900 lifts; MSY = a2/(4b) = 141,000
pounds
E; = 123,000 pounds;‘?£ = 3,800 lifts

 

80

with observed mean catch, Ct’ and mean effort, ft’ during
this period. Figure IV-1 depicts the equilibrium
relationship between catch and effort generated by
application of equation IV-21. Also shown in Figure IV-1 is
the short term production function obtained by applying
equation III-19 to this time series of catch and effort
data. The surplus production model appears to be a
reasonable approximation of the relationship between catch
and effort data from 1930 to 1955. From this evidence, one
could infer that the Saginaw Bay commercial sucker fishery
was in a state of equilibrium during this period. If this
calculated relationship reflects potential stock dynamics,
then an upper bound (from the standpoint of MSY) for annual
sucker harvest from Saginaw Bay is approximately 1,000,000

pounds.

4. 1961 to 1967
Both estimation procedures failed to generate parameter
estimates with the expected signs. Realistically, this
period is represented by too few data points to warrant the

application of surplus production modeling techniques.

5. 1968 to 1978
Both estimation procedures yielded parameter estimates
with the expected signs. The estimates of fMSY and MSY
obtained from both methods are within the neighborhood of

observed mean catch, Ct’ and mean effort, ft.

81

 

   

 

 

 

32
l2 36 .
3|
34
, 4—0, = 29.”,
'54 33
IO-
. ‘36
. 37
43
. 45
4| ’ .
8- as 35
Conn 39 44
(lo5 lbs) .
42
6.
54.473414, - 0.00071726f?
4 .
2 l.
O J L 1 1 1_
o lo 20 30 4o 50
Efflxi '
(lo3 lifts)

Figure IV-l. LogiStic Surplus Production Model
(1930-1955) .

82

Interpretation of these results is complicated by the fact
that this period is represented by relatively few data
points. Figure IV-2 depicts the relationship between catch
and effort data generated by the multiple regression
linearization procedure» Also shown in Figure IV-2 is the
short term production function obtained by applying equation
'III-19 to these catch.and effort data. Three observations
can be made about the relationship portrayed in Figure IV-2:
(1) the fishery was not in equilibrium; (2) the fishery was
overexploited; or (3) the logistic surplus production model
is an inappropriate descriptor of stock dynamics for this

period.

6. 1961 to 1978

 

Both estimation procedures failed to generate parameter

estimates with the expected signs.

D. Saginaw Bay Sucker Fishery Economics

 

The preceding analysis of catch and effort data via the
logistic surplus production model is an attempt to determine
a workable production function which can be used in an
economic analysis of the fishery. This section describes a
simple bioeconomic fishery management model based upon the
logistic surplus production model. The following economic
analysis is patterned after the presentations by Fullenbaum
and Bell (1974) and Anderson (1977). This analysis depends
upon a reasonably accurate determination of the cost and
revenue curves for the Saginaw Bay commercial sucker

fishery. Cost and revenue data for 1979 were obtained from

83

  
    

 

 

 

304
e—c, = 23l.7374t, - 0.0647482le
20 .
Catch ,
(104 lbs) 6' 73
74 °

l0 . 7"
o l l l l J

0 IO 20 30 4o 50

Effort
(lo2 lifts)

Figure IV-2. Logistic Surplus Production Model
(1968—1978).

 

84

an eastern Saginaw Bay commercial fishery operation which
encompassed two full time fishing vessels.

Table IV-2 contains a description of a typical Saginaw
Bay commercial fishery operation for the year, 1979.
The total costs of $88,639 represent the operating cost
exclusive of operator labor for the two fishing vessels.
This operation must earn at least this amount plus a
moderate return to the operators' labor and capital or, in
the long run, it will be forced to cease fishing. Since
this operation fishes 6O trap nets, the average opportunity
cost of a trap net is ($88,639/60 nets) or approximately
$1,477. During 1979, each vessel was fished an averagelof
175 days. Approximately 3 nets were lifted per boat day.
An estimate of the total amount of nominal fishing effort,

f produced by this operation is:

t’
2 boats x 175 days fished/boat x 3 lifts/day fished :
1050 lifts.

An estimated yearly average of 17.5 lifts per trap net is
obtained by dividing 1050 lifts by 60 nets. The average
cost per net lift is ($1,477/17.5 lifts) or approximately
$84. In 1979, suckers comprised about 6% of the total
landings (in pounds) of this operation. Assuming that the
cost of landing a given species is directly proportional to
its percentage composition in the total landed catch, the
cost of one trap net lift for suckers isffiflltimes .06 or

about $5.04. The total cost of fishing for suckers can be

expressed as a function of nominal fishing effort, ft:

 

85

Table IV-2
Cost and Revenue Data

Description: 2-40 ft. steel-hulled trap net boats; 190 h.p.
engines; 60 nylon trap nets (6 ft. and 10
ft.); power spools; net pullers; ship to
shore radios; total est. market value:
$70,000.

1979 Production (2 boats)

 

Species Total Lbs. Avg. Price/Lb. Total Return
($) ($)
Catfish 115851 0.50 57925.50
Carp 58162 0.10 5816.20
Quillback 38150 0.15 5722.50
Sheepshead 33158 0.15 4973.70
Perch 22547 0.55 12400.85
Suckers 17655 0.05 882.75
Whitefish 10296 0.90 9266.40
Other Spp. 5000 0.30 1500.00
-Totals 300819 Lbs. $98487.90

Variable costs

 

crewshare; net materials and repair; vessel maint. and
repair; fuel and oil; ice; boxes; misc.
$41365.00

Fixed costs
depreciation; license fees; mortgages; utilities;
insurance; misc.

 

$47274.00

Total costs $88639.00

 

898487.90
-a6_32_09

Returntm>labor, management and investment $ 9848.90

 

86

IV-24 TC : $5.04(ft).
In section C of this chapter we obtained the following
estimate of sustainable yield in terum of nominal fishing
effort for the 1968 to 1978 period:
IV-25 ct = 281.7374(ft) - .06474821(ft2).
This relationship was used to describe the dynamics of the
Saginaw Bay sucker fishery during 1979.

.Average cost in terms of’physical yield is calculated
by dividing equation IV-24 by equation IV-25;
IV-26 AC : $5.04/(281.7374 - .06474821ft).
If we solve equation IV-25 for effort, ft, by means of the
quadratic equation, we obtain:
IV-27 ft: {-281.737 1 (79375.9626 - .259oct)‘/2)/(-.1295).
If we substitute equation IV-24 into equation IV-26 we
«obtain an expression for average cost in terms of physical
yield:
IV-28 At = $10.08/(281.7374 s (79375.9626 - .259oct)‘/2).
By substituting equation IV-27 into equation IV-24 and
differentiating the result with respect to Ct we obtain the
following expression for marginal cost in terms of physical
yield:
1v-29 MC : $5.04/(79375.9626 - .259oct)1/2.

Table IV-3 and Figure IV—3 show the relationship
‘between total physical yield, Ct’ and average and marginal
costs per pound. As the amount of effort applied to the

fishery increases, total physical yield eventually falls

lflnile average cost per pound increases since more money is

 

87

Table IV-3

Average and Marginal Cost Data

Total Catch Nominal Effort Average Cost Marginal Cost

(lbs) (trap net lifts) per Pound per Pound
($) ($)

66400 250 .019 .020
124700 500 .020 .023
174900 750 .022 .027
217000 1000 .023 .033
251000 1250 .025 .042
277000 1500 .027 .058
294700 1750 .030 .091
304500 2000 .033 .223
306500* 2176 .036 infinity
306100 2250 .037
299700 2500 .042 -
285100 2750 .049 -
262500 3000 .058 -
231700 3250 .071 -
192900 3500 .091 _
146000 3750 .129 -
91000 4000 .222 -
27900 4250 .769 -

* MSY

88

 

 

 

 

 

 

CL304
0.20 «
Con
'($/Ib)
Marginal Cost
0 O '0‘
.mmnnerfl
I
OJDO ' L ‘ L
O IOOOOO 200000 300000 400000

Quantity (lbs)

Figure IV-3. Average and Marginal Cost of the
Saginaw Bay Commercial Sucker Fishery, 1979.

89

spent to obtain less yield (Anderson, 1977).

We know from the theory of open access fishery
economics that the open access equilibrium or zero economic
rent point will occur when the demand curve for suckers
intersects the average cost curve. Due to the relative
unimportance of the current Great Lakes sucker fishery,
adequate information is not available to estimate a reliable
sucker demand curve. In such instances, according to
Anderson (1977), a simple fixed price model can be used in
place of an explicitly derived demand curve. According to
data obtained from the Michigan Deparment of Natural
Resources' Fisheries Division, the ex-vessel pricewof
suckers from Michigan waters of Lake Huron has remained
relatively constant at about $0.05 Per pound over the past
twenty years. The fixed price of $0.05 Per pound will be
used in all subsequent bioeconomic calculations concerning
the Saginaw Bay commercial sucker fishery.

The P = $0.05 line intersects the average cost curve
shown in Figure IV-3 on its backward bending portion where
approximate fishery yield is 282,500 pounds. Substituting
this figure into equation IV-27 we find that this
corresponds to an effcrt level of approximately 2,800 trap
net lifts. This is very close to the 1979 observed level of
2,590 trap net lifts. The expected fishery yield (i.e.
282,500 pounds), however, is more than two and one half
times the reported catch of approximately 108JMM1pounds.

Conversations with commercial fishermen indicate that only

 

9O

one-fourth to one-third of the current sucker catch is
landed due to the lack of profitable markets for these fish.
If the reported 1979 sucker landings for Saginaw Bay
represented approximately one—third of the actual catch,
then anl estimate of the 1979 total catch is approximately 3
x 108,000 pounds or 324,000 pounds. This figure is very
close to the above estimated equilibrium catch of 282,500
pounds.

Maximum economic yield (MEY) occurs at the intersection
of the P = $0.05 line with the marginal cost curve shown in
Figure IV-3. According to Figure IV-3, MEY occurs when
yield is approximately 267,000 pounds. Substituting into
equation IV-27 we find that this corresponds to an effort
level of approximately 1,400 trap net lifts. The total
economic yield at this point is the difference between the
selling price (i.e. $0.05/lb) and the average cost (i.e.
$0.03/1b) multiplied by the total yield or about $5,340.

The total revenue curve for the 1979 Saginaw Bay
commercial sucker fishery is obtained by multiplying
equation IV-25 by $0.05:

IV-3O TR

.05(281.737ft - .06474821ft2)
14.0869ft - .00323741ft2.

Recall equation IV-24:

IV-24 TC : 5.04(ft).

Open access equilibrium occurs at the level of effort where
total revenue, TR, equals total cost, TC. Equating

equations IV—24 and IV-3O and solving for effort, ft, we

 

91

find that open access equilibrium occurs when total fishery
effort equals approximately 2,800 trap net lifts. Figure
IV-4 depicts the relationship between total revenue and
total cost. Maximum economic yield occurs at that effort
level where the slope of the total revenue curve equals the
slope of the total cost curve. Differentiating equation IV-

30 with respect to effort, f and setting the result equal

t?
to 5.04, we find that maximum economic yield occurs when
effort equals about 1,400 trap net lifts. Substituting this
effort level into equations IV-24 and IV-3O and subtracting

equation IV-24 from equation IV-30 yields a maximum economic

rent estimate of approximately $6,320.

E. Discussion

 

The preceding abbreviated bioeconomic analysis of the
Saginaw Bay commercial sucker fishery is fraught with
complications. Recall the multispecies nature of the
Saginaw Bay commercial fishery. Any analysis of a single
exploited population within a multispecies fishery is
contingent upon a number of sweeping assumptions. The
assumptions of biological equilibrium conditions, essential
to bioeconomic analyses, refer not only to the exploited
population of interest, but also to its total environment.
The components of an exploited population's environment
include: the genetic composition of the population's gene
pool; the physical environment; the biological elements of
the ecosystem and the commercial fishery. The interactions

between an exploited population and its environment

92

 

  

 

 

 

<-TC = 5.04t,
l5;
lo-
6057/ Revenue
003 3)
5 4
TR = l4.0869t, - 0.0032374”? :
O . i l L
0 I0 20 30 4o
Effort
(lo2 lifts)

Figure IV-4. Cost and Revenue in Terms of Effort.

93

determine the aggregate dynamic characteristics of that.
stock. Bioeconomic analyses of exploited fisheries are
often conducted with limited information. Armed only with
commercial fishery catch, effort, cost and revenue data plus
certain broad generalizations concerning aggregate
population biomass growth (e.g. logistic growth), biologists
and economists routinely construct simple lumped parameter
fishery management models. While such models may reasonably
describe the dynamics of single species fisheries in non-
fluctuating environments, they almost certainly fail to
explain the complex dynamics of multispecies fisheries.
Multispecies fisheries require complex methods of analysis.
Unfortunately, techniques for biological and economic
analyses of multispecies fisheries are poorly developed (cf.
Clark, 1976 and Anderson, 1977).

Most efforts to obtain dynamic biological and economic
information from multispecies fisheries involve attempts to
determine sustainable yield curves for each species within
such fisheries (c.f. Walter and Hogman's (1971) study of the
Green Bay, Lake Michigan fishery). Frequently, the catch
and effort data from one or more exploited species fail to
fit the analytical model (e.g. surplus production model)
used to describe the fishery. In such cases, independent
estimates of MSY and allowable effort obtained by other
means are necessary to avoid gaping holes in the analysis.
In some instances, individual sustained yield curves

calculated for different exploited stocks are<xmuflned to

94

generate a single sustained yield curve whose often complex
shape defies analytical description. Optimization procedures
(e.g. linear and nonlinear programming, linear and nonlinear
least squares analysis, gradient analysis) can be utilized
in efforts to maximize aggregate biomass yield from
multispecies fisheries.

The costs involved in obtaining the requisite
biological and economic information necessary for
construction of more accurate and reliable fisheries
management models are, at least in the case of the Great
Lakes commercial fisheries probably out of proportion to the
value of the fishery. Obviously, the analysis and
evaluation of multispecies fisheries isznlincredibly
complex task. With this in mind, what can we say about the
reasonableness of our bioeconomic analysis of the Saginaw
Bay commercial sucker fishery?

The catch-effort relationship described by equation IV-
25 and shown in Figure IV-2 is the basis for our analysis.
The reliabiliy of this relationship insofar as it describes
assumed equilibrium fishery dynamics is questionable due to
the relatively small number of data points used to estimate
the equation. The cost and revenue data shown in Table IV-2
are assumed to reflect the average economic condition of the
Saginaw Bay commercial fishery. Commercial fishery catch,
C and nominal effort, f

t’ t’
since they do not reflect actual catch and effort

data are not entirely reliable

conditions. The propriety of'analyzing a single exploited

95

species apart from the other principally exploited species
may be questioned.

It is apparent that our bioeconomic analysis of the
Saginaw Bay sucker fishery contains several weak links.
However, it is possibly one of the few means available to
obtain biological and economic information about this
fishery, given the type and quality of existing data.

The logistic surplus production model, applied to catch
and effort data from 1930 to 1955, indicated a potential
sustained annual harvest of approximately 1,000,000 pounds.
This estimate should be considered as an upper limit to
sucker exploitation since environmental and biological
changes which have occurred sinceithe mid-1950's have
probably altered the potential productivity of Saginaw Bay
suckers. The logistic surplus production model, applied to
catch and effort data from 1968 to 1978, indicates a
potential sustainable annual harvest of approximately
300,000 pounds. Current annual sucker landings are about
100,000 pounds. Based upon these considerations, and
assuming that the market for suckers or sucker products can
be expanded, a reasonable projection of annual sucker
harvest from Saginaw Bay is about 500,000 pounds.

Assume that an increased market for suckers drives up
the ex-vessel price received by fishermen from $0.05/lb to
$0.10/lb. This would have the effect of raising the total
revenue curve shown in Figure IV-4. In the short run,

assuming that costs do not change appreciably and that

96

effort remains at approximately 3,000 trap net lifts, the
fishery will experience a net economic rent of approximately
$11,000. This represents an additional net annual income of
from $1,000-2,000 for each of the current Saginaw Bay
commercial fishing operations. A $0.05/lb ex-vessel price
increase should result in relatively substantial monetary
benefits, in the short run, to Saginaw Bay commercial
fishermene Ihl the long run, under the assumptions of open
access equilibrium, this economic rent should be dissipated.
The above economic analysis ignores the discount rate, in
effect attaching a discount rate of zero to future
anticipated fishery revenues. A more thorough dynamic
analysis of the Saginaw Bay sucker fishery would consider
the maximization of discounted net future revenues. Such
considerations are beyond the scope of our simple
bioeconomic analysis. The principal reason for the present
analysis is to obtain an estimate of expected short term
gains to the commercial fishery resulting from a modest
(i.e. $0.05 per pound) ex-vessel price increase.

The preceding analysis assumes that it costs the same
amount to catch each pound of fish, regardless of species.
The principal species currently sought by the Saginaw Bay
commercial fishery are channel catfish, yellow perch, lake
‘whitefish and carp, from which most of the fishermen's
income is derived. ‘It is reasonable to assume that most or
all of the costs associated with fishing should be allocated

to these species. Suckers, at present, are always an

97

incidental part of the commercial catch. The additional
(marginal) costs associated with landing incidental species
are probably negligible. The ex-vessel price for suckers is
approximately $0.05/1b. At this price fishermen are just
willing to land suckers. Below this price, fishermen would
probably not deem it worthwhile to bring in any portion of
their sucker catch. A plausible estimate of the marginal
<mcst associated with the Saginaw Bay sucker fishery, then,
is $0.05/lb. If current levels of sucker landings are
maintained (i.e. from 100,000-150,000 pounds per year), then
no economic rent will be generated by the sucker fishery. A
$0.05/lb ex-vessel price increase would generate an annual
economic rent ranging from $5,000 to $7JMML assuming
current levels of sucker use.

As long as suckers remain an incidental part of the
total commercial catch (i.e. they are not actively sought by
fishermen) the costs associated with landing these fish
should remain negligible. If the fishery actively seeks
these fish, then the costs associated with their harvest
will rise noticeably, necessitating a reevaluation of sucker
fishery bioeconomics. The marginal cost of landing suckers
is not likely to increase until annual landings approach

400,000 pounds.

CHAPTER V

PCB's and DDT in Saginaw Bay Suckers

A. Introduction

 

Sporadic attempts to monitor environmental contaminant
levels in underutilized Great Lakes fish have been conducted
'by various state and federal governmental agencies over the
past decade. Two of the most prominent ongoing
environmental contaminant monitoring programs are conducted
by the State of Michigan's Great Lakes Environmental
Contaminant Survey (GLECS) and the UJl.Fbod and Drug
Administration. These efforts typically involve obtaining a
small runnber of fish (e.g. one or two individuals) from the
commercial catch within a given catch district and analyzing
them for a broad spectrum of common organic and heavy metal
contaminants. Due to the limited nature of these
surveillance sampling programs, resulting data represent
litte more than statistical point estimatesznuh as such,
are not amenable to meaningful statistical analyses. More
intensive sampling programs are needed to reliably monitor
chemical contaminant levels in underutilized fish.

In order to ensure compliance with existing public
health standards and regulations, it is necessary to

determine the levels of environmental contaminants in

98

99

underutilized fish prior to increasing the SCOpe of fishing
operations for such species. Two of the most widespread
environmental contaminants for which allowable tolerance
limits have been established are PCB's (polychlorinated
biphenyls) and DDT (dichlorodiphenyltrichloroethane).
According to the U.S. Food and Drug Administratunh Great
Lakes fish marketed for human consumption may contain no
more than 5 ppm PCB's and 5 ppm DDT and its analogs, DDE and
DDD. This investigation included provisions for
establishing background.information on the levels of PCB's
and DDT compounds in Saginaw Bay sucker fillets. The sum of
the concentrations of DDE, DDD and DDT is designated by ZDDT
in the following discussions.

Since most halogenated hydrocarbon environmental
contaminants are fat soluble, they are usually found
concentrated in the fatty tissues of fish and other animals.
Accordingly, fish with high amounts of body fat (e.g. lake
trout, channel catfish) contain relatively greater amounts
of such contaminants than fish with low amounts of body fat
(e.g. yellow perch, suckers). Previous investigations
(GLECS,1974 and 1975; Dawson, et al, 1978; Zabik et a1,
1978) revealed that suckers obtained from Lake Huron had an
approximate fat content cfi'2%. Fat content is not uniform
throughout the organism and is often found concentrated in
the belly flap tissues and the medial muscle of the lateral

line.

 

100

B. Sampling Procedures ~

 

A total of 25 suckers (4 immatures, 9 females and 12
males) were obtained from the Fall 1979 and Spring 1980
Saginaw Bay commercial sucker landings. These fish were
selected to represent thetrange of sizes commonly found in
the commercial catch. Table V-l contains the size and sex
composition of this sample. Fish were transported in ice to
the laboratory for processing, usually arriving the day
after the catch. Fillets were homogenized in an Osterizer
blender, frozen and held in glass jars at ‘23O.C prior to

being thawed for contaminant analysis.

C. Analytical Procedures

 

Analyses of Saginaw Bay suckers for PCB's and EDDT were
conducted by the Department of Food Science, Michigan State
University. The ensuing analytical procedures, described by
Zabik, et a1, (1978), were followed. Two samples from each
fish were extracted separately with hexane-acetone (2:1),
partitioned with acetonitrile, and subjected to Florisil—
Celite column cleanup. Solids were determined by drying 2-g
samples under vacuum at 90°C to constant weight; lipid was
estimated by evaporating an aliquot of the hexane extract to
dryness at 700 under vacuum. Gas chromatographic analyses
were performed with a Tracor 560 gas-liquid chromatograph
(GLC) equipped with a 63Ni electron-capture detector and
interfaced with a Digital PDP-8e-Pamila GC data system.

Instrument parameters and operating conditions follow.

101

Table V-1

PCB's and EDDT in Saginaw Bay Suckers

Sampling Date Sex TL(cm) WGT(g) PCB's(ppm) zDDT(ppm)
9-6-79 M 32.0 400 0.08 0.010
9-6-79 F 47.9 1220 0.04 0.043
9-26-79 IM 27.5 200 0.03 0.000
9-26-79 F 50.5 1115 0.05 0.020
10-28-79 IM 32.3 550 0.08 0.001
10-28-79 M 36.3 535 0.01 0.003
10-28-79 M 40.5 740 0.03 0.010
10-28-79 F 45.8 1145 0.18 0.010
11-10-79 M 41.5 660 0.02 0.017
11-10-79 F 49.7 1150 0.05 0.007
4-9-80 M 36.8 560 0.04 0.003
4-9-80 M 45.2 985 0.03 0.044
4-9-80 F 44.0 925 0.05 0.012
4-9-80 F 45.7 1100 0.17 0.030
4-26-80 IM 39.6 590 0.06 0.002
4-26-80 M 41.7 830 0.05 0.015
4-26-80 F 44.1 1075 0.05 0.007
5-15-80 IM 38.6 580 0.01 0.005
5-15-80 M 45.8 960 0.02 0.007
5-15-80 M 46.5 940 0.03 0.040
5-15-80 F 49.6 1225 0.01 0.010
6-14-80 M 31.5 355 0.03 0.016
6-14-80 M 33.4 350 0.03 0.004
6-14-80 M 40.9 755 0.03 0.033
6-14-80 F 41.9 625 0.01 0.014
mean 0.05 0.015
std. error 0.0086 0.0027
range 0.01- 0.000-
0.18 0.044

102

Column: 1.83 m x 4.0 mm ID Pyrex, packed with 3
percent OV-l on 80-100 mesh Chromosorb
W-HP

Temperature: column 190°C

injection post 2300C
detector 300 C

Carrier gas: nitrogen flowing at 40 ml/minute

Standards were prepared with 99+ percent pure recrystallized
p,p'-DDT and p,p'-TDE, and Aroclor 1248 in Nanograde hexane.
Quantitations were based on peak area for zDDT; the area of
three peaks was used to quantitate the PCB's. Standards
were run every morning and after every eight or nine
samples. Recoveries with this method of extraction and
quantitation were 85 + 2 percent for PCB'sani92 + 1
percent for DDT compounds; limits of detection were 0.01 ppm

for PCB's and 0.001 ppm for DDT compounds.

D. Results

Table V-l presents the levels of PCB's and ZDDT found
in Saginaw Bay sucker fillets during this investigation.
PCB's ranged fhwnn13.01 ppm to 0.18 ppm. ZDDT ranged from
(3.000 ppm to 0.044 ppm. All observed levels are below the
tolerance limits for these compounds established by the U.S.
Food and Drug Administration. Approximate 95% confidence
limits for PCB's are (0.03 ppm, 0.07 ppm). Approximate 95%
confidence limits for EDDT are (0.009 ppm, 0.021 ppm). In
1974, the Great Lakes Environmental Contaminant Survey
analyzed four suckers from Saginaw Bay ranging in total
length from 30.5 cm to 45.7 cm and in weight from 320 g to

1,140 g. PCB's ranged from less than 0.02 ppm to 0.84 ppm.

103
XDDT ranged from 0.01 ppm to 0.17 ppm. Zabik, et al, (1978)
sampled suckers from Saginaw Bay and Lake Huron from 1975 to
1976. These workers found 0.03 ppm to 0.20 ppm ZDDT and
0.17 ppm to 0.70 ppm PCB's in mechanically deboned sucker
flesh. From this evidence it appears as though the levels
of PCB's and EDDT are decreasing in Saginaw Bay suckers.

The data presented in Table V-l can be partitioned into
the following categories or "treatments": fall immatures;
fall males; fall females; spring immatures; spring males and
spring females. For both the PCB and zDDT data, one-way
analyses of variance were constructed to test for
differences between mean contaminant levels. For both
contaminants, no significant differences (p = .05) were
detected between treatment means. The absence of
significant differences between the treatment means allowed
for pooling of fall and spring data which, in turn, enabled
lcalculation of approximate 95% confidence limits about mean
PCB and ZDDT values. No significant correlations were
detected between contaminant levels and fish size (measured
by length or weight).

The results obtained from the present investigation,
along with previous work conducted by the Great Lakes
Environmental Contaminant Survey and by Zabik, et a1, (1978)
indicate that Saginaw Bay suckers do not contain harmful
amounts of'the environmental contaminants, PCB's and ZDDT
'relative to the upper tolerance limits for these substances

mandated by governmental agencies. If other environmental

104

contaminants are present in comparably low amounts, then the
enhanced development of the Saginaw Bay sucker fishery
should not be hindered by high levels of potentially harmful
pollutants. Continued monitoring of a broad spectrum of
environmental contaminants in all Great Lakes fish marketed
for human consumption is essential for the health and

welfare of the fish-consuming public.

CHAPTER VI

Saginaw Bay Sucker Population Dynamics

A. Introduction

 

Historical catch records and commercial fishery catch
sampling reveal that white suckers comprise more than 90% of
the total Saginaw Bay sucker landings. Accordingly, the
present study was designed to evaluatelfluepopulathnl
dynamics of Saginaw Bay white suckers. Sampling from the
commercial catch was restricted to the east central Saginaw
Bay fishery and was conducted from September 1979 through
April 1981. A sample consisted of the entire sucker catch
from one or two shallow trap nets. Samples were preserved
in crushed ice until they could be worked up (usually the
day after the catch). Each fish was sexed, measured to the
nearest 0.1 cm (total length), weighed to the nearest5 grams
on an Accu-Weigh spring loaded single pan scale, and
examined for obvious morphological anomalies (e.g. sea
lamprey scars). In addition, the state of maturity of each
fish was noted and a pectoral fin was taken for age
determination via the fin ray technique. Sampling was
concentrated during the Spring,(April through early June)
and fall (September through November) fishing seasons when

suckers were most readily available to the commercial

105

106

fishery. All sucker samples were landed at the facilities
of the Bayport Fish Company, Bayport, Michigan. The pot end
mesh in the shallow trap nets used to obtain the sucker

samples was approximately 1 1/2 inches (stretch mesh).

B. A Brief Review of White Sucker Biology

 

Concise and informative reviews of white sucker
biology are provided by Scott and Crossman (1973) and Eddy
and Underhill (1974) from whom much of the following is
taken. White suckers are widely distributed throughout
North American lakes and rivers and, when present, are
usually found in abundance. Spawning occurs from late April
to early June in tributary streams and, at times, in the
margins of lakes adjacent to principal spawning streams.
Geen, et al, (1966) observed the beginning of spring
spawning migrations when stream temperatures reached 100 C.
Barton (1980) found that both water temperature and strewn
discharge were important factors contributing to spring
spawning migrations. From 20,000 to 50,000 demersal eggs
are scattered along the stream substrate. Spent females
return to the lake before spent males. Fry begin migrating
downstream to the lake about one month after spawning.
According to Geen, et al, (1966), survival from egg to
migrating fry may be as little as 0.3%.

Male and female growth rates are comparable up to the
time of maturity, after which there is a marked difference
in growth rates with females growing faster and larger than

lnales. After the first year of life, growth is extremely

107-

variable, ranging from 0-2 cm per year. Males usually
mature one year before females. In Ontario, according to
Scott and Crossman (1973), sexual maturity is normally
attained by the third or fourth year. Both sexes begin
spawning in appreciable numbers after reaching sexual
maturity. Spawning mortality is often low, usually less
than 20%. Maximum total length approaches 60J3cml while
maximum weight approaches 3,000 g. Maximum age is 17-19
jyears. Females usually live longer, on the average, than
males.

According to Lalancette (1977), the diet of white
suckers varies with the size of the fish, time of the year
and size of the food organism. Young of the year feed
mainly on plankton and small copepods and crustaceans.
After fish reach about 7.5 cm in total length, they change
to adult (bottom) feeding habits. Adults consume a variety
of crustaceans, insects, vegetation and detritus. Maxnmnn
food consumption and growth rates occur during June through
August. Adults do not feed during spawning.

White sucker adult survival is high. Olson (1963)-
reported a 0.87 annual survival rate for adults in a 695
hectare Minnesota lake while Cable (1967) calculated a 0.70-
0.75 annual survival estimate for adult white suckers in
South Bay, Lake Huron. Cable (1967) correlated a steady
decrease over time in numbers of'large white suckers with
increasing numbers of sea lampreys in South Bay. In a study

of 552 white suckers obtained with experimental trap nets in

 

108

Northern Lake Huron, Hall and Elliott (1954) reported that
'71%lof all fish over 38.0 cm (total length) bore at least
one sea lamprey scar.

Until white suckers reach approximately 30.0 cm (total
length) they serve as an important source of food for large
piscivorous fish and fish eating birds. After reaching
adult size, white suckers appear to have few natural enemies

other than man and the sea lamprey.

C. Saginaw Bay White Sucker Characteristics

 

1. Fall 1979-

 

During the Fall 1979 fishing season (September through
November) five samples consisting of 45 immature, 392 male
and 358 female white suckers were obtained from the east
central Saginaw Bay commercial fishery. Table VI-1 contains
summary statistics for the composite sample. The total
length (cm) and weight (g) statistics are self-explanatory.
Females are larger and more variable in size than males
which, in turn, are larger and more variable in size than
immatures.

Figure VI-1 shows the length-frequency distributions
of immature, male and female white suckers. For the purpose
of this analysis, fish are grouped into 1.9 cm length
classes (i.e. 28.0-29.9 cm, 30.0-31.9 cm, etc.). Figure
VI-1 depicts the size distribution differences between male
and female white suckers.

The bottom portion of Table VI-1 shows the calculated

weight-length relationships for immatures, males and

109

Table VI-1

Fall 1979 Summary Statistics

 

 

Statistic Immatures Males Females
Sample size (n) 45 392 358
T1 (cm) 32.3 37.3 45.3
71 (cm) 32.0 38.0 44.6
STL (°m) 2.980 3.719 5.121
TL range (cm) 25.7- 29.9 32.8
37.0 52.7 58.2
751 (g) 380 635 1030
SWGT (3) 105.149 169.560 314.700
WGT range (g) 170- 340- 370-
570 1480 1795
Fr 0.94 0.99 1.02
SKr 0.094 0.132 0.110
Kr We 1:33’ 3:38‘ 1:31”?
WGT : a(TL)b
a 0.013 0.144 0.038
b 2.968 2.300 2.680
r2 0.90 0.78 0.90
I
TL = median total length (cm)

'TL = mean total length (cm)

STL : standard deviation total length (cm)

 

1.10

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

60. Females “‘
4o n = 358 ‘7‘
I l—’ .
TL = 45.3 cm "
204 ——‘
O "—'
801
Moms
32
g 60‘ ' I'I - 392
2 1—~ TL = 37.3mm
404 ,1,
20.‘
O ,5 A==I:‘ - j=__=—._
@9292
20‘ , n=45
——i 71. = 32.3cm

 

 

 

 

 

 

 

0 ‘ $ L _
25.0 29.0 33.0 37.0 41.0 45.0 49.0 53.0 57.0
Length Class Midpoint (on)

Note. TL denotes medial total length (cm).

Figure VI-l. Fall 1979 Length—Frequency
Distributions.

111

females. The weight-length relationships are described by,

v1-1 WGT = a(TL)b,
where: WGT = weight (g)
TL = total length (cm).

The Fall 1979 population weight-length regression (which
includes all 795 fish from the composite sample) is given
by,

v1-2 Ln WGT = -3.5140 + 2.7379 Ln TL; r2 = (192.
.Analysis of covariance (Snedecor and Cochran, 1967) reveals
significant differences (p<.005) between the slopes and
elevations of the male and female weight-length regressions,
underscoring the disparity between male and female growth
patterns indicated in Table VI-1 and Figure IV-1.

Carlander (1969), discusses condition factors used by
biologists to describe the condition, plumpness or well-
being of a fish. The plumper the fish, the higher its
condition factor. The condition of a given fish is subject
to change from season to season (e.g. due to factors
associated with growth, maturity, spawning and temperature-
induced feeding changes). The following index of condition,
K, can be used to compare weight and length between
individuals of roughly similar lengths within a given sample

or season,

v1-3 K = (WGT)/(TL)3 x 100,
where: WGT = weight (g)
TL = total length (cm).

112

The following relative condition factor, Kr , is used
in this study to eliminate possible trends in condition
associated in increases in length,
VI-4 Kr = (WGT)/(aTLb).
The relative condition factor,Kr, is calculated by dividing
the observed weight of’a given fish by its expected weight
(as computed from the papulation weight-length regression).

Table VI-1 shows an increase in mean relative
condition, E; , from immatures to males to females. Figure

‘VI-2 depicts mean relative condition, E. as a function of

r’
length for males and females. Both sexes exhibit a general
decrease in Er with increasing length. Males exhibit a
more precipitous decline in Er. toward the end of their
length range than females.

Age determinations were made by examining 0.5 mm cross
sections of pectoral fin rays under a compound microscope.
The cross sections were made by a specially designed
microtome provided by the Michigan Department of Natural
Resources. Fin ray annuli are detected as regions of lesser
optical density than surrounding bony tissue. Only the
first three pectoral fin rays are suitable for aging by this
method. Cross sections must be obtained from as close Us
the base of the fin ray as possible in order to provide
readable sections. This method of aging white suckers is
detailed by Beamish and Harvey (1969) and Beamish (1973) who

observed gross inconsistencies in the scale method of sucker

aging. Scale samples from all immature and smaller (inc.

113

 

 

 

 

l.20‘ Females (-- -)
lMde( )
IJOi
(.5?ng2 \
- (33)
Kr 1 .OO‘ \\
\, 5”
(15)
0.90:
(6)
0.80 ‘ *7 5* ‘ “—

25.0 29.0 33.0 37.0 41.0 45.0 49.0 53.0 57.0

Length Class Midpoint (cm)
N010. Numbtthmmrmmthomnwwhmhmmdm.

Figure VI-2. Fall 1979 Relative Condition
Factors.

114

less than 35.0 cm total length) fish were obtained to verify
the position of the first fin ray annulus. Scales were
obtained from the lateral line region directly below the
anterior base of the dorsal fin.

Approximately one out of every three fish from the Fall
1979 sample were aged via the fin ray method. This
subsample was selected to represent the size-frequency
distribution of male and female fish (i.e. the number of
fish aged from each 1.9 cm length class was roughly
proportional to the total number of fish in that length
class). Appendix Tables A1 through A3 describe the size-age
characteristics of the Fall 1979 age sub-sample (n = 260).
IFall ages, designated by'n+, indicate that a fish of age n
is in its n+1 year of life. For the purpose of this
investigation it is assumed that Saginaw bay white suckers.
undergo annulus formation during the winter season (i.e.
from January to March). A fish of age n+ in the fall will
become age n+1 by the fcllowing spring. Ages greater than
or equal to 13+ for fall samples and greater than or equal
to 14 for spring and summer samples were pooled because of
the slow growth exhibited by these older fish and also
because fish older than 13 years were extremely difficult to
age.

The Fall 1979 female age subsample showed a progressive
increase in mean total length (TI) and mean weight (WET)
with age. The male age subsample, however, exhibited a more

irregular size-age increment, especially after age 7+.

 

115

Figure VI-3 depicts the Fall 1979 male and female length-age
relationships. The vertical lines are approximate 95%
confidence limits about the mean total lengths (T1) at each
age. The male length-age relationship was calculated by
omitting age 12+ due to the small.sample size (n = 2)
representative of this age. The E},- age relationship for

both males and females did not exhibit any obvious patterns.

2. Spring 1980

 

During the Spring 1980 fishing season (April through
May) four samples consisting of 41 immature, 95 male and 206
female white suckers were obtained. Table VI-2 contains
summary statistics for the composite-sample. The number of
lnales in this spring sample was considerably less than the
number of females. This phenomenon could be attributed to
the observation by Geen, et al, (1966), who noted that spent
females return to the lake before Spent males,lflm>have a
tendency to linger in the spawning streams. As was observed
in the Fall 1979 sample, females are larger than males
which, in turn, are larger than immatures. Comparison of
the length and weight statistics contained thables VI—l
and VI-2 reveals that the Spring 1980 fish are larger than
the Fall 1979 fish. This could be due to springtime
concentrations of large spawning individuals.

0n 4-9-80, none of the 15 males or 21 females sampled
had spawned. 0n 4-26-80, none of the 16 males but 12 of the
28 females sampled had spawned. By 5-3-80, 12 of the 15

males and all 92 of the females sampled had spawned. From

 

60.0

50.0

It (cm)

40.0

30 .0

60.0

50.0

It (cm)

40.0

30.0

 

 

116

Females

 

7: = 30.5930 (405)“978

r2 = 0.99

Males

’rT. = 31.2644 (AGEYI422

- . . (2: 0.95 ,
2+ 4+ 6+ 8+ I0+ I2+ I4+

AGE (years)

Figure VI-3. Fall 1979 Length—Age Relationships.

 

117

Table VI-2

Spring 1980 Summary Statistics

 

 

Sample Size (n)

T1 (cm)
TL (cm)
sTL (cm)

TL range (cm)

WGT (g)
SWGT(g)
‘WGT range (3)
15

r"

SK
1‘

Kr range

WGT : a(TL)b
a

b
2

Statistic Immatures Males Females
41 9 95 206
36.4 40.5 47.4
36.7 40.2 47.3
2.589 3.634 3.563
29.1- 32.6- 37.7-
42.6 46.5 55.1
530 670 1045
92.783 152.437 207.171
255- 400- 620-
720 1000 1730
0.98 0.98 1.02
0.068 0.089 0.103
0.86- 0.78- 0.80-
1.14 1.18 1.34
0.060 0.132 (0.161
2.516 2.305 2.274
0.88 0.85 0.77

l"

 

118

these observations, it appears that Saginaw Bay white
suckers spawn during the latter week of April and the early
portion of May.

Figure VI-4 shows the length-frequency distributions of
immature, male and female white suckers. Note that the
largest Fall 1979 length class for females (i.e., from 44.0-
45.9 cm) has moved into the 46.0-47.9 cm length class, which
represents the largest springtime female length class.

Weight-length relationships for immatures,nmfles and
females are shown at the bottom of Table VI-2. The Spring
1980 population weight-length regression (which includes all
343 fish from the composite sample) is,
v1-5 Ln WGT = -2.9235 + 2.5545 LnTL; r2 = 0.86.

Mean total length (TL) and mean relative condition (3})
of unspent females were 44.6 cm and 1.14, respectively. The
unspent female (n = 41) weight-length regression is,
v1-6 Ln WGT = -2.5745 + 2.4964 LnTL; r2 : 0.92.

Mean total length (TL) and mean relative condition (E;)
of spent females were 47.9 cm and 0.99, respectively. The
spent female (n = 165) weight-length regression is,
v1-7 Ln WGT = 3.1285 + 2.6031 LnTL; r2 = 0.83.
Analysis of covariance reveals a significant difference
(p<.01) between the lepes of the unspent and spent female
weight-length regressions. This is reflected by the

difference between unspent and spent femalermun1relative

condition factors.

60

40

20

60

40

20

20

 

119

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

annhs
n = 206 r—fir—
. _.
TL=47.4cm
'1
Nhhs
n==95
T',_ = 40.5cm
fil—T
,__.
Immatures
'n=4l
1.7— TL = 36.4cm
==E_‘ R
25.0 29.0 33.0 37.0 41.0 45.0 49.0 53.0 57.0
Length Class Midpoint (cm)
Figure VI—A. Spring 1980 Length-Frequency

Distributions.

120

Mean total length (TL) and mean relative condition (5;)
of unspent males were 38.3 cm'and 1.05, respectively. The
unspent male (n = 35) weight-length regression is,

VI-8 Ln WGT = -3.1170 + 2.6200 LnTL; r2 = 0.92.

Mean total length (TL) and mean relative condition (5;)
of spent males were 41.3 cm and 0.94 respectively. The
spent male (n : 60) weight-length regression is,

VI-9 Ln WGT = .2.6311 + 2.4600 Ln TL; r2 = 0.86.
Analysis of covariance indicates no significant difference
(p):: .05) between the slopes of the unspent and spent male
weight-length regressions.

Figure VI-5 depicts mean relative condition, fr, as a
function of length for males and females. Males and
females exhibit a similar E; pattern for length classes
marked by the 39.0 cm to 47.0 cm midpoints.

Appendix Tables A4, A5 and A6 describe the size-age
characteristics of the Spring 1980 age subsample (n = 313).
All of the 41 immature fish were aged as were all but one of
the 95 males. Females were aged until the length
coefficient of variation for each age class approached a
value of approximately 0.05.

The female age subsample (n = 178) demonstrated a
progressive increase in mean total length (T1) with age.
The male age subsample (n: 94), however, exhibited a more
irregular size-age relationship, similar to that represented
by the Fall 1979 male age subsample. Figure VI-6 depicts

the Spring 1980 male and female length-age relationships.

121

 

 

I.20-
Females (—— —)
Mobs( )
|.|0q
Kr 1.003
CANT
0.80

 

 

25.0 29.0 33.0 37.0 41.0 45.0 49.0 53.0 57.0
Length Class Midpoint (on)

Note. Nuttbenhpwthuuntertothewgthdmlneaehlangthdan

Figure VI-S. Spring 1980 Relative Condition
Factors.

 

122

    

 

 

 
   
 

 

60 .0-
5000‘“I
71: (cm)
Famous
40.04 1'7. = 32.5910(AGE)'|658
r2 = 0.98
3043' 4 4*
60.0‘
50.0‘
It (cm)
40.04 I
' I Males
fl: = 31.8536 (Aoer'm
r2: 0.82
30.0 A A 1 A 1 1
3 5 7 9 ll 13
AGE (year)

Figure VI—6. Spring 1980 Length-Age Relationships.

123.

The vertical lines are approximate 95% confidence limits
about the mean total length (T?) at each age. The male
length-age relationship was calculated by omitting age 13
due to the small sample size (n = 2) representative of this

age. The Kr - age relationships listed in Appendix Tables
A4 through A6 indicate that the younger age classes (iue.
from 3-7 years old) were in better condition than the older

age classes.

3. Summer 1980

 

During the Summer 1980 fishing season (June through
August) two samples consisting of 31 immature, 71 male and
'71 female white suckers were obtained. Table VI-3 contains
summary statistics for the composite sample. Once again,
the previously observed size relationship among immatures,
male and females is maintained. The fish from the summer
sample, however, are considerably smaller than fish from the
Fall 1979 and Spring 1980 samples (c.f. Tables VI-l and VI-
2).

Figure VI-7 shows the length-frequency distributions of
immature, male and female white suckers. The diminished
sizes of these summer immature, male and female fish may
reflect the fact that larger white suckers move offshore
into deeper, colder waters during the summer months where
they are less available to the commercial fishery (which
operates in relatively shallow waters during this time).

‘Weight-length relationships for immatures, males and

females are shown at the bottom of Table VI-3. The Summer

124

Table VI—3

Summer 1980 Summary Statistics

 

 

Sample size (n)

l
TL (cm)
TL (cm)
STL (cm)

TL range (cm)

Ill-5'1" (g)
SWOT (3)

WGT range (g)

‘E
1"
SK
1"

Kr range

WGT : a(TL)b
a

b

r2

Statistics Immatures Males Females
31 71 71
30.5 35.3 38.9
30.4 35.4 39.8
3.351 2.605 4.662
24.3- 29.0- 30.8-
39.2 42.5 55.8
300 450 650
95.567 96.536 219.909
145- .240- 310-
580 .755 1380
1.00 1.00 1.04
0.135 0.087 0.105
0.42- 0.82- 0.77-
1.22 1.18 1.33
0.065 0.043 0.047
2.458 2.593 2.578
0.72 0.83 0.90

Number

125'

 

 

 

 

 

 

 

 

 

60‘
Females
404 n = 71
1
TL = 38.9 cm
_,
0 ,
50' Males
n = 7|
404 .
TL = 35.3 cm
20- fi—1
0 =1 “ 'j—L
Immatures
20- 'n = 3|
__.[—— TL = 30.5cm
o ‘ l

 

 

 

 

 

25.0 29.0 33.0 37.0 4|.0 45.0 49.0 53.0 57.0
Length Class Midpoint (cm)

Figure VI-7. Summer 1980 Length-Frequency
Distributions.

126

1980 population weight-length regression (which includes all
173 fish from the composite sample) is,

VI-10 Ln WGT = -3.6797 + 2.7429 LnTL; r2 : 0.92.
Analysis of covariance reveals no significant difference (p
= .05) between the slopes of the male and female weight-
length regressions.

Figure VI-8 depicts mean relative condition, E; , as a
function of length for males and females. No obvious
trends in K; are evident.

Appendix Tables A7, A8 and A9 describe the size-age
characteristics of the Summer 1980 age subsample. Fin rays
were not Collected during the second of two summer sampling
periods (which accounts for the poor age class
representation for both males and females). Length-age

relationships were not calculated for the summer subsample

due to this inadequate age distribution.

4. Fall 1980

 

During the Fall 1980 fishing season (September through
November) five samples consisting of 76 immature, 113 male
and 153 female white suckers were obtained. Table VI-4
contains summary statistics for the composite sample. The
size statistics calculated from this sample are similar to
the Fall 1979 size statistics (c.f. Table VI-1).

Figure VI-9 shows the length-frequency distributions of
immature, male and female white suckers. The Fall 1979
sample (Figure VI-1) indicated that the dominant male size

class was from 36.0 -37.9 cm. Figure VI-9 shows the

127

 

 

 

 

1.20“
Females (-——)
Mahs( )
LIO‘
Kr 1.00-
0.90.
0.80

25.0 29.0 33.0 370 41.0 45.0 49.0 53.0 57.0
Length Class Midpoint (cm)
mmmma letttothemmberothclmlneodllutgthclw.

Figure VI-8. Summer 1980 Relative Condition
Factors.

128

Table VI-u

Fall 1980 Summary Statistics

 

 

Statistic Immatures flalgg Females
Sample size (n) 76 113 153
Ti (cm) 33.2 39.2 43.7
¥i (cm) 32.6 39.6 40.0
STL (cm) 4.179 3.367 4.288
TL range (cm) 23.4- 32.6- 34.6-
40.“ 50.8 53.1
W's—f (g) 365 6115 920
SWGT (3) 124.396 199.493 254.190
WGT range (g) 125- 350- 460-
635 1230 1580
fir 0.96 0.99 1.02
SKr 0.076 0.099 0.097
K range 0.65- 0.68- 0.75-
1.12 1.30 1.35
WGT = a(TL)b
a 0.017 0.077 0.033
b 2.85” 2.u51 2.701
r2 0.96 0.83 0.89

129

60- .
Females

40‘ n = I53

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

'l"L = 43.7crn
201
0
E 66.
g Males
Z
401 n = ”3
I
TL = 39.2 cm 7—
o a
We;
20* "—11—7 'n = 76
__ __ —“ TL= 33.20tn
O I ‘ _—L=

 

 

 

 

 

 

 

25.0 29.0 33.0 37.0 4|.O 45.0 49.0 53.0 57.0
Length Class Mldpolnt (cm)

Figure VI-9. Fall 1980 Length-Frequency
Distributions.

130
dominant Fall 1980 male size class to be from 38.0-39.9 cm.
This could reflect the movement of a dominant year class
through the fishery.

The bottom portion of Table VI-Ll shows the calculated
weight-length relationships for immatures, males and
females. Analysis of covariance indicates no significant
difference (p = .05) between slopes of the Fall 1979 female
and Fall 1980 female weight-length regressions. Comparison
of the slopes of the Fall 1979 male and Fall 1980 male
weight-length regressions reveals a significant (p = .05)
difference. The Fall 1980 population weight-length
regression (which includes all 3142 fish from the composite
sample) is given by,
v1—11 Ln WGT = -u.u351 + 2.9629 LnTL; r2 = 0.96.
Analysis of covariance revealed no significant difference
(p: .05) between the slopes of the male and female weight-
length regressions.

Figure VI-lO depicts mean relative condition, K}, as a
function of length for males and females. Figure VI-lO
shows that K; decreases with increasing length for both
males and females.

Appendix Tables A10, A11 and A12 describe the size-age
characteristics of the Fall 1980 age subsample (n : 3H1).
Figure VI-A‘lriepicts the Fall 1980 male and female length-
age relationships. The vertical lines are approximate 95%

confidence limits about the mean total lengths (T1) at each

age. Analysis of covariance reveals no significant

131

 

 

 

l.20-
Females (—--)
Mahs( )
IJO-
Kr l.00~
O o 90"
‘0»
0.80

 

25.0 29.0 33.0 37.0 4|.0 45.0 49.0 53.0 57.0
Length Class Midpoint (cm)
Nam MmbusmpaumuunnhrNMMmuunr«humunhhemmhmmhalm

Figure VI—lO. Fall 1980 Relative Condition
Factors.

132

   
 

 

 

 

 

 

 

 

60.0«
50.01
it (cm)
Females
4°°°‘ TL = 3|.2447lAeEr'885
r2 = 0.97
30.0 ' . * . ‘
60.0
7: (cm) 7 ’1
40.03 ,_ ._
Mans
TL = 3|.5594lAGE)"3°°
r2: 0.95
30.0 I ‘ l L g ‘ 1
2+ 4+ 6+ 8+ 10+ l2+ 14+
AGE (years)

Figure Vl-ll. Fall 1980 Length—Age Relationships

133

differences (p = .05) between the slopes or elevations of
the Fall 1979 male and Fall 1980 male length—age
regressions. Similarly, analysis of covariance reveals no
significant differences (p = .05) between the slopes or
elevations of the Fall 1979 female and Fall 1980 female
length-age regressions. The Er - age relationships for both
males and females (Appendix Tables A10 and A11) indicate

that the younger fish are generally in better condition than

the older fish.

5. Spring 1981

 

During April 1981 two samples, consisting of 12
immature, 33 male and 106 female white suckers were
obtained. Table VI-5 contains summary statistics for the
composite sample. This sample was obtained to provide
additional information about spring spawning periods and
female size-fecundity relationships. Meaningful comparisons
between Spring 1980 and Spring 1981 summary statistics are
difficult due to the limited scope of the Spring 1981
sampling program. On ”-8-81, none of the 21 males or 22
females sampled had spawned. on 4-24-81, 6 of the 12 males
and 77 of the 84 females had spawned. These observations
parallel similar trends noted during the Spring 1980 season.
Saginaw Bay white sucker spawning occurs_during the latter
portion of April and the early part of May. Females return
to the bay before males (as is reflected by the Spring 1980

and 1981 sample sex composition).

134

Table VI-5

Spring 1981 Summary Statistics

 

 

Sample size (n)

1
TL (cm)

Tl (cm)
sTL (cm)

TL range (cm)

WET (g)
chr (g)

WGT range (g)
'1?

r

r

K range
r

WGT : a(TL)b
a

b
2

Statistics Immatures flalgs Females
12 33 106
34.4 38.0 46.9
34.8 38.7 45.2
1.H80 3.72” H.668
32.8- 30.9- 39.1-
38.1 47.fl 55.7
#40 595 925
53.165 139.760 251.562
345- 330 400
570 975 1850
0.97 0.99 1.03
0.087 0.088 0.099
0.85- 0.73- 0.83-
1.18 1.18 1.31
0.376 0.174 0.077
1.990 2.221 2.460
O.A9 0.86 0.88

r

135

Figure VI-12 shows the length-frequency distribution of
immatures, males and females. The 46.0-47.9 cm length class
is the dominant female length class (as was also the case
during the preceding spring).

Figure VI-13 depicts mean relative condition, Er’ as a
function of length for males and females. Relative
condition is optimal for both sexes at an intermediate size,
after which it experiences a progressive decline.

The Spring 1981 population weight-length regression
(which includes all 151 fish in the composite sample) is,
VI-12 Ln WGT = -3.0070 + 2.5721 LnTL; r2 = 0.93.
Analysis of covariance reveals no significant difference (p:
.05) between the slopes of the spent and unspent female
weight-length regressions. No significant differences (p:
.05) were indicated between the slopes or elevations of
Spring 1980 and Spring 1981 spent female weight-length
regressions. Similarly, no significant differences were
indicated between the slopes or elevations offhndng 1980
and Spring 1981 unspent female weight-length regressions.

Appendix Tables A13, A14 and A15 contain immature, male
and female size-age statistics. Figure VI-l4 depicts the
Spring 1981 female length-age relationship. The vertical
lines are approximately 95% confidence limits about the mean

total lengths at each age. A length-age regression was not

calculated for Spring 1981 males due to the small sample

136

 

 

 

 

 

 

 

 

 

 

 

 

 

60-
Females
40. I11 = l06
TL = 46.9 cm
204
o r—r— “ ‘1...
L
g 601
:1 Males
z
40. n = 33
+1. = 33.0 cm
20-
0 Immatures
2m 0 = 12
+1.: 34.4 cm
0 Fr;

 

 

25.0 29.0 33.0 37.0 41.0 45.0 49.0 53.0 57.0
Length Class Midpoint (cm)

Figure VI-12. Spring 1981 Length—Frequency
Distributions.

XI

1'37

 

 

 

 

1.20‘
Females (—--)
Males ( 1
1.10“
1.004 (6)
0.901
0.80

25.0 29.0 33.0 37.0 41.0 45.0 49.0 53.0 57.0
Length Class Midpoint (cm)
Note. Whparsnthssss rst‘srtothsm'nbsrdhdvmalshsaohlongthcloss.

Figure VI-lB. Spring 1981 Relative Condition
Factors.

138

    

 

 

 

 

60.04
50001
TL (om)
Pandas
.. — .1961
40.0. A 11. = 29.8138 (AGE)
12 = 0.94
.L
30.0 . . . . . . 1
3 5 7 9 11 13 15
AGE (years)

Figure VI—lh. Spring 1981 Female Length-Age
Relationship.

 

139

size represented by this sex. Analysis of covariance
'revealed no significant differences (p = .05) between the
slopes or elevations of Spring 1980 and Spring 1981 female

length-age regressions.

6. Composite Length-Frequency Distributions

Figure VI-15 represents the composite female, male and
immature length-frequency distributions obtained by
combining all samples (n :.1803) obtained during this
investigation. This figure depicts the size selective scope

of the Saginaw Bay trap net fishery for white suckers.

7. White Sucker Fecundity

 

The size-fecundity relationship fer Saginaw Bay white
suckers was determined as follows. Ovaries from 45 ripe
females sampled during April 1980 and 1981 were removed and
weighed to the nearest 5 grams. The ovaries from 13 of
these ripe females were placed in glass jars and transported
in ice to the laboratory. In the laboratory, a subsample of
50 eggs from each of the 13 ovaries was weighed to the
nearest 0.1 mg on a Mettler single pan analytical balance,
with the result divided by 50 to obtain an estimate of mean
egg weight. The resulting 13 mean egg weight estimates
ranged from 3.6 to 5.3 mg and averaged 4.5 mg. An estimate
of the total number of eggs produced by each female was
obtained by dividing the total ovary-egg complex weight (in
grams) by 4.5 mg. The following size-fecundity regression

was calculated,

Number

1110

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

150‘ Females —1
n = 894
'20- TL = 45.7 cm
80" r.._.1—1_—
404 ~
“'1
O ?L— _1
160' "" Malgs
. ‘1— n = 704
1204 .
‘ '1 TL = 37.7 cm
80- .
F—_
40' "——
Immatures
40- __' n = 205
__. ‘
—-4"‘ TL= 33.70111

 

 

 

 

 

 

 

 

25.0 29.0 33.0 37.0 41.0 45.0 49.0 53.0 57.0
Length Class Midpoint (cm)

Figure VI—l5. Composite Length—Frequency
Distributions.

141

VI-13 m = a(TL)b,
where: m = total number of eggs
TL = total length (cm).

The fecundity-size relationship for the combined sample of
45 ripe females sampled during April 1980 and 1981 is,

v1-14 Ln m = -0.0195 + 2.7355 LnTL; r2 = 0.49.
Fecundity computations obtained by applying equation VI-14
overestimate actual fecundity since the fraction of the
total ovary-egg complex which is ovary is not subtracted
from the total ovary-egg weight in the estimation process.
This positive bias in estimating fecundity is not considered
important for current descriptive purposes since more than
90% of the total ovary-egg weight consists of eggs.

The ovary-egg mass of the above 45 ripe females
averaged approximately 14.9% of the total female weight and
ranged from 9.8-20.5% of the total weight. In contrast, the
testes-sperm mass of 22 ripe males sampled during April 1981
averaged 5.7% of the ’total male body weight, ranging from
3.7-9.1% of the total weight. The reproductive investment
of female white suckers is considerably greater than that of
males and is consistent with similar observations made by
Lalancette (1975) in his investigation of white suckers in a

12.3 hectare Quebec lake.

142

8. Miscellaneous Observations

Previous investigations of white suckers in Lake Huron
(i.e. Hall and Elliott, 1954 and Cable, 1967) indicated that
the sea lamprey was an important factor regulating the size
composition of white sucker populations. The current study
found 11 (Nit of 894 sampled females bore noticeable lamprey
wound scars. All but 1 of these scarred females were
greater than or equal to 45.0 cm in total length. Only 1
out of the 704 males sampled bore a noticeable lamprey scar.
The sea lamprey does not appear to play a major role as a
predator of Saginaw Bay white suckers.unless it kills the
vast majority of the individuals it attacks.

A number of large fish of both sexes were afflicted
with a severe degenerative fungal-like disease of the anal
or caudal fins. Several larger fish collected during the
fall and spring seasons had parasitic roundworms embedded in
their paired fins. A few individuals contained massive
abdominal tumors which constituted up to 20% of their total
body weight.

The maximum observed ages for females and males were
17+ and 19+ years old, respectively. Aging of individuals
greater than 13+ or 14 years old was very difficult since
fin ray annuli formed during later years were extremely
close together and often undiscernible.

Food habit studies of Saginaw bay white suckers were
precluded in this study by the fact that most samples,

although well-preserved on ice, were more than 24 hours old.

143

A rapid degeneration of the digestive tract contents
occurred within 24 hours after the fish were captured.
Those digestive tracts observed contained large amounts of
detritus and plant material and, especially in the fall,
were often engorged with amphipods.

Perhaps the most notable aspect of’Saginaw Bay white
sucker biology, as inferred from samples obtained during the
course of the present investigation, is the marked
difference between male and female growth patterns as
evidenced by the. weight-length and length-age regressions.
Analysis of. covariance reveals no significant differences
(p: .05) between slopes or elevations of the Fall 1979,
spring 1980 or Fall 1980 male length-age regressions.
Covariance analysis indicates no significant differences (p:
.05) between slopes of the Fall 1979, Spring 1980, Fall 1980
or Spring 1981 female length-age regressions. The
<iifferences between male and female length-age regressions,
however, were significant (p = .05) for the Fall 1979,
Spring 1980 and Fall 1980 samples. Insufficient numbers of
Spring 1981 males were collected to justify calculation of
the length-age regression, therefore no comparisons could be

made between male and female Spring 1981 growth patterns.

9. Saginaw Bay Adult White Sucker Survival

 

A catch curve is a plot of the loge frequency of
occurrence of fish in each age class, from a given sample or
samples, versus age. The typical form of a catch curve

(Ricker, 1975) has a relatively steeply ascending left limb,

144

a somewhat dome-shaped or flat upper portion, and a long
descending right limb. The difference between natural log
frequencies of fish in successive age classes in the
descending portion of the catch.curve is the negative
quantity, -Z, where Z equals the instantaneous total
mortality rate. The annual survival rate, S, between age
classes is equal to e-Z. Estimates of annual survival rate,
3, between age classes can be calculated from the descending
portions of catch curves. Ricker (1975) outlines the
following assumptions upon which catch curve analyses
depend: (1) annual survival rate, S, is uniform with age,
over the range of age classes in question; (2) the
instantaneous rates of fishing mortality, F, and natural
mortality, M, are uniform; (3) mortality rate is tnme
invariant; (4) the age sample is taken randomly from the age
groups involved; and (5) the age classes in question were
equal in numbers at the time each was recruited into the
fishery.

The assumption of uniform recruitment from year to year
is unrealistic since most fisheries experience frequent
fluctuations in recruitment patterns, resulting in
differences in year class strength. Differences in year
class strength (i.e. unequal recruitment) alter the shape of
the descending right limb of a catch curve. Ricker (1975)
suggests that a good way to reduce catch curve
irregularities caused by unstable recruitment is to combine

samples of successive years. Male and female white sucker

M6

age-frequency distributions observed during each seasonal
sampling period exhibited marked irregularities.
Accordingly, the combined male and female Fall 1979 and Fall
1980 observed age-frequency distributions were used to
construct a catch curve (Figure VI-16) to estimate adult
white sucker survival.

The ascending left limb of the catch curve shown in
Figure VI-16 reaches a peak at age 3+, which indicates that
fish are fully recruited into the fishery by this age. From
age 3+ until age 7+ the curve experiences a steady decline,
after which it rises until age 10+. The gradual rise in the
curve from age 7+ to age 10+ could be due to non-random
sampling or could reflect the influence of 3-4 consecutive
strong year classes. - After age 10+, the combined male and
female catch curve declines more or less regularly until age
16+. The slope of the curve from age 3+ to age 7+ equals
approximately -O.322 which- yields a 0.72 survival rate
estimate. The slope of the curve from age 10+ to age 16+ is
about -O.422, which results in a 0.66 survival rate
estimate. Due to the irregular ascending central portion of
the curve (i.e. from age 7+ to age 10+), survival rate
estimates could not be obtained for ages 8+ and 9+.

Figure VI-17 is the catch curve obtained by combining
the male and female Spring 1980 and Spring 1981 age-
frequency distributions. The ascending left thDCM‘this

catch curve reaches a peak at age 4. From age 4 until age 7

146

5.00-

4.00‘

3.00'

Loge (Number)

2000'

 

LOO-

 

 

0'00 2+4 4+ 6+ 8+ 10+ 12+ 14+ 16+

AGE (years)

Figure VI-16. Combined Fall 1979 and Fall 1980
Catch Curve.

5.00-

Loge (Number)

4.001

147

3.00'

2.00'

 

1.00-1

 

P

P
(71
a b
q

0.00 . .
3 5 7 9 ' 11

AGE (years)

Figure VI—l7. Combined Spring.1980 and Spring 1981

Catch Curve.

148

the curve declines steadily, after which it rises until age
11. After age 11, the curve declines steadily until age 17.
The slope of'the curve from age 4 to age 7 is approximately
-O.533 which gives a 0.59 survival rate estimate. The slope
of the curve from age 11 to age 17 is about -0.377, which
results in a 0.69 survival rate estimate. Survival rate
estimates could not be obtained for ages 8 - 10.

Estimates of annual survival rate, S, obtained by
analyzing the catch curves shown in Figures VI-16 and VI-17
indicate that annual survival of adult white suckers in
Saginaw Bay ranges from approximately 0.59 to 0.72. Coble
(1967) in his study of white suckers in South Bay, Lake
HUTTHI, estimated that annual adult survival, S, ranged from
0.70 to 0.75. These investigations indicate that the
survival rate of white suckers in Lake Huron is relatively
high, reflecting the longevity and absence of signficant

commercial exploitation of this species.

D. Estimation of Yield from a Given Recruitment

 

The decline in numbers of individuals after hatching is

given by,
VI-15 dN/dt : -M(N), with solution given by,
_ -Mt
VI-16 Nt - Noe ,
where: N0 = the initial number of fish (i.e.

fertilized eggs)
M = instantaneous natural mortality

rate.

149

The number of individuals which survive to recruitment,
R, is given by,
VI-17

-Mt
R = Noe R,

where tR = age of
recruitment.

After the age of recruitment, fish are subject, not only to

the probability of death due to natural causes, M, but also

to the probability of death due to fishing, F. At time, t

(where t>tR), the original number of recruits, R, is reduced

to,

VI-18

Rt = Re'Z(t ' ta),

instantaneous mortality rate

where Z = total

(i.e. M + F).
The mean number of individuals from a given year class
present during a given time period is obtained by

integrating equation VI-18 with respect to time,

t=tmz(t _

VI-1 N = R -
9 t t=tre

attained. The catch in numbers from a year class after

tR)dt, where tm = maximum age

recruitment is obtained by multiplying the instantaneous
rate of fishing mortality, F, times equation VI-19. By
multiplying equation VI-18 by an expression for weight as a
function of age and integrating with respect to time we
obtain an expresSion for the sum of the yearly mean biomass
of all fish in a year class, for all the years that it

contributes to the fishery,

__ t=tm
VI-20 B = J/’ (WGT)tRe‘Z(t‘tR) dt,
t=tr

150

where WGTt = weight as a function of age. The total yield

in weight, Ct’ from a year class after recruitment is

obtained by multiplying the instantaneous rate of fishing

mortality, F, times equation VI-20,

t=tm
_ -Z(t-t )
VI-21 Ct - F tztéWGT)tRe R

To avoid specificiation of‘a recruitment function

dt.

(which is usually much more complex than that implied by
equation VI-l7) equation VI-21 is often divided by R, which
results in an expression for average yield per recruit,
Ct/R, as a function of fishing mortality rate, F,

t=tm -Z(t - t
VI-22 Ct/R = E/:;tr (WGT)t e R) dt.

Equation VI-22 is referred to as the Beverton-Holt
dynamic pool model and was introduced by Beverton and Holt
(1957). iIf an investigator has information concerning: age
at recruitment; maximum attainable age; instantaneous rate
of natural mortality, M; and the relationship between weight
and age, then equation VI-22 can be used to calculate the
effects of different rates of fishing on the average yield
per recruit. A given fishery can then be managed so as to
produce a specified optimal yield per recruit by adjusting
the rate of fishing, F. Of course, if the relationship
between spawning stock size and subsequent number of
recruits is known, then equation VI-21 can be used to
estimate the effects of different rates of fishing on the
total yield from the fishery.

When using either equation VI-21 or equation VI-22 to

calculate equilibrium yield or yield per recruit, it is not

151

necessary to know the form of the growth function prior to
the age of recruitment. All that is necessary is to find a
function which describes weight as a function of age from t
= tR to t : tm. The historical development and use of the
dynamic pool model assumed that fish growth was described by
3 Von Bertalanffy-type growth equation, which assumes
asymptotic growth in length. There is no 3 priori reason to
assume that the pattern of growth demonstrated by any given
fish population is necessarily asymptotic. A better growth
equation, which is directly applicable to the population
under investigation, can often be obtained by relating the
observed sizes of fish to their corresponding ages via some
form of linear functional relationship. This is the
suggestion put forward by Roff (1980) who advocates
abandoning the use of the Von Bertalanffy growth function in
fisheries applications.

Weight at age data for ages 2+ through >13+:fi%m1the
Fall 1979 samples were used to determine a linear weight-age
relationship for both males and females. Fish greater than
13+ years old demonstrated very little growth, therefore
they were pooled with the age 13+ fish. For the purposes of
these and other size-age calculations, fish aged as n+ were
designated as age n.5 (i.e. a fish aged 3+ was treated as
though it were age 3.5, etc.). The following simple linear
regression model was used to describe the relationship
between mean weight and age for both males and females,

v1-23 Wfi’ = a + b (AGE).

152

The female weight-age relationship is given by,

VI-24 RE? = 426.3141 + 81.3462(AGE); r2 = 0.99.
The male weight-age relationship is given by,
VI-25 WET = 475.6294 + 38.6713(AGE); r2 = 0.87.

By substituting equation VI-23 into equation VI-22 we

get:

. t=tm
VI-26 Ct/R = F~//’ (a +bt)e-(F * ”)(t ’ tR) dt.
t=tr

where t = AGE.
Integration of equation VI-26 results in,
aF
VI-27 Ct/R = :XTZWU [e (F + M)(tm tR) - 1]

m [true
[e'(F + M)(tm - t

+

-(F + M)(tm - tR)

- t +

R]
R) - 1].
As mentioned previously, growth (in terms of length and
weight) is negligible for both males and females after about
age 13+. Therefore, for the purposes of the present
analysis, tm = 13+ (i.e. 13.5 years) for both sexes. Catch
curve analyses indicate that adult white sucker survival, S,
ranges from 0.59 to 0.72. If all mortality were
attributable to natural factors, then these annual survival
rates would indicate that instantaneous natural mortality,
M, ranges from about 0.329 to 0.528. The results of Chapter
III indicate that the current rate of fishing, F, is small
(i.e. less than 0.1), implying that most of the Saginaw Bay
adult white sucker mortality is probably attributable to
natural causes. Assume a constant natural mortality rate of

0.4 for both males and females. The weight-age regression

parameters, a and b, are obtained directly from equations

EB

‘VI-24 and VI-25 for females and males, respectively. Catch
curve analyses show that, although.significant numbers of
fish of age 2+ in the fall and of age 3 in the spring are
captured by the fishery, full recruitment does not occur
until fish are about age 3+ (fer fall fish) and age 4 (for
spring fish). We now have all of the parameters necessary
to estimate equilibrium yield per recruit via equation VI-
27.

Figure VI-18 depicts equilibrium yield per recruit for
1nales and females at different rates of fishing, F, for two
possible ages of recruitment, tR = 2.5 and tR = 3.5. The
yield per recruit for females is higher than that for males
at all levels of fishing, F, reflecting the differences
between male and female growth rates. Optimal utilization
of Saginaw Bay white suckers (in terms of maximization of
equilibrium yield per recruit) under the conditions implied
by Figure VI-18 should occur at rates of fishing, F, ranging
from 0.6 to 1.2. This would involve a substantial

increase in nominal fishing effort, f from current levels,

t’
which are on the order of‘3,000 shallow trap net lifts per
year. The equilibrium yield per recruit curves shown in
Figure VI-18 indicate that, at present rates of fishing
(i.e. F less than or equal to 0.10), the potential
productivity of Saginaw Bay sucker stocks is greatly
underutilized. Recall the relationship between rate of
fishing, F, and nominal effort, ft’

III-9 F : qf where q : catchability.

t,

 

154

'6004

 

 

 

 

 

 

 

 

 

 

2.0

Femams__
Mobs
400'
culR (a)
200-1
0
Females —'—
600-1
Mahs #—
400~
C? IR (0) M = 0.4
t = 315
200- R
1 =13.5
0' . A 1* i
0.0 0.4 0.8 1.2 1.6

Rate of Fishing (F)

Figure VI—18. Equilibrium Yield Per
Different Rates of Fishing.

Recruit at

35

If catchability remains constant, then F can only increase
141th increasing nominal effort, ft' If the current rate of
fishing, F is approximately 0.1, then a six-fold expansion
in nominal effort, ft’ would be necessary to raise the rate
of fishing to.0.6. It is doubtful that the present
commercial fishery has the capability to expand its effort
by the amount necessary to attain optimal utilization of the

sucker fishery.

E. Estimating Mortality Rates of Pre-recruits

One of the most important aspects of population
dynamics research is the determination of age-specific rates
of survival. Representative sampling (e.g. of the
commercial catch) of’the adult portion of a population
enables an investigator to obtain post-recruit survival
estimates via catch curve analyses. Unfortunately,
difficulties and costs associated with sampling yoUnger life
stages usually preclude direct survival rate estimation of
pre-recruit age classes.

Perhaps the most fundamental problem facing fisheries
managers and investigators is the determination of the
spawning stock - recruit relationship (i.e. the so-called
recruitment curve). Knowledge about the recruitment process
helps fisheries managers exert more precise control over the
fisheries relegated to their stewardship. Compensatory
recruitment mechanisms, if properly delineated, describe
potential fluctuations in year class strength. Knowledge

about year class strength enables fisheries managers to

156

adjust the rates of fishing, F, to take advantage of
.exceptionally strong year classes or to protect the fishery
during a series of poor recruitment years. Recruitment
curve delineation depends upon knowledge of pre-recruitment
survival rates, particularly of the age 0 year class.
Vaughan and Saila (1976) describe an indirect method
for determining age 0 survival rates via application of the
Leslie Matrix. This method depends upon the following
assumptions: (1) mortality rates are constant over time; (2)
fishing mortality is zero for the age10 year class; (i.e.
the population age structure is stable); and (3) the
population is in equilibrium (i.e. is neither growing nor
declining in size). Any one element of’the Leslie Matrix
can be indirectly determined if the remaining age-specific
fecundity and survival relationships are known. The
assumption of an equilibrium population implies that the
dominant eigenvalue of the Leslie Matrix is equal to 1. Let
Fx be the number of females born in the interval, t to t +
1, per female of age x to x + 1 at time t who will be alive
in the 0th age class at time t + 1 or,
VI-28 F = S m

x x x+1’

.where: SX : the probability that an individual

reaching age x will survive to age x + 1

x+1 the number of females born per female

of age x + 1.

Define the Leslie Matrix as follows,

157

 

 

F0 F1 F2 ° Fk-2 Fk—1
S0 0 0 . . 0 0
0 S1 0 O 0
.E= 0 O 32 . . . 0 0
_0 0 0 . . . Sk_2 0 _j
Define the population state vector, N(t) as follows:

no(t)
n1(t)

.———h

M(t):
nk_1(t)'
L

 

 

where nx(t) equals the number of individuals (females) of
age x at time t. The population state vector at time t + 1
is,

VI-29 NZt+1> :

Equation VI—29 is a special case of the following first

2L
A
c1.

- >.

order linear homogeneous difference equation,

VI-30 KT?) :.gtNCO).

The dominant positive eigenvalue, R, of L.is obtained by
solving,

v1-31 |;-BI|= 0, where ;_= the identity matrix.
A scalar R is called an eigenvalue of'L_if there exists a
non-zero vector i such that,

v1-32 g? = R3.

158

If R =:‘l (i.e. the p0pulation is in equilibrium) age 0

survival rate, 80’ is determined from,

VI-33 [4:11 = 0, or
VI-34 s0 = (1 - FO)/(F1 + S1F2 + $15293 +
. . . + S1 . . 'Sk-2Fk _ 1).

Table VI-6 contains the basic data necessay for constructing
a Leslie Matr'ix for female Saginaw Bay white suckers.
Analysis of covariance indicates nosflgmificantdifferences
(p = .05) between slopes of the Fall 1979, Spring 1980,or
Fall 1981 female total length-age regressions. The length
at age column in Table VI-6 is derived by pooling the Fall
1979,3pring 1980, Fall 1980 and Spring 1981 female length-
age regressions which results in,

v1-35 TL = 31.1073(AGE)‘1862.
Catch curve analysis of the combined Fall 1979 and Fall 1980
age samples indicates that annual survival rate, S, from age
3+ to age 7+ is about 0.72 and from age 10+ to age 16+ is
approximately 0.66. These survival rate estimates were used
as guidelines for the age-specific survival rates listed in
Table VI-6 for ages 3+ to 13+. Since survival rates of pre-
recruits are almost certainly less than adult survival
rates, estimates of age 1 and 2 survival are S : 0.30 and S
= 0.50, respectively. Recall the size-fecundity
relationship described by equation VI-14.

2.7355.

v1-14 mx .9807(TL)

Assuming a 1-1 sex ratio (i.e. the number of male and female

eggs produced are equal) the size-fecundity relationship

159

Table VI-6

Female Age-Specific Fecundity and Survival Rates

Age TL(cm) SX mx
0 0

1 0.30 0

2 0.50 0

3 0.70 0

4 40.3 0.70 12073
5 42.0 0.70 13518
6 43.4 0.70 14786
7 44.7 0.70 16030
8 45.8 0.65 17132
9 46.8 0.65 18175
10 47.8 0.65 19257
11 48.6 0.60 20151
12 49.4 0.60 21072
13+ 50.2 0.00 22018

 

160

shown in Table VI-6 is given by,
2.7355

VI-36 mx c (.5) x .9807(TL)
The Leslie Matrix constructed from the data contained
in Table VI-6 is a 13 x 13 square matrix. Age 0 survival
rate is obtained by application of equation VI-35 to the
Table VI-6 data,
VI-37 s0 = (1 - 0)/(4816.08) = 2.076 x 10‘”.
According to equation VI-37, approximately 2 females out of
10,000 fertilized eggs survive the first year of life.
The preceding Leslie Matrix calculations assume that Saginaw
Bay female white suckers first spawn at age 4. The weakest
links in the age-specific survival and fecundity data listed
in Table VI-6 are the estimates of age 1 and age 2 annual
survival rates, for which no concrete data could be
obtained.

In Chapter IV, application of the logistic surplus
production model to commercial fishery catch and effort data
from 1968 to 1978 resulted in an estimated carrying capacity
of approximately 3,913,000 lbs (i.e. 1,779,000 kg). This is
an estimate of the total catchable biomass of white suckers
in Saginaw Bay. The percentages of the combined weight of
all the samples obtained during this study as immatures,
males and females were 5.85%, 31.52% and 62.63%,
respectively. If we multiply these percentages by 1,779,000
kg, we obtain rough estimates of the total weight of

catchable immatures, males and females (i.e. 104,072 kg

immatures; 560,741 kg males; 1,114,188 kg females). The

161

mean weights of immatures, males and females obtained during
this investigation were 395 g, 620 g and 970 g,
respectively. By dividing total weight estimates of
catchable immatures, males and females by mean individual
weights, we obtain estimates of the total numbers of
catchable immatures, males and females (i.e. 263,000
immatures; 904,000 males; 1,149,000 females). Assume, for
the present purposes, that white suckers are catchable once
they reach age 2. Also assume that age 13 is the maximum

attainable age. The total number of exploitable females is

given by,
13
VI-38 M(t) = .22ni(t), where ni(t) = no. in age class i.
1:
Now, the absolute numbfr of females in age class i is,
1....

VI-39 ni(t) = nO jIEO Sj’ where Sj = prob. of survival
from age 5 to
age t-+1.

Substituting equation VI-39 into equatipg V1738 we obtain,

. l:

v1-40 N(t) = 1,3 (n if s.) =n 2 II 3..

._ 0 '-0 3 01:2 j=0 J

Solving equationJVfL40 fog no we get,

. 13 i-l
v1-41 no = N(t)/( X IIs.).
i=2 3:0 J

With the estimate of age 0 survival obtained from equation
VI-37 coupled with the annual survival rate estimates listed
in Table VI-6 we obtain the following value for the

denominator of equation VI-41,

13 i—l
2 II S. : 0.0001622982943.
i: 2 J=O J

The numerator for equation VI-41 is our previous estimate of

to total number of catchable females (i.e. 1,149,000).

162

Dividing 1,149,000 by 0.0001622982943 we obtain an estimate
of the total number of female eggs produced (i.e.
'7,079,557,000). Assuming a 1-1 sex ratio, the total number
«of eggs produced would equal about 14,159,114,000. If the
.average female (TL = 44.8 cm) produces .9807(44.8)2‘7355 =
32,300 eggs, then the total number of'spawning females is
approximately 438,000.

Jensen (1974) describes a Leslie Matrix approach to
fishery yield estimation as follows,

.as t...)
VI-42 C = Efll‘ 11(0) ,

t
where: ‘0; = total catch (harvest) vector

F = a diagonal k x k matrix with age-specific
rates of fiShing on the diagonal and zeroes
elsewhere

W = a diagonal k x k matrix with age-specific
weights on the diagonal and zeroes elsewhere

L_= the Leslie Matrix

._..

N(O) = the initial population state vector.

Equation VI-42 and similar discrete time population
harvest models could be useful predictors of future harvest
levels'from age-specific rates of fishing, Fi, provided a
reliable estimate of the population state variable, M(B3, is
available. The present investigation obtained a rough
estimate of the age class distributions of males and females
greater than or equal to 2+ years old. The contribution of

2+ year old fish to these age class distributions is

underestimated since fish are not fully recruited into the

163

fishery until they reach age 3+. Indirect estimates of the
size of age class 0 depend upon the assumption of an
equilibrium population. Insufficient data is available to
substantiate this assumption although it seems reasonable as
a first approximation, in light of the absence of
information to the contrary. No estimates of the size of
age class 1 are available. In short, the lack of a reliable
measure of the Saginaw Bay white sucker population state

variable, N( ), precludes meaningful application of equation

VI-42.

CHAPTER VII

Summary and Recommendations

A. Recapitulation

 

This investigation was undertaken to determine
potential limits on the commercial exploitation of Saginaw
Bay suckers. Three species of suckers (i.e. white, longnose
and redhorse suckers) occur commonly in the waters of
Saginaw Bay. The white sucker dominates the species
composition of the commercial sucker catch, accounting for
more than 90% of all landings. Accordingly, the major
portion of this study involved deterndning the population
dynamics of Saginaw Bay white suckers.

Population dynamics data (e.g. catch and effort records,
age and growth information) were considered representative
of a single homogeneous stock of fish. This may be a rather
sweeping assumption, considering the large area (2960 km2)
of Saginaw Bay and the observations by Olson and Scidmore
(1963) and Coble (1967), which indicate that white suckers
typically exhibit little in the way of long range movements.
Refer to Figure II-l which shows at least 10 tributary
streams and rivers emptying into Saginaw Bay. Each of these
tributaries may accommodate the spring spawning run of a

distinct stock of white suckers. For present purposes,

16h

165

however, it is assumed that available data reflect the
average characteristics of several possible white sucker
stocks.

The description and analysis of commercial fishery catch
and effort data is central to this investigation's study of
white sucker population dynamics. Chapter III contains an
in-depth analysis of Saginaw Bay sucker fishery data from
1929-1956 and from 1960-1979. In general, these data
reflect decreasing rates of exploitation and increasing
levels of abundance of Saginaw Bay sucker stocks. Current
annual sucker harvests approximate 100,000-150,000 pounds,
which is about an order of magnitude less than historical
maximum harvest levels. The analysis of recent (1960-1979)
catch and effort data is complicated by the fact that, in
recent years, only a fraction (i.e. from one-fourth to one-
third) of the actual catch has been landed, reflecting the
absence of a remunerative market for Great Lakes suckers. A
cautious interpretation of these data indicates that up to
500,000 pounds of suckers could be harvested annually
without overexploiting the stocks.

Chapter IV contains a simple bioeconomic analysis of the
current Saginaw Bay commercial sucker fishery. The logistic
surplus production model is used as the production function
for this analysis. The reliability of this production
function may be questioned since it was derived for a period
of time (1968-1978) represented by relatively few data

points. A similar production function, representing a

166.

longer time frame (1930-1955) was not used since it was
assumed these earlier data no longer describe current
fishery dynamics. According to this analysis, an ex-vessel
price increase from $0.05/lb to $0.10/lb would generate a
short term economic rent of approximately $11,000 to the
commercial fishery.

According to the U.S. Food and Drug Administration,
Great Lakes fish marketed for human consumption may contain
no more than 5 ppm PCB's and no more than 5 ppm EDDT (i.e.
DDT and its analogs, DDD and DDE). Chapter V summarizes the
results of PCB and XDDT analyses on samples of 25 white
suckers obtained from the Fall 1979 and Spring 1980
.commercial landings. These analyses indicate that the mean
levels of PCB's (0.05 ppm) and XDDT (0.015 ppm) in the
edible fish (fillets) of Saginaw Bay white suckers are well
below the FDA tolerance limits for these compounds. These
results are encouraging with respect to future development
of the Saginaw Bay sucker fishery since they indicate that
organic chemical contaminants should not be limiting factors
to enhanced sucker utilization.

One of the primary objectives of this investigation was
the establishment of a preliminary data base regarding: the
size, age and sex composition and age-specific growth and
mortality rates of Saginaw Bay white sucker stocks. This
information is presented in Chapter VI and constitutes,
perhaps, the most important aspect of this study. Adult

white suckers in Saginaw Bay experience low growth rates

167

(from 1-2 cm in total length per year), high annual survival
rates, S (from 0.59 to 0.72) and long life Spans (up to 19+
jyear old). Females grow faster and larger and, on the
average, live longer than males. These characteristics are
diagnostic of an underexploited fishery. An increased rate
of fishing, F, should result in increased growth rates as
the stocks compensate for increased mortality rates.
Increased levels of exploitation should change the age and
size composition of the stocks as older and larger
individuals are removed from the fishery.

In Chapter VI, a modification of the Beverton-Holt
dynamic pool model was used to estimate the equilibrium
yield per recruit at different rates of fishing, F. In this
model, the age-weight relationship was described by a simple
linear function, which allowed for a simple closed form
solution of the dynamic pool equation without the assumption
of a Von Bertalanffy-type growth fomn. This analysis
indicates that optimal utilization (in terms of maximum
yield per recruit) of the Saginaw Bay sucker fishery can
only occur with a comparatively large increase in the rate
of fishing (i.e. from 6 to 10 times current levels).

The Saginaw Bay sucker fishery is a greatly
underexploited resource. Current harvest levels could
probably be safely increased by as much as 500% without risk
of overexploitation. Increased sucker utilization, however,
depends upon establishment of dependable and profitable

markets for suckers and sucker products.

M%

B. Recommendations

Future investigations of the dynamics of Saginaw Bay
suckers stocks should include a comprehensive mark and
recapture study to obtain estimates of population size and
information about fish movements and distribution within the
‘bay. Catchseffort methods of population size estimation as
used in Chapter III are not the most reliable means of
determining the magnitude of the population state variable
4.55. In Chapter VI, a Leslie Matrix approach to yield
estimation (Jensen, 1974) was suggested, but was not
utilized due to the lack of a reliable estimate oflr15.
This could prove to be a fruitful approach in future
investigations of the dynamics of Saginaw Bay sucker stocks.

This investigation concentrated on white sucker
population dynamics. Future studies should include efforts
to obtain population dynamics information about the other
commercially harvested sucker species, longnose and redhorse
suckers. This information is needed to help insure the
‘rational exploitation of’all sucker species within Saginaw
Bay.

Management of the Saginaw Bay sucker fishery solely by
means of commercial fishery catch and effort statistics
should be avoided, if possible, since these data are not
entirely reliable. Catch-effort data analyses depend upon
the assumption of equilibrium conditions with respect to the

various factors influencing population growth. In the short

term, fishery expansion upsets these equilibrium conditions,

169

thereby destroying the predictive utility of the catch per
unit effort ratio, Ct/ft’ at least until the fishery reaches
a new equilibrium. Increased development of this fishery
should be monitored by a spring and fall commercial catch
sampling program designed to ascertain the size, age and sex
distribution of the exploited populations. Information
obtained from such sampling efforts could be used to monitor
‘the response of the sucker stocks to changing rates of
fishing, F. For example, marked shifts in stock size or age
composition toward smaller sizes and younger ages could
indicate overfishing, in which cases appropriate controls on
fishing effort could be instituted.

The difficulties associated with the economic analysis
of a single species fishery within the framework of a multi-
species fishery are discussed in Chapter IV. Additional
economic information (e.g. costs and revenues associated
with harvesting all commercially utilized species) would be
helpful. This information would be used to place the
developing sucker fishery in its proper perspective relative
to the entire fishery.

Periodic assessments of the levels of potentially
harmful environmental contaminants should be undertaken to
insure continuing compliance with applicable governmental
tolerance limits. Such efforts should probably be conducted
every second or third year, funding permitted.

Management expenditures incurred to monitor the size,

age and sex composition of Saginaw Bay sucker stocks should

170

be relatively minor, particularly if similar information on
other exploited species (e.g. yellow perch, channel catfish,

lake whitefish) is concurrently obtained.

C. Critique

 

Often, when analyzing dynamic ecological and economic
systems, 1“? presuppose the existence of more or less stable
equilibria around which the behavior of these systems is
centered. All of the analyses of Saginaw Bay white sucker
population dynamics contained in this report assume the
existence of equilibrium conditions. In light of available
information, we have no way of knowing whether or not such

assumptions are justified.

APPENDICES

Table A1

Fall 1979 Female Size-Age Statistics

Age n TL(cm) STL(°m) TL range(cm)
2+ 20 37.3 1.443 35.0 - 41.7
3+ 12 39.4 2.003 36.5 - 42.6
4+ 10 40.4 2.334 37.2 - 44.0
5+ 7 42.7 1.718 39.8 - 44.4
6+ 12 43.9 2.139 38.8 - 47.8
7+ 4 45.8 2.765 42.5 - 49.0
8+ 7 46.0 1.049 44.8 - 48.0
9+ 8 47.0 2.047 44.3 - 50.5

10+ 9 49.5 0.866 48.4 - 50.7

11+ 11 49.8 3.281 44.6 - 54.5

12+ 10 50.7 2.057 47.3 - 53.3

313+ 10 51.7 2.017 48.6 - 55.6

Age n WGT(g) SWGT(g) WGT range(g)
2+ 20 590 84.722 450 - 830
3+' 12 715 111.017 570 - 910
4+ 10 780 151.416 550 - 995
5+ 7 905 105.503 720 - 1010
6+ 12 960 151.130 590 - 1220
7+ 4 1095 95.656 970 - 1195
8+ 7 1115 104.100 995 - 1260
9+ 8 1150 142.422 925 - 1385

10+ 9 1305 181.311 1045 - 1550

11+ 11 1370 271.605 970 - 1700

12+ 10 1435 200.042 1125 - 1795

213+ 10 1505 136.541 1260 - 1795

Age n Kr SKr Kr range
2+ 20 0.98 0.096 0.75 - 1.21
3+ 12 1.02 0.067 0.93 - 1.17
4+ 10 1.03 0.102 0.92 - 1.26
5+ 7 1.03 0.050 0.95 - 1.10
6+ 12 1.01 0.067 0.88 - 1.09
7+ 4 1.04 0.081 0.94 - 1.12
8+ 7 1.04 0.069 0.95 - 1.15
9+ 8 1.01 0.125 0.81 - 1.23

10+ 9 0.99 0.119 0.82 - 1.16

11+ 11 1.02 0.107 0.80 - 1.21

12+ 10 1.02 0.087 0.94 - 1.22

£fl3+ 10 1.02 0.070 0.88 - 1.11

ix_

0
momO—IQOOU') 3031-”

TABLE A2

Fall 1979 Male Size-Age Statistics

_ I
Age n TL(cm) sTL(cm) TL range(cm) TL(cm)
2+ 34 35.6 1.863 32.0 - 40.5 35.7
3+ 17 37.6 3.110 34.4 - 48.5 37.7
4+ 7 39.3 1.670 37.4 - 42.4 38.8
5+ 5 39.9 2.066 37.5 - 42.2 40.8
6+ 5 40.0 2.064 38.0 - 43.0 39.6
7+ 5 41.1 1.199 39.7 - 42.0 41.8
8+ 5 41.0 1.598 39.4 - 43.5 40.5
9+ 6 43.7 1.819 41.5 - 46.2 43.8
10+ 6 44.7 0.857 43.5 - 45.6 44.9
11+ 5 44.6 1.903 42.5 - 47.6 44.0
12+ 2 42.9 42.5 - 43.2
213+ 7 45.2 1.649 43.8 - 47.8 44.4
Age n WGT(g) SWGT(g) WGT range(g)
2+ 34 520 80.597 375 - 740
3+ 17 570 67.064 450 - 655
4+ 7 660 75.986 600 - 815
5+ 5 720 126.125 545 - 840
6+ 5 725 158.493 485 - 900
7+ 5 800 81.670 690 - 870
8+ 5 805 82.417 730 - 910
9+ 6 935 148.683 790 -1180
10+ 6 915 93.113 810 -1030
11+ 5 965 75.033 860 ~1060
12+ 2 845 825 - 860
213+ 7 960 123.741 820 -1190
Age n Kr SKr Kr range
2+ 34 0.97 0.058 0.82 - 1.07
3+ 17 0.93 0.115 0.51 - 1.01
4+ 7 0.94 0.044 0.90 - 1.04
5+ 5 0.98 0.067 0.89 - 1.07
6+ 5 0.98 0.121 0.76 - 1.05
7+ 5 1.02 0.031 0.96 - 1.04
8+ 5 1.03 0.043 0.99 - 1.09
9+ 6 1.01 0.114 0.79 - 1.10
10+ 6 0.92 0.071 0.83 - 1.01
11+ 5 0.99 0.112 0.79 - 1.06
12+ 2 0.96 0.95 - 0.96
213+ 7 0.94 0.079 0.84 - 1.04

Age

1+
2+
3+

Age

1+
2+
3+

Age

1+
2+
3+

Table A3

Fall 1979 Immature Size-Age Statistics

T1(cm)

28.5
33.5
35.7

WGT(g)
265

430
500

sTL(cm) TL range(cm)
1.634 25.7 - 30.8
1.903 30.8 - 37.0
31409 " 36.5

SWGT(3) WGT range(g)
57.641 170 - 350
80.668 315 - 570
455 - 545

SKr Kr range
0.100 0.76 - 1.06
0.110 0.81 - 1.35
0.90 - 0.96

xi

TL(cm)

28.5
33.0

 

Table A4
Spring 1980 Female Size-Age Statistics

1
TL(cm)

xii

Age n TL(cm) STL(°m) TL range(cm)
3 9 39.5 1.218 37.7 - 41.6 39.1
4 6 39.9 1.256 38.0 - 41.5 40.1
5 11 43.1 2.787 38.9 - 48.5 43.0
6 7 44.6 1.172 42.5 - 45.7 44.3
7 3 45.3 0.723 44.5 - 45.8 45.7
8 17 45.4 2.814 40.6 - 50.4 45.0
9 21 46.5 1.821 43.3 - 50.8 46.
10 18 47.6 1.768 44.3 - 50.2 47.3
11 27 48.2 2.154 42.9 - 52.0 47.7
12 17 49.0 1.813 45.1 - 51.7 49.3
l3 14 50.4 1.924 47.3 - 54.4 50.5
.14 28 50.5 2.306 46.7 - 55.1 50.6
Age n WGT(g) SWGT(8) WGT range(g)
3 9 705 ' 65.865 620 - 830
4 6 745 58.195 640 - 820
5 11 895 162.500 720 - 1140
6 7 945 172.354 740 - 1200
7 3 980 169.730 830 - 1165
8 17 915 160.438 685 - 1300
9 21 940 130.512 720 - 1230
10 18 1050 135.916 820 - 1360
ll 27 1070 144.571 850 - 1500
l2 17 1160 125.716 1010 - 1470
l3 14 1220 155.599 1060 - 1630
>14 28 1240 200.988 865 - 1730
Age n r SKr Kr range
3 9 1.09 0.093 0.93 - 1.22
4 6 1.12 0.073 1.04 - 1.21
5 11 1.10 0.118 0.93 - 1.26
6 7 1.08 0.175 0.87 - 1.28
7 3 1.07 0.176 0.89 - 1.24
8 17 0.99 0.079 0.80 - 1.09
9 21 0.96 0.060 0.87 - 1.04
10 18 1.01 0.083 0.91 - 1.21
11 27 1.00 0.101 0.84 - 1.34
l2 17 1.02 0.078 0.90 - 1.19
13 14 1.01 0.083 0.83 - 1.13
234 28 1.02 0.124 0.87 - 1.34

Table A5

Spring 1980 Male Size-Age Statistics

Age n TL(cm) sTL(cm) TL range(cm)
3 29 36.3 1.789 32.6 - 39.8
4 8 37.3 1.441 35.2 — 39.6
5 5 41.6 1.369 40.1 - 43.5
6 4 38.1 0.854 37.0 - 39.0
7 6 41.3 0.967 39.8 - 42.8
8 4 42.5 2.347 39.5 - 44.5
9 6 140.8 00937 3907 - ”202

10 12 43.3 2.322 40.0 - 46.5

11 6 43.1 1.479 41.5 - 45.2

12 5 44.2 1.440 42.7 - 45.8

13 2 46.2 45.8 - 46.5

214 7 44.4 1.419 42.0 - 46.0

Age n WGT(g) SWGT(8) WGT range(g)
3 29 530 80.017 420 - 740
4 8 575 102.913 470 - 750
5 5 700 76.518 600 - 780
6 4 600 53.600 530 - 660
7 6 690 79.587 635 - 840
8 4 820 149.129 635 - 1000
9 6 675 69.839 600 - 790

10 12 770 139.069 595 - 1000

ll 6 755 118.677 640 - 985

12 5 790 ' 109.396 625 - 900

13 2 950 940 — 960

214 7 860 95.250 700 - 960

Age n Kr SKr Kr range
3 29 1.01 0.081 0.84 - 1.16
4 8 1.02 0.126 0.78 - 1.16
5 5 0.95 0.083 '0.84 - 1.08
6 4 1.01 0.056 0.97 - 1.09
7 6 0.95 0.067 0.88 - 1.06
8 4 1.05 0.095 0.95 - 1.14
9 6 0.96 0.074 0.89 - 1.07

10 12 0.94 0.101 0.75 - 1.07

11 6 0.93 0.086 0.86 - 1.08

12 5 0.91 0.091 0.79 - 1.01

13 2 0.99 0.96 - 1.02

214 7 0.97 0.053 0.90 - 1.06

xiii

I
TL(cm)

36.

37.
41.

41.
43.
40.
42.
42.
44.

44.

U1 WONO‘OWWOHN

A

A

on NOJ‘JWN on NO-DUON

«maroon:

:3 NNNCDN

NNNCDN

Spring 1980 Immature Size-Age Statistics
Ti(cm)

3

36.7

3

36.2

n

WET(g)

3.7
6.4
1.7

395
525
530
535
695

‘E

Ol—‘I—‘OO

.91
.97
.01
.03
.94

Table A6

STL(°m)

2.407
2.307

SWGT(g)

83.826
90.501

SKr

0.068
0.070

xiv

TL range(cm)

35.5

3

1.9

29.1

3

3
u

WGT range(g)

OI—‘OOO

4.2
4.9
0.7

340
255
450
480
670

K
r

.91
.86
.89
.02
.92

range

.92
.14
.09
.04

.96

40.3
40.7

37.5
6

42.

450
690
695
590
720

cahuahao

I
TL(cm)

36.6
35.9

Age

\ONOU'l-I‘JUON

Iv
00H

>
GO
0

H
I—‘QNO‘WJL'WN

zl3

.‘3 WNI-‘UOI-‘N-t-QH 3 WNI—‘WHN-fi'fll—J :3

WNI—‘WE—‘N-ZNH

Table A7

Summer 1980 Female Size-Age Statistics

OHHHHu—JHHH

71(cm)

33.
37.
39.
42.
42.
42.
43.
47.
52.

WET(g)

ONENUTUTJL’UTUUO

420
550
655
825
805
730
840
1125
1185

K

r
.15
.06
.09
.11
.10
.00
.10
.13
.90

8
6
24

9

19

S

0 OOO

sTL(cm)

1.900

1.239
2.480

2.339
3.601
SWGT(3)

6.674

2.099
1.451

0.875
7.547
Kr

.103
.032
.180

.038

.122

XV

TL range(cm)

34.0 - 40.8
38.0 - 40.5
39.1 - 46.5
3908 - ”309
46.8 - 48.0
48.7 - 55.8
WGT range(g)
410 - 690
600 - 710
570 - 1220
630 - 805
1055 - 1190
985 - 1380
Kr range
0.85 - 1.29
1.04 - 1.11
0.89 - 1.33
0.96 - 1.03
1.10 - 1.16
0.77 - 1.01

l
TL(cm)

.=un»
pus—q

.1:
LA)

0 o o o
(I) 0400

 

Table A8

Summer 1980 Male Size-Age Statistics

Age n TL(cm) STL(°m) TL range(cm) TL(cm)
2 7 31.9 1.490 30.0 - 33.9 31.7
3 35 35.2 1.713 29.0 - 37.9 35.0
’4 2 3900 38.6 - 39.”

5 3 39.5 1.217 38.7 - 40.9 38.9
7 1 40.0
9 2 41.0 39.5 - 42.5

11 1 41.0

Age n WGT(g) SWGT(3) WGT range(g)
2 7 360 45.539 300 - 420
3 35 445 80.844 240 - 620
u 2 580 565 - 595
5 3 625 132.571 490 - 755
7 1 580
9 2 630 580 - 680

11 1 650 '

Age n Kr SKr Kr range
2 7 1.08 0.036 1.04 - 1.14
3 35 1.01 0.101 0.84 - 1.22
4 2 1.00
5 3 1.04 0.150 0.87 - 1.15
7 1 0.94
9 2 0.95 0.93 - 0.97
11 1 0.98

xvi

Age

Age

WM

Age

DUN

Table A9

Summer 1980 Immature Size-Age Statistics

n TL(cm) STL(°m) TL range(cm)
21 29.3 3.055 24.3 - 39.2
2 37.0 36.7 - 37.2
n 957(g) SWGT(g) WGT range(g)
21 ' 260 65.938 145 - 385
2 540 495 - 580

n Er SKr Kr range
21 0.98 0.150 0.45 - 1.22
2 1.08 0.97 - 1.18

xvii

I
TL(cm)
29.0

Table A10
Fall 1980 Female Size-Age Statistics

I
TL(cm)

Age n TL(cm) sTL(cm) TL range(cm)
2+ 4 36.7 1.555 34.6 - 38.0 37.0
3+ 37 39.9 1.596 36.2 - 43.0 40.0
4+ 24 40.9 2.137 36.7 - 45.8 41.1
5+ 21 44.2 1.762 38.9 - 46.4 44.5
6+ 8 44.7 2.340 41.7 - 49.2 43.8
7+ 9 45.1 2.398 40.1 - 48.7 45.5
8+ 8 46.8 1.962 43.6 - 50.4 46.7
9+ 10 49.0 1.623 46.5 - 51.2 49.4
10+ 12 48.8 2.093 45.9 - 52.8 48.8
11+ 9 48.2 1.869 45.4 - 51.4 47.9
12+ 4 49.7 1.593 47.4 - 51.0 50.1
213+ 6 51.5 1.375 49.4 - 53.1 51.9
Age n WGT(g) SWGT(g) WGT range(g)
2+ 4 545 99.875 460 - 685
3+ 37 680 89.570 530 - 845
4+ 24 755 133.274 530 - 1070
5+ 21 935 130.274 650 - 1140
6+ 8 965 141.911 740 - 1210
7+ 9 955 134.508 775 - 1150
8+ 8 1085 172,730 885 - 1310
9+ 10 1270 179.679 1010 - 1580
10+ 12 1160 151.544 870 - 1350
11+ 9 1185 128.916 1000 — 1360
12+ 4 1165 179.141 1000 - 1410
213+ 6 1310 192.319 1040 - 1510
Age n Kr SKr Kr range
2+ 4 1.04 0.108 0.95 - 1.19
3+ 37 1.02 0.068 0.90 - 1.18
4+ 24 1.05 0.064 0.90 - 1.19
5+ 21 1.03 0.097 0.82 - 1.22
6+ 8 1.04 0.068 0.94 - 1.11
7+ 9 1.00 0.077 0.92 - 1.15
8+ 8 1.02 0.169 0.78 - 1.35
9+ 10 1.04 0.127 0.84 - 1.17
10+ 12 0.96 0.120 0.78 - 1.18
11+ 9 1.02 0.104 0.90 - 1.17
12+ 4 0.91 0.080 0.84 - 1.02
213+ 6 0.92 0.140 0.75 - 1.14

xviii

Age

2+
3+
4+
5+
6+
7+
8+
9+
10+
11+
12+
213+

Age

2+
3+
4+
5+
6+
7+
8+
9+
10+
11+
12+
213+

Age

2+
3+
4+
5+
6+
7+
8+
9+
10+
11+
12+
213+

n

I—‘
N

l-‘ w
HO‘O)

CDU'ILAJO‘SNCDUT

:3

LA)!"-I
CDN

H
(DU'ILAJCh-PNICDUTI—‘O‘

:3

oomwozfioom

TL(cm)

35.
37.
38.
no.
no.
41.
141.
112.
1:3.
112.
43.
us.

tNI—‘OP—‘P—‘NNWWUTO

ma)

455
565
620
720
610
685
760
755
765
765
790
850

.00
.01
.06
.06
.89
.94
.02
.95
.93
.97
.91
.87

OOOOOI—‘OOHI—‘l—‘l—J

Table A11

sTL(cm)

OOOOOOOOOOOO

NUOONNNI-‘NI-‘Ol-‘H

.310
.313
.756

.082
.124
.134
.073
.072
.025
.126
.127
.083
.055

xix

32.5
3u.u
37.1
37.9
38.0
39.3
38.4
39.2
39.8
41.2
no.5
n2.2

350
440
585
650
390
605
600
620
705
660
550
700

K
r

000000000000
\0
0"

oc>tho+Jo+4Hr4H+4
H
N

Fall 1980 Male Size-Age Statistics

TL range(cm)

37.
40.
39.
42.
43.
43.
46.
44.
46.
43.
49.
50.

GDU'II-‘OWOOUUQDzOU'I

WGT range(g)

620
690
645
850
700
790
970
840
820
920
1230
1120

Age

1+
2+
3+
4+

Age

1+
2+
3+
7+

Age

1+
2+
3+
7+

n

22
45

22
45

Table A12

Fall 1980 Immature Size-Age Statistics

'

{£7

TL(cm)

27.4
34.2
37.1
38.0

WETXg)

215
415
500
405

STL(°m)

2.536
2.154
3.071

SWGT(3)

59.978
77.029
118.864

SKr

0.049
0.066
0.086

XX

I
TL range(cm) TL(cm)

23.4 - 33.1 27.0
29.0 - 3809 34.0
32.4 - 40.4 37.8
36.2 - 39.7

WGT range(cm)

125 - 345
265 - 635
355 - 645
385 - 425
Kr range
0.86 - 1.06
0.81 - 1.13
0.80 - 1.04
0.65 - 0.77

Table A13

Spring 1981 Female SiZe-Age Statistics

._ l
Age n TL(cm) STL(°m) TL range(cm) TL(cm)
3 3 36.0 2.074 34.1 - 38.2 35.6
4 28 40.1 1.667 36.0 - 43.0 40.1
5 4 40.0 2.319 37.8 - 42.7 39.8
6 3 44.3 1.617 42.6 - 45.8 44.6
7 1 43.5
8 5 43.8 1.404 ' 41.5 - 44.7 44.7
9 5 46.9 1.404 45.3 - 48.1 47.7
10 6 47.7 1.550 45.9 - 49.6 47.7
11 11 47.0 1.863 44.6 — 51.0 47.0
12 12 47.7 2.298 45.7 - 52.3 46.7
13 11 48.5 1.577 46.5 - 51.5 48.3
214 17 50.6 2.944 46.5 - 55.7 51.1
Age n WGT(g) SWGT(g) WGT range(g)
3 3 460 62.517 400 - 525
4 28 710 105.457 470 - 930
5 4 660 127.435 490 - 795
6 3 890 62.517 820 - 935
7 1 740
8 5 790 75.614 675 - 865
9 5 950 81.056 860 - 1040
10 6 960 117.757 830 - 1090
11 11 960 137.647 770 - 1170
12 12 1045 153.669 910 - 1340
13 11 1075 114.342 890 - 1310
214 17 1245 255.615 930 - 1850
Age n Kr SKr Kr range
3 3 0.94 0.121 0.84 - 1.07
4 28 1.09 0.096 0.93 - 1.26
5 4 1.01 0.106 0.88 - 1.13
6 3 1.06 0.145 0.96 - 1.23
7 1 0.92
8 5 0.97 0.025 0.95 - 1.01
9 5 0.97 0.026 0.93 - 1.00
10 6 0.94 0.040 0.90 - 1.00
11 11 0.97 0.070 0.86 - 1.12
12 12 1.02 0.102 0.92 - 1.31
13 11 1.01 0.075 0.89 - 1.11
314 17 1.04 0.084 0.87 - 1.26

xxi

Table A14

Spring 1981 Male Size-Age Statistics

_ I
Age n TL(cm) STL(°m) TL range(cm) TL(cm)
2 1 30.9
3 6 35.2 0.668 34.7 - 36.5 35.0
4 12 37.0 1.221 35.0 - 39.1
5 1 38.4
6 2 41.7 40.8 - 42.6
8 1 40.7
9 3 41.0 1.079 40.2 - 42.2 40.5
11 2 42.5 41.0 - 44.0
12 1 44.7
13 1 45.5
214 1 47.4
Age n WGT(g) SWGT(g) WGT range(g)
2 1 330
3 6 480 36.560 440 - 540
4 12 535 50.142 435 - 640
5 1 610 -
6 2 750 600 - 900
8 1 685
9 3 605 106.888 485 - 690
11 2 735 720 - 750
12 1 690
13 1 795
214 1 975
Age n Kr SKr Kr range
2 1 0.99
3 6 1.03 0.047 0.97 - 1.09
4 12 1.01 0.059 0.91 - 1.12
5 1 1.05
6 2 1.03 0.88 - 1.18
8 1 1.01
9 3 0.87 0.127 0.73 - 0.96
11 2 0.98 0.91 - 1.05
12 1 0.80
13 1 0.88
214 1 0.97

xxii

Age

Age

Age

Table A15-

Spring 1981 Immature Size-Age Statistics

n TL(cm) STL(cm) TL range(cm)
12 34.8 1.480 32.8 - 38.1
n WGT(g) SWGT(8) WGT range(g)

12 440 53.165 390 - 570
n 1;, SKr Kr range
12 0.97 0.087 0.85 - 1.04

xxiii

I
TL(cm)
34.6

REFERENCES

REFERENCES

Anderson, A.M. 1973. Developing markets for unfamiliar
species. J. Fish. Res. Board Can. 30: 2166-2169.

Anderson, L.G. 1977. The economics of fisheries
management. The Johns Hopkins University Press.
Baltimore. 214 p.

Barton, B.A. 1980. Spawning migrations, age and growth,
and summer feeding of white and longnose suckers in an
irrigation reservoir. Can. Field-Nat. 94(3): 300-304.

Beamish, R.J. and H.H. Harvey. 1969. Age determination in
the white sucker. J. Fish. Res. Board Can. 30: 607-
638.

Beamish, R.J. 1973. Determination of age and growth of
populations of the white sucker (Catostomus commersoni)
exhibiting a wide range in size at maturity. J. Fish.
Res. Board Can. 30: 607-616.

 

Beeton, A.M., et a1. 1967. Physical limnology of Saginaw
Bay, Lake Huron. Great Lakes Fish. Comm. Tech. Rep. No.
12: 56 p.

Berst, A.H. and G.R. Spangler. 1973. Lake Huron: the
ecology of the fish community and man's effects on it.
Great Lakes Fish. Comm. Tech. Rep. No. 21: 41 p.

Beverton, R.J.H. and S.J. Holt. 1957. On the dynamics of
exploited fish populations. U.K. Min. Agric. Fish.,
Fish. Invest. (Ser. 2) 19: 533 p.

Borgeson, D.P. (ed.) 1972. Status of Michigan's fisheries
management, 1971. Mich. Dept. Nat. Res. Fish. Div.
Fish. Manage. Rep. No. 4: 31 p.

Braaten, D.0. '1969.' Robustness of the DeLury population
estimator. J. Fish. Res. Board Can. 26(2): 339-355.

Buettner, H.J. 1968. Commercial fisheries of the Great
Lakes, 1879-1966. U.S. Bureau Comm. Fish. Stat. Digest
No. 60: 557-576.

Carlander, K.D. 1969. Handbook of freshwater fishery
biology. The Iowa State University Press, Ames. 752 p.

xxiv

..6gu-=‘

 

 

Carr, I.A. 1962. Distribution and seasonal movements of
Saginaw Bay fishes. U.S. Fish. Wildl. Serv. Spec. Sci.
Rep. Fisheries, No. 417: 13 p.

Christie, W.J. 1974. Changes in the fish species composi-
tion of the Great Lakes. J. Fish. Res. Board Can. 31:
827-854.

Clark, c.w. 1976. Mathematical bioeconomics: The optimal
management of renewable resources. John Wiley and Sons,
Inc., New York. 352 p.

Coble, D.w. 1967. The white sucker population of South
Bay, Lake Huron, and effects of the sea lamprey on it.
J. Fish. Res. Board Can. 24: 2117-2136.

Crutchfield, J. A. 1973. Economic and political objectives
in fishery management. Trans. Am. Fish. Soc. 102: 481-
491.

Deriso, R.B. 1980. Harvesting strategies and para-
meter estimation for an age-structured model. Can. J.
Fish. Aquat. Sci. 37: 268-282.

Dawson, L.E., et a1. 1978. Stability of freshwater sucker
(Catostomus spp.) flesh during frozen storage. J. Fish
Res. Board Can. 35(2): 253-257.

 

Eddy, S. and J. C. Underhill. 1974. Northern fishes.
University of Minnesota Press, Minneapolis: 414 p.

Eshenroder, R.L. 1977. Effects of intensified fishing,
species changes, and spring water temperatures on yellow
perch, Perca flavescens, in Saginaw Bay. J. Fish. Res.

Board Can. 34(10): 1830-1838.

 

Fox, W.W., Jr. 1971. Random variability and parameter
estimation for the generalized production model. Fish.
Bull. 69(3): 569-580.

Fox, W.W., Jr. 1975. Fitting the generalized stock produc-
tion model by least squares and equilibrium approxima-
tion. Fish. Bull. 73(1): 23-27.

Freedman, P.L. 1974. Saginaw Bay: An evaluation of
existing and historical conditions. EPA Great Lakes
Cont. Prog. Rep. No. EPA-905/9-74-003: 137 p.

Frick, H.C. 1965. Economic aspects of the Great Lakes

fisheries of Ontario. Bull. Fish. Res. Board Can. No.
149: 160 p.

XXV

 

Fullenbaum, R.F. and F.W. Bell. 1974. A simple bioeconomic
fishery management model.' A case study of the American
lobster fishery. Fish. Bull. 72(1): 13-25.

Galloway, J.E. and N.R. Kevern. 1976. Michigan suckers,
their life histories, abundance and potential for
harvest. Michigan Sea Grant Program Tech. Rep. No.
53: 46 p.

Gatto, M. and S. Rinaldi. 1980. On the determination of a
commercial fishery production model. Ecol. Modelling.
8: 165-172.

Geen, G.H., et a1. 1966. Life histories of two species of
catostomid fishes in Sixteenmile Lake, British Columbia,
with particular reference to inlet stream spawning. J.
Fish. Res. Board Can. 23(11): 1761-1788.

Great Lakes Environmental Contaminant Survey. 1974.
Michigan Department of Agriculture, Lansing, MI p. 35.

Great Lakes Environmental Contaminant Survey. 1975.
Michigan Department of Agriculture, Lansing, MI p. 25.

Hall, A.E., Jr. and 0.R. Elliott. 1954. Relatonship of
length of fish to incidence of sea lamprey scars on

white suckers. Catostomus commersoni, in Lake Huron.
Copeia 1954(1): 73—74.

Hilborn, R. 1979. Comparison of fisheries control systems
that utilize catch and effort data. J. Fish. Res. Board
Can. 36(12): 1477-1489.

Hile, R. and H.J. Buettner. 1959. Fluctuations in the
commercial fisheries of Saginaw Bay, 185-1956. U.S.
Fish. Wildl. Serv. Res. Rep. No. 51: 38 p.

Hile, R. 1962. Collection and analysis of commercial
fishery statistics in the Great Lakes. Great Lakes
Fish. Comm. Tech. Rep. No. 5: 55 p.

Jensen, A.L. 1974. Leslie Matrix models for fisheries
studies. Biometrics. 30(3): 547-551.

Jensen, A.L. 1976. Assessment of the United States lake
whitefish (Coregonus clupeaformis) fisheries of Lake
Superior, Lake Michigan and Lake Huron. J. Fish. Res.
Board Can. 33(4): 747-759-

 

 

Laarman, P.W. and J.R. Ryckman.. 1980. Size selectivity of
trap nets for eight species of fish. Mich. Dept. Nat.
Res. Fish. Div. Fish. Res. Rep. No. 1880: 16 p.

xxvi

 

Lalancette, L-M. 1975. The seasonal cycle in the germinal
cells (testes and ovaries) of the white sucker,
Catostomus commersoni) from Gamelin Lake, Quebec. Nat.
Can. 102(6): 721-735}

Lalancette, L-M. 1977. Feeding in white suckers
Catostomus commersoni) from Gamelin Lake, Quebec, over
a twelve month period. Nat. Can. 104: 369-376.

Latta, H.C. 1959. Significance of trap-net selectivity in
estimating fish population statistics. Pap. Mich. Acad.
Sci. Arts Lett. 44: 123-138.

Leslie, P.H. and D.H.S. Davis. 1939. An attempt to
determine the absolute number of rats on a given area.
J. Anim. Ecol. 8: 94-113.

Michigan State University. 1975. Increasng the economic
use of Upper Michigan's commercial fishery. First Year
Report to the Upper Great Lakes Regional Commission.
Tech. Assist. Proj. No. 10520239. December 1975.

Michigan State University. 1978. Increasng the economic
value of Upper Michigan's commercial fishery. Second
Year Report to the Upper Great Lakes Regional Com-
mission. Tech. Assist. Proj. No. 10620305. March 1978.

McGaw, R.L. 1980. Confidence invervals for optimal effort
estimates from the Schaefer Production Model. Can. J.
Fish. Aquat. Sci. 37: 288-289.

Mohn, R.K. 1980. Bias and error propagation in logistic
production models. Can. J. Fish. Aquat. Sci. 37: 1276-

Olson, D.E. 1963. Role of the white sucker in Minnesota
waters. Proc. Minn. Acad. Sci. 31(1): 68-73.

Olson, D.E. and W.J. Scidmore. 1963. Homing tendency of
spawning white suckers in Many Point Lake, Minnesota.
Trans. Am. Fish. Soc. 92(1): 13-16.

Patriarche, M.H. 1968. Rate of escape from trap nets.
Trans. Am. Fish. Soc. 97(1): 59-61.

Regier, H.A. 1973. Sequence of exploitation of stocks in
multi-species fisheries in the Laurentian Great Lakes.
J. Fish. Res. Board Can. 30: 1992-1999.

Ricker, W.E. 1975. Computation and interpretation of

biological statistics of fish populations. Bull. Fish.
Res. Board Can. No. 191: 382 p.

xxvii

Roff, D.A. 1980. A motion for the retirement of the Von
Bertalanffy function. Can. J. Fish. Aquat. Sci. 37:
127-129.

Rybicki, R.W. 1979. Assessment of undertutilized
anadromous fishes in the Great Lakes. MDNR - Proj. -
AFC - 12. Compl. Rep. Mich. Dept. Nat. Res. Fish. Div.:
82 p.

Schaefer, M.B. 1968. Methods of estimating effects of
fishing on fish populations. Trans. Am. Fish. Soc. 97:
231-241.

Schneider, J.C., et a1. 1969. The distribution and abun-
dance of benthic fauna in Saginaw Bay, Lake Huron. Proc.
12th Conf. Great Lakes Res.: 80-90.

Schuytema, G.S. and R.B. Powers. 1966. The distribution of
benthic fauna in Lake Huron. The Univ. of Mich. Great
Lakes Res. Div. Pub. No. 15: 155-163.

Scott, W.B. and E. J. Crossman. 1973. Freshwater fishes of
Canada. Bull. Fish. Res. Board Can. No. 184: 966 p.

Scott, J.A. 1974. A historical review of the productivity
and regulation of Michigan's commercial fisheries,
1870-1970. Michigan Fisheries Centennial Report, 1873-
1973. Mich. Dept. Nat. Res., Lansing, MI: pp. 77-79.

Seber, G.A.F. 1973. The estimation of animal abundance and
related parameters. Charles Griffin and Co., Ltd.,
London. 506 p.

Smith, S.H. et a1. 1961. Fishery statistical districts of
the Great Lakes. Great Lakes Fish. Comm. Tech. Rep.
No. 2: 24 p.

Smith, S.H. 1968. Species succession and fishery exploita-
tion in the Great Lakes. J. Fish. Res. Board Can.
25(4): 667-693.

Smith, S.H. 1973. Application of theory and research in
fishery management of the Laurentian Great Lakes.
Trans. Am. Fish. Soc. 102(1): 156-163.

Snedecor, G.w. and W.G. Cochran. 1967. Statistical
methods. The Iowa State University Press, Ames. 593 p.

Talhelm, D.R., et a1. 1979. Current estimates of Great

Lakes fisheries values. Great Lakes Fish. Comm. Status
Rep. 18 p.

xxviii

Uhler, R.S. 1980. Least squares regression estimates of
the Schaefer production model: Some Monte Carlo
simulation results. Can J. Fish. Aquat. Sci. 37: 1284-
1294.

Vaughan, D.S. and S.B. Saila. 1976. A method for deter-
mining mortality rates using the Leslie Matrix. Trans.
Am. Fish. Soc. 105(3): 380-383.

Walter, G. and W.J. Hogman. 1971. Mathematical models for
estimating changes in fish populations with
applications to Green Bay. Int. Assoc. Great Lakes Res.
14: 170-184.

Walter, G.G. 1973. wDelay-differential equation models for
fisheries. J. Fish. Res. Board Can. 30(7): 939-945.

Walter, G.G. 1975. Graphical models for estimating para-
meters in simple models of fisheries. J. Fish. Res.
Board Can. 32(11): 2163-2168.

Wright, A.T. Great Lakes Section Chief. Mich. Dept. Nat.
Res., Lansing, Michigan.

Zabik, M.B., et a1. 1978. Dieldrin, DDT, PCB's and mercury

levels in freshwater mullet from the Upper Great Lakes
Pestic. Monit. J. 12(1: 36-39.

xxix